Datasets:

mimiklee

/

masterthesis_clean_text_20k

like 0

Introduction Recent advances in fertility technologies and legislation are allowing more sexual minority people to become parents than ever before . Correspondingly, studies on the psychological adjustment of children and parents in sexual minority parent families are increasing . Most prior research in this area has shown that neither sexual minority parents nor their children differ from their heterosexual counterparts, in terms of mental health . However, some studies have provided evidence of better mental health outcomes for sexual minority parents relative to heterosexual parents ). Common explanations for these differences emphasize the deliberate, intentional, and planned nature of sexual minority parenthood, as well as personal psychological resources associated with the ability to contend with the financial and bureaucratic challenges of pursuing parenthood as a sexual minority individual . To date, no research has investigated the specific role played by parents' positivity in determining better mental health outcomes among sexual minority parents, relative to heterosexual parents. Moreover, most research on the mental health outcomes of sexual minority parents through assisted reproduction has focused on gay fathers, in comparison with heterosexual fathers ; in contrast, research on the potential differences between lesbian mothers and heterosexual parents has received less attention . The present study aimed at filling this gap in the literature by exploring the role played by positivity in mediating mental health differences between lesbian mothers and heterosexual parents. This investigation can shed light on the potential mechanisms explaining the resiliency manifested by better mental health outcomes among lesbian mothers. So doing, the study echoes the call to further investigate resiliency factors among sexual minorities as a research agenda . The outcome variable of mental health was indicated by both a negative and a positive aspect, as is typical for this field of research , and in which the use of both polarities of mental health may serve as a comprehensive point of view of one's well-being . --- Mental Health Differences Between Same-Gendered and Different-Gendered Parents Minority stress may adversely affect the mental health of sexual minority people . Most research on minority stress within sexual minority people has involved child-free participants. However, the studies that have focused on sexual minority parents have shown that these parents do not report worse mental health outcomes than heterosexual parents . In fact, some have even reported better mental health outcomes among sexual minority parents ). To explain this result, it has been proposed that parenthood may serve as a potential protective factor for mental health among sexual minority people . The present study was conducted in the familistic and pronatalist context of Israel, where parenthood implies entry into the social consensus, linked with positive psychological outcomes . For sexual minority people, the process of overcoming legal, financial, and bureaucratic hardships to become a parent may result in a sense of triumph, growth, and accomplishment, which might protect against adverse mental health . Additionally, it may be linked to better psychological well-being . Furthermore, in Israel, motherhood is a core identity characteristic and often seen as a "national mission" . Therefore, lesbian mothers in Israel may benefit from increased social support, which may contribute to protecting their mental health . Moreover, the tendency to strike an equal division of labor between parents in lesbian mother families and to focus on the satisfaction of basic needs in the couple relationship may also support lesbian mothers' positive psychological welfare . Personal psychological resources may additionally explain the optimal functioning and better mental health that tend to characterize lesbian mothers, in comparison with heterosexual parents. Positivity, or the tendency to view oneself, one's life, and one's future with a positive outlook, is an individual characteristic that promotes well-being and mental health and helps to manage adversity . To the best of our knowledge, no prior research has explored the role of this psychological asset in the context of mental health disparities between lesbian mothers and heterosexual parents. --- Positivity, Mental Health, and Parents' Sexual Orientation Research has shown links between higher positivity and greater optimism, resilience, self-esteem, life satisfaction, and positive health outcomes . Some studies have even shown that positivity may protect against mental illness-specifically depression . Several theories posit that negative views of the self, the world, and the future are the major characteristics of depressive symptomatology . Positivity enables individuals to reflect on their experiences and hardships while trying to give meaning to their lives; this, in turn, may mitigate depressive symptomatology . Considering the major stressor of COVID-19 and its connection to anxiety and depression, recent findings have also pointed to the protective role played by positivity in reducing individuals' anxiety and depressive symptomatology during the pandemic . While differences in positivity as a function of sexual orientation have been scarcely examined, a recent study found that lesbian mothers reported higher positivity than heterosexual parents ). This result was then interpreted in the context of more favorable outcomes for lesbian mothers, relative to heterosexual parents. In this vein, lesbian motherhood may represent a victory over the prejudice against sexual minority parents and possible social approval within the Israeli pronatalist and familistic context, which may support better mental health outcomes ). However, it remains unknown whether positivity might mediate the relation between parents' sexual orientation and mental health. Evidence on the role played by positivity in counteracting adverse mental health among lesbian mothers may contribute to the literature on the strengths and adaptation of sexual minority parents , as well as the literature on the protective mechanisms of positivity with respect to mental health vulnerability in challenging times, such as the COVID-19 pandemic . The current study was framed by positive psychology and minority stress theory , which focus on factors associated with individual mental health. Positive psychology emphasizes the contribution of personal psychological resources to mental health and optimal functioning. In a similar vein, minority stress theory emphasizes that sexual minority individuals' mental health depends on the interaction between stressful social environment created by experiences of stigma, prejudice, and discrimination and possible stress-ameliorating factors, divided to personal and group resources . In the current study, we suggest positivity as a protective personal resource that may buffer against stigma while enhancing mental health. Notably, our study was conducted in Israel. The sociocultural setting of Israel offers a unique context for comparing the mental health of lesbian versus heterosexual parents. On the one hand, Israel is a familistic and pronatalist society that sanctifies parenthood, and especially motherhood. This is evidenced by its fertility rate , which is the highest among all countries in the Organization for Economic Co-operation and Development . Moreover, all women in Israel, irrespective of their marital or family status, are entitled to wide health maintenance organization coverage for multiple in vitro fertilization cycles, up to the age of 45 years . On the other hand, until recently, Israeli legislation restricted access to parenthood for sexual minority people . Moreover, the Israeli context is considered patriarchal, machoistic, and influenced by the Orthodox Jewish law, which tends to disapprove of sexual minority people. This context can maintain a hostile atmosphere for sexual minorities . Research into the mediating role of positivity among lesbian mothers in this context, particularly during the COVID-19 pandemic, could contribute important insights. --- Research Hypotheses The hypotheses were derived from the abovementioned literature on mental health differences between lesbian and heterosexual parents, as well as the literature on the impact of positivity on mental health. Specifically, it was hypothesized that: H1: positivity would mediate the association between sexual orientation and life satisfaction, such that higher life satisfaction among lesbian mothers, relative to heterosexual parents, would be accounted for by lesbian mothers' greater positivity; H2: positivity would mediate the association between sexual orientation and depressive symptomatology, such that lower depressive symptomatology among lesbian mothers, relative to heterosexual parents, would be accounted for by lesbian mothers' greater positivity. --- Method --- Sample The participants were 72 Israeli lesbian mothers through donor insemination and 72 Israeli heterosexual parents through assisted reproduction . All parents self-identified as cisgender and had at least one child born through assisted reproduction, aged 3-10 years . Most participants were highly educated, worked full-time, were secular, and were in a relationship defined by marriage, civil partnership, or contract. The two family types significantly differed on several socio-demographic variables. Specifically, relative to heterosexual parents, lesbian mothers were in relationships of shorter duration, estimate = -.747, SE = .220, p = .001, younger, estimate = .439, SE = .210, p = .041, and less likely to be married or in a civil partnership, χ 2 = 6.89, p = .009. Table 1 displays the full demographics. Notably, due to the relatively small sample size, when analyzing indirect effects, we have calculated a post hoc power analysis based on the findings. --- Procedure The participants were originally recruited for a larger, ongoing, cross-cultural study on socialization about family diversity in families formed by assisted reproduction . Israeli parents were recruited via announcements placed on Israeli Internet forums pertaining to IVF and LGB parenting, social media, and the Web sites of fertility agencies and local LGBTQ + organizations. The study only included parenting couples. To ensure that members from the same couple could be identified, each participant was asked to create a password using the first letters of the target child's name and the numbers of the target child's birth date. This allowed the reports of each parent in a family to be matched. The inclusion criteria for parents were as follows: selfidentified as a lesbian or heterosexual cisgender parent; parent to a child born through assisted reproduction , aged 3-10 years, with no physical and/or mental illness or disability; and residing in Israel. Participating couples who filled in the questionnaire considering different target children in the family were omitted, since their responses could not be matched. Parents who indicated sexual intercourse as a method of conception were also excluded. The final sample was comprised of 144 parents, nested in 72 families. The questionnaires were administered in Israel between December 2019 and February 2021, using two main sources: personal email, with the questionnaire included as an attached Word document to be completed and emailed back to the researcher; and a Qualtrics link, which 443 people accessed. The questionnaire completion method determined no differences across family types on any of the study variables. Most parents completed the questionnaire during the first or second wave of COVID-19 and its related lockdowns in Israel. However, no differences were found across family types on the number of parents who completed the questionnaire before or during the COVID-19 pandemic, χ 2 = 3.04, p = .081. Similarly, the period of completion determined no differences across family types on any of the study variables. All participants were informed that their participation was voluntary, and all consented to take part in the research. Prior to data collection, the study was reviewed and approved by the Institutional Review Board of the Reichman university , for compliance with the standards for the ethical treatment of human participants. --- Measures --- Demographics Participants supplied demographic data on their family type, sexual orientation , gender identity , age, religiosity, relationship duration, marital status, working status, educational level, number of children, child age and gender. See Table 1. --- Life Satisfaction Parents' global life satisfaction was assessed using the Satisfaction with Life Scale . The SWLS is comprised of 5 items , which are rated on a 7-point Likert scale ranging from 1 to 7 . In the current study, a total score was calculated for each participant, with higher values representing greater life satisfaction. Cronbach's alphas were 0.81 and 0.83 for lesbian mothers and heterosexual parents, respectively. The instrument was proven to have very good psychometric properties and was widely used worldwide and in Israel . --- Depressive Symptomatology Self-reported depressive symptoms were assessed using the Center for Epidemiologic Studies Depression Scale . This 20-item measure asks respondents to report the frequency with which, over the prior week, they experienced symptoms associated with depression on a scale ranging from 1 to 4 . In the current study, a total score was calculated for each participant, with higher values representing more depressive symptoms. Cronbach's alphas were 0.86 and 0.88 for lesbian mothers and heterosexual parents, respectively. This instrument has been extensively used for research and clinical purposes and was widely used in Israel . --- Positivity Parents' tendency to view their life and experiences positively was assessed using the Positivity Scale . The POS is comprised of 8 items , which are rated on a 5-point Likert scale ranging from 1 to 5 . In the current study, a total score was calculated for each participant, with higher values reflecting greater positivity. Cronbach's alphas were 0.82 and 0.83 for lesbian mothers and heterosexual parents, respectively. This instrument was proven to have very good psychometric properties and was used worldwide and in Israel . --- Individual, Couple, and Family COVID-19 Experiences Parents who completed the questionnaire during the first or second wave of COVID-19 and the associated lockdowns in Israel were asked to indicate the extent to which COVID-19 had impacted eight individual, couple, and family domains on a 4-point Likert scale ranging from 0 to 3 . A total score was calculated for this 8-item scale, with higher scores indicating more negative experiences. Cronbach's alphas were 0.74 and 0.82 for lesbian mothers and heterosexual parents, respectively. --- Data Analysis All analyses were performed using the statistical software R . Effects that were significant at p < 0.05 were interpreted. The preliminary outlier analysis was conducted with the interquartile range method; no outliers were identified. Data distribution was checked with skewness and kurtosis values, with all study variables falling into the acceptable ranges , indicating a normal distribution. Preliminary chi-square tests and mixed models were run to compare potential differences in individual and family socio-demographic variables between lesbian mother and heterosexual parent families . Multilevel correlations were also performed to identify associations among sociodemographic variables and parents' depression, life satisfaction, and positivity, while accounting for within-couple dependency. Mixed models and multilevel correlations were used to account for the nested data structure . Multilevel modeling also served to enhance the estimation of models and standard errors, providing a more accurate estimation of between-and within-couple variance . Finally, the two mediational hypotheses on the influence of parents' sexual orientation on parents' depression and life satisfaction through positivity were tested using two mediational models -one for each outcomewith 95% CIs for bootstrap percentiles and 5,000 resamples . A post hoc Monte Carlo power simulation was computed to obtain the statistical power of the results for the indirect effects. Given that parents' age, marital status, and relationship duration significantly differed between the two groups, these variables were controlled for in the model. Also, as most data were collected during the first or second wave of COVID-19 in Israel, the individual, couple, and family COVID-19 experiences score was entered as a further covariate. --- Results --- Associations Among Parental, Family Socio-demographic, and Study Variables Table 2 displays the full significant and non-significant multilevel associations among parental, family socio-demographic, and study variables, by parents' sexual orientation. Lesbian mothers reporting higher depressive symptoms demonstrated significantly lower life satisfaction and lower positivity , and a significantly higher negative impact of COVID-19 on individual, couple, and family levels . Also, lesbian mothers who reported higher life satisfaction demonstrated significantly higher positivity and a significantly lower negative impact of COVID-19 on individual, couple, and family levels ; additionally, those demonstrating higher positivity reported a significantly lower negative impact of COVID-19 on individual, couple, and family levels . Similarly, heterosexual parents reporting higher depressive symptoms perceived significantly lower life satisfaction and lower positivity ; additionally, those demonstrating higher positivity showed significantly higher life satisfaction . --- Mediation of Parents' Positivity in the Association Between Parents' Sexual Orientation and Parents' Depression or Life Satisfaction The mediation analysis with confidence intervals computed using the Bootstrap percentiles method and 5000 resamples indicated that the indirect effect of parents' sexual orientation on parents' depression through positivity was significant, point estimate = 0.095, SE = 0.042, 95% CI [-0.188, -0.022], t = -2.271, p = .023. As shown in Fig. 1, compared to heterosexual parents, lesbian mothers reported higher positivity, which was, in turn, associated with lower depressive symptomatology. However, the direct effect of parents' sexual orientation on depression was not significant, point estimate = -0.062, SE = 0.065, 95% CI [-0.189, 0.065], p = .347, indicating a complete mediation of parents' positivity. A Monte Carlo power analysis for the indirect effects showed a large power of 80% . Similarly, the indirect effect of parents' sexual orientation on life satisfaction through positivity was significant, point estimate = 0.269, SE = 0.108, 95% CI [0.071, 0.492], t = 2.480, p = .013. As shown in Fig. 2, lesbian mothers reported higher positivity, which was, in turn, associated with greater life satisfaction. Again, since the direct effect of parents' sexual orientation on life satisfaction was not significant, point estimate = 0.113, SE = 0.151, 95% CI [-0.063, 0.527], p = .142, there was evidence of a complete mediation of parents' positivity. A Monte Carlo power analysis for the indirect effects showed a large power of 80% . Figures 1 and2 display the full statistics. --- Discussion The present study investigated the mediating role of positivity in the association between sexual orientation and mental health among lesbian mothers and heterosexual parents, in light of evidence of significant mental health differences between these groups . Consistent with the first hypothesis, a significant mediation effect for positivity emerged, explaining the association between sexual orientation and life satisfaction. Similarly, and in line with the second hypothesis, positivity also had a significant mediation effect on the association between sexual orientation and depressive symptomatology. Lesbian mothers reported greater positivity, higher life satisfaction, and less depressive symptomatology in comparison to heterosexual parents. These results align with prior findings suggesting that sexual minority parents do not tend to report worse mental health in comparison to heterosexual parents, and sometimes even report better mental health outcomes . It has been proposed that a more egalitarian and equal division of labor and greater attention to the satisfaction of basic psychological needs in lesbian couples may support lesbian mothers' better mental health . Moreover, in the Israeli pronatalist context, which highly cherishes childrearing yet presents hardships for sexual minority individuals who wish to become parents , the achievement of parenthood may be experienced as a triumph . This sense of victory, alongside entry to the social consensus via parenthood, may be associated with elevated mental health indicators . Such an explanation may be especially relevant to lesbian mothers, as motherhood is a cardinal characteristic of Israeli women that is likely to elicit greater social support . The mediating role of positivity in the mental health differences between lesbian mothers and heterosexual parents may explain prior evidence of higher life satisfaction and lower depressive symptomatology among lesbian mothers, in comparison with heterosexual parents. It seems that positivity, as a personal psychological asset, is characteristic of Israeli lesbian mothers, and may serve as a force of resiliency related to greater life satisfaction and less depressive symptomatology. This new evidence of the protective and strengthening role of positivity for lesbian mothers in Israel contributes important insights to the growing literature on diverse family forms in diverse societal contexts , building on positive psychology and minority stress theory . --- Strengths and Limitations The main strength of the current study was the identification of positivity as a mediating mechanism in mental health disparities between lesbian mothers and heterosexual parents. Additionally, the multilevel modeling analyses enabled us to accurately analyze the reports from both parents in a couple, while accounting for within-couple dependency . Yet, several limitations should also be noted. First, as the study relied solely on self-report questionnaires, it was vulnerable to self-presentation biases. For example, a "positivity" bias could be attributed to participants filling the questionnaires of positivity in a "good day." Future studies should consider the use of other sources of data to prevent this possible bias. Second, the sampling was not based on a random or representative sample. Unfortunately, current Israeli representative national surveys do not collect data on respondents' sexual orientation, therefore such representative samples of sexual minority individuals are currently not feasible. Third, although the mediation analyses suggested a possible specific mechanism concerning the results, the cross-sectional correlational design did not allow for causal inferences to be drawn. Thus, it could also be suggested that parenthood in the Israeli context could improve relationships with family members , leading to more life satisfaction and less depressive symptomatology, which could, in turn, associate with more positive views of the world. Longitudinal study design could clarify the issue of causality. Fourth, most participants reported themselves as university educated and employed full-time, and these characteristics may not be representative of all lesbian mothers and heterosexual parents in Israel. Fifth, the comparison of lesbian women and heterosexual parents may have overlooked family dynamics or processes that are specific to sexual minority parent families , as well as particular gender differences in the study variables may have been undetected. Sixth, although post hoc power simulations were used to obtain satisfactory statistical power of the results, the relatively small sample size should be acknowledged. Seventh, while the distinctiveness of the Israeli pronatalist and familistic society may be considered one of the strengths of the study, it also restricts the generalizability of the results to other social contexts. Many of the abovementioned methodological limitations reflect common difficulties of research with sexual minority populations . Future research should examine the suggested mediation mechanism in other socio-cultural contexts and among diverse forms of sexual minority parent families, such as families headed by gay, bisexual, or trans parents. Such exploration among diverse sexual minorities could contribute to the visibility of other subjectivities in the LGBTQ + acronym . Further possible mediation effects of adjacent variables, such as parents' optimism , should also be explored in the association between parents' sexual orientation and mental health. --- Social-Policy Implications and Conclusion The current study found that, compared to heterosexual parents, lesbian mothers reported higher positivity, which was, in turn, associated with greater life satisfaction and lower depressive symptomatology. Positivity mediated disparities in life satisfaction and depressive symptomatology between the parental groups. The results have practical implications for counselors, clinicians, and mental health professionals, as they highlight the importance of improving positive orientation toward oneself, one's life, and one's future as a chief strategy for reducing depressive symptomatology and supporting life satisfaction. These results seem especially relevant for parents with young children, who may be more likely to experience significant anxiety and depression . Moreover, the results are relevant to family and mental health researchers, highlighting the importance of considering positivity as a key research variable. Policymakers and social agents should also be aware that parenthood among sexual minorities is associated with heightened reports of mental health, as also reported by previous research . These results are different from most studies on childless sexual minority individuals, who tend to report worsen mental health than heterosexual counterparts . These findings shout out the need for policy makers and legislators to ease the access to parenthood for LGBTQ + individuals and to promote supportive and inclusive policies for diverse family forms. This may be very relevant in the wide context of modern societies, where sexual minority individuals still face prejudice and discrimination quite often , and also in contemporary Israel, where legislation regarding sexual minority rights is currently changing and is in constant social debates. --- Availability of Data and Material This study was not preregistered. Data are available upon reasonable request from the corresponding author. There was no prior dissemination of the ideas and data appearing in the manuscript. Code Availability Not applicable. --- --- Author Contribution G.S., H.B., S.D., and N.C.: conception and design. Data collection was completed by G.S. Analyses of data were conducted by N.C. First draft was written by G.S. and N.C. Following drafts were reviewed and revised according to comments by S.D. and H.B. All authors read and approved the final manuscript. --- Conflict of Interest The authors declare no competing interests. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author or other rightsholder; author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Introduction Previous evidence has shown better mental health outcomes for lesbian mothers, in comparison with heterosexual parents. The present study explored the mediating role of positivity (i.e., the tendency to view life with a positive outlook) in determining these mental health differences. Methods Seventy-two Israeli lesbian mothers by donor insemination (from 36 families) were compared with 72 Israeli heterosexual parents by assisted reproduction (without donated gametes; from 36 families) on positivity, life satisfaction, and depressive symptomatology. All parents were cisgender and had at least one child born through assisted reproduction, aged 3-10 years. Data were collected between December 2019 and February 2021. Results Mediational analyses indicated that lesbian mothers reported greater positivity, which was, in turn, associated with greater life satisfaction and less depressive symptomatology. Conclusions The results pinpoint the possible protective and strengthening role of positivity for lesbian mothers in the pronatalist and familistic environment of Israel. The research contributes to the literature on the mental health of parents with diverse sexual orientations living outside of the USA and Europe.Policymakers and mental health professional should be knowledgeable about the importance of improving positivity as a strategy for reducing depressive symptomatology and promoting life satisfaction among lesbian mothers. Moreover, policy makers and legislators are called to ease the access of sexual minority individuals to parenthood, as in this population parenthood associates with heightened mental health outcomes.

Resum. Investigació-acció i transformació: lliçons de tres dècades de pràctica En aquest article repasso tres dècades de recerca-acció participativa i n'extrec alguns ensenyaments que tenen a veure amb els problemes urgents del nostre temps: com actuar davant de trajectòries de creixement socioeconòmic insostenibles i una degradació ambiental en cascada. Començo amb el meu tortuós viatge cap a la investigació-acció i la importància de la subjectivitat humana en els processos de transformació. A continuació, traço el pensament que va donar forma a una agenda de recerca social feminista postmoderna. A continuació, exposo el desenvolupament d'un disseny de recerca per a la investigació-acció participativa postestructuralista que posa l'accent en la comprensió de les «formes de vida» en el lloc. A partir de projectes reals de recerca-acció, les lliçons que se'n desprenen fan referència a la importància de 1) inventariar i donar testimoni de la diferència, 2) noves tècniques de resubjectivació que fomentin la transformació i 3) desplaçar el centre d'atenció de l'acció dels subjectes humans als conjunts sociotècnics en què s'entrellacen els éssers humans, les infraestructures materials, els hàbits i l'experimentació. Al llarg de les meves tres dècades de participació en la investigació-acció, he après sobre els límits en la durabilitat del canvi de subjectivitat i la necessitat de treballar en el suport d'ecosistemes més amplis de governança, infraestructura, salut i expectativa social. Paraules clau: subjectivitat; inventari; diferència econòmica; resubjectivació; postestructuralista; assemblatge sociotècnic --- Resumen. Investigación-acción y transformación: lecciones de tres décadas de práctica En este artículo repaso tres décadas de investigación-acción participativa y extraigo algunas enseñanzas que tienen que ver con los problemas acuciantes de nuestro tiempo: cómo actuar ante trayectorias de crecimiento socioeconómico insostenibles y una degradación ambiental en cascada. Comienzo con mi tortuoso viaje hacia la investigación-acción y la importancia de la subjetividad humana en los procesos de transformación. A continuación, trazo el pensamiento que dio forma a una agenda de investigación social feminista posmoderna. A continuación, expongo el desarrollo de un diseño de investigación para la investigaciónacción participativa posestructuralista que hace hincapié en la comprensión de las «formas de vida» en el lugar. A partir de proyectos reales de investigación-acción, las lecciones que se desprenden se refieren a la importancia de 1) inventariar y dar testimonio de la diferencia, 2) nuevas técnicas de resubjetivación que fomenten la transformación y 3) desplazar el centro de atención de la acción de los sujetos humanos a los conjuntos sociotécnicos en los que se entrelazan los seres humanos, las infraestructuras materiales, los hábitos y la experimentación. A lo largo de mis tres décadas de participación en la investigación-acción he aprendido acerca de los límites en la durabilidad del cambio de subjetividad y la necesidad de trabajar en el apoyo de ecosistemas más amplios de gobernanza, infraestructura, salud y expectativa social. Palabras clave: subjetividad; inventario; diferencia económica; resubjetivación; posestructuralista; ensamblaje sociotécnico --- Résumé. Recherche-action et transformation : Leçons tirées de trois décennies de pratique Dans cet article, je passe en revue trois décennies de recherche-action participative et j'en tire quelques leçons qui se rapportent aux questions urgentes de notre époque : comment agir face à des trajectoires de croissance socio-économique insoutenables et à une dégradation environnementale en cascade. Je commence par évoquer mon parcours tortueux dans la recherche-action et l'importance de la subjectivité humaine dans les processus de --- Introduction In the first action research project I ever worked on there was a community researcher whom I'll call Lila, who frustratingly answered in response to any question posed to her: 'Stuffed if I know.' The emphasis was on the 'I'. And in Australian slang this phrase indicated that she had no idea. It was a kneejerk reaction to being placed on the spot and asked for an opinion on something that perhaps she had no formulated thoughts ready to expound. It was apparent that Lila refused to play the game of the social research process, which relies on the fiction of a centred, knowing subject who can report on feelings, opinions or action at the drop of a hat or the upward inflection of the interviewer's voice. As our training workshop for community researchers proceeded, I discovered that this initial response was just an opening. In fact, Lila knew a great deal, but the process of research needed to allow time and space for knowledge to be put into words and voiced . After a while, and with encouraging prompts from others, she readily shared her thoughts and feelings. Lila helped to give shape to what 'I' came to refer to as post-modern feminist social research and 'action research in a poststructuralist vein', an approach that I pioneered in the early 1990s. 1 This response from Lila sprang to mind when I was asked to address the Danish Action Research Network's annual conference in early 2021, where the theme was "The role of action research in the Great Transition". The questions the conference organizers posed to me were: 1. How can we democratically renew our way of living and the social-and nature-relations it implies? 2. How do we act upon root-causes rather than symptoms of plural crisis? 3. How can critical research and exemplary practices jointly nurture sustainable futures? I could assume that attendees at this conference all agreed that action research, as one of the pedagogies of the oppressed, democratises and decolonizes research, speaks truth to power and makes sure that non-experts have a say in what is studied. But it seemed from the questions being posed that there was some uncertainty about where this radical research positioning stands today in the face of voracious socio-economic growth trajectories and cascading environmental degradation. The Great Transition referenced in the conference title refers to what is needed if we take seriously growing income inequality, racial injustice, gender violence, environmental destruction and health threats associated with the Great Acceleration . I was being asked what could a poststructuralist approach to participatory action research -that is, an approach that challenges dominant monocultural knowledge systems and accepts the decentered nature of subjectivity and the power of not knowing-contribute? After an initial 'Stuffed if I know!' reaction, these questions became a provocation to reflect on my style of action research. In this paper I look back over three decades of doing participatory action research, and I extract some lessons that bear upon these questions in turn. Reader beware! I am not aiming for any kind of systematic overview here, but merely to explore in retrospect what my experience of action research might contribute to the urgent calls to transform our way of living on this planet. The paper starts with my circuitous journey into action research and to the importance of human subjectivity to transformation processes. It then traces the thinking that shaped a postmodern feminist social research agenda. Next follows an account of developing a research design for PPAR that foregrounded understanding 'ways of living' in place-in this case, ones shaped by undemocratic social relations and extractive environmental relations. The 1. The quotes around I signify that this process of development took place in conversation with key others, specifically Julie Graham, with whom from 1992 I shared a joint authorial name, J.K. Gibson-Graham, and Jenny Cameron, who co-designed and ran my second main PPAR project and helped to clarify this approach . action research project examined here engaged critically with the status quo and raised the question of how to renew the ways of living of shift workers and their families. Lessons learnt about the importance of inventory and bearing witness to difference are relevant to question 1, about how we can democratically renew our way of living and the social-and nature-relations it implies. The next discussion centres on a project that began in the ruins of deindustrialization and decline, where existing ways of living had been destroyed and where there was some appetite to act upon the deep causes of regional crisis. Bearing on question 2 about how to act upon root-causes rather than symptoms of plural crisis, there are lessons to be learnt about new techniques of resubjectivation that encourage transformation, and other lessons about limits to the durability of subjectivity change without support from wider ecosystems of governance, infrastructure, health and social expectation. Finally, the last project engages with transition design and processes of commoning to nurture sustainable futures by shifting the action focus from human subjects to socio-technical assemblages in which humans, material infrastructures, habits and experimentation are interwoven in built environments that are under pressure to change. The lessons are still emerging from this project, in which new ways of living with each other and with rising urban heat must be developed as a matter of survival. --- Pre-history: from 'researching on' to 'researching with' human subjects I started as an undergraduate researcher with a training in survey design and questionnaire-based social research, but gravitated towards qualitative methods including historical and participatory research. My honours thesis research on a remnant working-class community in Glebe, a rapidly gentrifying area of inner Sydney, involved using electoral rolls to seek out long-term residents of more than 40 years and contacting them to talk about the changes they had seen in the neighbourhood over their lifetimes . I joined a social group of elderly residents who went on monthly bus trips to Sydney's environs to spend the day together enjoying lunch and companionship. I learnt to crochet from one bus tripper, and I visited another wheelchair-bound resident in between day trips. As a resident in the area I was researching, I learnt to see the urban environment from the perspective of now-retired working-class people for whom Glebe had once been a tightly connected Anglo-Celtic community with distinct zones differentiated by income and occupation. Soon after finishing this research, I came under the powerful spell of Marxist political economy and cast aside my interest in human subjects, their perceptions and life stories. At that time, it seemed that radical urban political economy had no need for the thoughts and feelings of people. The rent books and records of banks and building societies were all that was needed to grasp the exploitative dynamics of disinvestment and reinvestment in the city. Research for my PhD thesis was theoretical and his-torical . Thinking back, it was executed in the historical materialist mode of Marx's method, as laid out in Capital, and drew on archives, company records and newspaper reports. Perhaps it was from Marx's three volumes that I gained a suspicion of subjects and their false consciousness, not to be trusted as sources of insight into the structure of the capitalist world. It took a good decade for me to accept that talking to real, live people about their experiences of economic and social change might be useful to the research process. This realization was linked to the desire to be more involved in political change and the process of organization and 'conscientization'. My research on new forms of indentured labour, in the form of international contract labour migration, led me to talk to workers from the Philippines who had travelled to the Middle East as construction labourers, or to Hong Kong as domestic servants . Their harrowing experiences of loss of any control over their lives stayed with me as I later came to recognize that the 'capitalist system' hosted slave class processes in the midst of a modern construction industry and within global cities. The insights into this new global workforce, with its distinctive motivations and tolerances, had implications for working-class unity and the hoped-for global mobilization that was advocated by radical social scientists as a vehicle of change. In a world where millions of people were signing up to indenture as a way of making ends meet for their families, a centred, working-class subject was hard to find. Returning to face-to-face research still seemed kind of weak and unimportant. Especially when it was revealing the disorganization of a global working class. It was at this stage in my intellectual evolution that I was relieved from a host of dilemmas by poststructuralist philosophy, anti-essentialist Marxism and 'postmodern' feminism. I was attracted to the idea of a decentred subject, of a self who occupied multiple subject positions, of a subject who participated in more than one class process and thus had contradictory 'class interests'. This provoked a rethinking of what class politics might be, and more generally what transformation in a society where multiple class processes coexisted might involve. My interest in working with people in the research process, not on people, had drawn me to participatory action research as a method. Of course, this approach as developed by politicized educators working in the Global South had strong connections with liberation theology and Marxian thinking. Working with marginalized people to name, understand and confront their exploitation, that is, to develop a pedagogy of the oppressed, this was a pathway towards liberation from oppression. But it seemed that this approach relied on the possibility of activating a radical political subjectivity and awakening a centred subject of resistance. How could this approach cope with the idea of decentred subjects and multiple political subjectivities? Feminist research pointed to the very differ-Documents d'Anàlisi Geogràfica 2024, vol. 70/1 33 ent interests of men and women in the same living situation, and the possible suppression of women's concerns within a working-class-focused program of change. These were problems that needed to be addressed if participatory action research was to be conducted 'in a poststructuralist vein'. These questions about subjectivity and social transformation accompanied me on my journey to rethink action research. --- Postmodern feminist social research: co-researching ways of living The opportunity to experiment with a different mode of action research was provoked by an issue that was raising concerned interest in the coal mining towns of Central Queensland in the early 1990s. These townships had been built by global mining companies to house workers associated with huge open cut coal mines producing coking and steaming coal for export to the rapidly growing economies in Asia ). Many towns were built from scratch close to their mine, while others took existing small settlements and expanded them into small-sized towns by appending large clusters of companyowned residences. A distinctive feature of the housing supplied was that the hierarchy of the mine site was replicated in the housing styles, with managers and technicians in 'high set' houses up off the ground and workers in 'low set' houses. With each municipality in Central Queensland wanting their 'own' mining town and each corporation interested in maintaining maximum control over 'their' workforce, the Queensland State Government allowed multiple small towns to be built, rather than one or two medium-sized towns servicing different mining operations. This decentralized town infrastructure was subsequently to become a regional development disaster, but in the early 1990s these communities were thriving, despite being small and with only limited services. Much of the social life of the community revolved around sporting activities on the weekends, when families would come together to participate or watch team sports, with the regulation BBQ to follow. A matter of concern that significantly affected livelihoods in these communities arose in the early 1990s. A change in work arrangements was introduced by the companies to increase productivity. It involved miners working a sevenday roster which, in effect, destroyed the temporality of social activity in the towns. This roster meant that miners worked through the weekend and were 'off work' on only one weekend every month. The implications for families were keenly felt by those with children. Wives could no longer rely on some respite from childcare and other household tasks on the weekend. The implications for townships were dire. Social and sporting clubs could no longer rely on patronage, with two thirds of the workforce working on the weekend at any one time. It was not long before it became known as the 'divorce roster'. The decision to accept the new roster had been supported by many miners as it meant an increase in pay. While many workers had previously done overtime work at double pay on some weekends, it was now a mandated part of the work contract. Women, however, had not been consulted about this change, as they were not members of the mineworkers' union and were not included in any company-initiated roundtables. They were, in effect, locked out of a decision that had huge implications for their workload in the household and community. They had not raised their voices in any organized opposition. This issue that my research had stumbled across revealed a significant flaw in a class-based politics focused on the workplace . In older coal mining communities, wives of miners often belonged to organizations that were companions to the union, the Miner's Women's Auxiliaries. These organisations came to the fore especially at times of crisis, for example where mine closures caused unemployment, or a mine disaster occurred. Women came together to support the miners and their families, providing food, clothing, emergency accommodation and other assistance. They also took to the streets in demonstrations alongside their menfolk during strikes or other forms of political action. Women were expected to align their economic interests with their partners. In the case of the seven-day roster, there was no easy alignment of interests between miners and their partners. The identity and interests of the miner's wife was challenged. The action research started from a recognition that women were not centred subjects. There was no collective subject to awaken by breaking the oppressive chains of false consciousness. Women occupied multiple economic and social subject positions. They were miners' wives, yes, but they were also mothers, homeworkers, carers and community volunteers, or had previously been employees in other occupations before moving to the mining region, where employment for women was almost non-existent. These women lived heterogeneous lives, laboured in diverse ways, identified with multiple images of womanhood. So, was it possible to conceive of a collective action to improve livelihoods led by women in these towns? This question was the motivation behind my experiment in post-modern feminist action research. --- Evolving a research design for poststructuralist participatory action research Central to the design of action research is the involvement of research subjects in the research process itself. In this case my focus was on women and how they could be more influential over processes in the mine site that affected them. I made the decision to work only with women, and to employ them as 'Community Researchers' . The lack of paid employment opportunities for women in these remote, single industry towns meant that most were confined to the home. Expending the project funds to train and pay women as researchers was a political act in itself-it was one way of addressing the economic power imbalance of these towns. It also meant that recruitment was relatively easy. I was assisted in the process of CR selection by a regional social worker who lived at some distance from the towns selected for study. There were few requirements for selection. Women had to be living in a household with a miner on the seven-day roster and they had to be able to Documents d'Anàlisi Geogràfica 2024, vol. 70/1 35 attend a two-day and two-night training session away from home in another town, separate from their own. The research was sited in four different mining towns. Three women from each town were recruited as CRs. The plan was that they would be trained to conduct interviews with women in their social circle about the impact the roster was having on their lives. To get the perspectives of women at different stages of the life course, I recruited one woman from each town with preschool children, one with school-age children, and one with no children at home. The project thus enrolled individuals in a process of both first-person research and second-person research . In the process they reached out to others in their circle of connection to gather information and conduct conversations. Once recruitment had been completed, we began a first cycle of research, which involved a training workshop facilitated by my colleague, the regional social worker, and myself. This involved the women traveling to Tieri, another mining town, where research was not being conducted. The reasons for this were multiple. I was interested to see if it would be possible to build solidarity between women across places. I was conscious that by taking people out of their familiar comfort zone and inviting all the CRs to come together away from their homes would instil a certain freedom, and perhaps curiosity. The training process and the women's reactions to it was the research in many ways. Ostensibly the task was to get familiar with the interview schedule, ground test the questions and learn interviewing skills. The interview questions aimed to develop an inventory of the workload of women and men in the household and the timetable of this work over a 24-hour period and over a week. During the introductions at the outset of the training many women had pointed to differences that set them aside from others . However, as the training proceeded and concrete experiences were shared, there developed more of a sense of recognition of similarity and connection. Training involved learning how to conduct and record an interview. Inexperience with this kind of activity brought everyone to the same level. Listening and prompting skills were developed. The community researchers practiced on each other, and suggested modifications to the interview schedule. This constituted one cycle of action-reflection-modification focused on method. The event was significant for the participants-it provided time and space for individual self-inquiry and the beginnings of group reflection, as familiarity and trust started to build. After some nervousness at the outset, it increasingly became a social success. The workshops were held in a community centre and the women were put up in a nearby motel, with all expenses paid. Staying away from home and family, having all the food provisioning and travel arrangements done for them, preparing meals together in a big kitchen and eating together, hanging out as a group of motivated and adventurous women: 36 Documents d'Anàlisi Geogràfica 2024, vol. 70/1 all these factors contributed to making the training a rather festive event. I remember observing and participating in pleasures shared, as well as serious and sometimes confronting exploration. After the two-day training the community researchers returned to their towns and were tasked with interviewing five women in their social circle who were at a similar stage of the family life course as themselves. In this second cycle of the research process, in depth interviews were recorded. Completed survey schedules were mailed to me for compilation and analysis. A second meet-up was planned, for the group to report on the process, review their findings and discuss possible future actions. This third cycle of research, reflection and action took place over two days and one night. The 12 women from four towns reviewed the information they had gathered about how women in similar situations were coping with the new roster arrangement. This produced insights into how the household division of labour between women and men was affected, how children were coping with less access to their fathers on weekends, and how each town's social life was impacted. The community researchers discovered the depth of concern that women had about how their family and their own lives had been negatively affected. Some came across coercive situations where women felt unable to speak up about the impact because of an overbearing husband. They reported a sense of incredulity at the conditions in which some women were living. There was some brainstorming about possible actions that could be taken based on their findings. The community researchers wanted their research to be put in front of their husbands, other miners and company management. They suggested interventions such as more access to childcare for families, especially on weekends when fathers were no longer present. The research was written up in a formal report and in a small booklet in which the findings and quotes from the interviews were illustrated in cartoons . The United Mineworkers, Queensland District Branch contributed funds for 500 copies of the booklet to be printed. In a last cycle of the research third-person action research took place whereby we attempted to bring wider circles of people into a community of inquiry. The community researchers agreed to take the booklet and talk to people in their communities about the findings. The miners' union circulated the booklets to their membership. I travelled around to the mining towns where the women had come from to disseminate the research results and meet up with the community researchers to find out how involvement in the research had impacted their lives. Stepping back from the process to conduct theoretical reflection, I found myself viewing the research as an exercise in 'discursive destabilization' of the constraining identities of 'miner's wives' and 'mining town women'. Through the postmodern social research process which spanned months, the women assumed the collective identity of 'shift workers' wives'-a new subject position borne of the context. My tentative hope was that by creating a discourse of the mining shift worker's wife, the women could be empowered to step into industrial policy making, or at least able to intervene in their partners' workplace politics. But the research design did not include a time frame and budget to support a public action phase. Some women did indeed take individual actions in this role. One woman who was interviewed asked the community researcher for the recording and played it to her husband. This helped her to talk about her concerns and broach issues she had not been able to before. The CR reported that the couple were 'like a pair of newlyweds now'. Other CRs reported that they had become more motivated around community issues and begun to take a role where previously they had sat back. In one community this resulted in lobbying for a childcare centre and its subsequent building. Another CR was encouraged by the experience to return to higher degree study. Private change took place, but not public transformation. We had achieved a great deal by naming the 'problem' of the seven-day roster and reflecting the 'problem' back to men, to the union and to the companies. The 'problem' was that company profitability and men's desire for higher income and more time off were being valued more highly than family dynamics and community well-being. The research revealed the fiction that increased monetary income is correlated with a better life. It revealed the injustice of families and communities in remote regions on the resource frontier having no say in decisions that greatly affected their quality of life. These important revelations stopped short, however, of informing a concerted intervention at an industry-or region-wide level. As such, this project marked for me the limits of what might be achieved from 'outside the tent'. The research had been conducted by women and placed in the lap of the companies and the union. But the power to act collectively on the results was not in the reach of the women themselves. A year or two after completing this research I was contacted by a miner and his wife from one of the smaller mining towns. He told me that the company was introducing a 12-hour shift, with a work cycle that allowed miners to live on the coast and commute to work, living in barracks when they were at the mine. This miner and his wife could see that this meant the potential end to the town where they had raised their family and where they had intended to stay in retirement. He was anxious to see if my research could be used to inform the debate and help shore up the argument of those who wanted to maintain viable communities. As it turned out our community-based research was indeed too insignificant in the scheme of things-the powerful coal mining industry pushed ahead with the 12-hour shift, with support from the workforce and the fly-in, fly-out model of working became increasingly prevalent. Some 30 years later, the mining towns of Central Queensland are very changed places, with outsourced contract labour doing most of the mining . This postmodern social research worked with women who wanted a better life for their families within the confines of an extractive capitalist regional economy. The action research did not broach the consequences of rampant capitalist coal growth and corporate greed, it did not ask participants to take heed of the environmental degradation caused by coal mining-whether global atmospheric impacts or more local impacts on coral bleaching on the Great Barrier Reef, or on water tables in the mining basins-it did not ask them to question 'The Economy'. They were already part of it, benefiting from the high incomes it afforded their households. They had no reason to abandon the system-except for the fact that daily life was hard, emotionally demanding and socially isolated. It was possible to enroll postmodern feminist participatory action research in co-producing new subjectivities, but more would be required to make other worlds possible. --- Lesson 1: Implications for "democratically renewing our way of living and the social-and nature-relations it implies" We need to take further steps within action research processes to provoke the desire to democratically renew 'ways of living' and engage critically with the status quo. The shiftwork research showed how PPAR could enroll actors and shift subjectivity but in the context of the Great Transition more collectively transformative action is needed. We took up this challenge some time later when writing Take Back the Economy . The gendered workloads that informed the cartoon on the cover of the booklet showed the different work rhythms of miners and households. This research gave inspiration for the 24-hour clock exercise in Take Back, in which the inventory of work is juxtaposed with 1) a self-assessment of various aspects of individual wellbeing and 2) the ecological footprint of the way of living described by this work/life profile. This tool links an inventory of human labour with an analysis of consumption of earth's resources, and exposes what the global community impact of individual 'choices' to live well are. It encapsulates the ethical decisions and ecological negotiations that cluster around a work profile. Within an action research context this tool lends itself to taking a group approach to ways of living, so that the idea of lifestyle options can be collectivized and responsibility can be shared. . To a large extent, this research contributed to the theoretical and political rupture that gave rise to that book. The difficulty of working within the constraints of 'traditional working-class' politics had become blatantly clear. Coal companies wanted to expand their production to meet growing global demand, with little concern for the environments they destroyed or the families they impacted. Union members earning six-figure salaries by excavating coal justified their acceptance of more unhealthy shifts by saying they were doing it 'for their families'. Solidarity with workers meant, at the end of the 20th century, condoning a growth-at-any-expense model of economy. In Australia, mining unions and unions associated with manufacturing were eager to see their membership not be disadvantaged by globalization. It was at this time that the economic 'bottom line' was frequently referred to as the determining rationale for all policy and action. 'The Economy' was to be kowtowed to lest it smite us all down. There was, it appeared, no alternative. --- Action research in a poststructuralist vein-reimagining The Economy --- My work in Central Queensland predated the writing of The End of Capitalism : A Feminist Critique of Political Economy This forced a rethinking of participatory action research. The 'oppressed' working class subject was interested in betterment and in addressing the 'symptoms' of crisis such as retrenchment, unemployment and lack of opportunity, not necessarily in the social transformation needed to change ways of life. How could action research ignite the desire for something different? It was at this point that I turned my attention to people and places that were being bypassed by capitalist growth and globalization. The method of conducting postmodern social research with CRs that I had developed in Central Queensland was transferred first to areas of deindustrialization where there was little hope of capitalist revitalization , and then to rural communities in Asia lacking capitalist economic development except as exporters of contract labour migrants. In all cases there were those in the community who wanted to see economic 'development' take place, so the opportunity arose for a conversation around what kind of development and an opening to think beyond capitalist growth as the only acceptable or viable form of development. The choice of sites was determined somewhat by happenstance. As noted above, the places in Australia and the US were close to where the two of us who comprised JK Gibson-Graham lived. We were each familiar with the dominant neoliberal narratives Documents d'Anàlisi Geogràfica 2024, vol. 70/1 41 that were beginning to dominate discussions about regional development in each place. We were eager to insert a discourse of a diverse economy into each scene and to experiment with processes of re-subjectivation that awakened the potential for new kinds of non-capitalist subjectivity. In the Australian context, where Katherine was situated, the Latrobe Valley east of Melbourne was a place where thousands of brown coal miners and electricity workers had lost their jobs when the Victorian State Government privatized the State Electricity Commission by selling off their power station assets to global operators . In the Pioneer Valley of western Massachusetts, where Julie Graham was, there was long term unemployment dating from the exit of manufacturers in the 1970s-80s to greenfield sites in the US South or overseas. Councils in both locations were being courted by businesses who wanted significant financial and regulatory incentives to relocate or set up in each region. 2 In the Latrobe Valley, forced council amalgamations had seen the formation of a regional Latrobe Shire Council, which incorporated many of the coal mining towns of the valley that had once had their own municipal government. A new set of councillors appeared to be interested in engaging grass roots opinions and charting a more community focused pathway for social and economic development. One councillor had come across Asset-Based Community Development, as developed by Jodie Kretzmann and John McKnight in the US, and he agreed to back an action research project to see if this new approach to regional development could be fostered. With funding procured from the Australian Research Council and with input from research partners the Latrobe Shire and Energy company, in 1999 Jenny Cameron and I embarked on a three-year participatory action research project . This funding meant that we were able to recruit three community researchers for a much longer duration than in the Central Queensland project. In this project the aim was to enrol retrenched industrial workers, unemployed young people and women who were largely excluded from the dominantly male regional employment profile into a conversation about what kind of pathways for development they were interested in seeing eventuate. Accordingly, we recruited a retrenched crane driver in his 50s, an unemployed young woman in her 20s and a single mother with two children in her 30s as our community researchers. For each person the project offered part time employment initially for six months. Jenny Cameron was employed full time as project leader, and she moved to the Latrobe Valley to live in the community. My visits were of a week at a time at certain junctures, as travel was somewhat complicated by my having taken up a new job interstate in Canberra, some 600 kms away. The project began with great enthusiasm, especially given the support we initially enjoyed from the Shire Council. Office space was allocated to the project and council officers facilitated access to other resources as needed. We were 2. For more on the Pioneer Valley Rethinking Economy Project see Community Economies Collective . still focused on shifting subjectivity of people who felt they had no role in 'The Economy' but we operated from a standpoint that imagined a diverse economy in which they were actors, not outsiders. We designed a raft of strategies to bring people together, and to help them reframe their economic identity and begin to see themselves as contributing to a diverse more-than-capitalist economy . Photo essays made by the CRs about their inherited economic identities were put on public display to prompt reflection. CRs visited little known community initiatives-a saw milling operation, a cooperative pub , a craft network-that ran according to people-centred, not profit-centred, logics. The CRs visited neighbourhood centres and employment hubs to talk to valley residents, inviting them to document their gifts of the hands, head and heart, and to reframe self-identity. The people assets of the Valley were documented. Community members were invited to pizzamaking events to attract interest in economic experimentation. These strategies are documented in the training manual Shifting Focus: Alternative Pathways for Development . Unlike in the Central Queensland research, this project had the capacity to generate a wider community conversation about what an economy could be, and then to experiment with actual interventions in building a people-centred economy. Four initiatives were settled on as potential projects: a Santa's workshop, a youth circus training group, a community garden and a reuse and repair shed. At the time in 2000, community enterprises were largely unknown in the Valley, and the community garden revolution that was soon to take over all urban areas in Australia had yet to materialize. Men's sheds were known about but there were none in the Valley. The CRs enrolled community members who were interested to volunteer and experiment with the initiatives they had suggested. The research team offered support to expand their understanding of what was possible. For example, for those interested in a community garden, a field trip to Melbourne was organized to visit a pioneering community garden occupying an old tip site in inner city Brunswick . Representatives of other initiatives were invited to come and talk about what they did and how they did it. There were many challenges to surmount in setting up these initiatives and the Council, our research partner, was initially a great help. They identified underutilized public assets which could be made available and repurposed. An abandoned public caravan park became the site of the community garden, a closed pre-school building was converted into Santa's workshop, a relatively new but no longer functional factory space was repurposed as a reuse and repair shed. All these assets were accessed free of charge, in that the Council allocated use rights to these sites. Each site was unutilized property, owned by the public sector , but no longer servicing the community. The community groups that formed around each initiative were able to 'common' these resources. They became those who had access to and the right to use the site. They were the ones to care for the site and benefit from its repurposing. They endeavoured to take responsibility for the management of the site-but this was difficult given the lack of experience in group governance, lack of independent financial resources and ultimately lack of legal property right to the site. The council as the formal owner had to be called in to reconnect water, remove rubbish, fence the sites, remove concrete caravan settings, secure the premises etc. These were small things, but they produced a lot of delays and undermined the enthusiasm of people not skilled in waiting or building slowly. One of the lessons of this phase of the participatory action research was that group social and governance skills take time to develop . Most participants had not been involved in collective activities prior to this project. They may have been church goers or members of unions, but their roles had been largely passive and certainly had not involved democratic group problem solving and dispute resolution. Assuming new economic subjectivities in a diverse community economy was one thing, but initiating a community venture involved a difficult learning curve around things such as meeting procedure, record-keeping, honouring differences of opinion, maintaining focus, coping with delays, keeping spirits up. There were some wonderful highs in the early stages of development and the energy people gained from their involvement in the projects was palpable. But it was also a strain to be involved in something so new, especially on those with mental health concerns and financial worries. The timeline for learning collective practice was very short, and at the end of the project, when financing for Jenny's position as Project Manager ended, the groups were left to cope on their own. It was at this stage that we had hoped that the Council might step in to offer occasional support to the community initiatives. Unfortunately, a change of leadership at the mayoral level saw a diminution of enthusiasm for Community Partnering. Added to that, competition between elected officials associated with different wards filtered down to the enterprises, leading to undermining of the projects in sites without strong political patronage. Lurking in the shadows was the belief that these projects were not really needed, or not really 'economic solutions' to the Valley's woes. A negative pragmatism was voiced when, for example, councillors argued that a community garden was not needed in a region where everyone had 'their own garden'. This kind of intervention was fine for inner city Melbourne, where people live in apartments or didn't have much backyard space, they intimated, but in the suburban context of Valley towns, this kind of initiative seemed needless. These judgements were internalized by some of the gardeners, and they started to have doubts about their venture. Delays were caused by a decision to employ a landscape designer to make a masterplan for the garden site and replicate what they had seen at CERES. Rather than starting with what they had at hand to get growing, time was wasted waiting for elements of the design to be implemented that they could not do themselves. In hindsight, some gardeners reflected that they might have been better off cultivating collective plots of vegetables for the local food bank, rather than sticking to the individual allotments and aesthetic features of the original plan. The garden project folded after three years. The project that lasted the longest was Santa's workshop. This was an initiative that was spearheaded by a retrenched electricity industry worker who had a personal interest in designing and making large Christmas-themed garden installations with lighting attached. Keith's front yard was well known in the Valley as a place to visit with kids at Christmas, to see his displays. He offered to teach others how to make these wooden cut outs, and the Santa's Workshop was set up a few weeks before the Christmas of 2000. Keith teamed up with Carmen, a businesswoman who came to the workshop one year and stayed on to help organize and fundraise. With Carmen's organizational skills and Keith's creative and practical skills, the workshop continued to thrive for ten years, until sadly Keith passed away. Over time they provided a context for many people to learn new skills, produce low-cost decorations, and connect with others in the process of doing purposeful work. 3 Our action research project aimed to gain credibility for a different imaginary of economy-one focused on people as the primary resource in the Latrobe Valley and on for-purpose community enterprises that could support the wellbeing of people and the planet directly. The project demonstrated the potential for new subjects of a post-extractive post-capitalist economy to emerge and take collective action. But left to their own devices, without support in the wider region from local government and State institutions the advances made by small community associations were vulnerable. Lesson 2: Implications for "How to act on root causes rather than symptoms of plural crisis?" Poststructuralist participatory action research processes can work with participants on resubjectivation to develop non-capitalist ways of living, sharing wealth and caring for country. By working in the cracks to generate new economic identities associated with democratic wealth generation and collective approaches to property ownership, PPAR addresses the 'root causes' of capitalist crisis. But what will contribute to the durability of these experiments? And how do practitioners respond to those who attribute any failure for community economies to thrive as indications of the superior nature of capitalist power? In my experience it was the absence of an ecosystem of support that led to the demise of the experiments our action research initiated, not any form of co-option by a 'capitalist system'. Some 20 years later the support ecosystem is much more developed. Social enterprises are now seen as strong actors in local economies-not just welfare supports. Today the Latrobe City now has a Community Gardens Policy that states that community gardening contributes "to public health and wellbeing, learning, amenity of the municipality social connections and can improve 3. The PPAR method developed in the La Trobe Valley Project was replicated and further developed in two rural municipalities in the Philippines . local food security." A further change is the heightened degree of community mobilization around transition away from coal. In 2014 an out-of-control fire at the Hazelwood mine polluted the environment for 46 days. A community organization called Voices of the Valley formed to address questions of transition and health, and has successfully lobbied the State Government for assistance. The Valley hosts the active and high profile Earthworker workerowned cooperative formed in 2013, for which one of my graduate students is the secretary. Perhaps the LTV Community Partnering Project helped to prepare the ground for these developments. --- Action research, commoning and transition design My most recent venture into action research is one that is informed by transition design and the process of commoning. 4 These represent two different approaches to transformation. Transition design activates relationships at the intersections of human and more-than-human concerns. It shifts focus away from the human subject alone and pays attention to the material 'element' of social practices, that is on the durable infrastructures that channel and nudge human behaviour. Transition design has an eye for "long horizons of time and visions of sustainable futures" . It attempts to relieve people of having to make conscious decisions as part of everyday practice. The emphasis is less on shifting subjectivity directly and more on providing contexts in which new social practices can emerge. Commoning, on the other hand, is the name for collective actions that make and share community by orchestrating access to, use of, responsibility for, benefit from, and care for a commons-a property that is crucial to livelihood and wellbeing. How a property, whether tangible such as land or intangible such as knowledge, is commoned is an ongoing outcome of deliberation and negotiation. So, by bringing commoning and transition design together we are working at an interface where human subjectivity, more-than-human community and material assemblages meet. This project focuses on urban heating and the challenges of living with rising temperatures without exacerbating global warming by increased reliance on expensive and heat-releasing air-conditioning during summer months. The research participants are community housing provider organizations and residents living in social housing in Sydney's west, where the impact of global warming and climate change is affecting liveability. The residents are people who have low incomes or live on social benefits-the 'left behind' in terms of a housing market in which home-ownership is privileged and increasingly out of reach of many middle 4. The research project is led by Stephen Healy and Abby Mellick-Lopes and comprises a multi-disciplinary team of economic, social and urban geographers, design practitioners, urban planners, a landscape architect and community housing providers . to low income households. As the COVID pandemic highlighted, those living in Western Sydney are often 'essential workers' keeping the health, aged care, transport and goods distributions systems going. This region is also where public housing has traditionally played a role in housing the most vulnerablethose on health or disability pensions, long term unemployed, aged or infirm. Pilot research on the topic involved interviews with residents of Western Sydney to find out how they coped when the temperatures rose during summer. We found that younger residents had few strategies for keeping cool other than staying indoors with the blinds down and the air-con on or visiting the local shopping mall and hanging out in McDonalds, for as long as the management would tolerate. In contrast, we found that older residents retained a repertoire of cooling strategies that they had 'grown up with' and still relied upon. These included carefully orchestrated opening and closing of windows to catch cool breezes, strategic placement of wet sheets across doorways to cool the breeze, affixing outdoor window blinds, and cultivating garden plantings to maximize shade during hot summer day. For those whose houses had air conditioning, there was careful switching of the power on and off to make the most of off-peak and thus more affordable prices. Clearly there was tacit cooling 'know-how' possessed by the older generation of residents, which involved the manipulation of the immediate infrastructural environment. This initial research alerted us to the fact that the many ways of keeping cool in urban settings had been reduced to just one technologically dependent practice-switching on the air-con. The once common knowledge of how to produce low energy 'coolth' was not being passed on generationally. Meanwhile the ecological impacts of this energy-guzzling practice were hidden, as were the physical bodily impacts of normalizing the 22-degree indoor comfort standard. Modern bodies are losing the ability to regulate their temperature and adjust to colder or hotter weather. Comfort is defined as not having to experience cold noses poking out from under blankets on a sleeping verandah , or hot houses where babies were placed under tables covered by wet sheets with the fan on to make a cooler interior microclimate . The premise of our action research is that this kind of know-how can be shared between generations and between groups from different cultural backgrounds, or it can be learnt anew, and that learning to become part of a cooling commons assemblage will be an important way of living with global warming and 'nurturing sustainable futures' . Of course, this is not to displace the urgent need to curb temperature rises; but for residents who are at some distance from the political process where decision-making takes place, we are focusing on what can be done with what's at hand. The research agenda includes inventorying heat profiles in households and the practices of residents during urban heat events. At follow-up workshops, residents, housing providers and technicians will devise experiments to curb heat rises and increase coolth. This might include retrofits in the individual household or collective adaptations in the environs of the housing. These experiments will be one context where action research will provoke shifts in subjectivity as residents learn to become part of a cooling commons assemblage. The present of housing providers who will be involved in future builds is crucially important if these cooling assemblages are to become embedded in future built environments. Lesson 3: Implications for "How critical research and exemplary practices can jointly nurture sustainable futures?" This project is not trying to reshape the economy directly, but is aimed at intervening in how buildings and cities are designed and thus how lives are lived. The starting point is a critical analysis of the problem of urban heating and its connection to global warming on one hand, and on the other, the private housing market that puts profits before liveability in growing urban areas. Action research with community housing providers who are focused on housing as a human right and a social good is focused on protecting and restoring a living environment that affords comfort and wellbeing. Insofar as design can influence contexts that live on beyond the time span of a single action research project, connections can be forged between local interventions that offer new ways of living together and larger visions of transition. Embedding design solutions in plans, policies, regulations, buildings and other infrastructures is a way of amplifying what was at one time a vulnerable, emergent, grassroots experiment. Design can transform something that was ephemeral into something that has durability. --- Conclusion I started this paper with the story of Lila and her knee-jerk negative reaction to any question that put her on the spot. She pleaded a lack of knowledge and then proceeded to reveal fragments of understanding that were invaluable. If her story has a parallel here, then the fragments of understanding that have emerged from this very personal review look something like this. I am convinced that poststructuralist participatory action research has the power to enrol people in processes of resubjectivation, and that subjectivity change is a necessary step towards marshalling the people power we need to make any transformation of our way of living to meet the current dire crises. Is this enough? Probably not. Design theorist Clive Dilnot warns us that design "is everywhere seen and everywhere remains invisible" . Others, such as Tony Fry, warn that if design designs, then we are both the designers and the designed, and thus we have inherited a world that has designed us as complicit in 'the unsustainment' . Action research has a role to play in cracking open this nut to reveal the discourses and infrastructures that invisibly channel our actions and thoughts. Practices of inventory and collective reflection are invaluable tools 48 Documents d'Anàlisi Geogràfica 2024, vol. 70/1 for making the invisible visible, for differentiating the world, for revealing the stark outlines of un-liveability and unsustainment, and provoking a shift in subjectivity. Clearly action research cannot rely only on human subjects changing without drawing in the supportive ecosystems that surround any community of actors to ensure the durability of a shift in subjectivity. If the ontological power of design resides in its capacity to achieve ubiquity and to 'disappear' into the culture at large, then action research must delve into that culture to identify what infrastructures of support need to be alerted and activated, and what institutions can be enrolled in transformational practice. If we are to contribute to the Great Transition our action research must enrol institutional actors as well as fellow citizens. The intractability of infrastructures such as laws, regulations, built forms, powerful barriers and walls, both figurative and material, cannot be underestimated. What we know is that the ethical force of design means that designed things have the power to alter the condition and behaviour of other things . So, experimenting with new ways of living and the infrastructures of support that enable these ways will have its own creative momentum. Today the 'economic common sense' of modernity, of The Great Acceleration, of the capitalist era, the unsustainment, , is starting to fragment. It is no longer, if it ever was, unquestioned. PPAR has a play to a role in The Great Transition by experimenting with new designs and testing and prototyping different ways of living that might become invisible in another possible world. --- Bibliographical references

In this paper I look back over three decades of doing participatory action research and extract some lessons that bear upon the pressing issues of our time-namely, how to act in the face of unsustainable socio-economic growth trajectories and cascading environmental degradation. I start with my circuitous journey into action research and to the importance of human subjectivity to transformation processes. I then trace the thinking that shaped a postmodern feminist social research agenda. Next follows an account of developing a research design for poststructuralist participatory action research (PPAR) that foregrounds understanding 'ways of living' in place. Drawing on actual action research projects, the lessons that emerge concern the importance of 1) inventory and bearing witness to difference; 2) new techniques of resubjectivation that encourage transformation; and 3) shifting the action focus from human subjects to socio-technical assemblages in which humans, material infrastructures, habits and experimentation are interwoven. Throughout my three decades of involvement with action research I have learnt about the limits to the durability of subjectivity change and the need to work on support from wider ecosystems of governance, infrastructure, health and social expectation.

experiencing low social network support . In particular, risk factors for depression among this group have been linked to having fewer perceived future opportunities ; low neighborhood social capital, kinship social support, and social network size ; older age ; and greater exposure to delinquent peers and violence . While treatable, there are significant racial and gender differences concerning who gets treatment for depression. Although research finds that depression is highly amenable to treatment , few children and adolescents, especially African American adolescents with a depressive disorder, receive care . In general, prior studies have shown that adolescents experiencing mental problems typically talk about these problems with informal sources, such as friends and family, prior to and/or in lieu of seeking professional help . The underutilization of formal mental health services among African American youth with depression has spurred an interest in examining factors that underlie their symptoms and subsequent help-seeking behaviors . For example, Scott, Munson, McMillen, and Snowden found that mental health problem identification and eventual help-seeking behaviors for African American adolescent boys may be compromised by genderbased notions of help seeking or masculine norms, that is, socialization reinforced by "macho messages" that suggest that males who show their emotions are weak. These help-seeking patterns are important to consider given that males account for more than 80% of suicide among African American adolescents . Given these prior findings and the underutilization of mental health services among depressed African American adolescent boys, the purpose of this study was to examine the influence of mental health stigma and social support on depressive symptoms and subsequent help-seeking behaviors for this population. --- Stigma and Depression Much of what is known about mental health stigma among children and adolescents is based on adult studies of mental illness, relies on adults' perception of childhood stigma rather than children's own perceptions, and often does not include race-specific information . In a recent study, Perry, Pescosolido, Martin, McLeod, and Jensen found that adults perceived that children with depression may be more vulnerable to stigmatization than adults. Childhood depression was also viewed as more serious than adult depression, requiring immediate formal treatment . In a study examining perceptions of mental health stigma among youth aged 8 to 18 years, Walker, Coleman, Lee, Squire, and Friesen found that depression and attention deficit hyperactivity disorder was more stigmatized than asthma, and depression elicited the most negative stigmatizing responses. That is, youth perceived that peers who were depressed were potentially dangerous, and these youth reported higher levels of social distance with depressed peers . Mental illness stigma is reflected in how young people describe their depressive symptoms and gender appears to play an important role. When presented an opportunity to describe their depressive symptoms, Moses found that ethnic minority youth, in particular, will not "self-label" or conceptualize their problems in pathological terms. In fact, males may be particularly sensitive to stigma associated with depression and mental health service treatment. Chandra and Minkovitz found that boys experienced more stigmas concerning mental illness and service use than girls. Results from this investigation also suggest that boys preferred to address their mental health struggles with their family. What remains unclear, however, is the nature of the direct influence of mental health stigma on depressive symptoms among adolescents, particularly among African American adolescent boys. According to Link, Cullen, Struening, Shrout, and Dohrenwend's modified labeling theory, labeling or stigma may either induce a state of vulnerability that increases the likelihood of a person experiencing repeated episodes of a disorder; or, labeling and stigma may be the cause of negative outcomes that place mentally ill persons at risk for recurrence or prolongation of disorders that emanated from other causes. In a study examining the association between mental health stigma and depressive symptoms in a sample of adolescents, Moses describes two forms of stigma, societal devaluation and self-stigma, and both were associated with higher levels of depressive symptoms. Interestingly, findings from the Moses study also reflected those youths' perceptions that their network members would not stigmatize them for having an emotional/behavioral disorder. Rather, they felt their peers would tease them about using mental health services, and shame was accorded to receipt of formal mental health treatment . It is unclear from this study whether these findings differed based on ethnic identity. --- Social Support and Depression Social support has received considerable attention in the adolescent literature as a main and interactive effect on adolescent responses to stressors . As a main effect, research findings among African American adolescents have indicate that higher levels of social support predict lower levels of depressive symptoms for this group . Research on the interactive effects of social support suggests that this construct may interact with other factors, for example, stress, to buffer or moderate the deleterious effects of stress on an outcome such as depression . For example, in the presence of stress, individuals who have a high level of social support might be protected from experiencing a related negative reaction to the stressor. Consistent with this proposition, Zimmerman et al. examined the "stress-buffering" effects of social support on depressive symptoms among African American adolescent boys. They found that parental support had a buffering effect on stressful events for predicting depressive symptoms crosssectionally; however, depressive symptoms did not activate more parental support for African American adolescent boys over time . While the study by Zimmerman et al. offers support for the stress-buffering effect of social support, findings on the interactive effect of social support among adolescent samples have been mixed. For example, McCarty et al. found a direct effect of social support on depressive symptoms but did not find support for a moderating influence of social support by gender among a sample of delinquent youth. In a study examining dating violence victimization and depression among African American and Caucasian adolescents, Holt and Espelage found a moderating effect of maternal social support on the association between victimization and anxiety/depression for African American males. Thus, the stressbuffering effect may only hold for certain types of behavior , specific sources of support, and the type of problems evaluated . --- Help-Seeking Pathways and the Influence of Social Networks Concepts derived from the network-episode model are well suited to exploring how African American adolescents' networks might influence their response to depression. The NEM suggests that help seeking is not an isolated, rational action; rather, it is a process involving influence from individuals in one's social network: family members, friends, and close acquaintances . As social beings, individuals who experience psychiatric illness may have people in their lives that confirm the presence of a problem and help them to negotiate a resolution. This may include identifying or providing emotional support, reminding them of appointments, offering transportation, or ensuring that they adhere to appropriate aftercare. Social network members might also transmit stigma and negative messages about formal services, thereby influencing both an individual's mental health symptoms and their helpseeking behaviors. Examining the social processes within an adolescent's social network will help clarify the mechanisms that underlie their problem identification, symptom expression, and help-seeking behaviors. Previous studies have failed to account for the unique cultural and contextual influences on help seeking among ethnic minority youth, including African American adolescents . Understanding these social network influences may provide insight into why contingencies such as age, race, and gender do not fully explain what precipitates their psychological needs and help-seeking behaviors. --- Current Study The extant literature on mental health stigma, social support, and depressive symptoms among African American adolescents is limited. Given the elevated risk for untreated depression among African American adolescents , the higher risk for suicide among African American adolescent males and the strong relationship between depression and suicide risk , a focus on depressive symptom expression and its underlying factors may be particularly salient to early identification of mental health need for this group. Thus, the purposes of this study were to examine the ability of mental health stigma and perceived social support to predict depressive symptoms among urban African American adolescent boys and explore the social network influences on initial problem recognition and eventual help-seeking attitudes/behaviors among adolescents who exhibited elevated depressive symptom levels. Although this study relied on a small convenience sample, it is one of a limited number of studies directly examining the help-seeking behaviors of depressed, African American adolescent boys. Given the disproportionate underutilization of mental health services for depression among African American adolescents and its cost to society , it is of interest to clarify how social and familial factors influence their depressive symptoms and help-seeking behaviors. Based on extant literature, we expected to find direct effects for mental health stigma and social support on depressive symptoms for this sample. We also expected to find support for the buffering effect of social support on the relationship between mental health stigma and depressive symptoms. Furthermore, through a qualitative exploration of perceptions of social network influences among African American adolescents with higher levels of depressive symptoms, we expected a greater understanding of the processes and mechanisms by which symptoms were initially identified and the subsequent influence of social support on helpseeking behaviors. --- Method Study Design The current study applied mixed-methods research, in particular a triangulation design, to address the research aims. Specifically, a variant of the triangulation design, validating quantitative data model, was employed whereby quantitative and qualitative data were concurrently collected to best understand the substantive area . Thus, quantitative methods were used to identify factors associated with depressive symptoms and help-seeking behavior, while qualitative methods were used to document critical processes and mechanisms through which adolescents understand and respond to the phenomenon of depression. The aim of the qualitative interviews was to describe rather than verify phenomena. In particular, the investigators sought thick description via the use of qualitative interviews to gain understanding of the unique features of support, as well as content about depressive symptom recognition and subsequent behavioral responses by participants, including their interactions with social networks when experiencing depressive symptoms. --- Description of Setting Participants were recruited from mental health treatment and community-based settings, including two outpatient community mental health centers, a psychotherapist in private practice, and three community-based programs for high-risk youth in a large Northeastern city. These programs were chosen as sample sources because the youths served in these settings were predominantly African American, they met age requirements for inclusion in this study, and many of the clients served in each of the treatment/ nontreatment settings were identified as being at risk for depression based on their residence in poor, highly stressed communities, multiple family disruptions, and other associated stressors. This study received institutional review board approval . Depending on the setting, therapists or program staff identified potential participants according to established criteria on age, ethnicity, and gender. Informed consent of the parent/guardian and child assent was secured prior to initiating any data collection procedures. Once consented/ assented, participants were then administered a survey that included the Centers for Epidemiologic Studies Depression Scale as an indicator of depressive symptoms. Additional survey items assessed mental health stigma, social support, and demographic characteristics. Eighteen of the 69 participants had elevated depressive symptoms based on the CES-D measure and agreed to participate in the qualitative interview. More information on the CES-D measure is described in the measures section below. --- Data Collection Data were collected from participants in one interview session, which lasted 45 to 60 minutes or up to 90 minutes for those participating in quantitative and qualitative studies. Interviews were conducted in private rooms located within the mental health centers, community-based organizations, or homes of participants. Participants received a $25 gift certificate for participating in the quantitative study and a $35 gift certificate for participating in the quantitative and qualitative studies. --- Quantitative Measures Mental health stigma-The Attitudes Toward Seeking Professional Help Scale assessed respondent's personal attitudes about seeking professional help. In particular, the construct was operationalized via the measure stigma tolerance, the extent to which participants might be concerned about what others might think if he visited a mental health professional for treatment. The original measure contains 29 items that represent four subscales: in addition to stigma tolerance, recognition of need for psychotherapeutic help, interpersonal openness, and confidence in mental health practitioner. Given its relevance to the original study aims, only the stigma tolerance subscale was administered to participants. Study participants rated on a 4-point Likert-type scale, ranging from 1 to 4 . Scores ranged between 0 and 30: Lower scores indicated sensitivity to what others would think if he visited a mental health professional for treatment; higher scores represented less concern about what others might think . Based on reliability statistics, one item from the stigma tolerance subscale was removed from the final scale used in final analyses . The internal reliability estimate for the stigma tolerance subscale was .65. Social support-The Social Support Scale assessed the amount of perceived availability of emotional and practical support from individuals comprising the participant's social network . Participants rated on a 5-point Likert-type scale, ranging from 1 to 5 , each network member relative to the amount of support they received. If a respondent did not perceive the availability of support from the indicated network member, "not applicable" may be selected. Ratings, summed for each network member across both emotional and practical support, yielded a final total support score. The internal reliability estimate for this sample was .90. Depressive symptoms-The CES-D is a 20-item scale that assessed depressive symptoms among adolescents in the study. Participants were asked to indicate on a 4-point scale how often during the past week they felt depressed, lonely, sad, unusually bothered by things, or could not get going; higher total scores indicate more depressive symptoms. Total scores on the CES-D have a range of 0 to 60. The internal reliability estimate obtained for African American adolescents in past studies ranged from .64 to .89 . Although several of these studies have used different cutoff points as indicators of being at risk for later development of clinical depression, a cutoff point of 16 was used as a risk indicator and an inclusion criteria for the qualitative interviews in this study. Prevalence of depressive symptoms using CES-D caseness criteria of 16 or greater is in the range of 45% to 55% . Twenty-six percent of this sample scored 16 or better on the CES-D. The internal reliability estimate was .84 for this sample. Demographics-To explicate a profile of the background and family characteristics of participants in the study, demographic information was collected from each participant, including age, current living arrangements and family structure, parent/caregiver level of education, frequency of religious service attendance, and social network size. Examples of questions included the following: Are you currently living with your biological family? Who are your primary caregivers? Who are the people in your life right now who you feel you can depend on for help if you need it? How often do you attend religious services? --- Qualitative Interview Procedures Open-ended questions based on the NEM constituted the qualitative interview. In particular, questions pertaining to the NEM concept network content assessed the degree of support from network members, as well as members' attitudes and beliefs toward mental illness and mental health treatment. In addition to network content, questions were also derived from the literature on help-seeking behaviors among adolescents , as well as the literature on mental health service utilization among African Americans. The first author and a trained research assistant conducted the interviews. Participants were encouraged to talk at length about their help-seeking behaviors in relation to their depressive symptoms, with detailed accounts regarding the ways their network influenced their behaviors. They were also asked how they conceptualized and defined mental health and associated depressive symptoms . --- Analytic Strategies Quantitative Data Analysis-The quantitative component involved univariate and multivariate analyses. First, univariate analyses were performed to provide a demographic background for the sample, including a description of their social network characteristics. Second, hierarchical regression analyses were performed to determine the degree to which predictor variables were associated with depressive symptoms and whether social support moderated the association between mental health stigma and depressive symptoms. The hierarchical regression analyses were performed in three steps. In Step 1, age and network size were entered. In Step 2, mental health stigma and social support were entered as predictors. Finally, in Step 3, an interaction term, mental health stigma × social support, was entered to determine the potential moderating influence of social support. In accordance with recommendations by Cohen, Cohen, West, and Aiken , the predictor variables were centered to their respective means to adjust for multicollinearity. Regression analyses were conducted with the centered values of mental health stigma and social support. R 2 statistics across the three steps were evaluated to determine the degree to which predictors were associated with depression. A significant interaction was plotted following guidelines using a word processing application. The y-axis reflects predicted depression values for each group from the full regression equation including unstandardized beta weights and the constant . Analyses were conducted using SPSS version 16.0. Qualitative Data Analysis-Qualitative interviews were tape recorded, transcribed, and analyzed using inductive coding techniques . Three readers, including the first author and two research assistants, independently reviewed and coded transcripts to identify patterns and themes emerging from the data. After review and designation of codes, the readers convened consensus sessions to determine the categories and subcategories of themes and to establish interrater reliability. Through an iterative process involving all reviewers, the research team developed a final coding matrix. The final coding matrix denoting the category and subcategory of themes provided a definition clarifying the meaning for each category and subcategory and identified corresponding sample quotes that best captured the theme. --- Results --- Descriptive Analyses Participants in this study had an average age of 15.3 years . Eighty-five percent of the participants were living with their biological family at the time of the interview. Fifty-five percent were being reared in mothers-only families, while 29% lived with both their mothers and fathers. The participants, on average, lived with 3 to 4 siblings. The mean CES-D score for the sample was 12.01 , and scores ranged from 0 to 45. See Table 2 for overall sample characteristics. --- Multivariate Regression Analyses Predicting Depressive Symptoms Hierarchical regression analyses were performed to determine the unique associations with depressive symptoms among the study variables. In Step 1, age and network size were not significant predictors of depressive symptoms. In Step 2, both mental health stigma and social support were significant negative predictors of depressive symptoms, accounting for 22% of the variance. In the final step, the interaction term, mental health stigma × social support , was a significant positive predictor of depressive symptoms, accounting for 35% of the variance; an increase of 12% from the variance explained in Step 2. See Table 3 for full results. --- Qualitative Results Questions and probes in the qualitative interview assessed social network influences on helpseeking behaviors among participants with elevated depressive symptoms . Qualitative findings illustrated the behaviors engaged in when initially experiencing depressive symptoms, how family members recognize/identify depressive symptoms, and turning to family members as a first option for help. --- Initial Behavioral Responses to Depressive Symptoms Eleven of the 18 participants shared that they initially spent time alone when dealing with depressive symptoms. One participant noted: "I just stay in my room and lock the door. That's it … I don't want to be bothered." While in isolation, participants might engage in reading, listening to music, playing video games, or watching television to help ease the pain. These leisure activities might also temporarily serve as a distraction from ruminating about problems. Time alone might additionally provide the opportunity to "self-reflect," a behavioral response reported by 6 of the 18 participants. As one adolescent male noted, "I try to go within myself so I pretty much get the answers. It's like a self-conscious." On the other hand, depressive symptoms might conjure up negative feelings regarding the emotional pain, which then, as reported by 7 out of 18 participants, precipitates anger as a coping response. As noted by one participant, "I might take my anger [about being depressed] out on others because I want them to hurt like I do." --- Family Members First to Recognize/Identify Depressive Symptoms The majority of participants reported it was a family member who helped them to initially identify their depressive symptoms. One participant noted: Well, my mom, and my aunt, and stuff like that, or granddad, they could actually sit there and tell like when I'm not, when I [don't] want to be bothered. I'll sit there and like I block myself off from everybody else. Usually I'm like [an] active talkative type person. Laugh, comedy person. But then when I start becoming-the antistuff then they know that something is wrong [Referencing the experience of having depressive symptoms]. Among participants who did not have strong familial support , identifying their depressive symptoms was something they did in isolation, as one participant noted: "It [identifying the problem] was something I did on my own." --- Help Seeking: Go to Family Members First In terms of help-seeking behaviors, the majority of participants discussed turning first to family members for help with depressive symptoms and felt comfortable with taking their problems to family members before going to "outsiders" . Most often mothers played a particularly prominent role among participants who indicated that their mothers would have the "best" perspective regarding their well-being and would provide useful feedback on what they might do to resolve the problem. The following quote captures the theme that family members are the best source and first resort for help when dealing with depressive symptoms: My first resort [when dealing with a depressive symptom] would be to go to my family … you really don't want to skip over them because they really have a good input on you and they really look at you everyday and know a lot of things about you so they could also help you. --- Distrusting of Mental Health Professionals and Peer Networks Participants were likely to experience the combined effects of family support and lack of trust in professional providers. In particular, they shared their negative perceptions of mental health professionals-perceptions that were often transmitted to them by their family members. Participants questioned the authenticity and genuineness of mental health professionals and indicated that professionals would not be able to break down perceptual barriers associated with accessing treatment-barriers family members would more easily penetrate: They [MH professionals] can't-I mean they don't really know you until you actually spread yourself out to them… . Like something to break the ice because everybody has defensive mechanisms. They have to defeat them first to get to know the problem. And when it's someone you know like your family sometimes you may put up defensive mechanisms against them, but also they know your tendencies well. That's why they're there helping you. Eight out of the 18 participants associated the use of formal mental health treatment for depression with feeling "shame." For participants in this study, having depressive symptoms meant that friends would generally not be as supportive as family members. Even expressing an emotion such as "crying" is seen as a sign of weakness among friend networks. Crying as a result of some emotional pain might precipitate feelings of vulnerability and weakness, and these feelings might be conveyed to the peer referent group. If conveyed to the peer referent group, the potential for antagonism increases by peers due to the expression of emotional pain. Thus, peer influence can play a huge role as the one experiencing depressive symptoms engages in risk-aversive behaviors and therefore attempts to save face by masking his symptoms under the guise of being tough and strong among peers. --- Discussion Overall, findings indicate that mental health stigma and social support both influence depressive symptoms among African American adolescent boys, albeit in different ways. Both constructs also seem to have some influence on eventual help-seeking behaviors. In particular, our quantitative findings support a main and potential moderating effect of social support on the relationship between mental health stigma and depressive symptoms. As a main effect, social support was negatively associated with depressive symptoms. This finding is consistent with previous literature suggesting the positive influence of social support on depressive symptoms . Interestingly, in the Hall et al. study, this main effect was only found for girls. As an interactive effect, regression coefficients indicated that the negative relationship between mental health stigma and depressive symptoms increased with every increase in level of social support. Thus, the existing negative relationship between mental health stigma and lower depressive symptoms was enhanced by the presence of social support, suggesting the potential buffering effect of this construct. Given the exclusive gender focus of this study, findings reflect that social support among African American boys may play an important role in lowering depressive symptoms, particularly when these youth face stigma regarding mental illness and service use. Qualitative findings in this study helped elucidate the quantitative findings and provide a context for the observed pattern of relationships. In particular, participants reported that their network members were able to "see through" or discern their depressive symptoms. Family members, in particular, were a first source for help. Consistent with this proposition, Scott and Davis noted that African American males "are particularly averse to sharing personal vulnerabilities" outside of their family members . They further note that help seeking for African American males is bounded within contextual stressors and that these larger systemic stressors influence their interactions with and unwillingness to connect with "professionals" . Findings from this study suggest, at least partially, that participants felt it was safe to be vulnerable among family, if at all. Draucker noted a particular pattern among African Americans relative to Caucasians in a retrospective study of help-seeking behaviors among young adults who were first diagnosed with depressive disorders as adolescents. African Americans in the Draucker study reported heavy reliance upon their social networks. Mental health research has also consistently documented this reality for African Americans. For example, prior research highlights African Americans' reliance on family and informal help sources when experiencing emotional/ psychological problems , their stigma of mental health service use , and fear of being perceived negatively by friends and family . Based on findings in this study, similar help-seeking patterns and reliance upon the family for emotional/practice support seem to exist for African American adolescent boys. Peers may be a particularly challenging peer referent group for African American adolescent boys with depressive symptoms. Extant research on African American males' psychological response to distress and illness also suggests that their display of vulnerability is, in fact, considered weakness among their male counterparts . In this study, not only were friends generally viewed as unsupportive, but for the adolescent boys in this sample, it was also important that they did not exude or display any depressive symptoms among their male friends. For example, asking for help to deal with depressive symptoms was viewed as weak, and shame was accorded to seeking or using professional mental health services to address symptoms. Consistent with previous research on gender and help-seeking behaviors , our findings suggest that boys may be negatively influenced by peer perceptions of mental illness and help seeking. --- Linking Qualitative and Quantitative Findings Similar to previous studies regarding social network influences on help seeking and mental illness , quantitative and qualitative findings indicate that networks play a particularly important role regarding stigma associated with depressive symptoms and eventual help-seeking behaviors. In particular, quantitative findings suggested that social networks may buffer potential negative perceptions associated with depressive symptoms. The qualitative findings provided further details regarding what was learned quantitatively by revealing the ways in which both families and friends induced, transmitted, and endorsed stigmas regarding help seeking for depressive symptoms or how one might respond to symptoms. Stigma is a powerful influence and is based on both actual and perceived experiences, which may be largely based on experiences depressed, African American adolescents boys have within their social network. Family members and peers, however, differentially influence help-seeking behaviors related to depression. Family members were the first point of contact for depressive symptoms, offering support for depressive symptoms, whereas peers incited both negative feelings about symptoms and a reluctance to ask for or seek help. Professionals were similarly viewed with skepticism. --- Limitations Confirmatory and comparative analyses from the perspective of actual network members would have been desirable, but limitations of time and funding restricted this study to the adolescents' perceptions of their depressive symptoms, perceived mental health stigma and social network support, and social network influences on help-seeking behaviors. For example, if the investigators had data from parents and peers, further triangulation and verification of the data could have been done. Sample size was expected to be restricted because of the small number of African American adolescent boys who were participants in the community-based programs that provided participants for the study. Thus, this expected small sample size contributed to the decision to use mixed methods that would yield more robust data than would be derived from one method alone . An interview procedure was chosen as the data collection strategy because personal contact enhanced the comfort level and facilitated the cooperation of the adolescents who participated in the study. A limitation, however, of this strategy is that participants may have provided socially desirable responses to the survey items and open-ended questions. Our measurement of mental stigma was based on measure previously developed for adult mental health services research and may not be generalizable to the unique experiences of African American adolescent boys. What the measure lacked in cultural and developmental sensitivity, the qualitative study, as part of the mixed-methods design, attempted to augment important gaps in the measurement of stigma. Additionally, researchers have differentiated public versus self-stigma in studies examining this construct . For this study, however, we were only able to assess public stigma, that is, stigma tolerance. A more robust measurement of stigma among the participants in this study would provide a richer understanding about which forms of stigma are more influential in both depressive symptom expression and help-seeking behaviors. The CES-D, as a measure of depressive symptoms, is not equivalent to a clinical diagnosis of depression. A study of ethnic minority adolescents, however, found the CES-D to be predictive of later diagnosis of a depressive disorder, including major depressive disorder or dysthymic disorder ; the positive predictive value of MDD based on CES-D cutoff scores of 16 was 17%, 13% for DD, and 15% for having both MDD and DD . Other studies, which included African American adolescent participants, similarly indicate the predictive ability of the CES-D regarding later diagnosis of MDD or DD among adolescents . Thus, the depressive symptoms described by participants in this sample may be more consistent with apparent moodiness and not clinically diagnosed depression for which treatment might be required. Given the success of the CES-D in predicting later clinical diagnosis of depression among adolescents, however, this measure may be best suited as an initial screen to confirm the need for more intensive follow-up to confirm a clinical diagnosis of depression. Confirmation of a depressive disorder was not the purpose of this study. Therefore, our findings may be limited to African American youth who exhibit early signs and symptoms of depression and not actual clinical depression. Finally, given the cross-sectional nature of the data collected, we were not able to determine causality among the study variables. Future research is warranted to examine the influence of mental health stigma and social support on depressive symptoms and help-seeking behaviors longitudinally to determine the true effect of these constructs for this population. --- Implications for Research and Practice Given the findings of this study, future investigations regarding the design of engagement strategies with the families of African American adolescent boys might improve depression symptom identification, stigma reduction, and formal care seeking for this population. In particular, our findings suggest that mental health practitioners should make concerted efforts to increase their connections to the family social support system of African American adolescent boys with depression to remediate symptoms and enhance the help-seeking behaviors of this group. Moreover, agencies might increase opportunities for families to be involved in the treatment process by providing services at convenient hours or offering other supportive services such as child care and transportation. Future research should examine the efficacy of antistigma mental illness interventions to help foster a more supportive peer environment for African American adolescent boys with depression. Because of their close proximity to youth, school-based mental health practitioners have a unique opportunity to engage adolescents in the development of schoolwide antistigma campaigns to encourage sensitivity to the mental health challenges faced by students and to educate students and school personnel about the signs and symptoms of mental health problems such as depression. To develop such campaigns, practitioners might pay particular attention to the rich narratives provided by participants in this study regarding their perceptions of mental illness, services, and providers. Finally, when treating a population such as African American boys with depression, it might be especially important for providers to display an openness and willingness to explore perceptual barriers held by this group concerning depression and formal service use. Having an honest conversation about the perceptual barriers that have the potential to negatively influence therapeutic engagement and alliance demonstrates the provider's sincerity and promotes an interaction based on trust. Given this study's findings regarding the positive effects of family support, practitioners might also explore the adolescent and caregiver's convergence regarding perception of depressive symptoms and identify cultural/gender-specific preferences and potential barriers to help seeking and formal service use.

African American adolescent boys underutilize mental health service due to stigma associated with depression. Gaining an increased understanding of how depressed, African American adolescent boys perceive their mental health needs and engage in help-seeking behaviors might play an essential role in efforts to improve their symptoms and access to care. Using a mixed-methods design, this study examined the influence of mental health stigma and social support on depressive symptoms among African American adolescent boys. Findings indicated the protective effects of social support in decreasing depressive symptoms, especially when participants experienced mental health stigma. Results also revealed the pivotal role of family social support over both professional and peer support for participants who struggled with depressive symptoms. The primacy of family support among the sample, combined with the frequent distrust of professionals and peer networks, would indicate that working with families may improve initial identification of depression among African American adolescent boys and decrease their barriers to care.

Background Sex and reproductive health issues affecting adolescents in developing countries have received more and more attention in the recent 30 years since 1994 [1]. There are increases in policies and programs to improve their situation, but most focus on older adolescents. Early adolescence is generally considered one of the healthiest life stages [2]. However, their status -dealing with the emerging romantic interest and sex desire in the context of rapid puberty changes -has been overlooked [3]. SRH contributes significantly to the health status of young adolescents even while the majority has not engaged in sexual intercourse. Unsafe sexual behaviors are major contributors to disability-adjusted life years lost for very young adolescents in nearly every region of the world [4]. As young people transition from the early to late adolescent years, sexual and reproductive behaviors contribute to diverging mortality and morbidity patterns by gender, with young girls facing an increased risk of experiencing sexually coercive interactions, contracting STIs , as well as suffering the gender-specific consequences of unintended pregnancies and psychological trauma [4]. Studying very young adolescents' romantic and intimate sexual practices could help inform the context in which most sexual behaviors occur [5,6]. Researches on adolescents' sexual trajectories from western and African countries indicated that normative heterosexual behavior among older adolescents develops in an orderly, progressive sequence: from noncoital sexual interactions, such as kissing, hugging, fondling or petting, generally precedes engagement in the next [7][8][9]. Recently, there is a study in Kenya explored young adolescents' intimate sexual practices before sexual intercourse through latent class analysis [10]. It identified three subgroups of youth characterized by involving, observing or being naïve to romance, suggesting merely looking at the prevalence of sexual intercourse would miss the point for early prevention. However, most of the above studies were in the US, Europe, or Sub-Saharan Africa. The information on the normative romantic and sexual-related experiences from very young Chinese adolescents is scant. China is quite different from Western and African countries both in sex behavioral patterns and sex-related norms [7,11,12]. Some researchers argue that Chinese populations are strongly influenced by Confucianism and believing that women should stay virgins until they get married, and would rarely think of having sexual intercourse before age 15 [13,14]. However, significant and rapid social changes have taken place in China's coastal regions, majorly because of the urbanization and westernization during the past four decades [7,15]. Attitudes towards romance and sex also changed dramatically [16]. The sexual interest among middle school students is considered normative now [17]. Some middle school adolescents have begun to explore intimate relationships. The present study is guided by Blum's conceptual framework regarding early adolescence research based on the ecological model [18]. Young adolescents' behaviors could be influenced by both the characteristics of individuals , and the nature of the contexts -such as families, peer groups, schools, and neighborhoods' environments -within which they socialize, study, and reside , as well as the policies and norms they were exposed to [3,18]. For instance, a review indicates that older age, early onset of puberty, being male, and using alcohol were some of the key risky individual factors for ever had sex among youth in developing countries [19]. No parental care and less family stability are considered risky for early sex [19]. Related research points out that befriended adolescents are likely to be similar to each start romance before college. This study used the Global Early Adolescent Study Shanghai site baseline data collected among1776 teenagers to explore their romantic and sexual experiences patterns and identify the influencing factors. There were significant gender differences in teenagers' romantic and sexual-related behaviors. According to the multi-group latent class analysis, we classified the teenagers into three classes: the general group, the early romance group, and the sex exploratory group. The multi-nominal logistic regression showed youth in the early romance group were more likely to had friends of both boys and girls, ever had boyfriend/girlfriend and were freer to go out than the general group; while boys in the sex exploratory group were older, ever had a boyfriend/girlfriend, believed it was normal for boys to be sexually active than girls, had more freedom to go out, and had less perception of care from school adults and neighborhood. Girls in the sex exploratory group were older and less likely to make decisions on their own. Given the result we found, we call for health researchers and educators to conduct sex education programs promoting more equal gender norms and provide services to help young adolescents get right information and strengthen their skills in communication, decision making and critical thinking. other in behaviors and characteristics [20]. Care from school adults is a protective factor for youth in developing health risky antisocial behaviors [21]. Neighborhood disadvantages are correlated with risky adolescent sexual behaviors such as early first intercourse and abortion [21,22]. Macroly, Interactive media use has become ubiquitous among adolescents [23]. Given the everyday use of smartphones and related technology in adolescents' lives, it is also essential to include sexually related media use and electronic communication within the macro-context of romantic and sexual development [24]. Harmful gender norms, such as sexual double standards , are reported to be correlated with more permission for casual sex among adolescent boys in both white and Asian Americans [25]. However, such correlations tend to work through adolescent peer acceptance in different gender [26]. As cultural differences and contextual changes exist in the developmental stage of sexuality [7,27], it is valuable to explore very young adolescents' romantic and sexual behavioral patterns in the present context of China. This paper aims to add our knowledge mentioned above using the GEAS baseline cross-sectional data collected among young adolescents ages 10-14 years in a less developed community of Shanghai. We hypothesize that the proportion of romantic and experiences among young adolescents varied among boys and girls while the patterns would be identical at some point but would also be different due to the micro-, meso-, and macro-factors [18,28]. --- Methods --- Study design, participants, and procedure Data were extracted from the baseline study of the Global Early Adolescent Study. GEAS is a 5-year longitudinal study implemented through a collaboration of university and research institutions globally, focusing on gender norms in early adolescents(aged [10][11][12][13][14] and their relation to adolescent health, especially sexual and reproductive health, in disadvantaged urban environments. The GEAS seeks to estimate the frequency of different events among boys and girls, respectively, with sufficient precision. Given the lack of information on romance and related behaviors among youth aged 10-14, we assume a frequency of 50 % and calculated that inclusion of 450 boys and 450 girls would allow a precision of ±4.6 %. Considering the annual attrition rate of 10 % requires the initially enrolled sample of our site to be approximately 1450 adolescents aged 10-14. A stratified cluster sampling procedure was adopted for the selection of participants in the Shanghai site. Three primary public middle schools in two less-developed sub-districts of the Jing'an district were selected. The STROBE cross-sectional reporting guidelines were used as a checklist to draft the manuscript [29]. All study procedures in Shanghai were approved and monitored for ethics by the Medical Ethical Committee of the Shanghai Institute of Planned Parenthood Research . Parental consent forms of the study were sent to their parents and then collected through class headteachers. Youth assent was collected before the data collection. Data were collected through tablets using the Computer-Assisted Self-Interview method during November and December in 2017. The fieldwork was implemented with the coordination of key informants from the local teachers' organization. The eligible students in grades 6th to 8th were organized by their teachers in the class units to independently fill in the electronic questionnaire. In each class, 1-2 trained investigators were present in case the participants need assistance with the tablet using or clarification of the unclear question they encountered. The survey took approximately 40 to 60 min to finish. The tablets were returned after the process and checked by the investigators to ensure that all necessary questions were answered before submission. Each student was compensated for their participation with a small gift valued at 20-30 CNY after the process. Ultimately 1776 adolescents completed the baseline survey using tablets. We excluded 16 participants who were over 14 years old and 47 respondents who had a high share of missing data. The final analytic sample was 1714 . --- Measurements --- Romantic and sexual related experiences We asked "Have you watched pornography before?" by adding an explanation of "movies or videos that show people's private parts during sexual scenes" to help young adolescents distinguishing them from other romantic scenes like kissing or hugging. Options "rarely", "sometimes", and "often" were combined and coded as ever, while "don't know" and "refuse to answer" were coded as missing. We included ever flirted using social media as another indicator reflecting the romantic or sexual related experiences via media, and ever held hands, ever hugged/cuddled, ever kissed, ever touched/ fondled private parts, ever had sexual intercourse/oral sex/anal sex as indicators reflecting the romantic and sexual related experiences of direct contact. Options with either "yes, with a boy", " yes, with a girl" or "yes, with both" were coded as ever. --- Covariates We included individual factors such as age, gender, personal autonomy on freedom of movement and decision-making [30]. For the above two scales, the mean score of each scale was calculated, with a range from 1 to 4. We then included meso factors like parental monitoring, peer structure, peer drinking behaviors, romantic engagement, as well as school connectedness and neighborhood cohesion . For the macrolevel factor, we included one aspect of the gender norms that could influence adolescents' behavior selection: the scale of sexual double standard developed by the GEAS research group [31]. A mean score and a median score across items were calculated; participants whose mean score fell below the median were classified in the lower group, whereas those who scored equal to or above the median were organized in the higher group. --- Analysis strategy We used a person-oriented approach, the exploratory latent class analysis, as it is well-suited for studying romantic and sexual-related behaviors because of their complex, multidimensional nature [32]. Besides, analyzing the romantic and sexual-related behavior patterns, rather than individual behaviors, can bring additional insight into consideration and find the difference with respect to the influencing factors among each pattern of the behaviors [33]. Multi-group Latent Class Analysis is used because the probability of being in a given class could vary by the observed variable group . In terms of LCA, a solution of classes is optimal when classes are as homogeneous as possible while differences between classes are as large as possible. Classes are added until the model fits the data well. Fifty random sets of starting values were used. Models ranging from 2 to 5 classes were evaluated. Model fit was evaluated with the following commonly-used statistics: Akaike information criterion, the Bayesian Information Criterion, sample size adjusted BIC , and Entropy values. Small values of AIC, BIC and aBIC, and larger values of Entropy, indicate better model fit. LMR and BLRT indicate if this model is better than the model fitted using the previous one. This paper's primary purpose is to use LCA and MLCA through Mplus 7.5 to examine the heterogeneity, or variation, in romantic and sexual related behaviors among the sample and identify different types of subgroups of youths across gender assuming heterogeneity across gender exists. The secondary objective was to examine how the identified subgroups are differentially related to the demographic, environmental, and contextual covariates. Thus, following the LCA, an additional multi-nominal logistic regression was performed among each gender group and the total sample, respectively. The latent categorical variables were regressed on the covariates using Stata SE 15.0. --- Results --- Sample description A total of 1714 respondents were included in the analysis. Male and female youth were distributed equally. Most adolescents have experienced puberty onset, with females showed a higher percentage than males . Over 80 % of the adolescents were living with both parents. Though not so many, about 17.2 % of male respondents and 10.1 % of female respondents reported ever had engaged in a romantic relationship, with significant gender differences . Table 1 listed the details of the sociodemographic characteristics of the analytic sample. Table 2 shows the percentages of male and female adolescents utilizing each romantic and sexual related behavior and the comparisons between gender using the chi-square test. The prevalence was relatively low, and gender differences could be observed in the romantic and sexual related behaviors, including watching porn, holding hands with the one you loved, kissing, having the sexual touch, and penetrative sexual behaviors. Because the response rate of ever sexting and ever had penetrative sexual behaviors were too low, we decided to exclude these items in the latent class analysis. --- Identifying the latent class of romantic and sexual related experiences We compared models with 2 through 5 latent classes in both LCA models and MLCA models to determine the optimal latent class solution. Table 3 shows the result of the latent class analysis. The lowest AIC and aBIC was research in the 4-class LCA model, while the lowest BIC was reached in the 3-class LCA model. Latent class analysis with the grouping variable was used to examine the invariance in the latent class structured by gender. Similarly, the lowest AIC and aBIC was research in the 4-class MLCA model, while the lowest BIC was reached in the 2-class MLCA model. Taking the Entropy into consideration, the highest Entropy is in the 2-class multi-group LCA model , while the Entropy for 3-class and 4-class MLCA model is 0.91 and 0.88, respectively. Thus, we decided to select the 3-class MLCA model as the final best-fitting model. Figures 1 and2 showed the behavioral patterns among male and female groups based on the 3-class multi-group LCA model, respectively. Table 4 depicted the observed class memberships and the romantic and sexual related experiences' endorsement frequencies in the best fitting LCA solution. Table 5 listed the item-response probabilities indicating the likelihood of individuals in each latent class engaging in a particular behavior; the probabilities help interpret the classes produced by multi-group LCA. For example, Class 1 of both male and female were characterized by very low endorsement rates for most of the 6 sex and romancerelated behaviors, thus the group of both boys and girls was termed as the general group. Class 2 of both male and female were characterized by endorsing romance-related behaviors such as flirting, holding hands, and cuddling. Thus the group of both boys and girls was termed as the early romance group. Class 3 of male and female showed a different behavioral pattern: the male group was characterized by a very high endorsement of all the romantic and sexual related behaviors while a higher endorsement of pornography-watch characterized the female group. This group of both boys and girls was termed as the sex exploratory group. --- Latent class membership comparison on covariates A multi-nominal logistic regression was performed to assess covariates' influence in the predicted membership compared to the general group in the gender-specific sample and the total sample. As shown in Table 6, a number of significant effects were found in the covariates. Among males, youth in the early romance group were significantly more likely to have friends of both gender, had friends who drink, ever had a romantic relationship, and had more autonomy in deciding where to go, than the general group. In contrast, male youth in the sex exploratory group were older, ever had a romantic relationship, perceived more sexual double standards and had more autonomy in deciding where to go, while perceived less school connection and neighborhood cohesion. Among females, youth in the early romance group were significantly more likely to report having friends of both gender, ever being in a romantic relationship, and having more autonomy in deciding where to go, similar to that group of boys. Female youth in the sex exploratory group were of a --- Discussion We used cluster samples of young adolescents aged 10-14 in three middle schools of Shanghai to explore the patterns of romantic and sexual-related behaviors distributed in this age group. Unlike most prior work investigating a single sexual behavior, the present study used a multi-group latent class analysis to examine intricate patterns of the sexual and romantic experience. We also examined the factors correlated with different memberships in the latent class. The findings suggest that though most young adolescents have not started any romantic and sexual experiences yet, it is a natural process that needs more attention. In the membership classification, we found that compared to the general group, individuals in the early romance group were less likely to experience kissing and fondling but were more likely to experience holding hands and hugging. In western countries, kissing is a sign of a relationship and serves as relationship maintenance and potential romantic partners' assessment for further sexual attachment [34]. However, despite widespread acceptance that dating is becoming increasingly popular among adolescents, kissing and fondling are still somewhat taboo in Chinese culture [7]. While there is no clear definition of an appropriate age for individuals to begin a romance, those who showed their sexual interest at early ages will typically have to go against the adults [35]. Not surprisingly, Chinese parents and teachers tend to discourage young adolescents from becoming sexually active. Many adults are opposed to adolescent students being involved in relationships until they enter college. It might explain why we found that perceived no care from school adults was associated with the sexual exploratory group, and adolescents being classified in early romance and sexual exploratory group perceived more freedom of movement . Another aim of this study was to determine if there was a similar pattern of sex and romantic experiences among very young adolescent boys and girls. The findings suggested gender differences in their romantic and sexual behavior patterns as revealed in adolescents from other Asian countries [36]. While in the sex exploratory stage, males tend to experience all the possible ways to explore their sex interests; Females showed an entirely different tendency. For instance, girls were more reluctant to explore this sex interest through the interaction with the boys; instead, they resort to pornography to satisfy their curiosity about sex. The high rate of pornography use and flirting through social media but the absence of kissing and fondling in the girls' sex exploratory latent class group was noteworthy. Kissing and fondling are always relationally or romanticly meaningful. Therefore, the absence of kissing may connotate a lack of affection or emotional intimacy in this group. It echos the multi-group logistic regression results that the covariate romantic engagement was not significantly associated with this group compared with the general group. In the Chinese culture, where the Confucian philosophy is valued, women are expected to maintain the three rules of obedience: obeying their fathers and brothers prior to marriage, obeying their husbands within marriage, and as a widow, obeying their adult sons [12]. This set of beliefs, while seemingly outdated in contemporary society, nonetheless still has its impact. Indeed, several studies have suggested that even in the face of modernization and Western culture's influence, traditional gender attitudes may persist. Sexual double standards often refer to notions that people judge sexual behavior differently depending on whether the sexual act is a woman or a man. With traditional sexual double standards, it is more socially acceptable for men to engage in sexually permissive behavior. At the same time, women tend to receive stigma for expressing or pursuing their sexual desires [25]. Past research on female sexual development reveals that sexually active girls always are considered as "bad girl" or "sluts", while sexually non-active girls are thought as "good girls" or "virgins" [37,38]. The social changes of China, such as the urbanization and migration, which brought the income growth of citizens and the popularity of consumer culture, create conditions for materializing the female [39]. With the widespread use of mass media platforms and personal communication devices such as smartphones, the notion of sexual attractiveness and the obedience of females are strengthened by media [11]. Good girls are expected to be not sexual and then deserving of respect. Adolescents are highly vulnerable and susceptible to the information conveyed by the pervasive media. Thus, female adolescents might feel expected to prove or seek attractiveness through online activities or relatively private activities while simultaneously managing the perception of few sexual and romantic behaviors offline. Echoing the influence of stereotypical gender attitudes imposing on girls, we also found a significantly higher sexual double standard score among the boys in the sex exploratory group. According to social structural theory, the gendered division of labor and gender difference in power place men in a dominant role and women in a subordinate role [40]. Boys are expected to be more sexually active and show their prowess through more sexual experiences. That might explain why we see the more apparent differences in romantic and sexual experiences between gender. Sexual Scripting Theory would suggest that people are sexually socialized throughout the script they perceived. Previously conducted studies indicated that female adolescents and emerging adults who use pornography tend to have more sex partners, report copying behaviors seen in pornography, and engage in sexual behaviors at a younger age [41]. Our result showed that girls in the "sex exploratory" group were more likely to watch pornography. It may signal a critical intervention point for health researchers and educators. In addition, female adolescents may benefit from sex education programming that addressing gender equality and empowerment and promoting critical thinking skills for girls to mitigate downside effects from pornography viewing. Despite providing insights into the classification of romantic and sexual related behaviors among very young adolescents living in Shanghai's urban-poor setting, our study also provides some implications for intervention: First is cultural appropriateness, for instance, in the norms of Chinese adolescents, the freedom of movement means the autonomy of deciding where to go literally, which might be quite different of what Western or African adolescents think of as the freedom to have multiple romantic/sexual relationships/partners. Thus we should consider the cultural background when looking at the risk factors. Second, future interventions and sex education should target young adolescents aged 10-14 to mitigate the pervasive influence of stereotyped gender attitudes in sex development. Third, since peers and the media place more and more impact on youth, it is essential to equip them with the comprehensive knowledge and skill on sex and relationship as well as providing necessary caring and services: not only the knowledge about the biological differences of male and female, but also teaching them about how to see themselves and others as equal members in their relationships, the ability to protect their health, and as individuals, be capable of engaging as active participants in society [42]. Besides, it is also essential to acknowledge that it is normative for adolescents to be curious about sex and seek out sexual information or seek interaction online and offline. Therefore, experiencing romance and sexual behaviors should not be pathologized. This paper is not without limitations. First, we used self-reported data in the analyses, which could be biased as previous studies indicated that boys might overreport their sexual behaviors while girls tend to underreport such behaviors. Second, although we included several indicators shown in the literature to assess the patterns of adolescents' romantic and sexual-related behaviors, additional components regarding feelings were not included . Future works could examine how such uncovering factors inform group identification for better-tailored intervention. Third, the study used an urban poor, nonrepresentative sample to analyze, which prevents its generality to a broader range of adolescents in rural areas or country levels. However, it does show some unique characters of Chinese adolescents as the sexual scripts in China are quite different from western or African countries. Finally, as classes revealed gender and other ecological factor differences, researchers should carefully select covariates that were important to their research questions when planning the LCA analyses. There could also be other protective mediating or moderating factors not included. Nonetheless, this study's classified behavioral patterns helped identify the potential indicators of an increased risk for very young adolescents in China. --- Conclusions This study used a data-driven approach to reveal the romantic and sexual-related behavior patterns among very young adolescents. Our analysis showed three behavioral patterns using gender as a grouping variable: general, early romance, and sex exploratory. More disparities and different influencing factors were found in the sex exploratory group between boys and girls. Current sex education needs not only to be culturally appropriate but also to address the harm of gender inequality and stereotypes, as well as to provide accessible and supportive services to help young adolescents personalize their received information and strengthen their skills in communication, decision making and critical thinking. • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year --- • At BMC, research is always in progress. --- Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: --- --- Abbreviations GEAS: The Global Early Adolescent Study; SDS: Sexual Double Standard Scale; LCA: Latent class analysis; MLCA: Multi-group latent class analysis. --- --- --- --- Competing interests The authors declare that they have no competing interests. ---

Background: Studies on very young adolescents' romantic and sexual experiences would help inform the context in which early sex arises. However, such studies are scant in China due to sparse data and cultural issues. Method: This study used the GEAS baseline data conducted among1776 adolescents in Shanghai. Multi-group latent class analysis was used to explore adolescents' romantic and sexual experiences patterns and subgroups. Multinominal logistic regression was performed to identify the factors distinguishing different subgroups subsequently. Results: There were gender differences in the lifetime prevalence for very young adolescents' romantic and sexualrelated behaviors. The Multi-group latent class analysis indicated that the participants could be classified into three classes: general group, early romance group, and sex exploratory group. Multi-nominal logistic regression showed youth in the early romance group were more likely to had friends of both gender, ever had a romantic relationship, and had more autonomy in deciding where to go than the general group; while male respondents in the sex exploratory group were older, ever had a romantic relationship, believed that boys should be more sexually active and more proactive than girls, had more autonomy on deciding where to go, and perceived less school connection and neighborhood cohesion. Female respondents in the sex exploratory group were older and less empowered in decision-making than the general group.The result provides a picture of romantic and sexual behavior patterns among both gender of very young adolescents in China. Current sex education needs not only to be culturally appropriate but also to address the harm of gender inequality and stereotypes, as well as to provide accessible and supportive services to help young adolescents personalize their received information and strengthen their skills in communication, decision making, and critical thinking.Studies on teenagers' romantic and sexual experiences would help to prevent early sex arises before it really takes place. However, such studies are limited, especially in China. The traditional culture thought that youth should not

second category is further sub-divided into Baniharas or Dudhi Gujjars , and the Bakerwals .Baniharas or the Dudhi-Gujjars tend Buffaloes, sell milk and milk products and for this reason they are known as Dudhi . They live in Bans . Bakerwals keep herds of goats owing to which they are called as Bakerwals. --- Paharis: The Paharis are comprised of castes, creeds, sects and religions. They are Hindus, Muslims and Sikhs and are bounded to each other only through their language, Pahari. While social, cultural, rituals, lifestyle, dressing patterns and religious practice of these pahari communities is totally different from each other. Pahari community are concentrated in Rajouri and Poonch districts of Jammu province and some parts of Baramula, Kupwara, Bandipura, and Badgam districts of Kashmir province where these communities are predominant. --- The Scheduled Tribes The Scheduled Tribes of India have lived in varied degrees of isolation for many centuries. The term 'tribe' here means a group of people that have lived at a particular place from time immemorial. Anthropologically the tribe is a system of social organization which includes several local groupsvillages, districts on lineage and normally includes a common territory, a common language and a common culture, a common name, political system, simple economy, religion and belief, primitive law and own education system. According to the definition given by constitution of Indian, the Scheduled Tribes or tribal communities who have been so declared by the constitution and Article 366 defines Scheduled Tribes as 'such tribes or tribal communities or part of or groups within such tribes or tribal communities as are deemed under Article 342 to the Scheduled Tribes for the purposes of this Constitution'. In Article 342, the procedures followed for specification of a Scheduled Tribes are prescribed. According to lokur committee contain the following criterion for the specification of any community as scheduled tribe. a) Geographical Isolationthey live in cloistered, exclusive, remote and inhospitable areas such as hills and forests. b) Backwardnesstheir livelihood is based on primitive agriculture, a low value closed economy with a low level of technology that leads to their poverty. They have low levels of literacy and health. c) Distinctive culture, language and religioncommunities have developed their own distinctive culture, language and religion. d) Shyness of contactthey have a marginal degree of contact with other cultures and people. Background of the Study: Gujjar and Bakarwal are third largest ethnic group in Jammu and Kashmir after Dogra and Kashmiri. Gujjar-Bakarwal faced social stigma because of their lifestyle, culture and other distinct practice. The Gujjar-Bakarwal are seen as "brutish, dirty and dim-wited" and the term 'Gujjar' is a loaded and frequently used swear word. Historical records, oral and written, describe the injustice meted out to this marginalized community by the upper class Muslims ; Jattas, Mirza, Rajputs etc mostly during the rule of Jarals, who identified themselves as Muslim Rajputs, from the 12 th century to 19 th century in Rajouri and adjoining areas. Due to the social exclusion and Nomadism they were notified Scheduled Tribe in 1991 after a long struggle of around four decades. Majority of their population is concentrated in Pir Panjal region but they are spread across the Jammu and Kashmir. Gujjar and Bakarwal are nomadic and keep moving from one place to another in search of fodder for their livestock. The Constitution of India provides for various measures and provisions to safeguard their interests. But, the tribal community of Jammu and Kashmir is an exception to such measures and provisions as the same constitution grants special status Article 370 to the State of Jammu and Kashmir, because of which the tribals of J&K are not entitled to the benefits granted to tribals of other states. This difference lies because this state has its own constitution and the laws of Centre do not apply directly to this state. After the abrogation of article 370 on 05 August 2019, political reservation and Forest Rights Act 2006 were extended to the Union Territory of Jammu and Kashmir. Out of total 90 Legislative Assembly seats, 09 seats are reserved for STs in the state of Jammu and Kashmir. In Rajouri and Poonch districts five out of eight seats are reserved for Scheduled Tribe, where both Gujjar-Bakarwal and Pahari are in good numbers. The delimitation commission report has not gone down well with politicians of upper caste communities because they cannot contest election on these reserved seats for STs. They started politicizing the whole issue. The Bharatiya Janta Party cashed in on the growing tension between the two communities and started making promise to include Pahari community in the ST list. Gujjar Bakarwal states, that the people who were responsible for historical injustices against marginalized community are demanding ST status today. They ask how oppressor and oppressed can sail in the same boat. Paharis are neither backward nor poor and have no traits that would qualify them to be notified as tribe. It will be historic blunder if the Paharis are granted the ST status as it will defeat the purpose of reservation envisaged in the constitution by the framers of that great document. --- Scheduled Tribes of Jammu and Kashmir: In Jammu & Kashmir and Ladakh there are twelve scheduled Tribes: Balti, Beda, Bot, Brookpa, Changpa, Garra, Mon, Purigpa, Dard, Sheen, Gujjar-Bakarwal, Gaddi and Sippi. First eight Tribes got scheduled Tribe status in 1989. Rest four namely Gujjar, Bakarwal, Gaddi and Sippi got it in 1991 vide constitution Act 1991. After the Abrogation of Article 370 Jammu and Kashmir State Bifurcated into two Union Territory of Jammu and Kashmir and Ladakh recognized separate Union Territory. In Union Territory of Jammu and Kashmir there are now six scheduled Tribe communities, these are Gujjar, Bakarwal, Gaddi and Sippi, Bot, and Dard Sheen. According to the census of 2011, there is about 1493299 Scheduled Tribe populations in Jammu and Kashmir, which is about 11.91 % of total population of Jammu and Kashmir and about 1.43% of total Scheduled Tribe population of India. Out of the total population of Scheduled Tribe in Jammu and Kashmir, Gujjar holds the largest share followed by Bakerwal and Gaddis. Majority of the Tribal people belongs to nomadic and semi-nomadic community. --- Research Methodology The present study makes an attempt to examine ethnicity conflict among tribal and non tribal upper castes. It is primarily a fact finding study based on the explorative method. The study is based on both primary and secondary sources. A primary source includes government records, reports, and interviews. Secondary sources of data which includes books, journals, articles, daily newspapers, assembly debates, periodicals, Manuals. Hence this study is based on the historical, descriptive, and analytical approach. --- Objectives of the Study. The first and foremost objective of the study is to examine the non tribal inclusion in Scheduled Tribe List and its long term impact on the social, economic and educational aspects of the real tribes. --- Inclusion of Non Tribals in Scheduled Tribe -Emerging Changes and its Impact on Real Nomadic Tribe. The Indian constitution ensures certain protection for communities as having Scheduled Tribe status. But if any non tribal upper cast will get to enjoy the same status ,the basic idea of reservation will be completely abolished .To have ST status means to enjoy highly desired tangible benefits such as reserved seats in schools, government jobs and political reservation as well. With the Inclusion of Upper cast in the same category means depriving the real tribes from their very basic fundamental right as mentioned in the Constitution of India. These upper castes are highly educated while the literacy rate of Gujjar-Bakarwal is very low and hence it is nearly impossible for these real nomadic tribes to compete this Upper Section of the society to get jobs, education and any other benefits as provided by the constitution of India for their upliftment. --- Gujjar-Bakarwal, Gaddi, Sippi Tribes Reaction on Inclusion of Upper Class --- G.D Sharma Commission Roll Back: Government of Jammu and Kashmir set up a commission in March 2020 named as " Jammu and Kashmir Socially and Educationally Backward Classes Commission" headed by Justice G.D Sharma along with two members of the commission ,Lal Baharti, former Indian Foreign service officer and Manir Ahmed Khan, serving Indian Police service. To examine the social and educational backwardness of various backward classes people, in the union territory of the Jammu and Kashmir. G.D Sharma commission vide its supplementary report dated 23.05.2022 amongst various recommendations, has proposed the grant of Scheduled Tribe status to Various Communities namely Pahari, Paddari, Koli and Gadda Brahmin. Gujjar-Bakarwal nomadic community have already scheduled tribe reservation, after the recommendations of G.D Sharma commission Gujjar Bakarwal started revolting against identifying Paharis as an ethnic group as opposed to a linguistic group because Paharis are already enjoying four percent reservation on the basis of linguistic group . Gujjar-Bakarwal showed resentment against G.D Sharma commission which proposed ST status to Pahari, Koli, Gadda Brahmins and Paddari. Gujjar Bakarwal people say how Brahmins, Rajputs and Muslim Upper castes, who held the highest offices and controlled economic activities in Rajouri and Poonch, who till yesterday claimed to be landlords, tax collectors, and rulers and proudly identified themselves as Rajputs and used titles like sardars and raja are now claiming to be socially isolated and marginalized. Owing to this aspect of the Pahari Community, the Schedule tribes revolted against giving the special status to upper caste or elite class people. Even today, Gujjar is used as a slur by the upper castes. Gujjar-Bakarwal who reside mostly on mountainous terrain and jungles in far-flung rural areas where basic amenities like electricity, public health centre and schools are still a distant dream. Gujjar-Bakarwal Joint Action committee alleged that G.D Sharma commission is a biased and political motivated commission. Tribal people are now on roads to roll back G.D Sharma commission report and reconstitute "Jammu and Kashmir Socially and Educationally Backward Classes Commission". --- Gaddi-Sippi Tribe Reaction against the dilution of STs Status: Gaddi-Sippi Student Union also protested against dilution of STs Status in Jammu and Kashmir demanding no such attempt should be made at the highest Level to dilute the existing ST status/ reservation. --- Joint Action Committee's Dialogue with Central Government: The foot march was started by Joint Action Commission from district Kupwara on 04 November, on reaching in Rajouri at Kalakote after 18 days, was withheld by the government and was offered a table talk. The committee on accepting the dialogue offer met with the Home Minister, Shri Amit Shah at Delhi under the chairmanship of Haji Mohammed Yousuf. Gujjar-Bakarwal Jiont Action Committee appraised Home Minster about their demands that upper castes should not be given scheduled tribe status and biased report of G.D Sharma should be rolled back and a new commission be constituted to deliver justice to all the sections of the society. Home Minister Amit Shah assured the agitating Gujjar Bakarwal Joint Action Committee that the existing 10 percent reservation to the Gujjar, Bakarwal Gaddi, Sippi and other tribe will remain intact and the Paharis as scheduled tribe would be accommodated in a separate block after passing a bill from the parliament. He assured that all due rights and benefits of Gujjar-Bakarwal will be protected. Release of funds, admissions, jobs, and political reservation will remain safe and intact. Joint Action Committee called off the Tribal Bachao March after meeting with Home Minister. --- Conclusion: The social and political clash between Gujjar and Pahari started when Union Home Minister Amit Shah on his visit to Jammu and Kashmir on 03 October, promised reservation to Paharis and assured Gujjar-Bakarwal that not even 1% from their quota will be reduced. In response to the government plans to give ST status to Paharis , tribal communities in Jammu and Kashmir started "Tribal Bachao March" from Kupwara Srinagar to Kathua Lakhanpur under the banner of "Joint Action Committee" headed by Haji Mohammed Yousuf. Gujjar-Bakarwal people showed resentment against the decision of the government and stated, if Paharis are given ST status, it will come out as historical blunder and the basic idea of reservation will be totally abolished. Pahari Speaking people who till yesterday were landlords, tax collectors, and rulers and who proudly identified themselves as Rajputs and used titles like sardars and Raja are now claiming to be socially isolated and marginalized. Even today, Gujjar is used as a slur by the upper castes. Gujjar-Bakarwal who reside mostly on mountains terrains and jungles in far-flung rural areas where basic amenities like electricity, public health centre and schools are still a distant dream. Gujjar Bakarwal revolt against identifying Paharis as an ethnic group as opposed to a linguistic group because Paharis are already enjoying four percent reservation on the basis of linguistic group . Gujjar-Bakarwal showed resentment against G.D Sharma commission which proposes special status to Pahari, Koli, Gadda Brahmins and Paddari.

The Contemporary Political and Social clash between Gujjar-Bakarwal and Pahari Speaking peoples perhaps one of the most serious challenges to internal peace at micro level and is difficult to handle at large. The Focus of the study is on the recent G.D Sharma Commission report and observations that has proposed granting Scheduled Tribe status to Various Communities namely Pahari, Paddari, Koli and Gadda Brahmin. Gujjar-Bakarwal nomadic community have already scheduled tribe reservation, after the recommendations of G.D Sharma commission, Gujjar Bakarwal started revolting against identifying Paharis as an ethnic group as opposed to a linguistic group because Paharis are already enjoying four percent reservation on the basis of linguistic group (Pahari Speaking People). Gujjar-Bakarwal had already been the victim of social stigma for ages and in such circumstances with the inclusion of Pahari Upper clan in ST would clearly lead to a long term conflict and imbalance between both the communities. Furthermore, if non tribal communities of upper castes for example Jattas, Mirza, Rajputs, Sayeed and Brahmins from Rajouri and Poonch are given STs Status, how will the government prevent a similar demand from others sections like Khan, Gupta and Sharma, a business class living in the plains of Jammu and Chenab Valley? This research paper is to examine the ethnicity conflict based on inclusion of upper caste(Non Tribal) in STs and its impact on real tribal groups; Gujjar, Bakarwal, Gaddi, Sippi, Bot and Dard Sheen.

A. Introduction Efforts to expand oil palm plantations by burning forests show that economic development through oil palm plantations is still at the conventional development level because it tends to see development in the scope of only one generation and ignores the preservation of the natural environment. Whereas the economic development that should be developed is one that is environmentally sound or environmentally friendly. That is, economic development that is responsible for preserving nature so that future generations can meet their needs . The Indonesian government's ambition to become the largest palm oil producing country in the world and the success of oil palm plantations in boosting the level of community welfare in areas where oil palm plantations are developed and managed are not without problems . Land clearing and conversion of forest areas for the expansion of oil palm plantations occurred on a widespread basis without prior environmental studies. Even agricultural land and rubber plantations were converted to oil palm plantations. As a result, the phenomenon of environmental damage that causes floods and landslides is inevitable . After the trees are cut down, usually forests that are considered less economically valuable are then burned. This phenomenon is interesting to observe with the haze disaster that occurs every year in certain areas in Indonesia . If this phenomenon is not immediately addressed or stopped firmly by the government then Hood's prediction is really difficult to deny. Hood predicts the bad impact of oil palm plantations on the people who live around his area into three points, namely: 1) people are threatened with losing their ancestral land; the community is experiencing economic pressure so that their socioeconomic status or position has deteriorated; the occurrence of changes in the environment due to deforestation drastically by destroying and replacing them with new plant species . What Hood was worried about is now happening in residential areas and oil palm plantations. Since oil palm plantations were opened in 1982 in West Kalimantan, especially in the districts of Pontianak, Sanggau, Sintang, Sambas, and Ketapang. The forest area in West Kalimantan has shrunk drastically . Now oil palm plantations are spread throughout the districts/cities of West Kalimantan. In short, we find oil palm plantations, owned by companies or privately owned by residents, in almost all areas of West Kalimantan. The belief that oil palm is quite promising for the improvement and improvement of the community's economic life has led to a change in the mindset of the indigenous Dayak people towards forests . Residents who initially opposed the presence of oil palm plantations have now changed their minds and chose to plant oil palm in their area. In fact, not a few members of the community have converted their rubber plantations into oil palm plantations. As a result, the expansion of oil palm plantations and the conversion of forests to oil palm plantations is getting out of control. Companies are also increasingly aggressively expanding their plantation areas by buying land owned by residents or sharing the results with an agreed pattern. As a result, the area of timber forests is shrinking. Meanwhile, the area of oil palm plantations is getting wider and dominating the land. This condition clearly affects the lifestyle of shifting cultivation farmers or makes it difficult for residents of rural areas who have relied on forests for generations to fulfill their needs. Apart from having an impact on changing the mindset of the indigenous Dayak people towards forests, the expansion of oil palm plantations also has a significant effect on the order of the values of the life of the indigenous Dayak people. Changes in the order of life values by itself affect people's mindset towards forests and each other in everyday life. Further implications are changes in patterns of employment relations, patterns of employment and employment opportunities which in turn affect changes in social status, social relations, and patterns of people's daily life. Apart from that, the expansion of the area of oil palm plantations also has an effect on a significant increase in the number of migrants which in itself creates competition in several aspects of life with the local community which has the potential to cause horizontal conflict. Here it appears that the influence of oil palm plantations is socio-cultural and ecological . --- B. Methods In this study, research data collection used qualitative research methods. In general, in qualitative research there are the following: first, data is treated as verbal data or as something that can be transposed as verbal data. Second, it is oriented towards understanding meaning both in terms of characteristics, systematic relationships, conceptions, values, rules, and understanding formulation abstractions. Third, prioritizing the direct relationship between the researcher and the thing being studied. Fourth, prioritizing the role of the researcher as a key instrument . --- Data Collection Technique In this study, the sources and types of data that will be analyzed inductively are: First, the main data sources are the words and actions of people observed or interviewed related to oil palm plantations. In this context, we conducted a Focus Group Discussion with experienced traditional and community leaders in several West Kalimantan districts as sample research data related to the existence of oil palm plantations. Meanwhile, we conducted interviews to obtain information from several figures who have insight and experience about the local wisdom of the Dayak tribe and about oil palm plantations. Information obtained through FGDs and interviews was recorded through written notes or through video/audio tape recording. The information to be explored is regarding the shift in mindset of the Dayak tribe towards forests due to oil palm plantations. The recording of the main data sources through FGD, interviews or participant observation is the result of a combined effort of observing, listening and asking questions. All three can be carried out by a qualitative researcher consciously, purposefully, and always aims to obtain the necessary information. second additional data sources are written sources . While other descriptive data can be in the form of photos . Third, the role of researchers as research instruments. The position of the researcher in qualitative research is quite complicated. He is at the same time a planner, executor of data collection, analyst, data interpreter, and in the end, he becomes the pioneer of his research results. The definition of instrument or research tool here is appropriate because it becomes everything in the entire research process. Fourth, field notes. Notes taken in the field are very different from field notes. The notes are in the form of very shortened doodles, containing key words, phrases, main points of discussion or observations, maybe pictures, sketches, diagrams, and so on. The notes are useful as an intermediary tool between what is seen, heard, felt, smelled, and touched and recorded in the form of "field notes". The sources and types of data above are all useful for this study in order to help researchers to analyze and make a qualitative description of the reasons for changing the mindset of indigenous Dayak people towards forests, which then influence behavior in social interactions . --- C. Results And Discussion --- Description of Changes in the Mindset of the Dayak Indigenous People in West Kalimantan towards Indigenous Forests as a Result of Oil Palm Plantations: --- Definition of Tribe Dayaks The term Dayak or Daya' is used by residents of the coast of Borneo to refer to or refer to communities or groups of people who generally live in the interior of the island of Borneo: including Brunei, Malaysia , and Indonesia . Basically, the living culture of the Dayak tribe is classified as 'maritime' or maritime because on average the community is founded in the upper reaches of the river. Even the names of each Dayak sub-tribe are closely related to the name of the river which is an important factor for the daily life of the community. Before the era of infrastructure development in the form of roads, the Dayak tribe relied on large rivers as a means of transportation. It is not surprising that in ancient times various types of boats and water motors were the mainstay of the Dayak tribe and other tribes as a means of transportation between villages or between cities. Now only a small number of Dayak tribes still use these traditional means of transportation . --- Indigenous Peoples and Forests custom Indigenous Peoples and Indigenous Forests are talking about the existence of an anthropological community and their traditional rights that have been passed down from generation to generation. The existence of Indigenous Peoples is important to ensure the continued existence of Indigenous Forests. Without Indigenous Peoples it is certain that Indigenous Forests will be exploited indefinitely until they become extinct . --- Public custom In this study, the term Dayak Indigenous Peoples is understood as a society that still lives and applies certain Dayak customary traditions in the praxis of daily life throughout West Kalimantan. The customary traditions in question are the expectations of the community which are the compass or guidelines for social life, which have been passed down from generation to generation through a long process of internalization so that they become the identity of each member of the community and automatically guide them in behaving, speaking, and behave. The intended expectations can be in the form of orders, prohibitions, recommendations, which must be followed or avoided in the behavior of daily life, both towards the forest and towards each other. In relation to this research, the Dayak Indigenous Community is a community that from the beginning has owned and applied its customary traditions to maintain the sustainability of nature and harmony and its relations with fellow community members in their daily lives daily . --- Indigenous Forest To understand the term customary forest, we refer to the formulation of the Forestry Law Number 41 of 1999 concerning Forestry that customary forest is a state forest that is within the territory of a customary law community. Meanwhile, according to the Ruling of the Constitutional Court Number 35 of 2012 the term customary forest is understood as "forest that is within the territory of customary law communities". Regarding this understanding of customary forest, especially related to the communal rights of indigenous peoples over customary forests, the Minister of Agrarian Affairs/Head of the National Land Agency has issued a technical regulation concerning Guidelines for the Settlement of Indigenous Peoples' Ulayat Rights Issues, namely: Regulation of the Minister of Agrarian Affairs/ Head of the National Land Agency No. 5 of 1999. In this regulation the definition of Ulayat Rights and Ulayat Land is explained as following: Ulayat rights are the authority according to customary law which is owned by certain customary law communities over certain areas which are the living environment of these citizens, for their survival and livelihood, which arises from an outward and inward relationship that has been passed down from generation to generation between said customary law community and which region concerned and Ulayat Land is a plot of land on which there are customary rights from a customary law community certain. --- Description of Changes in the Mindset of the Dayak Indigenous People in West Kalimantan towards Indigenous Forests as a Result of Oil Palm Plantations: In connection with this research, we present the following research results in accordance with the data obtained in the field and supported by information from informants interviewed or through Focus Group Discussion , field notes, and direct researcher observations. --- Journal of Social Work and Science Education Volume 4 2023, 345-355 E-ISSN 2723-6919, P-ISSN 2746-0827 --- Forests are Viewed as Objects of Economic Activities Only Since the presence of the People's Nucleus Company in West Kalimantan in 1982, which was engaged in the development and management of oil palm plantations, social change in a broad sense, which then influenced the people's mindset towards forests, was unavoidable. The people's image or traditional beliefs about nature also changed drastically. The existence of forests in a broad sense is threatened. Communities who previously had an " animistic and pantheistic" traditional attitude towards forests so that they tried to protect and pass it on to their children and grandchildren, are now in a situation of change and at the same time uncertainty in attitudes towards forests, except for seeing forests as mere business objects or economic activities. Now traditional views, which are " animistic " or " pantheistic " in nature, are almost hard to find and are even threatened with extinction . --- Sociological Change and Threats Ecological The presence of oil palm plantations brings about social change in a broad sense and its consequences for culture . Workers in oil palm plantations who come or come from various regions, as described above, automatically bring about a change in perspective towards the forest, others, and the value system of life. Differences in habits, customs, habits, and levels of education characterize relations and interactions in the workplace. Such a situation has the potential to change the " mind-set" of the Dayak tribe in West Kalimantan in general towards forests, each other, and customs. Those who have long viewed forests as a source of life and livelihood in various aspects so that their sustainability must be maintained have now changed drastically. Now in general, the Dayak tribe views customary forests as limited to their economic value. This change in perspective is really troubling because forests are increasingly being stripped due to exploitation and illegal logging as a consequence of business activities. So, besides being successful in shifting traditional beliefs and views of the Dayak tribe towards forests, oil palm plantations are also one of the reasons behind all the phenomena of ecological and social change in the broadest sense which have implications for changing patterns of thought and action towards forests . --- Society Becomes Pragmatic, Consumeristic, and Mental hedonistic Most of the people give up their land or customary rights for the sake of temporary economic benefits, without thinking long about the fate of their children and grandchildren in the future. The profit-sharing system enticed by oil palm plantation companies has made them able to forget the advice and advice of their ancestors to maintain customary rights. There is an impression that they are struck by instant thoughts and want to get rich quickly themselves so that they hand over their agricultural land or rubber plantations or forests of their ancestral customary rights to Journal of Social Work and Science Education Volume 4 2023, 345-355 E-ISSN 2723-6919, P-ISSN 2746-0827 351 be managed by plantations with a profit-sharing system, the pattern or mechanism for sharing which is sometimes difficult to understand and accept common sense, but it is agreed nonetheless. inhabitant. In fact, the guarantee for the smooth running of the management system and profit sharing offered by the company is unclear and in fact many are not in accordance with the initial agreement. Communities who have already surrendered their land are becoming more easily controlled and their lives are increasingly squeezed by the interests of the plantation company. They are powerless to assert their rights again. Some rural people who are innocent and simple are deceived and deceived by the company by luring them that they will get a quite fantastic and large profit share without having to work hard. People who lack knowledge, have little ecological awareness, and have irregular incomes are so vulnerable to being deceived, are reluctant to work hard, and want to get rich quickly are one of the reasons why they without thinking critically hand over their land to oil palm plantation companies. Besides that, lifestyle is also an important factor behind the handover of land to the company. That is for example seen when they really crave or look forward to profit sharing every month from the company. They don't think long that maybe there is an ulterior motive behind the profit-sharing offer, which is neatly wrapped up in a rule that is consciously agreed upon by both parties, especially residents who are still innocent and straight-minded. it . sources are decreasing, forests are becoming homogeneous, thousands of plant and animal species are disappearing that we will never know again, and of course our next generations will never see them because they have become extinct forever. Not only species such as mammals or various types of birds are threatened with extinction forever, but also various types of fungi, lichens, worms, insects, reptiles, as well as various kinds of micro-organisms needed by our ecosystem are also becoming extinct. Rivers that are filled up cause the extinction of certain types of fish and aquatic biota. It is truly worthy of regret for all of humanity because the benefits derived from oil palm plantations are not comparable to the losses shared by all local, national and international communities. Recovery from forest damage takes a very long time and costs beyond the economic or financial benefits that only a few people enjoy. Humans bear higher and heavier costs and impacts due to damage to nature caused by the selfishness and greed of a few people than to enjoy the benefits and economic benefits in a sustainable manner. The victims who suffer the most are the poor who are trapped by the performance of a handful of rich people driven by their selfishness. We should regret together with economic activities that have resulted in drastic and merciless ecological damage this Purwadi, 2016). --- The Crisis "World View" of the Dayak Tribe In ancient times, the Dayak tribe thought of all spirits or spirits. They believe that rocks, trees, and all things in the forest have spirits or are the dwelling places of spirits and ancestral spirits. They believe that there are other rulers besides Ponompa Potara, Jubata . They believe that certain forests, rivers, rocks, trees have a sacred dimension because they have a spirit or become a place for them to live spirits. The sacred perspective of the Dayak tribe on the forest is usually associated with myths. For example, the myth about a water ruler called nabau or gana, who is believed to inhabit a certain depth in a river. Myths about forest rulers in the mountains and jungles in certain areas, called Domamang, Kamang, Bunyi' , etc. These rulers are of course spirits who are believed by the Dayak tribe to live in certain mountains, rocks in certain areas, certain trees, certain ponds in large and small rivers. At present, the existence of Dayak culture is threatened with extinction. Expansion of oil palm plantations, the existence of local mines owned by local people who are managed without being friendly and respectful of the environment are the beginning of the destruction of the local wisdom of the Dayak tribe. They are now having difficulty reviving the traditional beliefs of their ancestors because plantations, mining followed by changes in mindset and lifestyle have crushed the sacred objects of their ancestors. Trees, rocks and rivers in certain forests and mountains have been damaged and polluted by exploitative actions, their existence has even been terminated or evicted with the approval of community leaders and some customary administrators who basically do not understand the essence of adat. Apart from that, the irresponsible behavior of today's young generation has also exacerbated the situation. They don't just deliberately neglect auxiliary tools in ritual ceremonies such as pontok/pantak, but make it a business object and sell it to collectors of goods antique. Now the existence of ponto'/pantak has disappeared, I don't know where and were. With the disappearance of ritual tools, contamination of sacred objects, and the extinction of most of the " axis mundi" domains of the Dayak tribe, the essence of ethnic identity is threatened with being unclear, eroded and buried by irresponsible behavior. Thus, most of the Dayak tribes today are actually experiencing a worldview crisis. Their lives do not adhere to ancestral culture on the one hand, and are not serious about pursuing their religious beliefs on the other hand. Because if they are serious about religion, they will certainly know and understand that nature is part of the divine footprint, God's sublime creation, which is sufficient to meet basic human needs but certainly not fit to accommodate human greed and desires without boundaries , --- D. Conclusion The success of oil palm plantations in improving the community's economy deserves attention from economic considerations. But what we also need to realize, reflect on and consider carefully is the social and ecological impact that is difficult to prevent. Socially, there has been a change in people's mindset and actions towards the forest, which leads to ecological problems in the form of an ecological crisis. On the other hand, this change in thinking paradigm changes their daily life behavior and social interactions. Customary arrangements, customary law and local wisdom are increasingly loose in influence and are no longer seen and used as a compass of life or even if they are upheld and implemented then their essence and motivation have a pragmatic face and are loaded with economic considerations that only benefit interested parties. The phenomenon that is evident today is that what was previously seen and believed to be taboo and sacred is immediately seen as outdated by the younger generation and a handful of older groups whose thoughts are contaminated by the climate of capitalism manifested in oil palm plantations, which suppresses their lives. A further impact of this reality is disorientation of social attitudes and behavior, which can trigger horizontal conflicts on a small to large scale, both among residents themselves and between residents and the oil palm plantation companies. Now the expansion of oil palm plantations has penetrated into the interior areas of Kalimantan and especially in the interior of West Kalimantan, grinding away everything that was guarded and "taboo" by our ancestors in ancient times. Profits from oil palm plantations are clearly not enough to finance the losses incurred as a direct consequence of ecological, sociocultural, and changes in people's "world view". Changes in the community's "world view" can be seen in the phenomenon of forest desacralization. They no longer experience or perceive certain mountains, trees, rivers or forests as "trails of the Divine". The result of this desacralization of forests is exploitative actions against nature which then cause ecological damage. Now the phenomenon of desacralization of nature is increasingly uncontrollable, for example in the form of expansion of oil palm plantations. The presence of oil palm plantations in general and widely has caused more losses, especially regarding the drastic change in people's mindset towards forests in the form of waning of appreciation and commitment to local wisdom regarding the existence of forests. The government and the community must sit together to reconsider the possibility of limiting or even closing the opportunity to develop oil palm plantations whose permits have expired, and seriously design efforts to overcome the ecological and social losses that have been caused and are sure to threaten the continuation of the front of humans and the environment. If not, then the future together is at stake. In that context, decisions for every economic development plan and business activity that are based on natural resource management must have a vision of local wisdom, in favor of environmental friendliness for the future of the people in Indonesia and surrounding. --- Conflict Horizontal It is not uncommon for land disputes to occur between oil palm plantation companies and residents. There are customary lands that were taken by oil palm plantation companies simply because village officials misinterpreted or interpreted the term customary land. Traditionally, every Dayak community has customary land that must be protected. What is referred to as customary land is land that belongs to the community of indigenous peoples. Another term is customary rights. Included in customary lands are jungle forests which are joint rights of the Dayak customary community. In addition, customary land also includes " tembawang ", namely forests on which there are fruit trees that have been owned for generations because they are the inheritance of the ancestors. In relation to oil palm plantations, potential conflicts usually arise when customary forests are not respected or managed by companies without regard to local customary laws. It could be that this happened because the company already had a permit from the local government, but it could also be because the local government and village officials misinterpreted the meaning of customary forest or did not know the meaning of customary forest or did not want to know for economic reasons wrapped in a pragmatic and stylish attitude. hedonistic and consumerist life. That is for example what Mr. Amrazi Zakso revealed at a glance by referring to one of several other cases in the West Kalimantan area, namely the case in Jagoi Babang, Bengkayang Regency . --- Crisis Ecology The ecological crisis due to oil palm plantations raises complex issues. Clean water Journal of Social Work and Science Education Volume 4 2023, 345-355 E-ISSN 2723-6919, P-ISSN 2746-0827

Since the opening of oil palm plantations in 1980 in West Kalimantan, which since 1982 has been managed under the People's Nucleus Plantation (PIR-Bun) pattern, slowly but surely changing the mindset of indigenous Dayak people towards jungle forests or ulayat lands. Its influence is increasingly felt in the behavior and social life of the local community. The Dayak community has experienced a change in their "world view" of the forest. Those who have been friendly with nature since ancient times and uphold traditional principles, now in their lives, on the one hand, are leaving their ancestral traditions and following the trend of oil palm plantations, but on the other hand, their lives are increasingly cornered by the existence of oil palm plantations. The permissive attitude and pragmatism of those in power have resulted in a change in the "world view" of the local community. They no longer respect nature or forests as "sacred lands". Society is also increasingly pragmatic, hedonic and consumptive. Local customs and wisdom in the form of family values, mutual cooperation, loyalty to the community, humility, sincerity, respect for nature and each other are increasingly fading. The jungle forest and "tembawang forest" as customary rights have mostly been just stories.

To reach voters, political candidates and parties increasingly use social media. The internet, and especially social media, offers political candidates and parties the opportunity to communicate directly to citizens and it allows them to disclose personal information . For citizens, social media offers a platform to come in contact with candidates and discuss politics directly with others. Because of the possibility to communicate in a more personal and interactive manner , and because almost every politician and party uses social media, it is often believed that the internet is an important tool for democracy and a crucial element for a successful election campaign . Online political campaigning via social media has also received increased academic attention ). Although scholars differ in their beliefs about the impact of citizens' internet use, there seems to be a growing body of research that shows beneficial effects. Several researchers have demonstrated that internet use has positive effects on citizens' involvement in politics and this, consequently, contributes to the quality of democracy . Recently, studies demonstrated that the uses of social media for political purposes during election campaigns have an important political impact . Particularly the content characteristics have positive consequences for citizens' involvement into politics . Political involvement is important and desirable for a healthy democracy. When citizens are participating in political activities or feel connected to societal and political issues, they can create an association with their representatives. If citizens feel more connected to and interested in politics and their representatives, they are more likely to engage in political activities, such as becoming members of political organizations, voicing their opinions, and voting . Despite the increase in interest in the effects of social media use, only a few scholars examined why social media is an effective media tool for getting citizens involved. Although a few important attempts have been made to investigate which mechanisms explain the beneficial effect of social media use , there is still very little understanding of why social media affects citizen involvement . The current study tries to fill this void. Specifically, this study will focus on two key style characteristics of social media: interactive and three different dimensions of personal communication. An interactive communication style refers to direct reciprocal communication between politics and citizens. Personal campaigning refers to a communication style that is more focussed on the individual politician and his or her private life, than on the party the politician represents. It is examined how these key style characteristics of social media affect political involvement and which underlying processes mediate such a relationship. This study examines two underlying processes simultaneously; source expertise and social presence. The latter mechanism is especially important as recent studies call the attention to the importance of especially social presence theory as an important psychological process that explains positive effects of social media . This study contributes to the existing literature in various ways. Previous work on the consequences of social media focussed on the effect of social media use in general . Although it is important to study social media effects in general, to advance theoretical understanding of the effects of social media, it is also important to investigate how citizens use the internet . By focussing on specific style characteristics , or the attributes of social media , it is revealed which communication characteristics of social media actually affect political involvement. While some research has been done to examine the effects of different attributes of social media on citizen involvement , no study has yet combined these attributes and examined the effect of two different key attributes of social media simultaneously. This is especially important, as social media is multidimensional. Examining the mix of content characteristics of social media will give us more understanding into the question why social media affect involvement . In addition, this study investigates to what extent two different psychological processes explain the relationship between interactivity and personalization, and citizens' involvement. The central question is: does an interactive and personal communication style lead to a stronger sense of interpersonal contact and perceived source expertise, and does this, in turn, lead to increased levels of political involvement? Third, as the consequences of political personalization are rarely studied, OIR 40,5 this study examines which dimension of personalization is most effective in increasing political involvement. Thus, the current study also answers the question: how personal should personal communication be to be beneficial? As previous studies mainly focussed on the effects of one dimension of personal communication, this study includes three different dimensions of personal communication . --- 674 --- Interactivity and political personalization: two key style characteristics Interactivity is often considered the most important style characteristic of social media, and is generally believed to be a key variable when studying the effects of new media . Interactivity is a multidimensional construct; researchers use different conceptualizations to define interactivity . The rich body of interactivity research commonly defines interactivity as two-way communication : an interaction between two peopleone person can communicate directly to another person and vice versa. The reciprocal communication can be horizontal and also vertical . This notion is supported by Stromer-Galley , who argues that interactivity is both a product and a process. Interactivity-as-product is communication between people and computers, while interactivity-as-process entails communication between people themselves . This study focusses on interactivity as a process and operationalizes interactivity as two-way communication. Interactivity makes social media different from traditional media. Generally, traditional media send out information without receiving immediate response , while interactive communication is the main feature of social media. For instance, Twitter is designed to facilitate interactive communication, as it enables it users the opportunity to send, read, respond, forward and repeat messages directly to others, which offers possibilities for reciprocal dialogues between two or more users . In addition, such dialogues are also visual and observable for a broader audience, even if users do not participate. Reading these dialogues can influence them. Studies focussing on the effects of interactivity in a political context point toward positive outcomes: exposure to interactive websites increases recall , and positively affects political attitudes and evaluations of politicians . Likewise, exposure to interactivity on a political blog influenced positive attitudes toward the website, candidate evaluations and voting intention . Furthermore, exposure to an interactive Twitter page and exposure to a social networking site that was similar to Twitter leads to more positive candidate evaluations and a stronger voting intention, but only for people who usually avoid social interaction . Another key characteristic of social media is personal communication. Personal communication refers to the shift of focus from parties and institutions, to politicians and their private life . This shift seems to be present in social media , as politicians are using social media individually to communicate to voters. Social media is suited for personal campaigning and politicians frequently use social media for self-promotion , to talk about themselves and to refer to information about, for example, facts, 675 Online campaigning and political involvement opinions or links to articles . In general, personalization has two distinct dimensions; the first is individualization and the latter privatization . Individualization refers to a focus on individual parties and candidates. Individualization is especially present in social media, because politicians increasingly use personally kept online media platforms to communicate with their electorate, such as Twitter and Facebook. Privatization refers to a focus on private life and personal interest of politicians. Privatization is also present in social media; the communication by politicians is often characterized by sending messages about her or his emotions, thoughts and private issues . Despite the increasing interest in political personalization, little is known about the effects of personalization, especially in social media. There are scholars who argue that personalization has positive effects, because "politicians then lend their party's policies a face and a voice" . In other words, a focus on politicians makes politics easily approachable. Han demonstrates that disclosing personal information has positive effects on policy support. Others show that personalization on political websites positively influences psychological involvement . Other scholars are more skeptical and argue that a focus on politicians, and specifically their private life, distracts voters from important political processes and the bigger political picture . Others found that privatized information generates political cynicism , distrust in politicians , and campaign loss for political candidates . To sum up, despite the growing literature, there have been few empirical investigations that identify different psychological processes that explain the relationship between interactive and personalized communication, and political involvement. This study examines two underlying processes simultaneously that might explain the effects: social presence and interactivity . --- Social presence Social presence, a concept that is often used to explain interpersonal communication effects in computer-mediated communication , can be defined as "a sense of being with another" in a computer-mediated environment . The social presence theory explains that the effect a medium has depends on the "social presence" it conveys . Hence, social presence increases when "a person feels as if he/she were 'with' the communication partner, engaging in a direct, face-to-face conversation" . It is assumed that interactive communication positively influences social presence. This view is supported by Thorson and Rodgers , who argue that "providing an opportunity to interact with the candidate, encourage a sense of intimacy between the participants and candidate, creating a facsimile of an interpersonal relationship" . The sense of being in an interpersonal relationship could thus foster perceived closeness with the communicator , which, in turn, may have a positive influence on the evaluations and intention to vote for the communicator. It is, therefore, expected that people, who are exposed to a dialogue on Twitter, have a heightened sense of interpersonal contact, than people who read a Twitter page without such conversations . Consequently, it is assumed that this conveyed sense of interpersonal contact has a positive effect on political involvement, because heightened presence with the political communicator instigates political involvement: H1. Exposure to interactive online communication styles in social media positively affects political involvement via social presence. Likewise, personalized communication might also influence feelings of social presence, as personalized communication gives the impression of an informal personal conversation. Communicating with an individual politician makes identification with the communicator easier, than communicating with a political party. The perception of being in contact with a communicator mimics a real experience and helps readers to draw a "vivid picture" of the communicator. Readers can create imagined intimacy and emotional closeness with the communicator . There is, to our knowledge, only one scientific study that examines the consequences of personalized communication . This study shows that personalized tweets heighten perceived presence for people who are positive about social interactions. Although this study shows important findings, it does not take into account the differentiation between the different dimensions of personal communication. It can be expected that differential effects exist for these distinctive dimensions of personalization. First, based on the aforementioned studies, social presence is an important mediator that explains the effect of "individualized" communication on political involvement. Second, when the communication is also privatized , the perceived intimacy will be even higher. Readers identify even more with the communicator when more privatized information is enclosed; the communicator becomes "real" by creating an emotional bond . In other words, the perceived social presence will have a stronger explanatory role in the effect of personalization when the communication is privatized. It is again expected that this conveyed sense of interpersonal contact has a positive effect on political involvement: H2. Exposure to individualized communication styles in social media positively affects political involvement via social presence. H3. Exposure to privatized communication styles in social media positively affects political involvement via social presence, and the effect will be stronger than the effect of individualization . --- 677 --- Online campaigning and political involvement Source expertise Another explanation for the effects of interactivity and personalization on political involvement can be found in the source credibility literature. Source expertise has often been studied as a psychological mechanism of persuasion. It can be defined as the extent to which the communicator is regarded as professional and competent . Studies show that competence, integrity and reliability are important personality traits whereupon voters evaluate politicians . Following Sundar , interactive communication is likely to affect expertise, because interactive communication creates goodwill regarding the communicator on the side of the reader. More specifically, interactivity suggests that it is possible to engage in a conversation and this opportunity gives readers the idea that the communicator has an open mind . Moreover, making use of the communication tool in an advanced way makes the communicator appear experienced and skilled . Such professionalized skills are desirable qualities in a political communicator . Subsequently, the accumulation of these skills leads to positive evaluations of the political communicator . For example, expertise of the source is found to have a positive effect on opinion agreement . Likewise, a survey study found that favorable evaluation of politicians' Twitter use leads to positive perceived credibility, which had subsequently a positive effect on how people evaluate the politician . Hence, the following hypothesis is formulated: H4. Exposure to interactive online communication styles in social media positively affects political involvement via source expertise. Personalized communication on Twitter might also enhance perceived expertise. More specifically, Langer argues that humanizing a private persona gives someone experiential authentication. It is, therefore, expected that individualized communication fosters the feeling among readers that an actual person or human voice is behind the tweets . This will enhance the impression that the communication is more professional. The communicator does not hide behind the party, but tries to be open for communication and transparent. In other words, by communication as a person, instead of a party, the communicator shows that he is personally responsible for the communication that is sent out. In contrast, source expertise might also explain the negative effects of personalization. Following Jebril et al.'s line of arguing, it is expected that privatized news has a negative effect on source expertise, because a focus on private life distracts readers from political content. Greater focus on private life, means less focus on political issues. This may affect the communicators' expertise. Readers could then think that the communicator believes that private issues are more important than politician issues. Dispersing information about one's private life makes the communicator, therefore, less competent in the eyes of the reader. Taken together, it is hypothesized: H5. Exposure to individualized communication styles in social media positively affects political involvement via source expertise. H6. Exposure to privatized communication styles in social media negatively affects political involvement via source expertise. --- OIR 40,5 Interaction effect of interactivity and personalization As proposed by the mix of attributes approach , the key attributes of social media might also interact with one another. Both key characteristics might indeed be simultaneously present in online communication. Since it is hypothesized that both interactivity and personalization enhance presence with and expertise of the source, and consequently, political involvement, it is expected that when these characteristics are combined, this might have an even stronger effect on citizens' political involvement. There is one study that combines the two key characteristics. It demonstrates that the conjoined effects of interactive and personal communication on political involvement are indeed stronger . This is particularly true for the indirect effect via social presence. Interactive and personal communication styles provoke higher levels of perceived intimacy with the communicator. It is not surprising that combining these characteristics will probably activate even higher levels of perceived social presence. This will, in turn, affect political involvement. For the indirect effect via source expertise, similar results might be found. Interactive and individualized communication both enhance perceived source expertise, which, in turn, positively affects political involvement. Since privatized communication negatively influences source expertise, it is unlikely to find a conjoined effect of privatization and interactivity. Everything considered, the following hypotheses are proposed : H7. Personalized and interactive communication styles on social media strengthen each other in their effect on social presence, which in turn, positively affects political involvement. H8. Individualized and interactive communication styles on social media strengthen each other in their effect on source expertise, which in turn, positively affects political involvement. --- Method --- Procedure, participants and design To test our hypotheses, an online experiment was created. In the experiment, personalization as well as interactivity was abundantly manipulated in six political Twitter accounts. The accounts were based upon the actual Twitter accounts of a Dutch political party and politician . D66 is a liberal progressive party in the middle of the political spectrum. By using an existing Twitter account, stimulus materials will be more realistic. Additionally, two pre-tests were conducted, with the aim to establish the effectiveness of the manipulations of interactivity and personalization. The results of the pre-tests demonstrated that both personalization and interactivity were effectively manipulated . A 2 × 3 in between-subjects design was used. The experiment was conducted among younger citizens who often use social media, especially Twitter . Social media has thus an added value for youngsters: they are heavy users of social medium. Furthermore, because this study is primarily interested in underlying mechanisms, it is necessary to include participants 679 Online campaigning and political involvement who recognize interactivity and are used to or can interpreted interactive features on social media. Only including Twitter users might lead to familiarity bias, because Twitter users might be positive toward communication on Twitter in any case. Therefore, this study targeted younger citizens. Hence, participants who completed the questionnaire were 243 college students with a mean age of 20.72 . Participants were recruited via an online message board of the university. The participants were randomly assigned to one of the six conditions and received five euro or course credits for their participation. Data were collected in the winter of 2013. --- Stimulus material The actual content of the Twitter account of Alexander Pechtold and D66 was downloaded and then modified for the experiment . The stimulus materials consist of the first page of a Twitter account . The three personalization dimensions were manipulated in line with studies conducted by Van Santen and Van Zoonen and Van Aelst et al. . The personalization dimensions were manipulated in two different ways: the source of the tweets was manipulated and the reference to private life. In the "depersonalization" condition, the political party was the source of the tweets and tweets covered no information about the private life of the politician used in the experiment. In the "individualization" condition, the politician was the source of the tweets and tweets contained no reference to private life. In the "privatization" condition, the politician was the source of the tweets and the politicians' private life was mentioned in a few tweets. The level of interactivity was manipulated in line with previous studies that studied the effects of interactive political communication on websites and on Twitter more specifically . In the "high interactive" condition, the Twitter account had tweets that contained mentions. Twitter users use the @ characteristic to communicate to other Twitter users. Posting a tweet that includes a mention, the @ characteristic, which is followed by a name, indicates that a one Twitter user directly sends a text message to another Twitter user . In other words, @mention calls for the other Twitter user's attention, and this is an important prerequisite for a conversation to emerge . In the "low interactivity" condition, such mentions were not present. The tweets used in the low interactive condition were presented as one-way communication. The tweets were only used to send information . Between conditions, the amount of information was kept equal. --- Measures As "[p]olitical participation arises from the interaction of citizens and political mobilizers" , social media can particularly influence offline as well as online political behavioral intentions. The latter form of participation is also important, as online participation could translate into offline political activities, which has been shown in previous research . Therefore, in this study, political involvement was operationalized as online and offline political behavioral intentions. Political talk refers to citizens' intended behavior to discuss politics or Twitter use with friends, family and colleagues. The variable was measured on a seven-point scale using two items . Twitter behavior intention refers to the intention to revisit the Twitter account in the experiment, other politicians' or parties' accounts and intention to follow politicians or parties on Twitter. The variable was once again measured on a seven-point scale using four items . The factor analysis revealed that the items load on one dimension . Intention to vote for the party was measured with one item . Intention to vote for the politician was also measured with one item . Social presence items were derived from previous research and adapted to this study. social presence was tapped by using three items . Answers were coded on a seven-point scale where one equals "totally disagree" and seven "totally agree" . A factor analysis revealed that the three items load on one dimension . Source expertise was derived from the dimensions of source credibility . One dimension of source credibility is used; expertise. Expertise was measured using four items on a seven-point semantic difference scale . A factor analysis revealed that the four items load on one dimension . Two control variables were included in all our analyses to test familiarity bias: "likelihood of voting for D66" and "likelihood of voting Alexander Pechtold." Answers were recorded on an 11-point scale where one equals "I would never vote for this party/politician" and eleven equals "I would certainly vote for this party/politician" . --- Results To test our hypotheses, multiple mediation bootstrapping analyses with 1,000 resamples were used, using Hayes' PROCESS macro for SPSS . Analyses revealed significant positive indirect effects of high interactivity on different aspects of political involvement through social presence . For example, the use of interactive tweets had a positive effect on the perceived social presence of the source , and social presence had, in turn, a positive effect on the intention to vote for the politician . In other words, the effect of interactivity on the intention to support the politician after reading the tweets is mediated by social presence, even when controlled for initial intention to support the politician or party, b ¼ 0.18, 95 percent bias corrected confidence interval [0.08; 0.34]. Thus, H1 was supported. Table II. 683 Online campaigning and political involvement Subsequently, the analyses showed no significant indirect effects of individualization on political involvement via social presence. Exposure to a politicians' Twitter account does not induce feelings of presence. Thus, H2 is not supported. Next, bootstrapping analyses demonstrate that a privatized communication style compared to individualization, had a positive effect on political involvement, because personal information induces feelings of social presence. Next, the analyses revealed a significant positive indirect effect of privatization on political involvement via social presence compared to depersonalization. Exposure to a Twitter account containing information about a politicians' private life, compared to a Twitter account from a political party, had a positive effect on the feeling that the source was "there." This induced social presence had, in turn, a positive effect on different aspects of political involvement. For example, exposure to privatized communication had a positive effect on the perceived social presence of the source , and social presence had, in turn, a positive effect on the intention to talk about politics , b ¼ 0.14, 95 percent BC CI [0.03; 0.30] and b ¼ 0.21, 95 percent BC CI [0.08; 0.40], respectively. Personalization has only an effect on social presence when information about the politicians' private life is enclosed. Thus H3a supported, but not H3b. H4 predicted that interactive communication positively affects political involvement via source expertise. The bootstrapping analyses found significant positive indirect effects of high interactivity on different aspects of political involvement through source expertise. For instance, the use of interactive tweets had a positive effect on the perceived expertise of the source , and source expertise had, in turn, a positive effect on the intention to talk about politics . So again, it seems that when an interactive way of communicating on Twitter is used, readers feel that the source of that communication is an expert, which in turn makes readers, for instance, more likely to talk about politics, b ¼ 0.12, 95 percent BC CI [0.05;0.24]. Conform H5, it appears that exposure to a Twitter account from a politician, heightens perceived expertise of the source , and this leads to, for example, increased levels of political behavior on Twitter , b ¼ 0.20, 95 percent BC CI [0.08; 0.39]. Thus, the results suggest that exposure to a politician's Twitter account results in higher political involvement via source expertise instead of exposure to a parties' Twitter account. Conform H6, disclosing personal information versus not disclosing personal information had a negative effect on political involvement, as the disclosure of such information reduces levels of source expertise. For instance, exposure to privatized communication gives the readers the idea that the source of the communication is not an expert , which will lead to less intention to vote for the party leader , b ¼ -0.09, 95 percent BC CI [-0.22; -0.01]. This is an interesting result, as the same indirect effect was positive for social presence. It seems that both significant mediators cancel each other out. Analysis confirmed this; the total indirect effect was not significant , indicating that a strong beneficial effect of communicating about one's private life was not found. Lastly, this study tested the effects of privatization and depersonalization. As one might expect, no significant effect of privatization on political involvement via source expertise was found. Apparently, the positive effect of communicating as a politician diminishes when a politician discloses information about his private life. Thus, the findings regarding personalization indicate support for H4-H6. --- OIR 40,5 To test the interaction effect, a moderated mediation analysis was conducted. Specifically, it is examined whether the different dimensions of personalization moderated the indirect effect of interactivity on political involvement via social presence and source expertise. In short, using Model 9 in PROCESS, the study examined whether there exists an indirect effect of interactivity on involvement via social presence and source expertise. In these analyses the relationship between interactivity and social presence/source expertise was interacting with personalization . The conditional indirect effect of interactivity on political involvement through social presence was only significant when the communication was depersonalized , or the communication was privatized , but not when the communication was individualized. This indicates that exposure to an interactive Twitter page that was depersonalized or privatized, resulted in higher levels of political involvement through social presence compared to individualized communication. In other words, combining an interactive and privatized style of communicating on social media leads to even higher levels of social presence, and in turn, into heightened political involvement. The conditional indirect effect of interactivity on political involvement through source expertise showed similar results. The tested indirect conditional effects were significant when the communication was depersonalized , or the communication was privatized . This indicates that exposure to an interactive Twitter page which was depersonalized or privatized, resulted in higher levels of all variables of political involvement through source expertise compared to individualized communication. So, in general, combining both characteristics does not simply result in higher levels of source expertise and consequently political involvement. For a visualization of the indirect conditional effects, see Figures 2 and3. --- Conclusion and discussion Taken together, the results of this study revealed that exposure to an interactive communication style on Twitter, will lead to a stronger sense of interpersonal contact with, and perceived expertise of the communicator, which, in turn, positively affects Online campaigning and political involvement political involvement. The current findings add to a growing body of literature that shows positive effects of interactive online communication . Another important finding is that personalized communication styles on Twitter have an effect on citizens' political involvement. Generally, exposure to a Twitter page from an individual candidate positively affects political involvement through source expertise. Exposure to a Twitter page from an individual candidate containing private information can also positively affect political involvement through social presence. Privatization on the contrary can weaken involvement, as privatization negatively affects political involvement via source expertise. Nonetheless, because both processes cancel each other out, our findings provide tentative support for optimistic scholars, who argue that personalized online communication brings citizens closer to politics . Despite the exploratory nature of the investigation of the conjoined effect, our results show that personalized and interactive communication, to some extent, interact. However, since this study only explored the relationship and in-depth theoretical foundations are not present, it is recommended that further research investigates how different styles interact and how this interaction affects political involvement. A remark should be addressed before we address the implications of this study. The use of Twitter by Dutch politicians is similar to politicians who use Twitter in other western democracies. However, some scholars pointed out that Dutch politicians embraced the interactive use of Twitter and also tweet much more often than, for instance, UK politicians . Thus, Dutch politicians are more likely to engage in online conversation. This partly assists the external validity of our study, because it is not perceived as "odd" if politicians in the Netherlands use interactive communication online. However, future comparative research should examine whether the effects of interactivity and personalization are more generalizable. Furthermore, this study focussed on younger citizens. Yet, we cannot assess how this age group can be compared to older citizens in terms of underlying mechanisms. We believe that further research needs to be conducted to establish whether the mechanisms also apply to other groups. The findings from this research have several theoretical implications. First, the findings show that the characteristics of social media are important when explaining the effects of social media on political involvement. It is not general social media use that positively affects involvement, but the specific content characteristics within such media. In many instances, previous studies examined how many and how often citizens use the internet . Although it is important to consider the effects of the use of internet in general and social media more specific, to advance our theoretical understanding about the specific consequences of social media, one should study what it is about social media that causes effects . This is especially important as social media are rapidly developing and changing and new social media platforms arise and disappear. Our study demonstrates that by focussing on the specific content characteristics or attributes of social media , this contributes to our theoretical understanding about why and under which circumstances social media affects citizens. Second, although there are studies that examine the relationship between social media and political involvement , this study is the first to investigate two different psychological processes simultaneously that explain the relationship. Uncovering such processes gives us important and intriguing theoretical insight into the question how social media contributes to citizens' political involvement. Overall, it seems that when an interactive and/or a personalized style is used on social media, citizens feel connected with politics and they have the feeling that the communication is more professionalized . These findings show that two important marketing and computer-mediated communication theories are highly applicable within the political communication context as well, and they should be considered in future investigations regarding the effects of online political communication. Lastly, as this is one of the first studies that examines the effects of different dimensions of personalization, this study demonstrates that different dimensions of personalization can have negative and positive consequences. Apparently, when studying the consequences of a focus on politicians rather than parties, one should take these different dimensions into consideration. Individualization is entirely different compared to privatization, and both have different consequences depending on different processes. In addition, previous studies examined the consequences of interactivity and personalization by focussing on psychological feelings of involvement . This study found that interactive and personalized communication could actually affect indented political behavior. This striking finding notes that social media cannot only affect feelings of involvement, but social media can actually change ones voting behavior. An important practical implication is that an interactive and personalized communication style should play an important role in political marketing strategies. If political organizations and candidates want to persuade voters, they should first and foremost use a more interactive communication strategy on social media. Furthermore, the use of individualized communication can also be valuable. Finally, using privatized information is only beneficial in specific cases, indicating that communication strategies that include private information are recommended if a politician also portrays him or herself in a professional way. Privatized communication might be less beneficial, because a focus on politicians' private life distracts citizens from the political content . Thus, when citizens are exposed to privatized communication, a lack of political substance might then be noticed. Citizens could then be swayed that the politician cares more about private than political issues, which makes him or her less competent. --- Online campaigning and political involvement Downloaded by Universiteit van Amsterdam At 07:23 18 September 2017 --- Appendix

Purpose -The purpose of this paper is to examine the causal relationship between interactive and personal campaigning on social media and political involvement, and the mechanisms that explain the effects. Specifically, this study examines whether personal and interactive communication on Twitter increases political involvement among citizens through social presence and perceived expertise. Design/methodology/approach -An experimental designa 2 (low vs high interactivity) × 3 (depersonalized vs individualized vs privatized communication) between-subjects designis used. Findings -The findings show that interactive communication leads to a stronger sense of social presence and source expertise, which positively affect involvement. The effects of personal campaigning differ. Individualized communication positively affects involvement via source expertise. Interestingly, privatized communication positively affects involvement via social presence, but negatively via source expertise. Originality/value -Although a growing body of work examines the political consequences of social media, there is still very little understanding why social media affect citizens. The current study fills this void by investigating how the use of social media affects political involvement among citizens.

Introduction Improvement in workplace safety is often dependent upon the engagement and involvement of workforce leaders and workplace designers. Their long-term approach and attitude to this issue may be influenced by early exposure to the identification, assessment, and control of workplace hazards as young workers themselves or as part of formal education [1]. In particular, the university sector plays a key role in the preparation of workplace leaders, operating as an agent of social and community reform [2,3]. Among its students will be the future designers, end-users, or influencers of the work environment. Thus, safety content in curricula can be seen to facilitate safe learning in undergraduate programs, but also to influence attitudes of future professionals. Safety-related learning is often delivered and assessed on an ad-hoc basis within university coursework programs. Rather than be a formally planned aspect of curricula, safety education is a "hidden curriculum" [4]. This is, at least partly, due to the lack of any such requirement by course accreditation bodies [5][6][7]. Furthermore, there appears to be no curriculum appraisal tool with which courses and programs of study can be evaluated from a safety content perspective. Attempts to gather information on the inclusion of safety content in university curricula have met with limited success. According to a recent EU Report [5], The greatest challenge is to mainstream occupational safety and health into university education in order to reach future engineers, architects, medical professionals, business professionals, managers, etc. The mainstreaming of OSH into university-level courses is the least well-developed area for various reasons, which include the lack of direct national government control over university teaching. Actions to include OSH in relevant courses such as engineering or business studies are therefore ad hoc, and often dependent on the interest of individual professors or particular advocates within professional bodies. Professors need convincing of the need to include OSH in courses. They also need relevant materials. There is a paucity of published research on the integration of safety-related content into the university curriculum, other than for specialist, safety-focused undergraduate and postgraduate courses. Jackson, de Munk, and Elms [3] considered the extent to which health, safety and environment education was implemented into Master of Business Administration courses throughout Australian universities. In response to the finding of a significant absence of safety-related components within MBA courses throughout Australia, and given that the inclusion of safety information is professionally relevant, the authors recommended that relevant content be integrated into existing tertiary coursework. In 2009, Stacey et al. [7], in conjunction with the UK Health and Safety Executive, the Health and Safety Laboratory, and the University of Liverpool, conducted a project to assess whether engineering students, as future workplace designers, had a basic understanding of safety and health risk issues relevant to their specific course of professional study [7]. Students' understanding of safety risk issues and key concepts were assessed and as a result of limited understanding new safety-related teaching materials were integrated into existing course content, in liaison with key stakeholders. An e-learning package was subsequently developed. More than half of the students exposed to new teaching materials improved their knowledge of safety-related concepts by more than 10% [7]. The work of Wachtler and Troein [8] potentially provides a framework for the evaluation of safety-related content in curricula. In their research on cultural competency, they reviewed learning outcomes, interviewed program directors and lecturers about course content, and conducted focus groups with students. The study reviewed curricula through three different perspectives: the intended curriculum as designed by the academic; the taught curriculum as delivered to students, and the received curriculum as reflected in student experience. With reference to the methodology of Wachtler and Troein [8], we undertook a previous pilot project, at the University of Adelaide, to assess safety content in undergraduate programs in the Schools of Nursing and Mechanical Engineering. These Schools were selected as dichotomous examples of the future workforce-designers and "end-users"-both of whom are critical to improvements in the field of workplace safety and injury prevention. Course handbooks were scanned for the inclusion of safety-related content and semi-structured interviews were conducted with academic program coordinators. While there was an absence of safety-related content in course handbooks, program coordinators considered that safety-related content was integrated in their programs. A Safety Content Assessment Tool -as a means to identify this "hidden" content-was then developed and subsequently piloted. Subsequent semi-structured interviews with the program coordinators indicated that the SCAT provided them with a means to identify existing safety content-or its absence-in their curriculum. Due to the absence of published literature on the inclusion of safety content in university curricula, or a tool with which program or course content can be evaluated from a safety content perspective, the aims of the study reported here were twofold. First, to expand on our pilot project by exploring the views of final year undergraduate students, academic program coordinators and representatives of professional organizations and accrediting bodies with regard to existing safety-related curriculum content and the subsequent level of preparedness of graduates entering the workforce. Second, to validate the SCAT-which was developed during the previous pilot project-by conducting usability testing in three areas, namely undergraduate nursing, mechanical engineering, and education programs at all three major universities in South Australia . The future "teachers" workforce was included for its facilitation of safety-related skills, knowledge, and attitudes in young and/or new workers, who are over-represented in injury statistics [9]. --- Methods A qualitative research approach was used to address the aims of the study within the framework described by Wachtler and Troein [8]. --- Interviews and Focus Groups Semi-structured interviews were conducted with representatives of engineering, nursing, and education professional organizations, university academic coordinators , and final year undergraduate students. One focus group, comprising academic nursing staff at Flinders University, was conducted in order to accommodate their preference not to attend individual interviews. All interviews were audio recorded and transcribed. The interviews were based on semi-structured questions and were conducted by the same researcher . The interviews were between 20-45 min in length. The focus group was of one-hour duration. The questions for representatives of professional organizations included their understanding of safety, how they felt it was related to their profession, how prepared university graduates were for entry to the workforce, and opportunities and barriers for improvement in safety-related content in university curricula. Questions for academic coordinators included their own understanding and qualifications in workplace safety, its inclusion in the existing curriculum, the sources of information they were aware of, and regularly accessed, and student preparation for work placements. Questions for students included their understanding and the relevance of workplace safety, their experience of safety issues as an undergraduate how they were assessed, and how well prepared they felt, from a safety perspective, for their first student placement and after graduation. --- Recruitment and Ethics The process of accessing participants for interviews was initially via email/telephone contact. University academic coordinators also assisted in approaching specific students for interviews. Students were also accessed by asking participants if they knew of any interested peers who would be interested in being interviewed. Ethics approval was granted by the University of Adelaide . --- Data Analysis Audio recordings were transcribed prior to thematic analysis as described by Braun and Clarke [10]. This consisted of analyzing the transcripts and creating a list of themes based on participant statements. The transcriptions of each interview were read, re-read, and coded independently by two members of the research team. Open coding was conducted by identifying and labelling each idea or concept. Where labels coincided, this allowed for comparison between transcripts. These were then grouped together to develop themes that were compared between researchers. In order to maintain analytical coherency, the three groups were analyzed separately. --- Safety Content Assessment Tool Program coordinators from nursing, mechanical engineering and education from each of the universities were asked to apply the SCAT to their existing curriculum. The responses were analyzed for each of the tool's components. For the purposes of this study, a term more familiar to program coordinators-Occupational Health and Safety-was used in the tool instead of the more generic term of "safety". --- Results Altogether, 20 interviews and one focus group were conducted. Three interviews were conducted with representatives of professional organizations-the Australian Nursing and Midwifery Federation, Australian Education Union, and Engineers Australia. Academic coordinators from the three participating universities-the University of Adelaide, Flinders University, and University of South Australia-were involved in five interviews and one focus group . The responses from the four-academic staff from the University of Adelaide who participated in the pilot study were also included. In relation to students, five interviews were conducted with nursing students from the University of Adelaide and Flinders University . Four interviews were conducted with education students from the University of Adelaide, Flinders University, and University of South Australia . Three interviews were conducted with mechanical engineering students from the University of Adelaide, Flinders University, and University of South Australia . --- Perspectives of Professional Organizations --- Safety in the Curriculum All representatives agreed that workplace safety was an essential component in university curricula-one referred to it as a "necessary evil". There was a consistent view that there was a general lack of awareness by students as to how safety is managed in the workplace. This was associated with discrepancies between the safety content in university curricula and actual workplace practices. They considered that universities are not adequately equipped to provide the optimal level of safety training which is needed in the workplace. However, they did acknowledge the limitations associated with university curricula. --- Safety Preparation for the Workforce There were contrasting views on how well prepared new graduates were. While the engineering representative considered that engineering students were sufficiently prepared the education representative considered that new teachers were not. The nursing representative referred to the preparation of new graduates as a "mixed bag" due to differences in the curricula between universities. All three bodies agreed on the need for improving the relevance of workplace safety content in existing curricula however they acknowledged the limitations to doing so: I am not certain that it's ever possible for any academic program to prepare a graduate that is absolutely work-ready without the employer ultimately having some responsibility at that point of the person commencing in work to do some top-up education and training in relation to workplace safety. . --- Perspectives of Academic Coordinators --- Safety in the Curriculum Academic coordinators considered that workplace safety content was embedded in all three curricula, both formally and informally. However, there were differences in opinion on the degree to which this knowledge was formally assessed. Nursing coordinators agreed that workplace safety was embedded in the curriculum with various forms of assessment. For example, students were assessed at the completion of nursing intensive work placements on their clinical practice and their interaction with patients. Moreover, students were assessed on workplace safety aspects when working in simulated environments prior to work placements: It's embedded in the assessments that they do, particularly the practical assessments. . Engineering coordinators were divided whether safety knowledge was formally assessed, with one view that it should be considered as common-sense: . . . for me, occupational health and safety is common sense . . . if it's really basic you assume other people around you have already considered it. . Education coordinators agreed that workplace safety was not formally assessed in the curriculum. One commented that they were unsure as to why this was the case: It's not, and I guess I'll be going away and having a little look at that to see why not. . Engineering coordinators were more inclined to focus on the legalistic nature of workplace safety in the University curricula: Engineering is certainly a discipline where there are potentially workplace hazards in the university environment, of course, and in industry as well, and that requires that we make sure that our students have an appreciation of health and safety issues when they leave the courses here. . Nursing and education coordinators placed greater emphasis on the more social aspects of workplace safety and the need to monitor individual stress levels and mental health: So when we're talking about health, safety and wellbeing we're talking about the wellbeing of the whole person not just the bits of the person-the physical self but also the mental self, the emotional self. . --- Safety Preparation for the Workforce Coordinators from all disciplines considered that graduates were prepared, from a safety perspective, upon entering the workforce. However, their perceptions on the degree of preparedness differed: No matter how much you educate them or try to get them to be alert to things... there are still people who will not recognize a situation as being something reportable . . . . . I don't think we could ever teach a student so they can hit the ground running and be fully competent. . --- Perspectives of Final Year Students --- Safety in the Curriculum There was general agreement among students that workplace safety is embedded within the curriculum rather than being presented as a separate or "stand-alone" topic. I think it was mingled throughout the entire three-year course. . Students across all three disciplines reported different levels of safety-related assessment in their curriculum. Engineering and education students reported less-formal assessment procedures, with some not recalling any form of assessment. There was a pass and fail if you didn't get a good enough score, whatever it was, and then you wouldn't proceed. But it was pretty relaxed, I would say. It wasn't so formal. . In contrast, nursing students reported both formative and summative competency assessment within their curricula: We also have a lab where we re-enacted the clinical situation and we were watched and observed and did formal assessments which did contribute to our summative assessment. . --- Safety Preparation for the Workforce Education students raised concerns as to whether they had sufficient workplace safety knowledge when entering the teaching profession: I guess a lot of that sort of preparation does get handled by the schools more so than the university because it is that natural workplace setting so I'm not sure how well this school will be preparing me for occupational health and safety issues. . Engineering and nursing students agreed that the university curriculum had prepared them for the workforce from a safety perspective: I think we'll be well prepared when we get into the workplace and then they'll sort of run us through all that sort of stuff anyway so I think-well, I'm pretty confident that it should be okay. . --- Suggestions for Change Students consistently commented on the need for improved specificity and relevance in workplace safety content. Nursing students considered they did not learn enough in relation to patient handling during their undergraduate training and felt that the content should be more holistic in nature. For example, one participant suggested that more attention needed to be paid to the physiological and lifestyle effects associated with shift work such as poor eating habits, lack of exercise, and disrupted sleep patterns that may impact on musculoskeletal injury and the increased prevalence of obesity. Similarly, education students suggested the presentation of more real case scenarios instead of the more general safety content which is currently delivered. They were particularly concerned about the management of workplace bullying and stress which had affected them during their undergraduate placements: I ended up being so stressed at the placement that I ended up not eating until I went home-like hadn't eaten breakfast, hadn't eaten anything at the school all day and just was so stressed. . Engineering students also emphasized the importance of making safety-related curriculum content more relevant to their profession. They suggested the development of a safety-related learning laboratory on campus as a potentially significant educational experience. --- Perspectives of Accrediting Bodies Accrediting bodies may be seen as drivers of curriculum change in universities. Despite numerous attempts by telephone and written communication, access to representatives was not gained. Thus the views of accreditation bodies on workplace safety integration remain unexplored. --- The Safety Content Assessment Tool Academic coordinators from nursing, mechanical engineering and education from each of the universities were asked to apply the tool to their existing curriculum. It was completed by nine academic coordinators. These were collated and the responses analyzed for the each of the tool's components. --- The Intended Curriculum While all course coordinators were affirmative in including workplace safety activities in their intended curriculum, the required actions for improvement differed between the professions. Nursing coordinators identified relatively few needed actions, while engineering coordinators focused on the need to ensure that technical aspects of safety were adequately covered. Education coordinators identified areas for potential improvement primarily related to preparedness for student placements.in the intended curriculum. --- The Taught Curriculum Course coordinators generally considered that their taught curriculum was appropriate-mirroring the intended curriculum-while acknowledging that this would change with any alterations to the intended curriculum. --- The Received Curriculum There were mixed responses by the course coordinators in relation to assessments which specifically include safety elements that were either intended or taught. Nursing and engineering curricula including more formal assessment components than education. --- Discussion Using a multi-stakeholder perspective, we sought to examine the integration of safety content in three key disciplines , perceived effect on work-readiness and to validate a tool which could be used by academic coordinators to identify present, absent, and hidden aspects of the coursework curricula pertaining to workplace safety principles. --- Safety in the Curriculum While the importance of embedding workplace safety content in university curricula was emphasized by professional organizations, and academic coordinators considered this to be the case, both of these stakeholder groups acknowledged that its effectiveness was limited. Professional organizations attributed this to the inherent limitations associated with university curricula and the inability of universities to fully prepare their graduates for work entry. These limitations were largely mirrored in the responses of academic coordinators. While academic coordinators did consider workplace safety content to be embedded in the curricula, there were notable differences in the degree to which this was formally assessed. Nursing coordinators reported a variety of formal assessment procedures in simulated laboratories and real work environments while engineering coordinators reported a focus primarily on the legal aspects. However, workplace safety was described by one engineering coordinator as "common sense." In contrast to these varying levels of assessment, teaching coordinators reported a general absence of formal assessment in their curriculum. The varying approaches to assessment were reflected in the views of final year students. While students from each of the groups considered that workplace safety content was embedded in their programs the depth of this differed markedly. Nursing students commented on the variety of assessments, both formative and summative, they were required to complete. Engineering and education students, however, were generally unable to recall any formal assessment and where they did, described it as "relaxed." The determination of competence in the absence of any formal assessment is anathema in university pedagogy, particularly where practical skills need to be demonstrated. While experiential learning is important, students traditionally pay particular attention to aspects of the curriculum on which they expect to be formally assessed. This can be likened to the inevitable student question, "Will this be in the exam"? --- Safety Preparation for the Workforce The varying levels of workplace safety content were reflected in the varying levels of students' preparation for the workforce. Perhaps unsurprisingly, academic coordinators from all three disciplines considered that their graduates were sufficiently prepared, from a safety perspective, to enter the workforce. However, they did acknowledge that the level of preparation may not be optimal. The views of the professional organizations were more variable. The engineering discipline considered that new graduates were sufficiently prepared for the workforce. This was not the view of the education discipline. Although nursing graduates had the most nuanced and comprehensive workplace safety content, the professional body viewed new graduates as a "mixed bag" from a safety perspective. This pattern of responses was similar among final year students. Education students raised concerns regarding their level of workplace safety preparedness, while engineering and nursing students considered themselves to be adequately prepared. --- The Safety Content Assessment Tool In contrast to previous studies [3,7], our intent was to validate a generic tool which could be used by academic program coordinators to identify present, absent, and hidden aspects of coursework curricula pertaining to safety principles. By piecing together this safety "jigsaw puzzle" they could identify existing content and current gaps. This served to provide positive feedback where existing content was identified but also to highlight future areas for improvement and integration. Because hazards and risks are occupation-specific, we did not seek to develop a package of information to be integrated into course content but rather to assist program coordinators in the identification and modification of material. We considered that this approach would provide maximum flexibility for the integration of workplace safety concepts. The tool was designed as a quick "self-check" tool rather than an external audit or research tool. Consequently, it was purposely designed with "closed" questions to reduce any perceived administrative burden and to facilitate its ease of completion. However, even with a group of highly motivated coordinators, compliance with its completion was difficult with several of the coordinators requiring multiple reminders. In common with Stacey et al. [7], we found that a project of this nature can only be undertaken with the close cooperation of academic staff with specific responsibilities, and the ability, to influence or modify existing course content. By default those who participated in our study had a pre-existing awareness of the importance of workplace safety information. Reviewing course content across an entire multi-year program allowed them to consider how workplace safety information could be integrated and expanded over time as the students' knowledge and profession-specific expertise increased. This progressive integration is supported by sound pedagogical principles, in accordance with Bloom's taxonomy [11] by reinforcing important safety principles as student knowledge and experience increases. In the absence of these close cooperative links, it is likely that academics would be reluctant to review and modify course content without an external driver such as a formal course review or accreditation process. The strength of this study is the multi-discipline and multi-stakeholder perspective. However, one must consider the implications of personal bias among academic coordinators undertaking a self-assessment of their own program. Studies of self enhancement bias, the tendency to describe oneself more positively than a normative criterion would predict, indicate that most individuals self-enhance and expect others to do so as well [12]. Self-assessment in education may differ from that of peers or mentors, or otherwise be less accurate when compared with actual performance [13]. Individuals may also be overconfident with newly learned skills [13]. A further limitation of the toolkit is that it does not specifically address the needs of students from non-English speaking backgrounds in the university setting. The impact of the formulation of the toolkit on students from NESB backgrounds in relation to comprehension and cultural differences must not be ignored. Future revisions of the toolkit could provide a focus on learning outcomes for these students. The comments on the assessment tool made by the course coordinators were positive and the self-assessment was seen as a useful framework. However, it is, in its current form, a relatively static document. The tool would benefit from further input from end-users both in its format usability and presentation. The development of an engaging multi-media interface may increase its usability, compliance and adoption by academics, as noted by Stacey et al. [7] in their development of an e-learning package for undergraduate engineering students. Multimedia elements could include a brief video introduction about the importance and relevance of the review, along with testimonials and completed case studies. In order to be widely applicable the tool was developed as a generic resource. For internal use the tool would benefit by the inclusion of more specific information by university faculties/schools following adoption to tailor the tool to their educational goals. In relation to location and monitoring, accessibility of the document by program coordinators is critical. The optimal location is likely to vary between universities. However, University Quality & Review Units and Health, Safety & Wellbeing Units may have the necessary infrastructure to host the document internally at a central location. Given their specialized focus on hazard management and risk assessment, it is imperative that the toolkit not be used as an audit tool by such central units, although be accessible and readily utilized by coordinators of varied disciplines within the university to evaluate their programs. It is vital that despite the currently heavily regulated and crowded curricula, that academics integrate workplace safety education into their coursework rather than adopt safety education as an 'add-on' to the curriculum. Finally, accrediting bodies have the potential for significant influence in university curricula and their views are important. Further work should explore their perspectives. --- Conclusions In our study, there was no lack of desire for the integration of workplace safety content but rather a lack of awareness of where such content was already incorporated in multi-year programs, how it could be included if absent, and how it is perceived by graduating students. A Safety Content Assessment Tool can support the recognition of the integration of workplace safety content in curricula and enhance its inclusion so as to help achieve its social purpose. --- Author Contributions: P.R. conceptualized the study, developed the survey tool, undertook data collection, and was the lead writer. A.S. undertook data collection and performed qualitative analysis. D.P. conceptualized the study and assisted with writing the manuscript and revising the manuscript critically. --- --- Appendix A -Safety Content Assessment Tool Element 1-The Intended Curriculum Scope: The relevance of occupational health and safety for a particular course or program can be determined through examination of the program and course information, particularly learning outcomes. Element 2-The Taught Curriculum Scope: Examination of the taught curriculum offers insight into more practical aspects of OHS. The taught curriculum includes online content, lectures and lecture notes, practical work and field placements. Each of these elements should be examined to identify existing OHS content and opportunities to integrate additional information.

Improvement in workplace safety is dependent upon the active engagement of workforce leaders and designers. The university sector plays a key role in the education of these future leaders, and there is an expectation that safety education in universities will encompass more than just a safe learning environment-that is the nurturing of broader safety attitudes and awareness. However, with the exception of dedicated safety training programs, safety education is often delivered and assessed on an ad-hoc basis and at academic discretion. This is partly due to the absence of a simple tool with which curricula can be evaluated from a safety perspective. In a qualitative approach, semi-structured interviews were undertaken with multiple stakeholders (academics, professional organizations, and students) to determine their views on existing safety content in university curricula and on the level of preparedness, from a safety perspective, for workforce entry. University participants came from nursing, mechanical engineering, and education schools at three universities. A simple curriculum evaluative tool was also validated. Results indicated there were divergent views on the level of preparedness for workforce entry both between schools and stakeholder groups. However, the limitations of university curricula were acknowledged. The evaluation tool was shown to provide positive feedback on existing, but previously unacknowledged, safety content and also highlighted areas for future improvement and integration. However, voluntary utilization of the tool was a challenge for busy academics.

Data sources Data were from 3 representative survey platforms. We combined 17 cross-sectional surveys from the Canadian Tobacco Use Monitoring Survey , the Canadian Alcohol and Drug Use Monitoring Survey , and the Canadian Tobacco, Alcohol and Drugs Survey . [12][13][14][15][16][17][18][19][20][21][22][23] A complete description of the surveys is available in Appendix 1 . --- Outcome The primary outcome was past-year cannabis use . The specific survey questions are available in Appendix 2 . --- Covariates The independent variables included were age, sex, education, tobacco smoking, province of residence and recent pregnancy. --- Statistical analysis Individual survey files were concatenated to form a combined data set. Sampling weights were used in all analyses. These weights adjust estimates for excluded populations and survey nonresponse. The weighted prevalence of past-year cannabis consumption was calculated for each survey year across sociodemographic characteristics and province. All analyses were stratified by sex. Logistic regression was used to adjust estimates for age, education and tobacco smoking. Analyses of the crude trends in cannabis consumption indicated an inflection point in 2011. Therefore, we fit segmented logistic regression models with separate intercepts and linear trends before and after 2011. Interaction models and Wald tests were used to determine heterogeneity in sex-specific trends by age group, level of education and province. Analyses by level of education and province were age-adjusted. All prevalence estimates are presented with 95% confidence intervals . The criterion for statistical significance was α = 0.05. Analyses were performed using Stata statistical software . --- Ethics approval --- Results Across the study period, men of all ages were more likely than women to use cannabis , equivalent to an odds ratio of 1.96 . Cannabis use was strongly associated with age. The prevalence increased with age to a maximum of 33.0% among men and 22.1% among women aged 20-24 years, before declining . Among those aged 65 years and older, the prevalence was 1.3% in men and 0.5% in women. Tobacco smoking was strongly associated with cannabis use. Current smokers were nearly 5 times as likely to report cannabis use in both men and women . There was wide variation in the prevalence of cannabis use across provinces, with the highest prevalence in British Columbia and Nova Scotia, and lower prevalence in the remaining Atlantic provinces and Saskatchewan. Data on past 3-month cannabis use were similar . --- Trends in the prevalence of cannabis consumption Between 2004 and 2017, the age-adjusted prevalence of pastyear cannabis use increased in men and women, with greater increases after 2011 . In 2004, the prevalence of cannabis use was 12.2% in men and 6.6% in women. By 2017, the prevalence of cannabis use increased to 18.7% in men and 11.1% in women. The OR for the trend associated with the annual change in cannabis consumption was 1.02 in men and 1.00 in women between 2004 and 2011, and 1.08 in men and 1.11 in women between 2011 and 2017. After adjustment for age, education, tobacco smoking and province, the OR for the 2011-2017 trend was 1.24 in men and 1.13 in women . --- Trends by age The prevalence of past-year cannabis consumption increased between 2004 and 2017 in men and women aged 20-24 years, 25-34 years, 35-44 years, 45-64 years and 65 years and older . The trends in prevalence in these age groups remained relatively stable between 2004 and 2011 and generally increased from 2011 to 2017. Cannabis consumption among those 65 years and older increased from 0.1% in 2004 to 2.2% in women in 2017, and from 0.4% to 4.0% in men, relative increases of 24 and 10 times, respectively. However, among the youngest age group of 15-19 years, there was a significant declining trend in cannabis use, from 25.2% to 21.1% among adolescent men and from 22.9% to 17.6% among adolescent women. An interaction test indicated significant heterogeneity in the annual trend across all age groups by sex . --- Trends by province The age-adjusted prevalence of past-year cannabis use was relatively stable across provinces until 2011, after which consumption started to increase, with 2 exceptions . In Quebec, trends in usage were variable between 2004 and 2017; men had only moderate increases , whereas women showed an overall decline . In New Brunswick, between 2015 and 2017, the prevalence slowly increased in men from 12.5% to 14.7% over the study period, whereas in women the trend was variable with a substantial increase from 6.2% to 15.0% . There was heterogeneity in cannabis use trends across provinces . --- Cannabis use during pregnancy and among women of reproductive age We used data from CADUMS between 2008 and 2012 to investigate the prevalence of past-year cannabis use among women who were recently pregnant, compared with all women of reproductive age . In both groups, cannabis use was more common among younger women aged 15-24 years. These women were 4.5 times more likely to use cannabis than women aged 35-44 years in the overall sample and 6 times more likely than women who reported a recent pregnancy in unadjusted analyses. In unadjusted analyses, there was a graded and inverse association between educational attainment and cannabis use in both groups, and this was stronger among women with a recent pregnancy. Adjustment for age and tobacco smoking attenuated this association, with tobacco smoking remaining strongly predictive of cannabis use. Current smokers had an adjusted OR of 6.90 for cannabis use among all reproductiveaged women 15-44 years and an OR of 7.90 for cannabis use among women with a pregnancy within the past 5 years. Between 2008 and 2012, the age-adjusted prevalence of cannabis use decreased from 15.3% to 10.2% among women aged 15-44. However, the trend was not statistically significant . Among women reporting pregnancies, the prevalence of past-year cannabis use decreased between 2008 and 2009, before increasing from 8.2% in 2010 to 13.0% in 2012. The adjusted OR for the trend associated with cannabis use in pregnancy between 2008 and 2012 was 1.18 . --- Trends by educational attainment Data on respondents' educational attainment were available between 2004 and 2012 and in 2017 . Analyses of age-adjusted trends in the prevalence of cannabis use by education level indicated heterogeneity in the rate of change by level of education between 2004 and 2017 in men and women . Cannabis use appeared to be increasing faster among the higher educated groups, although data were not available in 2013 and 2015. Cannabis use was inversely associated with the level of education in ageadjusted analyses, and this was more apparent in men . Associations were attenuated after adjustment for tobacco smoking. The adjusted association between education and cannabis showed a positive gradient in men in 2017 after adjustment for tobacco but was not statistically significant . --- Research --- Interpretation We found that the prevalence of past-year cannabis use varied by age, sex, education and geography in Canada and was strongly associated with tobacco smoking. More than 15% of Canadians reported using cannabis in 2017. Between 2004 and 2017, the age-adjusted prevalence of cannabis use increased by 5% in women and 8% in men, relative increases of 81% and 64%. Despite the overall rise in cannabis consumption, we found decreasing trends among adolescent men and women aged 15-19 years. Compared with women, men consistently reported higher cannabis use over time, age and province, although relative increases were larger in women. The prevalence of cannabis use increased considerably among the most educated groups between 2012 and 2017. Contemporary national data on cannabis consumption in Canada since legalization is limited and mainly presented in aggregate form. 3,[24][25][26][27][28] Data are available on youth prevalence, [26][27][28] and 2 studies have included provincial differences in cannabis use across Canada. 3,24 Data from the COMPASS study among grade 9-12 students in Ontario and Alberta suggest a slight increase in cannabis use between 2016 and 2018. 26 However, the use of a convenience sample makes these results less generalizable. Previous reports of national trends in cannabis use are consistent with our results by age, sex and province, 24,25 although data on educa-tion and pregnancy have not been previously reported. Between 2018 and 2019, the National Cannabis Survey found an increase in prevalence among people aged 15-24 years. 3,24 Consistent with our findings, this increase is driven by the 18-24 year age group and declines were found among those aged 15-17 years. 24 Research in the US shows similar trends and age distribution, and higher consumption in males than in females. 3,4,8,9,29 Recent national surveys indicate an increase in adult cannabis consumption, and a decrease among adolescents . Recent work by Anderson and colleagues suggests that there is a decline in the prevalence of youth cannabis consumption post-legalization of recreational cannabis. 30 These data suggest that recreational cannabis laws in the US may be associated with an 8% decrease in past-month use and a 9% decrease in heavy use among US students in grades 9-12. 31 These findings have some methodological limitations, particularly that few states have sufficient post-legalization data, limiting the ability to draw conclusive inferences. [32][33][34][35][36] In our study, we found a similar trend in this age group before legalization, which suggests an age or cohort effect in addition to any potential effect from the policy change. There are concerns that the legalization of recreational cannabis may increase social acceptability and accessibility of the drug. In the US, recreational cannabis legalization may have contributed to increases in positive attitudes toward cannabis use and, ultimately, increases in consumption. 4,6,37,38 Several policy mechanisms exist, including restricting the maximum potency, increasing tax and the introduction of minimum unit pricing, which may increase the safety of cannabis use while minimizing increases in prevalence. 39 These policies should be combined with adequate public communication regarding the risks of cannabis use. We found a substantial increase in cannabis use in older people over time, with about half of them using cannabis for therapeutic reasons. Little is known about the potential harmful effects of cannabis in senior populations. As such, cannabis legislation should prioritize public health messages for vulnerable populations, including adolescents, pregnant women and older people. In our study, we identified a substantial number of reproductive-age women who consume cannabis. Recently, up to 2% of pregnant women in an Ontario birth cohort reported using cannabis while pregnant, and this increased to more than 6% among women aged 15-24 years. 40 Substantial evidence exists to discourage the use of cannabis during pregnancy; 41 however, the trends we observed in our study suggest that women may be unaware of the potential risks of cannabis in pregnancy. --- Limitations There are limitations to the present study. Cannabis consumption was self-reported. Self-reporting may under-or overestimate the true prevalence because of social desirability biases. We assessed past 12-month cannabis consumption because this question was consistently found on surveys starting in 2004, but this may only reflect occasional use. Analyses of past 3-month use, available in more recent surveys, show similar results and better capture current use. The surveys used exclude the Canadian territories along with other segments of the population, including individuals without residential landline telephones, limiting generalizability. Residential telephone surveys indicate that 14%-16% of the Canadian population do not have landlines, 43 and characteristics of ownership may correlate with cannabis use. The survey sampling weights are designed to account for these exclusions, and our estimates align with other sources with different sampling methodologies. Although cannabis for therapeutic reasons has been available in Canada with a physician prescription since 2001, survey questions on medical use were only available in 2015 and 2017. Although most of cannabis use is recreational, about one-third of past-year use is related to therapeutic purposes, and we were not able to differentiate therapeutic use in earlier surveys. Surveys were conducted before the legalization of recreational cannabis, and thus, we were unable to assess the influence of the new legislation. In addition, data on pregnancy have not been available nationally since 2012. Finally, educational data was missing from 2 recent surveys in 2013 and 2015, limiting trends analysis by level of education. --- Conclusion The prevalence of cannabis consumption in Canadians varies by sex, age, level of education and geography. Cannabis consumption increased between 2004 and 2017 across most ages, remained more prevalent in men than in women and has been increasing among those with higher levels of education. Despite overall increases, there have been steady decreases in cannabis use among adolescents. Further analyses of Canadian cannabis consumption post-legalization and across socioeconomic status will be essential to advise public health policies and inform future policies. --- Affiliations: Ottawa Hospital Research Institute , University of Ottawa; Faculty of Medicine , School of Epidemiology and Public Health, University of Ottawa, Ottawa, Ont. Contributors: Dana Lowry contributed to the data curation and formal analysis. Daniel Corsi contributed to the conceptualization, data curation, formal analysis, investigation, methodology, software and supervision. Both authors drafted and revised the article, gave final approval of the version to be published and agreed to be accountable for all aspects of the work. --- Funding: The Canadian Institutes of Health Research provided funding for this study and was not involved in study design, analysis or interpretation of data. The funder was not involved in the writing of the manuscript or in the decision to publish. Data sharing: All data and program files are available from the corresponding author, Daniel Corsi. --- Supplemental information: For reviewer comments and the original submission of this manuscript, please see www.cmajopen.ca/content/8/3/ E487/suppl/DC1.

n October 2018, the Government of Canada legalized the recreational use of cannabis nationally. 1 Before legalization, cannabis was the most widely used illicit drug in Canada, 2 and consumption increased from 14% to 17% between 2018 and 2019. 3 This rise may be a result of the increases in cannabis availability and circulation, and decreases in the perception of harm. 4,5 Surveys suggest that in the United States, the social acceptability of cannabis has increased 6 and perceptions of harm have decreased after recreational use was legalized in several states. 7 Trends in cannabis consumption may vary by sex, age, ethnicity and socioeconomic status and may be influenced by changing social norms and new legislation. 8,9 Studies that assess the differences in cannabis use trends across relevant socioeconomic and demographic groups are essential to inform legislation and public health priorities around cannabis consumption in Canada and among vulnerable populations. In this study, we combined 3 national survey platforms in Canada to evaluate and compare trends in cannabis use between 2004 and 2017 by age, sex, level of education, recent pregnancy and province.We followed the recommendations of Burns and Kho 10 and the STROBE statement 11 to report the methodology used in this repeated cross-sectional study.The surveys included in this study were conducted in Canadian provinces between 2004 and 2017. The surveys covered all people in Canada aged 15 years or older residing in the 10 provinces. Exclusions included residents of the Yukon, Northwest Territories and Nunavut and populations living in

Introduction ACCORDING to the report entitled "Suicide Worldwide in 2019: Global Health Estimates," published by the World Health Organization, the global suicide rate is alarmingly high, with over 700,000 individuals succumbing to it annually. The report further highlights that suicide accounted for more than 1.3% of all deaths in 2019, indicating a significant public health concern. The phenomenon of suicide has emerged as a significant public health concern on a global scale. Of greater concern is the escalating annual incidence of suicide fatalities among adolescents, coupled with a decreasing average age of suicide victims . Empirical evidence indicates that in approximately 33% of nations globally, adolescents are at the greatest peril for suicide when compared to other age cohorts . According to recent statistics, suicide ranks as the third most common cause of death among adolescent females aged 15-19 and the fourth most common cause of death among males in the same age group, with tuberculosis being the only cause of death ranking higher. According to the recently released report, "The State of the World's Children 2021," published by UNICEF, suicide ranks as the fifth most common cause of mortality among adolescents aged 10-19, regardless of gender. Annually, approximately 46,000 youths aged 10 to 19 terminate their lives . The suicide rate in China has been trending downward over the past ten years or more, but it has been increasing among primary and secondary school students . A total of 73 occurrences of youth suicide were documented in 2010 , and that number rose to 79 in 2013 . For the time frame of October 2016 to September 2017, it was formally published as 267 . The suicide incidence among teens has been steadily increasing in recent years, according to China's Health Statistical Yearbook , and it is now the second biggest cause of death in this age group . As per research, the incidence of suicide behavior among Chinese youth increases steadily from the age of 10 to the age of 20 to 24, at which point it peaks . Additionally, suicides among younger people are increasing . The results of the study by Zhang et al. show that junior secondary students exhibit more suicidal-related behaviors than do their senior secondary counterparts. Moreover, there has been an increase in the occurrence of suicidal ideation, planning, and attempts among students in primary and secondary schools. According to a meta-analysis conducted on a sample of 288,354 secondary school students, a significant proportion of them, up to 16.3%, reported having seriously contemplated suicide during a particular year between 2010 and 2020 . The Blue Book of Education: Annual Report on China's Education , a collaborative effort between the 21st Century Education Research Institute and Social Sciences Academic Press , presents compelling data indicating that a significant proportion of secondary school students in China have contemplated suicide. Specifically, 17.7% of students have reported having suicidal thoughts, 7.3% have made plans to commit suicide, and 2.7% have attempted suicide unsuccessfully. To clarify, the statistics indicate that approximately 16.7% of secondary students have experienced ideation of suicide, while approximately 7.1% have formulated a plan for suicide, and approximately 2.7% have made an attempt on their own lives . --- Reasons for Adolescent Suicide in China Suicide is typically not a senseless, spontaneous behavior from a psychiatric perspective; instead, it involves a number of processes, such as suicidal propensity, ideation, planning, and attempt . This extreme human conduct is frequently motivated by complex social, cultural, and individualized factors. According to French sociologist Durkheim, external variables, including the living environment, ingrained social ideals, and moral standards, play a major role in determining attitudes toward suicide . He also stated that the reasons for suicide are more social than personal. Based on his findings, more in-depth research has been done to examine the biological, cognitive, psychological, and social risk factors for suicide . --- Social Conflicts Durkheim posited that the incidence of suicide was heavily influenced by social structures. In times of significant social upheaval, where the established order is disrupted and new norms and regulations have not yet been solidified, individuals within society may experience a loss of direction and sense of purpose. According to research, individuals face the dilemma of either repressing or succumbing to their nascent desires, which can heighten the likelihood of suicidal ideation . The phenomenon of suicidal tendencies is influenced by both social integration and regulation. Achieving a suitable balance between these factors can effectively mitigate suicidal behaviors, as evidenced by prior research . The growth and development of an individual are signifi-cantly influenced by external environments. The escalation in the prevalence of suicide among adolescents in China is indicative of the inadequacies inherent in the present social frameworks . China, being a nation in the process of development, is currently undergoing significant changes in its political, cultural, and economic spheres. The mental well-being of Chinese primary and secondary school students is inevitably influenced by social integration and regulation within their relatively unstable milieu. --- Influences of Social Transformation China is currently undergoing a reduction in its ability to achieve social integration as a result of its ongoing social transformation. The proliferation of diverse values and the shift from traditional moral norms to contemporary ones intensify social anomie, leading to heightened confusion among students. Simultaneously, the escalation of individualism and decentralization also serves as a contributing factor to the augmented susceptibility to suicide among adolescents . Furthermore, as industrialization advances, a considerable number of young rural inhabitants migrate to urban areas in search of job opportunities, leaving their underage offspring behind in rural regions, resulting in a significant population of "left-behind children." The survey data from the Ministry of Civil Affairs of China indicates that in 2018, there were an estimated 6.97 million children who were left behind in China. Notably, the provinces of Sichuan, Anhui, and Hunan each had over 700,000 left-behind children, as reported in reference 23. Left-behind children exhibit elevated levels of loneliness, anxiety, and depression when contrasted with their non-left-behind counterparts. These conditions subsequently lead to diminished levels of self-esteem, life satisfaction, and social adaptation. According to research findings, a significant contributing factor to the rise in suicidal tendencies among individuals is attributed to their compromised mental health status . --- Influences of Popular Cultures The ongoing societal transformation in China has resulted in corresponding changes in social cultures. The intersection of traditional and modern values, as well as the convergence of Eastern and Western cultures, exert significant influence on the psychological outlook of elementary and high school students . The internet has become an integral aspect of adolescents' lives due to the swift advancement of information technology. According to the Annual Report on the Internet Use of Chinese Minors released by the China Internet Network Information Center , the number of minor internet users in China in 2021 was 191 million. The report indicates that primary school students had a 95.0% penetration rate, while middle school students had a 99.4% penetration rate . The internet offers several advantages to young individuals; however, it is also inundated with superfluous, adverse, and inaccurate data that has the potential to significantly compromise the psychological well-being of elementary and high school students . A significant proportion of adolescents have exhibited a reliance on the internet, commonly referred to as cyber dependence. Empirical evidence suggests that individuals who suffer from internet addiction exhibit markedly elevated levels of suicidal ideation, planning, and attempts relative to non-addicted individuals. Furthermore, research indicates that child addicts are more likely to experience suicidal ideation than their adult counterparts, with a prevalence rate of 28%. The availability of "suicide websites" can potentially exacerbate the issue of suicidal ideation among adolescent populations by providing explicit guidance on various methods of suicide, thereby potentially facilitating the transition from ideation to action. The growing popularity of mobile technologies has considerably broadened the scope of bullying victimization, extending it from physical locations to the virtual realm. According to the Annual Report on the Internet Use of Chinese Minors , a significant proportion of minor internet users in China, specifically 16.6%, reported instances of online ridicule or abuse . According to research, minors who experience cyberbullying are more susceptible to experiencing heightened levels of fear and depression compared to adults. This can result in negative academic outcomes such as decreased performance, absenteeism, dropping out, and, in severe cases, suicide . --- Misleading Information from Mass Media One of the most important contributors to the rise in suicidal events is media coverage of suicides. Excessive exposure to suicide stories, especially inaccurate ones, may change audiences' perceptions of suicidal behavior or inspire them to imitate the reported suicides. David Phillips developed the "suicide imitation" or "Werther effect" concept, a theory about the suicide imitation behavior caused by the indirect exposure to or influence of suicidal incidents, as a result of his research into the correlation between publicity and suicide incidence in 1974, which revealed that the suicide rate would significantly increase in the areas reached by the reports of sensational suicide events. His research revealed that the subsequent spike in suicides was positivelycorrelated with the amount of media attention given to a suicide story . The link between the volume of media attention and the rise in suicides was validated in later research by other scientists. In order to provide a reference manual for ethical reporting on suicide, WHO produced Preventing Suicide: A Resource for Media Professionals Update 2017 in 2017 . Both contemporary and conventional media provide diverse perspectives on societal occurrences to the general populace. The perception of society held by the general public is heavily influenced by the artificial social setting constructed by the news media . Adolescents occupy a distinctive developmental phase characterized by a tendency to challenge conventional societal conventions while concurrently exhibiting a proclivity to emulate the conduct of their peers. The proliferation of media coverage on adolescent suicide has the potential to heighten the susceptibility of teenagers to suicidal tendencies and elevate their risk of engaging in suicidal behaviors. Certain media outlets prioritized the sensational and often comical aspects of suicidal incidents while neglecting logical and empathetic assessments of their devastating outcomes. This approach had a significantly detrimental impact on the promotion of mental health and suicide prevention across society . Some reports have been deemed irresponsible due to their pathological elaborations of suicide methods and glorification of suicide behavior. This has resulted in children developing the misconception that suicide is a viable solution to problems and a reasonable means of avoiding life's challenges . --- Skewed School Education Evaluation The school is frequently the first social setting a youngster experiences during the socialization process. K-12 education is the most extensive and significant education students will ever receive in their entire lives. The rising suicide rate among students in basic education is a clear sign that there are significant issues with China's current educational system. --- Overemphasis on Progression Rates At the moment, the only measure of a school's teaching effectiveness is its progression rates, which demonstrate how many students in the school, are able to advance to and enroll in the next level of education . The school usually crams students' schedules with countless tests and examinations, such as monthly exams, midterm exams, and terminal exams, to boost students' test aptitude and their development rate. Students are under a great deal of mental pressure because of their class and grade rankings. Exams with high stakes, like admission exams for higher education and senior secondary, are seen as crucial turning points in their lives. The causes of the 267 reported cases of student suicides in primary and secondary schools in 2017 that the 21st Century Education Research Institute thoroughly examined can be divided into six categories: bullying on campus, parent-child conflicts, excessive workloads, teacher-student conflicts, psychological issues, and affective problems. Three of them had a direct connection to academic strain, showing that learning stress is the main cause of suicide among students in basic education . --- Discrimination in Student Evaluation Recognition and encouragement from teachers are crucial to the character development of students who spend the majority of their daytime at school. The prevailing practice among Chinese educators is to employ a unidimensional criterion for assessing their students, whereby individuals who demonstrate exceptional academic performance and compliance are deemed exemplary, whereas those who exhibit subpar grades and nonconformist conduct are regarded as underachievers. According to the evaluation criteria, educators tend to exhibit a bias towards students who possess exemplary academic records while disregarding or subjecting those who are perceived as "underperforming students" to verbal mistreatment . Prolonged neglect and insufficient acknowledgement from educators may potentially trigger emotional turmoil and heightened exasperation among economically disadvantaged students. These students are more susceptible to the development of psychological disorders. Simultaneously, students who are deemed "good" and receive preferential treatment may develop a reliance on biased treatment as a result of their consistent receipt of praise, accolades, and distinctions, ultimately cultivating a perception of superiority. Individuals may readily acquire unfavorable personality traits, including but not limited to obstinacy, self-centeredness, reliance, apathy, and conceit. Individuals with low psychological resilience are vul-nerable to experiencing breakdowns when faced with academic or other forms of frustration, regardless of their severity . --- Inappropriate Home Education The psychological development and character formation of a child are significantly impacted by the home environment and parental behaviors. The phenomenon of suicidal ideation among adolescents is significantly associated with family structures, parenting practices, and educational ideologies. The likelihood of suicide among adolescents is considerably heightened by negative parenting styles . --- Authoritarian Parenting China has a long history of emphasizing rigid hierarchies. This legacy in family contexts is reflected in the unequal interaction between parents and children . A sizable percentage of parents view their kids as both their own property and the means by which they will accomplish their own objectives. The parent-child connection in these households is one of the controllers and the controlled, which raises the youngster to be obedient. The youngster may mature into an emotionally unstable, nihilistic, and low-self-esteem adult over time . --- Indulgent Parenting The implementation of China's family planning policy in the late 1970s has brought about a significant transformation in the conventional structure of Chinese households. The nuclear family, consisting of three members, has emerged as the predominant familial arrangement, particularly in urban locales. It is noteworthy that the allowance of three children per family was not implemented until the year 2021. In the absence of siblings, the only child is often regarded as the favored offspring by their parents and grandparents. The sole focus of the family's attention is directed towards a single child. Excessive parental indulgence poses a hindrance to the social, emotional, and behavioral growth of children. If a child is raised in a home environment that is excessively permissive, they may exhibit tendencies towards self-centeredness, emotional impulsivity, and dependence on others. Consequently, individuals with these characteristics exhibit reduced resilience, restricted social adaptability, and a diminished sense of accountability, rendering them more susceptible to adversities and setbacks . --- Academic Success-focused Parental Expectation Chinese parents frequently stress to their children the importance of achieving victory in the initial stages of the competitive "rat race" of life, specifically in the pursuit of academic excellence. It is common for children to adhere to predetermined trajectories towards achievement that are influenced by parental aspirations and societal norms rather than their own individual desires and inclinations. To enhance their academic performance and increase their likelihood of gaining acceptance into prestigious educational institutions, it is necessary for individuals to engage in supplementary education from an early stage in their development . These parents, who are described as "ambitious," tend to prioritize their children's academic achievements over their overall development and welfare. When children at-tain academic success, they are often bestowed with numerous compliments and rewards as a form of generosity. If not met, individuals may experience significant disappointment, and their children may be subjected to verbal or physical abuse. The arbitrary and unpredictable behavior of parents can have a significant negative impact on the psychological well-being of their children. Individuals in the latter category may consider suicide as a result of losing hope in their ability to receive genuine affection from their parents . --- Personal Factors Apart from environmental factors such as family, school, and social surroundings, distinct individual mental and psychological traits can also serve as predictors of suicidal tendencies among adolescent students. According to UNICEF's report, The State of the World's Children 2021, a significant proportion of adolescents between the ages of 10 and 19 worldwide experienced mental illnesses as defined by the World Health Organization. Additionally, in high-and middle-income countries, approximately 20% of individuals aged 15 to 24 reported experiencing frequent feelings of low mood or depression . --- Personality Traits The tendency to commit suicide is based on an individual's personality. The most significant correlations between personality traits and suicide behaviors are impulsiveness, paranoia, and introversion. Research has shown that students with a high impulsive disposition have a diminished capacity for executive control. When faced with sudden external stimuli, they are usually unable to control their emotions and behaviors, which can raise their risk of suicide . Moreover, adolescents with high levels of impulsivity are typically more aggressive and, as a result, are more likely to be shunned by others. Over time, the accumulation of experiences of rejection or distress can increase their likelihood of having suicidal thoughts . Students with paranoid tendencies frequently have inadequate emotional tolerance and emotion-regulation skills. The Classification and Diagnostic Criteria for Mental Disorders in China provides the following description of the paranoid personality: irrationally suspicious, frequently misinterpreting the unintentional, non-malicious, or even friendly actions of others as hostile or discriminatory; accommodating of conspiracy hypotheses; pathologically possessive; self-conceit-attributing one's own errors or failures to others; prone to offense; persistently and irrationally pursuing one's own interests; dismissive of objective evidence that contradicts one's own opinions . Typical characteristics of paranoid students are bigotry, narrow-mindedness, and fussiness, and they are easily provoked to go to extremes. Individuals with introverted tendencies may possess limited abilities to disentangle themselves from adverse emotions, thereby increasing their susceptibility to experiencing feelings of hopelessness. Based on pertinent survey data, it can be inferred that 53.8% of suicides among adolescents were committed by individuals belonging to the extreme introvert category . The individuals in question experienced a sense of hopelessness and inadequacy, perceiving themselves as lacking in ability and social acceptance and leading a monotonous existence. Furthermore, the individual experienced feelings of despair towards their parents, perceiving them as demanding and lacking consideration, ultimately viewing their own existence as a source of burden for their parents. Additionally, the individual expressed dissatisfaction with both the educational institution and the local populace, citing a lack of perceived regard for their personhood. Persistent exposure to adverse emotional states can ultimately lead to suicidal behaviors . --- Mental Disorders It is generally considered that the early stage of adolescence lasts from the ages of 10 to 13 and that the middle stage lasts from the ages of 14 to 17. Adolescents go through changes in their brain-neuroendocrine functions, hormone levels, and bodily morphological traits in both stages. While fostering physical and psychological development, these changes also significantly raise teenage risk for mental health issues, contributing to the high prevalence of mental disorders . The psychological tolerance of adolescents to stimuli from the outside environment is typically lower in those who are in a critical stage of physical and mental development. Due to a lack of coping mechanisms and abilities, it is easy for individuals to generate negative feelings, including disappointment, depression, regret, wrath, and envy, when confronted with difficulties in their academic pursuits, interpersonal interactions, and personal growth . The 2022 Blue Book on National Depression estimates that 95 million individuals in China suffer from depression, with 50% of those persons being students and 30% being under the age of 18 . The detection rates of anxiety and depression among 24,758 primary and secondary school children nationwide were found to be 31.3% and 17.9%, respectively, in a survey by the China Youth and Children Research Center in 2022. It was shown that 2.8% of fourth graders had serious depression, which is an increase of over 50% from 2020 . In their study, Xie et al. underlined that teenagers with depression are emotionally sensitive and vulnerable, and they frequently feel ashamed to share their inner sentiments and emotional experiences. Suicidal ideation is more likely to arise when unfavorable events take place . --- Low Psychological Resilience Psychological resilience refers to an individual's capacity to effectively manage and navigate through challenging circumstances and unexpected events, ultimately rebounding to their prior state of functioning in a timely manner. According to the source , utilizing this mechanism can aid in the resolution of adverse situations without incurring lasting negative effects. Additionally, it can function as an internal tool for managing depression and suicidal ideation. Adversities and failures are ubiquitous in reality and can elicit negative affective responses such as disappointment, frustration, and despair, which can have deleterious consequences for the affected individual. Therefore, possessing a sufficient level of resilience is imperative for both survival and achievement. Insufficient development of psychological resilience in young students can lead to the emergence of suicidal ideation even in response to minor challenges . --- Conclusion Children's suicide rates of Chinese elementary and secondary schools are on the rise, and this important social issue calls for prompt public action. The earliest study was carried out in the late 1990s, marking a very early beginning for suicide prevention and intervention research in China. Since then, efforts over the course of more than 20 years have improved psychological crisis intervention research and led to the development and use of suicide prevention and intervention techniques. The methods and processes used in China to prevent suicide, however, are still far behind those used in other countries. This study is intended to have some practical applications for teenage suicidal behavior intervention. Contemporary Chinese adolescents are confronted with intricate social surroundings, which encompass a decelerating domestic economy, an unpredictable international climate, and the emergence of the AI era. Adolescents' survival, social adaptation, and challenge coping are contingent upon a foundation of robust psychological principles in this particular milieu. The implementation and investigation of adolescent suicide intervention strategies are complex and enduring endeavors. The establishment of a robust framework for addressing psychological crises among adolescents requires collaborative efforts from various stakeholders, including the government, institutions, families, and individuals.■

Suicide has emerged as a severe global public health issue, ranking as the fourth most common cause of death for young people aged 15 to 19 years, according to the World Health Organization. Suicidal incidents among Chinese elementary and secondary school kids have increased recently, causing irreparable harm to families, schools, and society as a whole. To contribute to continuing efforts to strengthen suicide prevention strategies for Chinese adolescents, this paper focuses on studying the causes of teenage suicide in China from personal, familial, educational, and social viewpoints.

INTRODUCTION The epidemic of coronavirus disease 2019 began at the end of 2019. Because COVID-19 is extremely contagious, its rapid spread had an impact on people's lives, jobs, and opportunities for learning across the world. The extensive effects of the novel coronavirus have put the entire world to the test. Its effects are felt in all parts of life, including healthcare and educational services . It is well recognized that the health and educational systems, for instance in Sub-Saharan African countries face a number of difficulties that jeopardize the population's access to high-quality education and health services . This unpleasant situation was aggravated by the overwhelming effort to combat the COVID-19 epidemic that placed more burdens on these already frail systems . Because COVID-19 is so contagious, governments in a number of countries imposed lockdowns in an effort to stem the spread of the virus, and life became rather miserable as a result. Lockdown Measures include, but not restricted to numerous strict non-pharmaceutical interventions, such as hand sanitization, mask wear, the use of hand gels, and social distance measures, which were put into place to stop the spread of this highly contagious virus. The above measures notwithstanding, impacted negatively on a number of industries, including manufacturing , agriculture , and travel and tourism . Healthcare and education were not exempted from the negative impact of the COVID-19 pandemic, and as a result, saw significant upheaval . To combat the oncoming spread of the pandemic, the federal government of Nigeria undertook a number of policy steps to manage the virus, minimize disease transmission, and prevent economic and social disruptions. These measures were in line with the global effort to lessen COVID-19's impact. The government began by imposing a number of policy measures, such as social distancing in marketplaces, mosques, churches, schools, banks, shopping malls, airports, and beaches . All of this was done to slow the spread of the Covid-19 virus. On the other hand, the policy actions have a detrimental impact on the lives of the masses in a range of areas, such as unemployment, education, consumerism, income generation, health, and so on. From the foregoing, it is obvious that the extensive effects of the novel coronavirus have put the entire world to the test. Its effects are felt in all parts of life, including healthcare and educational services . However, most empirical studies neglected to investigate how its effects vary between the formal and informal sectors using health and education services, both of which are regarded as vital commodities by any standards. In other words, research on the impact of COVID-19 on education and health continues to emerge, but reports of the impact on health and education services are limited. The informal sector is the dominant sector of economic activities in Nigeria, which is very extroverted and vulnerable to external shocks . According to the Centre for the Study of the Economics of Africa [CSEA] , over 80% of working people in Nigeria are employed in the informal sector. While the formal sector of the economy is defined as those positions with regular hours and earnings that are documented, and acknowledged as sources of income on which income taxes must be paid . --- SAMPLE SIZE The sample size of the study is three hundred and sixty-three ( 363 --- HYPOTHESES H1: There is no significant variation in the mean of household health service uptake between formal and informal sector workers in Gwagwalada Area Council. H2: There is no significant difference in the mean of household educational service uptake between formal and informal sector workers in Gwagwalada Area Council. --- Empirical Review on the Study Variables This section examines the empirical literature on the study's two main variables. These variables, as the study's title suggests, are education and health. Education: COVID-19 uncovered the worldwide education environment's strengths and flaws in both developed and developing countries. Thus, this section empirically explores the impacts of the COVID-19 epidemic on Nigeria's educational system in order to help in a robust discussion of findings and incidentally provide a one-of-a-kind solution to the Nigerian educational sustainability challenges. A study by Spencer et al on challenges experienced by U.S. K-12 public schools in serving students with special education needs or underlying health conditions during the COVID-19 pandemic and strategies for improved accessibility reports that staff shortages and student compliance with prevention strategies were the two most frequently reported school-based challenges, and the two most frequently reported homebased challenges were a lack of learning partners at home and a lack of digital literacy among students' families . Similarly, Sonnenschein et al. used a survey sent via social media to examine parents' perspectives on PK-12 education for children receiving special education help during COVID-19 . Three main themes emerged from the results: Special education and related service hours were reduced during virtual learning; Parents reported that their children were unable to participate in virtual learning without a lot of adult support; and Parents frequently could not help their kids because of other obligations like childcare and work. In another related research, Lau and Lee reported that three weeks after in-school sessions were discontinued, information was obtained from 6702 parents of kindergarten and primary school students in Hong Kong. In the report, parents claimed that their kids struggled to finish assignments without heavy parental supervision. Additionally, parents desired increased school support. A study conducted by Debbarma and Durai to identify the areas of educational disruption due to the COVID-19 situation in Northeast Indian States showed that the educational system in some parts of northeastern India was disturbed and that many different things can interfere with children's education. Primarily, it was found that poor network connectivity made it difficult for students to communicate with their lecturers. According to a study conducted by Rwigema , COVID-19 had detrimental effects on education, including disruptions to learning and a decrease in access to resources for learning and research. In addition, it led to an increase in student debt and employment losses. Additionally, the findings indicated that many educators and learners relied on technology to maintain online education throughout the coronavirus outbreak. Online education was made challenging by a lack of infrastructure, which resulted in concerns with network connectivity, electricity, accessibility, and availability. Health: No doubt, the World Health Organisation proclaimed the coronavirus disease of 2019 pandemic on March 11, 2020, which had an impact on healthcare services with provider cancellations, delays, and patient avoidance or delay of emergency department or urgent care. Thus, the effects of the COVID-19 epidemic on Nigeria's health system are thus empirically explored in this section in order to aid in a thorough discussion of the findings and besides offers a novel solution to the problems with Nigeria's health system. Empirically, it has been found that over 56 million verified cases and more than 600,000 fatalities had been reported worldwide as of July 22, 2022 . Sharma et al. used secondary data analysis to examine how the COVID-19 pandemic affected the use of MCH provisions in India based on data that was routinely gathered from the HMIS for the quarter of April to June 2020 and 2021 relative to 2019. The results of the study demonstrate that the institutional and outreach maternal health services suffered during both COVID-19 periods that were taken into consideration. The finding is consistent with that of Sharma et al which confirms that the national lockdown, travel limitations, and increased accessibility to healthcare services had more significant effects of COVID-19's first wave in 2020. Additionally, a study that covers the pre-pandemic, lockdown and post-lockdown phases of the COVID-19 pandemic offers novel and relevant insights into the provision and consumption of crucial health and nutrition services. During the lockdown, the study found considerable delays in the delivery of facility-based services, but outreach-based activities were carried on as usual . The study reported that facility-based services were significantly impacted during the lockdown, mostly as a result of the state government suspending services in areas with positive COVID-19 instances. After the SARS-CoV-2 outbreak, inpatient and outpatient health service utilisation in China significantly decreased, according to research by Xiao et al. . This decrease was most likely brought on by changes in patient and provider behaviours, the suspension of health facilities' non-emergency services, severe mobility restrictions, and the SARS-CoV-2 epidemic itself. Despite some research suggesting increases in health service utilization during the Covid-19 lockdown, a study conducted by Moynihan et al on an evaluation of 81 studies covering more than 17.9 million services delivered across 20 countries revealed consistent evidence of significant decreases in the consumption of healthcare services during the pandemic period up to May 2020, compared to previous years. From the viewpoints of medical experts and community members, Assefa et al also assessed the effects of the COVID-19 pandemic on health service disruptions in three sub-Saharan African countries and found that the pandemic caused significant interruptions to vital health services. However, the study found that interruptions in health care in Nigeria were unusually high when compared to Burkina Faso and Ethiopia. The huge number of active COVID-19 cases in Nigeria may explain this disparity in impact. For example, in August, Nigeria reported 48,665 COVID-19 instances, Ethiopia had 27,242, and Burkina Faso had 1237.These findings imply that a country's pandemic curve influences health services that have been disrupted. When the weight of the COVID-19 caseload grows, so will the impact on access to other healthcare services. utilizing information gathered from in-depth interviews with thirty heads of household in the state capitals of Lagos, Enugu, and Anambra in Nigeria, Odii et.al. found that shortage of funds and an increase in the price of food and transportation were prevalent during the COVID 19 lockdown. Fundamentally, it was shown that many households self-medicated, employed home remedies, and ignored disease symptoms due to access issues to medical facilities. --- DATA PRESENTATION AND ANALYSIS --- Result Interpretation Table 2 shows the item-by-item descriptive analysis of health and how it differs between the formal and informal workers. According to the results produced for the formal sector workers, the mean scores for items 1 and 2 were higher than the average 5 -Likert scale measurement score of 3.0, while the mean scores for items 3, 4 and 5 were lower than the average. In contrast, the analysis of the workers in the informal sector revealed that the mean scores for items 1 and 2 were lower than the average measurement on the 5-likert scale, which was 3.0, while the mean scores for items 3, 4 and 5 were higher than the average. The results also revealed that the informal sector workers' sectional mean rating was higher than the formal sector workers' grand mean rating with a mean score of 3.42 compared to 2.84. Since the informal sector workers' grand means are higher than the average 5 Likert scale measurement score of 3.0, it can be inferred that they concurred that the covid-19 pandemic had an impact on health, in contrast to the formal sector workers, whose mean score of 2.84 was lower than the average 5 Likert scale measurement score of 3.0. Additionally, it was noted that there was a difference in opinion regarding health between the formal and informal workers in the Gwagwalada Area Council duringthe COVID-19 pandemic because the sectional mean of the informal sector workers was higher than the formal sector workers . --- Test of Hypotheses H0: There is no significant difference in the mean of household health service uptake between the formal and informal sector workers in Gwagwalada Area Council. --- Interpretation of Result An independent two-sample t-test was used to examine if there were any differences in opinion between formal and informal sector workers in Gwagwalada Area Council about how the COVID-19 pandemic lockdown affects household health service uptake. The test was found to be statistically significant because the mean response from the informal sector was statistically significantly higher than the mean response from the formal sector , with a mean difference of 1.56. This difference was significant since tcalculated = -18.45 was bigger than tcritical = -1.96, and P= 0.00 was less than 0.05 at 330 degrees of freedom. The alternative hypothesis was accepted, while the null hypothesis was rejected. This shows that there is a considerable difference in the means opinion of formal and informal sector workers in Gwagwalada Area Council over how the COVID-19 pandemic lockdown affects household health service uptake. --- Result Interpretation The item-by-item descriptive study of the education service uptake of formal and informal workers in Gwagwalada Area Council during the COVID-19 pandemic lockdown is shown in Table 4 above. The mean scores for items 1, 2, 4, and 5 among the formal sector workers were 2.68, 1.50, 2.60, and 1.48 respectively, which were below the average score of 3.0 on the 5-point Likert scale while item 3 had a higher mean score of 3.40. The examination of the informal sector workers revealed that, with the exception of item 2, all mean scores were above the average measurement of the 5-Likert scale. The result also showed that the sectional mean rating of the informal sector workers was higher with than the grand mean rating of the formal sector workers with . Because the informal sector workers' grand means are higher than the average of the five Likert scale measurements, or 3.0, it can be inferred that they agreed that the covid-19 pandemic had an impact on education levels, as opposed to the formal sector workers, whose mean scores of 2.33 were lower than the average of the five Likert scale measurements, or 3.0, implying that they did not. Similarly, during the COVID-19 pandemic, there was a difference in opinion regarding education level between formal and informal workers in Gwagwalada Area council because the sectional mean of the informal sector workers is higher than the formal sector workers . --- Test of Hypothesis H0: There is no significant difference in the mean of household education between the formal and informal sector workers in Gwagwalada Area Council. --- Interpretation of result An independent two-sample t-test was used to determine if there were any differences in the means opinion of the Covid-19 pandemic lockdown on household education in Gwagwalada Area Council between formal and informal sector workers. The test was found to be statistically significant because the mean response from the informal sector was statistically significantly higher than the mean response from the formal sector , with a mean difference of 0.74 found between the two sectors workers. This difference was significant since tcalculated = -30.92 was bigger than tcritical = -1.96 and P= 0.00 was less than 0.05 at the 330 degrees of freedom. The alternative hypothesis was accepted, while the null hypothesis was rejected. This means that there is a considerable variation in the mean perceptions of official and informal sector workers in Gwagwalada Area Council regarding how the Covid-19 pandemic lockdown affects household education. --- DISCUSSION OF RESULTS 6.1 Covid-19 Pandemic Lockdown and Household Health Service Uptake. The examination of health service intake revealed a substantial variation in the perceptions of formal and informal workers about their level of home health service consumption throughout the pandemic. The informal sector workers had a higher sectional mean rating than the formal sector workers. The null hypothesis was rejected as a result of this development. It follows that the informal sector workers at Gwagwalada Area Council agreed that their access to health care was hampered during the COVID-19 outbreak. This is in line with the finding of Moynihan et al. which found consistent evidence of large decreases in healthcare usage over the pandemic period up to May 2020, based on an analysis of 81 research spanning more than 17.9 million services given across 20 countries. This could be due to apprehension about contracting the illness or an inability to pay the medical bill due to a disturbance in their daily revenue creation. Furthermore, it is possible that the informal workers' healthcare budget was impacted by the lockdown, and they chose to rely on traditional medicine instead of seeking healthcare facilities. This finding aligns with the contribution of the World Health Organization regarding the effect of the Covid-19 pandemic . The impact of the COVID-19 pandemic has been felt worldwide in many different spheres of society. However, this affects mostly the people within the informal group, especially in access to health services for unrelated conditions. --- Covid-19 Pandemic Lockdown and Household Education. The analysis reveals that formal and informal worker opinions on household education vary. It is obvious that the informal sector workers' grand mean rating of the COVID-19's impact on household education was higher than that of the formal sector workers. This merely indicates that they have different opinions about Covid-19 and household education. The null hypothesis was rejected as a result of this development. Additionally, it can be said that during the Covid-19 lockdown, the informal sector workers in Gwagwalada Area Council had higher effects on their household education level than the workers in the formal sector. According to the result analysis, the informal sector was more negatively impacted by the consumption of educational services than the formal sector. This is so because the majority of them involved their families in online lectures, home lessons, etc. However, because it would have disrupted their livelihood, the workers in the informal sector could not afford to do it during the lockdown. Due to lost family income during the lockdown, the majority of children from the informal sector were not in school. The interruption of educational activities, job losses in the education industry, restricted access to laboratory facilities, loss of learning interests, restrictions on research, and decreased funding for education are only a few of the obvious repercussions of the lockdown. The results support the earlier claim made by Onyema et al. that the Covid-19 shutdown has made a number of issues in the education sector worse and limited the prospects of impoverished students to receive an education. The results are in line with those of a study by Susana Castro on the impact of the Covid-19 pandemic on household education, which unequivocally concluded that the Covid-19 pandemic had a significant negative impact on children and young people with special educational needs and disabilities as well as their families. --- CONCLUSION The economic and social shock caused by the COVID-19 pandemic lockdown has been revealed to be a universal phenomenon that has continued to disrupt the lives of people from all walks of life. The effects, however, differed between the formal and informal sector workers in Niger Gwagwalada Area Council, resulting in poor education and health service uptake, as well as disparity in education and health service uptake. COVID-19 was shown to have had a negative impact on education and health in Nigeria. Based on the foregoing, the study recommends the following: --- Household Health Service Uptake It is necessary to give more public funds to health and social protection programs. To guarantee the effective and efficient use of the resources at hand, a follow-up is necessary. By streamlining the administrative process, leveraging digital innovation, and equitably allocating costs between businesses, employees, and governments, we can create a more adaptable and inventive health system. The provision of basic medical care and initiatives to reach elusive informal laborers, especially undocumented migrants, are necessary. In hospitals, there should be access to medical supplies and medications. After COVID-19, all hands must be on the desk to guarantee that all medical supplies adhere to the necessary standards. --- Household Education To encourage educators and students to be committed when schools resume, the government should postpone the payment of school fees after Covid-19 and boost teacher welfare. Many parents lost their jobs and revenue generation decreased during the epidemic, making it difficult for them to pay for their children's school tuition following the pandemic. As a result, the government should step in and suspend students' school payments. Although it may not be relevant in private schools, the government can provide financial help to private schools in order to ensure the well-being of teachers and students. The government must construct broadband infrastructure that is affordable for all. To provide internet connectivity in all schools, the government should greatly support the cost of data packages. To bring it about, the government should create plans with mobile phone providers. Institutions can embrace distant learning platforms, technologies, and systems fast once they have access to affordable, dependable high-speed internet.As a matter of suggestion, the government should Lift or reduce taxes and dues on technology and media equipment and devices, including TV channels, mobile services, private radios, etc. A strong technological foundation should be supported by the private sector, and governments should work to provide a "enabling environment" for the educational system through fiscal policies and other incentives. The training of teachers in the use of technology and in delivering classes online should be a government priority. "I would urge the government to provide financial aid to institutions so they may build the foundation for an online queue in every institution. I would also suggest that the government require all teachers to complete a course on using ICT in both teaching and learning.

The study examines the impact of the COVID-19 pandemic lockdown policy on the welfare of workers in the formal and informal sectors in Gwagwalada Area Council, Abuja, Nigeria, during the COVID-19 period using health and education services as case study. The study was motivated by the fact that the outbreak had a negative influence on both formal and informal workers' well-being; nonetheless, concluding that the lockdown had an equivalent impact on both formal and informal sector workers in Gwagwalada Area Council without an empirical investigation may not have been reasonable enough. Thus, the need for this study to make a comparative analysis between formal and informal sector workers. For empirical purpose, independent two-sample t-test were used to compare perceptions of the two sectors regarding the effects of daily COVID-19 cases and the attendant lockdown stringency measures on health and education services uptake of workers in the two sectors understudy. A survey of 363 respondents was undertaken between the formal and informal workers in Gwagwalada Area Council of Abuja, Nigeria, to examine the effects of the pandemic lockdown policy measures on the uptake of education and health services. The study found that the informal sector workers were more severely affected than the formal sector during the lockdown. To address these challenges, the study recommends, among others, massive loans and grants for business owners with little or no interest rate to cushion the effects of the COVID-19.

Background Perinatal depression is widely reported to be the leading complication of childbirth [1]. Prevalence rates of depression in pregnant women have been estimated at between 6.5% to 12.9% [2]. As many as 19.2% of women have a depressive episode during the first 3 months postpartum [2]. Despite this level of need, the utilization of perinatal mental health services is low in terms of both accessing treatment and continuing treatment [3]. Untreated depression during pregnancy can lead to poor self-care, increased substance abuse [4], poor obstetrical outcomes [5], developmental delay in children [6,7], and increased risk of postpartum depression [8]. Despite its widespread prevalence, perinatal depression frequently goes undetected and disproportionately impacts disadvantaged women and their children. Access to help for perinatal conditions is also influenced by socioeconomic factors, with disparities negatively impacting perinatal mental health [9]. Despite the extent of perinatal mental health problems, much can be done to address mental health needs for women in the perinatal period. Cognitive behavioral therapy has been shown to improve depression [10]. Additionally, there is evidence that CBT depression treatments can be culturally adapted to improve their effectiveness with cultural minorities and low socioeconomic groups [11,12]. CBT and pharmacological interventions may offer effective treatment, but can increase costs of services [13], which may be a barrier for providing care [4][5][6][7][8]. Other approaches have also been explored, such as interpersonal psychotherapy [14,15], mindfulness [16] and the use of postpartum doulas [17]. The location of treatment delivery has also been researched, and shows that the delivery of evidencebased treatments by home visiting nurses can also be effective at addressing postpartum depression [18][19][20][21]. While the evidence base for effective treatment is growing, women are still not consistently accessing services for perinatal mental health problems. Research indicates that only a minority of women who are at risk for perinatal depression will receive referrals to address mental health issues, and fewer yet will ultimately engage in treatment [22]. The literature reports barriers to accessing help including a lack of knowledge or understanding about perinatal mental health issues [23], along with previous negative experiences with health care providers or a fear of being judged by health care providers [23]. Women have described how they value being able to talk with peers about their experiences [24]. Some studies indicate that accessing peer support was also potentially seen as the only acceptable option, due to negative views or experiences with more formal services [24,25]. Stigma related to accessing mental health services has also been identified as an important barrier for accessing help [26]. Despite existing research, there is a need to further our understanding of what kind of mental health needs women think they have, and what kind of support or help they would be likely to access to address those perceived needs. Integrating women's views and experiences into the development of services may help to overcome some of the barriers such as lack of trust and negative perceptions about accessing help [25]. This research focused on the geographic area of Minneapolis, Minnesota, a racially and culturally diverse large urban center with a population of over 380,000 [27]. Almost one-quarter of Minneapolis families with children under age five live in poverty. The investigators conducted a needs assessment to inform future development of public health systems approaches to reduce negative outcomes due to perinatal depression, which we defined as the time a woman conceives through the first year post-partum. Here we report on the findings from the focus groups conducted with women in the Minneapolis area as part of that needs assessment. The focus groups were specifically interested in exploring the following research objectives: A) What perceived needs do women describe they have in relation to their mental health through the perinatal period? B) What help do women describe current seeking in relation to addressing mental health concerns during the perinatal period? C) What support do women describe wanting for addressing mental health concerns during the perinatal period? --- Methods Women who were pregnant, or who delivered within the past year, were recruited to participate in focus groups intended to gain information on the mental health needs of women during the perinatal period. Women were recruited from three health care clinics, which were located in disadvantaged parts of the city and serving traditionally underserved populations. Staff in the clinic recruited women into the study from the population of women receiving prenatal or postnatal care in their clinic. Focus groups were conducted at the clinics from which the women were recruited. Groups were combined discussions of women receiving both prenatal and postnatal care, and focused on exploring a breadth of experiences. --- --- Procedures Institutional Review Boards of the University of Minnesota and Hennepin County Medical Center reviewed and approved this project as exempt for review. The study adhered to the RATS guidelines on qualitative research [28]. A moderator's guide was created with input from project stakeholders, which included members of the research team and community partners working in the area of perinatal mental health. The questions fell within the following categories: feelings and experiences of women after having a baby; how depression is viewed by these women and their families; help seeking behavior they have engaged in; current coping strategies used to deal with mental health challenges; and preferred types of mental health services they would like to access . The questions were semi-structured, and the focus group facilitators encouraged participants to share their views and experiences on these topics and interact with each other during the discussion. Participants were recruited from three sites by case managers at a federally qualified health care center in North Minneapolis and by prenatal care providers at Hennepin County Medical Center and a county clinic in south Minneapolis. Prospective participants were provided with a brochure describing the study and contact information for one of the two study coordinators. Recruits who were interested signed a release of information form to permit a study coordinator to contact them, or contacted the study coordinator directly. During the telephone conversation, the study coordinator verbally described the study and presented each element of the consent form. If the recruits chose to participate, they signed up for a specific focus group session at the end of the call. Seven focus groups were conducted with three to eight women in each group. Focus groups were conducted by Minnesota Department of Health staff who had specifically been trained in focus group facilitation and had experience conducting focus groups. Five groups were conducted in English and two groups were conducted in Spanish . The average age of the women participating in the focus groups was 27.5. At the beginning of each session, the facilitator went through the informed consent process with all of the participants, allowed time for questions, and obtained written consent from each participant. Participants were asked to complete a brief questionnaire, designed by the research team, describing socio-demographic characteristics and were told that this information would not be connected with their consent forms, in order to protect their confidentiality. All focus groups were recorded and transcribed verbatim by a professional transcription service. The groups that were conducted in Spanish were transcribed in Spanish and then translated to English by professional translators. Each transcript was systematically reviewed for themes. This entailed one member of the research team , working through each transcript, with the assistance of NVivo9 , and coding themes as they arose in the data. A series of themes and subthemes in the text were identified. The analysis was informed by a social constructivist version of grounded theory [29,30], where the researchers allow themes to emerge from the data, but acknowledge the influnce of additional sources on the analytic process, such as current literature, approach to the research, or knowledge on the topic. The research team met on two occasions to review the emerging codes alongside the data, and discuss any areas differences in interpretation until consensus was reached on the themes. Additionally the researchers discussed a summary of the findings with the research participants to further validate the emerging analysis. This discussion took place in the context of presenting key findings for further review and discussion at an additional focus group held with the same participants. --- Results The themes which emerged from the analysis are presented in relation to each of the research questions, namely a description of perinatal health needs, the current perception of available mental health resources, and women's desire for greater mental health supportive services. Here we discuss each of these main areas and their associated sub-themes, which are summarized in Table 1. --- Perinatal mental health needs Pregnancy was viewed as a time of experiencing mental health challenges Participants described a range of mental health needs, including dealing with changing moods, depression, isolation, worry and a sense of burden. Focus group participants described these needs as relating to pregnancy as a time of dealing with new and changing moods, hormonal changes and the challenges of life transitions. Some described living in a constant state of irritation. It's just a constant thing, it's all day every day with me. It's when I'm hungry, when I'm sleepy, when I'm tired, I just get irritated for -I don't even -I get irritated for no reason sometimes. Mental health concerns were exacerbated by life circumstances Some women described how some variations in mood were seen to be problematic and proceeded difficulties such as depression. Challenging life circumstances were seen by some as contributing to variations in mood and depression. For some this was associated with the challenges presented by the life transitions arising from having a baby, including disruptions to work or schooling. Focus group participants also identified that isolation and loneliness were serious issues facing women through the perinatal period. Some participants described feeling very alone, in general, with limited or no friend or family networks. For others this was an issue that was worsened by having an unsupportive, or absent, significant other. And you just feel alone. I mean like I have three older children and they have a dad and you know, he has a different dad, … this dude I mean he's not there, we don't communicate, it's no communication, none whatsoever. So it's just harder all around the board. --- There are a range of emotional and mood challenges in the perinatal period A range of other emotional issues emerged from the focus group discussion. While less prominent than the prospect of dealing with normal changes in moods which may grow into depression, other needs included dealing with feelings of anger, frustration, loss, grief, regret, anxiety and addressing the emotional impact of birth trauma. --- Social and emotional context impacts mental health The other prominent themes that were described as impacting mental health were economic and situational, rather than emotional. Many women in the focus groups expressed a persistent sense of worry about how to move through the life transition of having a baby, and all the responsibilities of providing for children. This was exacerbated by limited support. Many focus group participants were approaching parenthood with a great sense of burden about providing for their family, including the potential loss of income during pregnancy and after the child was born. Overall, the needs expressed illustrate that social and economic needs, such as difficulties with housing, and financial worries, along with emotional needs were all seen as having an impact upon mental health. This is particularly the case in circumstances in which women have limited or no support for facing these challenges. --- Mental health help currently available Women in the focus groups were asked to describe what they currently do to address their mental health needs, and they described a wide range of resources they access. --- Reliance on informal support Despite the social and economic barriers to accessing mental health services, participants exclusively focused on how they met their emotional needs. One prominent approach was to rely on informal networks, such as friends and family. Many participants described the importance of having people to talk to about pregnancy issues. Partners were seen as a valuable resource if the relationship was supportive; however many women did not have such a relationship to rely on, and drew on networks of friends for support instead. Participants described the benefits of sharing mutual experiences, and the usefulness of talking with someone who can empathize, someone who has shared the same, or similar, experience. This general support and advice was seen as a way to address mental health needs. I usually go to my mom or my sister, like my sister she's older than me and she's got three kids, this will be my first baby. So I go to women who have had kids before. That's where usually I go. This my first baby so I usually just take advice from women who has already been there. --- Reliance on self Many participants described relying mostly on themselves to get through difficult times. This included using strategies such as journaling. A certain degree of relying on the self was seen as important for identifying how to address one's mental health needs. Yeah, figuring it out by yourself is good too because ain't nobody else gonna sit here and have the answer for you, only you know. You know? --- Accessing services Some focus group participants had used therapy from a range of different professionals, to help address their mental health needs. Counseling and therapy was seen as a useful way to get support and identify strategies for coping. Some focus group participants used medications for depression and anxiety. This seemed to be particularly the case where there was a pre-existing issue with depression, or a previous experience of perinatal mood disorders. Aside from the main areas of mental health support, there was lesser mention of a very broad range of ways to address mental health needs, such as using acupuncture, going to church, using a doula, support from a hospital, peer support group, psychiatrist, midwife and social worker. --- Mental health support wanted Women in the study described the various forms of help or support they wanted to be able to access to address their mental health needs. --- Mental health needs where not sufficiently met Participants expressed a desire for a range of mental health support that was more comprehensive than the range of services that were currently available, indicating that there was a level of unmet need for this group. One key theme emerging from the analysis indicated that women wanted help that was accessible, including not being turned away for lack of health insurance and meeting childcare needs. --- Outreach services were considered important Women wanted the mental health support to include outreach to women in the community as a way to reach women who were having difficulties and to overcome access issues. For some women, they felt that home visits would also be important. Women described having a hard time seeking help, or even leaving the house, if they were suffering from depression. Home visits were seen as a proactive outreach approach to identifying women with mental health problems who may be reluctant to leave their home. --- Focusing on empowerment oriented support Women expressed the need to have mental health support that focused on supporting and empowering them to find solutions in a way that would build on their strengths and encourage resilience. One such approach that was seen as highly desirable by the focus group participants was peer support. They described the peer support as potentially being provided by other mothers, of all ages, including support from older women who had been through similar challenges. The prospect of being able to talk with and get support from someone who had been through a similar experience was seen as valuable, in part because that support would be seen as having credibility. I --- Peer support The value of peer support was evident as it was frequently used within the group sessions themselves, as participants shared their experiences and offered encouragement to each other. One participant gave an example of how she encourages other women. Yeah just being yourself and getting past it, trying to put it behind you, look forward, don't look back. If you look back you just gonna keep getting worse -that's why I tell people if something happen today then oh well, keep it pushing. The past is the past and the past and look to new beginnings, looking forward to new beginnings and new outcomes. --- Educational approach The prospect of mental health support being delivered in an educational format was described as being very desirable. Educational settings could be used to address broad issues such as preparation for pregnancy and childbirth, and early parenting. The setting was seen as an accessible way to bring together groups of parents to access support that would enhance mental health. Educational settings offered an advantage because some women felt it might more productively engage their partners in discussion and learning about perinatal mental health. --- Increased services Participants also described a range of specific services they thought should be offered such as being able to access a crisis phone line, faith-based support services, and assistance from doulas. There was a request for mental health services that could address broad social or economic difficulties by offering financial advice, housing assistance, and transport services. --- Discussion This study was designed to talk with women about what mental health needs they feel they have, what help they currently access and what help they would like to address their mental health needs. These needs include those that address emotional needs and ways to alleviate social and economic stressors that have a negative impact on their mental health. Women in the focus groups were asked to describe what they currently do to address their mental health needs. They primarily described relying on support from friends or family or their own strengths, although some acknowledged the usefulness of medication or therapy. This is consistent with other studies that have indicated women may turn to informal supports to meet their needs [24]. However the literature is mixed on the effectiveness of some forms of peer support for postpartum depression [31], although there is some growing indication that it may be an effective approach [17]. Women in this study were quite adamant that peer support would be a preferred option for mental health support. This may indicate that further work needs to be done to review the quality and nature of the peer support interventions offered in the perinatal field, as research on the use of peer support in general mental health indicates it can both be effective and very well received by those seeking mental health improvement [32,33]. Some women talked about the use of formal treatments, and this remains an important option. However our findings echoed other literature in highlighting the need for service providers to be cognizant of the fact that many women may have had negative experiences previously, and thus have negative perceptions of services [24,25]. Yet there remains a strong field of research about the usefulness of formal treatments [10], including their ability to be culturally adapted for use [11,12], and more work needs to be done to address the barriers women face in accessing services. A clear implication of this work is that the women want those helping them to address their mental health by using an empowerment orientated approach. They also preferred the use of an educational format. These findings highlight that a barrier to engaging women with current services might originate in the approach to services themselves, which use a medical model instead of an empowerment oriented approach to address mental health needs. One particularly notable finding in this research was how strongly women believed that social and economic challenges, such as difficult relationships, lack of adequate housing, financial issues or unemployment, contribute to mental health problems. Yet when women talked about what they currently do to address mental health needs, all actions focused on addressing emotional needs. Equally, the literature reflects this, as most interventions also focus on emotional needs [10][11][12][13]16,17]. There are some exceptions in the literature, with some interventions focusing on multiple risk factors [34], but these approaches appear to be primarily focused on improving health outcomes, rather than overall needs which may contribute to poor mental well-being. This signals the potential importance of addressing social and economic needs into services as one potential way to increase the perception of relevance and accessibility for women. --- Limitations This focus group study used a small group of women from one geographical area, and they may have self-selected into the study based on their interest and experience with mental health issues. This means there are limits to how broadly these findings can be generalized. Additionally we did not screen women for mental health problems, or ask them to disclose their experiences with mental health services, and this may limit generalizability of our findings. --- Conclusion Women in our study stated that they wanted access to a greater range of different kinds of support for mental health problems, such as educational groups or peer support. They also described significant social and economic needs which were rarely addressed in current services. These findings offer a challenge to further broaden the types of services offered to women to include peer support, resilience focused programming, and educational programming. Also, in order to increase relevance and accessibility for diverse women, mental health services should more clearly address social and economic challenges. --- Competing interests The authors declare that they have no competing interests. Authors' contributions NCR, PAH, HK, AG, and JK made substantial contributions to the conception and qualitative design of the study. They were also involved in the acquisition of the focus group data, RJP, NCR, PAH, HK, AG were involved in the analysis and interpretation of the data. They also were involved in writing and revising this manuscript. JMO and JK were involved in drafting the manuscript and revising it critically for important intellectual content. Everyone gave final approval of the version to be published.

Background: Perinatal mental health problems are common complications of pregnancy that can go undetected and untreated. Research indicated that mental health complications are more prevalent in women from disadvantaged communities, yet women from these communities often experience barriers to accessing treatments and interventions. Untreated depression during pregnancy can lead to poor self-care, increased substance abuse, poor obstetrical outcomes, developmental delay in children, and increased risk of postpartum depression. In this study we investigated the perceived perinatal mental health needs of our participants and they wanted to address their perceived needs. Methods: In this qualitative study, we invited women who resided in an underserved, urban community who were pregnant or who delivered within the past year to participate in focus groups. Results: Thirty-seven women participated in seven focus groups. Thirteen themes emerged which were described in relation to mental health needs, help currently accessed and the type of support wanted. The themes included the various mental health needs including dealing with changing moods, depression, feelings of isolation, worrying and a sense of being burdened. Women described using a limited range of supports and help. Participants expressed a preference for mental health support that was empowerment focused in its orientation, including peer support. Women also described the compounding effect that social and economic stresses had on their mental health. Conclusions: Participants wanted access to a greater range of supports for mental health than were currently available to them, including peer support, and wanted assistance in addressing social and economic needs. These findings offer a challenge to further broaden the types of services offered to women, and demonstrate that those services need to be responsive to the challenging contexts of women's lives. Integrating women's views and experiences into the development of services may help to overcome barriers to care.

Introduction The COVID-19 pandemic has a significant impact on various aspects of human life worldwide, including psychological health, as reported in almost all countries . The Arab nations are among the populations affected by poor mental health . As of June 2020, the Centers for Disease Control and Prevention reported that approximately one-third of individuals in the US suffer from depression or anxiety . Among young people, the incidence of mental health issues has increased dramatically over the last decade, with nearly 60% of those between the ages of 18 and 24 considered at risk for anxiety or depression, and 25% expressing suicidal ideation in the past month, compared to 11% of all individuals in 2019 . These rates represent a substantial increase in depression, coinciding with the implementation of stay-at-home orders, campus closures, and social distancing measures, which have significantly disrupted daily life and interactions . To mitigate the impact of the pandemic on mental health, the World Health Organization has recommended video games as part of its "Healthy at Home" program . However, excessive gaming may pose a risk to vulnerable individuals, including minors and those with gaming disorders . The pandemic has made it difficult to detect harmful effects such as harm to mental health or sleep patterns, as work and school have shifted to home-based practices . Moreover, prolonged stress during the pandemic may lead to significant emotional distress, including anxiety and depression . Although online games can help individuals deal with unpleasant emotions, greater involvement may make them more susceptible to Internet Gaming Disorder . According to the Diagnostic and Statistical Manual of Mental Disorders, 5th edition criteria, IGD is characterized by excessive and prolonged internet gaming, which leads to behavioral and cognitive symptoms such as progressive loss of control, tolerance, and withdrawal symptoms, similar to substance use disorders . IGD is associated with poorer mental health, sleep quality, interpersonal problems, and academic achievement, and is linked to lower quality of life . This aim of this study is to assess the impact of the COVID-19 pandemic on the prevalence of IGD and SPS among online gamers in the Arab world as well as dentifying risk factors for IGD negative consequences associated with these disorders. --- Methods --- Study design, setting This study aimed to assess the prevalence of internet gaming addiction and its association with social phobia among Arab adolescents and young adults aged 15 to 30 who played online games. The cross-sectional study was conducted between September 2021 and March 2022 in 11 Arab countries, namely Egypt, Syria, Palestine, Jordan, Tunisia, Iraq, Morocco, Algeria, Sudan, Libya, and Lebanon. --- --- Measurements The first section of the questionnaire gathered information on the participants' demographics, including age, sex, marital status, level of education, drug use history, level of smoking, country of residence, and gaming status. The Nine-item Internet Gaming Disorder Scale-Short Form was used to measure participants' online gaming addiction. The scale consists of nine items and assesses the symptoms of internet gaming addiction, including salience, mood modification, tolerance, withdrawal, conflict, relapse, problems, deception, and loss of control. Participants rated each item on a five-point Likert scale ranging from 1 to 5 . The scores ranged from 9 to 45, with higher scores indicating a higher degree of addiction . The Social Phobia Scale was used to measure social phobia. The SPS is a 20-item self-report measure that assesses the fear and avoidance of social situations. Participants rated each item on a fivepoint Likert scale ranging from 0 to 4 . The scores ranged from 0 to 80, with higher scores indicating a higher degree of social phobia . --- Sample size The sample size for this study was determined using the formula described by Swinscow and Cohen 1 [n = [2. P ]/d2]. Assuming a prevalence of 50% in the population, a confidence level of 95%, a margin of error of 5%, and a 5% addition for the non-response rate, the recommended sample size was 384 individuals. A total of --- Pilot study To ensure the survey's comprehensibility and to avoid errors, a pilot test was conducted with 50 participants. Cronbach's alpha values indicated strong internal consistency for the IGA and SPS scales . --- Statistical analysis The data were stored in an Excel sheet and analyzed using SPSS win statistical package version 26. Numerical data were summarized as means and standard deviations or medians and ranges as appropriate. Qualitative data were described as frequencies and percentages. Comparison between two groups for numerical variables was done using either the student t-test or the Mann-Whitney U test as appropriate. Comparison between more than two groups for numerical variables was done using ANOVA or non-parametric Kruskal-Wallis as appropriate. The relationship between qualitative data was analyzed using the Chi-square test or Fisher's exact test as appropriate. Binary logistic regression was used to determine the potential association between the risk of developing a high degree of internet gaming addiction and social phobia , based on the participants' baseline characteristics. --- Results --- Descriptive characteristics of sample by sociodemographic factors In this study, 2,237 individuals were included in the analysis, with a majority being male , and a mean age of 19.9 years. Most of the participants were from Egypt , single , and 2,102 of them were gamers. The study of gamers revealed that 73.8% played for less than 5 hours per day, and the majority started gaming before the age of 12 years. Approximately 75% of gamers reported watching streamers, with 35% indicating they watched "a lot. " Since the COVID-19 pandemic, 69% of participants reported playing more than usual since they were unable to leave their homes . --- Relationships between the social phobia scale and different sociodemographic factors The study found that social phobia scores were significantly associated with several sociodemographic factors, including national origin, gender, marital status, education status, age of first gameplay, gaming hours per week, COVID-19 impact on gaming attitudes, and increased dependence on gaming due to isolation . Being single, Egyptian, or male was associated with higher social phobia scores. Education level, age of first gameplay, and gaming hours also demonstrated significant differences . Additionally, the study found a significant difference in self-reported changes in gaming behavior during the pandemic , with those who reported "significantly increased" playing having a slightly higher SPS score than those whose behavior did not change . --- Relationships between internet gaming disorder scores and different sociodemographic factors The study found that gender, marital status, education level, and age at which the first game was played had no effect on scores for Internet Gaming Disorder . However, the study did observe significant differences in IGD scores based on nationality , changes in play behavior during the pandemic , duration of gaming per day , and following streamers . Specifically, participants who played games for five to 10 hours per day had scores almost 5 points higher on the IGD scale compared to those who played for half an hour. Moreover, gamers who watched "a lot" of streamers had scores two points higher on the IGD scale than those who did not watch. Additionally, Egyptian participants and those who reported a significant increase in gaming time due to the COVID-19 epidemic had the highest IGD scores . --- Potential predictors of a high level of IGD & SPS scores Of the nine variables examined in this study, only two were found to be statistically significant predictors of high levels of internet addiction: daily playing hours and starting age of gaming. Specifically, individuals who spent half to 2 hours per day on the internet were less likely to have a high level of internet gaming addiction compared to those who spent 5 to 10 h per day. Conversely, those who started gaming between the ages of seven and seventeen were more likely to have a high level of internet gaming addiction than those who began gaming before the age of six. Six factors were found to be statistically significant predictors of high levels of social phobia: gender, nationality, the impact of Covid-19 on gaming attitudes, social status, starting age of gaming, and daily playing hours. Males were less likely to have a high level of social phobia than females. Divorced individuals were 3.527 times more likely to experience severe social phobia than single individuals. Furthermore, those who spent 2 hours on internet gaming were less likely to experience . --- Discussion The present study investigated the relationship between sociodemographic factors and both Internet Gaming Disorder and Social Phobia Scale scores among gamers. The findings suggest that sociodemographic factors play an important role in predicting levels of IGD and SPS, indicating the complex and multifaceted nature of these conditions. In terms of IGD, the study found that daily playing hours and following streamers were significant predictors of high levels of internet addiction. These findings suggest that certain behaviors, such as excessive gaming or following online influencers, may lead to increased risk of developing IGD. It is important to note that excessive gaming has been associated with a range of negative outcomes, including poor academic performance, social isolation, and decreased physical activity. As such, clinicians should be aware of the potential risks associated with excessive gaming and work with patients to establish healthy gaming habits. The study also found that sociodemographic factors, such as nationality and the impact of COVID-19 on gaming attitudes, were significant predictors of IGD. Specifically, individuals from Egypt and those who reported a significant increase in gaming time due to the pandemic had the highest IGD scores. The COVID-19 pandemic has led to significant changes in daily routines and social interactions, and as such, it is not surprising that there has been an increase in gaming activity. However, clinicians should be aware of the potential negative impact of increased gaming on mental health and wellbeing. In terms of SPS, the study found that gender, nationality, and daily playing hours were significant predictors of high levels of social phobia. These findings suggest that there may be a complex interplay between gaming behavior, social anxiety, and sociodemographic factors. Social phobia can be a debilitating condition that can significantly impact an individual's quality of life. As such, clinicians should be aware of the potential risk factors associated with social phobia, such as excessive gaming and certain sociodemographic factors. Overall, the present study highlights the complex and multifaceted nature of IGD and social phobia. The findings suggest that sociodemographic factors play an important role in predicting levels of these conditions, and as such, clinicians should take into account the individual's sociodemographic background when assessing and treating these conditions. Additionally, it is important for clinicians to be aware of the potential negative impact of excessive gaming and work with patients to establish healthy gaming habits. The findings of this study have important theoretical implications for the field of psychiatry. The diagnostic criteria for Internet Gaming Disorder according to the DSM-5 are based on the presence of symptoms such as preoccupation, withdrawal, tolerance, loss of control, giving up other activities, continuation, deception, escape, and negative consequences . Our study found that the COVID-19 pandemic has led to an increase in dependency and tolerance towards gaming, which is consistent with the definition of IGD as a progressive loss of control over gaming, tolerance, and withdrawal symptoms. Additionally, our patients reported a cluster of behavioral and cognitive symptoms associated with IGD, including increased social phobia scores, anxiety, nervousness, and loss of self-confidence . Although gaming can be a healthy way to relieve stress in the moment, it should not be used as a long-term solution . Developing a gaming disorder or other related issues can arise if it becomes a habit rather than a means of relieving pressure . Due to the infeasibility of many contingency plans during the COVID-19 pandemic, the maladaptive use of gaming has increased in frequency . The increased dependency on playing video games during the pandemic observed in our study agrees with reports from other populations . We found that increasing playing hours was associated with increasing social phobia scores and aggravating the severity of social phobia . Moreover, the onset of playing video games at a younger age is strongly associated with an increased risk of developing social phobia and IGD . Our results also revealed that increasing playing time was associated with a higher IGD score, as reported previously . However, we observed a worsening in IGD scores in participants who played for a long time due to COVID-19 and in participants with increasing dependency on gaming due to isolation, which is consistent with observations from other studies in different populations . This study's practical implications highlight the need to monitor the increasing prevalence of IGD since the beginning of COVID-19, as it can cause further damage to the psychosocial situation, which is already damaged by the pandemic . The results of our study contradict the World Health Organization's advice to the general population to play video games to reduce COVID-19's psychosocial effects . --- Strengths and limitations There are several strengths and limitations to this study. It is the first study to investigate the impact of the COVID-19 pandemic on gamers in the Arab population, and the sampling method employed ensured that a comparatively representative sample was collected. Additionally, all related factors were assessed using validated questionnaires. However, the limitations of the study include the inherent nature of the data collection tool, which depends on the participant's ability to recall and complete the self-administered questionnaire. Despite using a validated Arabic version of the questionnaire, there is still a risk of recall bias. Furthermore, the response rate of males who completed the survey was much higher than the response rate of females, indicating that the number of males who are addicted to online gaming may be greater than the number of females. Finally, the study did not examine other variables that may affect IGD and social phobia, such as depression, anxiety, and stress, which may influence the prevalence of IGD . In conclusion, our study highlights the increasing prevalence of IGD during the COVID-19 pandemic in the Arab population and underscores the importance of monitoring its impact on mental health. The findings of this study may have implications for clinicians and policymakers working to mitigate the negative impact of the pandemic on mental health. --- Recommendations The current study highlights the importance of addressing the potential negative impact of COVID-19 on online gaming behavior and the development of IGD among the general population. To mitigate this impact, it is recommended to provide mental health support and counseling to individuals who are at risk of developing IGD due to prolonged isolation and boredom. This can be achieved through various means, such as online counseling, support groups, and educational campaigns to raise awareness about the risks associated with excessive gaming. Moreover, parents and caregivers should monitor and regulate their children's online gaming activities, especially during prolonged isolation periods, to prevent the development of IGD. --- Conclusion The present study adds to the growing body of literature on the impact of COVID-19 on mental health and online gaming behavior. Our findings suggest that the COVID-19 pandemic has led to an increase in online gaming activities, and this increase is associated with an increased risk of developing IGD and social phobia symptoms. Therefore, it is crucial to address this issue and provide adequate support to individuals who are at risk of developing IGD. Future studies should investigate the long-term impact of the pandemic on gaming behavior and the effectiveness of different interventions in mitigating the risk of IGD. Overall, this study highlights the importance of recognizing the potential negative impact of COVID-19 on mental health and emphasizes the need for proactive measures to address this issue. --- Data availability statement The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. --- Ethics statement The studies involving human participants were reviewed and approved by Ethics Committee, Faculty of Medicine, Alexandria University. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements. --- --- --- Publisher's note All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. --- Supplementary material The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/

Background: Gaming addiction is a compulsive mental health condition that can have severe negative consequences on a person's life. As online gaming has increased during the COVID-19 pandemic, studies have shown a heightened risk of mental health issues. This study aims to assess the prevalence of severe phobia and addiction to online gaming among Arab adolescents and identify risk factors associated with these disorders. Methods: This cross-sectional study was conducted across 11 Arab nations. Participants were recruited using convenience sampling through an online survey distributed on social media platforms in 11 Arab countries. The survey included demographic questions, the Nine-item Internet Gaming Disorder Scale-Short Form (IGDS-SF9) to measure participants' online gaming addiction, the Social Phobia Scale (SPS), and questions assessing the impact of the COVID-19 pandemic on the prevalence of internet gaming addiction. The data were analyzed using SPSS win statistical package version 26. Results: Out of 2,458 participants, 2,237 were included in the sample due to nonresponse and missing data. The average age of the participants was 19.9 ± 4.8 years, and the majority were Egyptian and unmarried. 69% of the participants reported playing more than usual since the COVID-19 pandemic, as they were confined to their homes. Higher social phobia scores were associated with being single, male, and Egyptian. Participants from Egypt and those who felt that the pandemic significantly increased their gaming time had higher scores for online gaming addiction. Several major criteria, such as playing hours per day and beginning gaming at an early age, were associated with a higher level of online gaming addiction with social phobia.The study's findings suggest that there is a high prevalence of internet gaming addiction among Arab adolescents and young adults who play online games. The results also indicate a significant association between social phobia and several sociodemographic factors, which may inform future interventions and treatments for individuals with gaming addiction and social phobia.

Introduction Pediatric obesity is a major health concern globally [1]. The prevalence of childhood and adolescent obesity has grown steadily in Spain in recent years. A meta-analysis of studies evaluating the prevalence of overweight/obesity among Spanish children demonstrated an increase among those aged 7-13 years, from 32.3% in 1999-2010 to 35.3% during 2011-2021 [2]. Preliminary results from Physical Activity, Sedentarism, and Obesity of Spanish youth 2022, a study on 2892 Spanish children aged 8-16 years, revealed the prevalence of overweight/obesity to be 33.4%, obesity 11.8%, and, notably, abdominal obesity 20.4%; this represented a 1.3% increase in obesity from 2000 to 2022 [3]. --- Materials and Methods --- Study Design and Participants ACTION Teens was a cross-sectional, online survey study that collated data from three groups of respondents in 10 countries/regions . The full methods for ACTION Teens have been published previously [14]. Participants from Spain were surveyed between August and December 2021 or between August and September 2021 . For the Spanish cohort, the recruited ALwO were 12 to <18 years old and living in Spain. Obesity was defined as a BMI ≥95th percentile for sex and age based on World Health Organization charts [15]. The recruited caregivers were ≥25 years old, resided with an ALwO in Spain, and participated in healthcare decisions for the ALwO. The recruited HCPs were required to be practicing in Spain and to have been in clinical practice for 2 or more years. They were also required to spend ≥50% of their time directly caring for patients and to see 10 or more ALwO per month. --- Survey Development Three distinct but overlapping surveys were developed. A steering committee comprising content experts and HCPs co-developed and approved all survey materials. Translation and back-translation from the English version were conducted and accuracy was fulfilled. The questionnaires have been published previously in the Online Supporting Information section of the global ACTION Teens study manuscript [14]. --- Procedures KJT Group Inc. oversaw data acquisition and reporting. An online survey utilizing Decipher Survey software was used for data collection. Online panels/databases were used to recruit caregivers and ALwO by targeting and screening adults from a stratified general population sample to identify caregivers of an ALwO. "Matched pairs" of ALwO and caregivers were maximized by asking verified caregivers for permission to enroll their ALwO. Following the maximization of the matched pairs, additional ALwO and caregivers were recruited to obtain the target sample size. HCPs were recruited from online physician panels/databases. Surveys were provided in Spanish and completed online by respondents or via in-person interviews . --- Outcomes As described previously [14], the primary outcome measures included the following: attitudes about people with obesity and obesity, and beliefs related to the impact of obesity; attempts to lose weight in the last year, barriers/motivations to losing weight, and definition of successful weight loss/management; frequency/history of weight conversations including the initiator, and responsibility for starting conversations about weight between HCPs and ALwO/caregivers; interactions between HCPs, ALwO, and caregivers, frequency of diagnosing obesity, reasons for not discussing obesity, and follow-up appointments; and information sources for learning about healthy lifestyles, weight loss/management, and obesity. Five-point Likert scales, yes/no responses, numeric responses, and single/multiple item selection were used to assess outcome measures, as appropriate. --- Sample Size The target number of completed surveys for Spain was 650 for ALwO, 650 for caregivers, and 250 for HCPs. These sample sizes were chosen to ensure statistical power while maintaining recruitment feasibility. --- Statistical Analysis As described previously [14], the full analysis set comprised all those who answered the survey in its entirety. KJT Group Inc. analyzed de-identified data using Stata , SPSS and Excel . Univariate descriptive statistics were used to summarize data. For continuous variables, where appropriate, outliers were truncated to the median. For generalizability and mitigation of selection bias, caregiver data were weighted to demographic targets representative for Spain for region, sex, age, household income, and education. Demographic targets were aligned with Spanish census data and were based on data from the United States Census Bureau-International Data Base, World Education News and Reviews , and Instituto Nacional de Estadística . The target adult general population sample was stratified to match these general population targets to ensure that the qualifying sample was largely representative of the adult population in Spain. The final incoming sample for the adult general population , which included those failing to qualify for the survey, was then weighted to match the demographic targets. To obtain the nearest possible target and sample balance, weights were calculated using a raking technique; individual respondent weights were capped at 0.5 and 5.00 to avoid extreme design effects. --- Results --- Participant Characteristics A total of 648 ALwO, 644 caregivers, and 251 HCPs were surveyed in Spain . The HCP response rate was 16.8%. Due to the recruitment methods, calculation of ALwO and caregiver response rates was not possible [14]. Table 1 shows the demographics and characteristics of the participants. Just over one-third of the ALwO were female, while almost half of the caregivers and HCPs were female. Approximately half of the ALwO and just over one-third of the caregivers indicated that the mother or father of the ALwO had overweight. Approximately 60% of the caregivers who answered the survey had overweight/obesity, and HCPs reported that 68% of their ALwO patients had family members with obesity in the household. HCPs' primary medical specialty and professional experience are reported in Table S1. --- Information Sources The information sources most commonly used by ALwO to learn about weight loss, healthy lifestyles, and weight management were YouTube, family and friends, information from a doctor, and social media, while for caregivers, they were information from a doctor, family and friends, YouTube, and dietitian/nutritionist . For HCPs, the most commonly used sources of information were medical education/continuing medical education programs, journal articles, and conferences . --- Perceptions of Obesity HCPs were more likely than caregivers or ALwO to believe that obesity has a strong/very strong impact on overall health and wellbeing . When evaluating the respondents' perceptions of the impact of obesity compared with other comorbidities, most respondents in the ALwO, caregiver, and HCP groups considered obesity to be as impactful or more impactful than several other health conditions, including diabetes, heart disease, and depression . The information sources most commonly used by ALwO to learn about weight loss, healthy lifestyles, and weight management were YouTube, family and friends, information from a doctor, and social media, while for caregivers, they were information from a doctor, family and friends, YouTube, and dietitian/nutritionist . For HCPs, the most commonly used sources of information were medical education/continuing medical education programs, journal articles, and conferences . --- Perceptions of Obesity HCPs were more likely than caregivers or ALwO to believe that obesity has a strong/very strong impact on overall health and wellbeing . When evaluating the respondents' perceptions of the impact of obesity compared with other comorbidities, most respondents in the ALwO, caregiver, and HCP groups considered obesity to be as impactful or more impactful than several other health conditions, including diabetes, heart disease, and depression . --- Figure 1. Perceived impact of obesity on overall health and wellbeing among ALwO, caregivers, and HCPs: impact of obesity; impact of obesity compared with other health conditions. Participants were asked how much of an impact they thought various health conditions had on a person's overall health and wellbeing using the following scale: 1 = no impact, 2 = slight impact, 3 = moderate impact, 4 = strong impact, and 5 = very strong impact . Panel A represents the proportion of respondents who indicated that obesity has a strong/very strong impact. Data were recoded to compare each participant's response regarding the impact of obesity with their response regarding the impact of other health conditions, and are shown in panel B; if the response was higher for obesity than for another health condition, it was coded as "Obesity is more impactful"; if equal, it was coded as "Obesity is as impactful". ADHD, attention deficit hyperactivity disorder; ALwO, adolescents living with obesity; HCP, healthcare professional. Figure adapted from [14]. --- Impact of Obesity Most ALwO considered their weight to be slightly above normal , while 25% of ALwO thought their weight was normal . However, caregivers were more likely to think that the weight of their ALwO was normal . Considering their overall health, 59% of ALwO believed it to be at least good. In contrast, 95% of caregivers believed that their ALwO's health was at least good . Only 1% or fewer ALwO and caregivers considered their/their child's health to be poor . Participants were asked how much of an impact they thought various health conditions had on a person's overall health and wellbeing using the following scale: 1 = no impact, 2 = slight impact, 3 = moderate impact, 4 = strong impact, and 5 = very strong impact . Panel A represents the proportion of respondents who indicated that obesity has a strong/very strong impact. Data were recoded to compare each participant's response regarding the impact of obesity with their response regarding the impact of other health conditions, and are shown in panel B; if the response was higher for obesity than for another health condition, it was coded as "Obesity is more impactful"; if equal, it was coded as "Obesity is as impactful". ADHD, attention deficit hyperactivity disorder; ALwO, adolescents living with obesity; HCP, healthcare professional. Figure adapted from [14]. --- Impact of Obesity Most ALwO considered their weight to be slightly above normal , while 25% of ALwO thought their weight was normal . However, caregivers were more likely to think that the weight of their ALwO was normal . Considering their overall health, 59% of ALwO believed it to be at least good. In contrast, 95% of caregivers believed that their ALwO's health was at least good . Only 1% or fewer ALwO and caregivers considered their/their child's health to be poor . Regarding how much the ALwO worried about their weight, 25% were very or extremely worried and 76% were at least somewhat worried . A similar percentage of caregivers reported that their ALwO was very or extremely worried about their weight , but fewer reported that their ALwO was at least somewhat worried . More ALwO than caregivers worried a lot or a little about the ALwO's weight affecting their future health . Regarding how much the ALwO worried about their weight, 25% were very or extremely worried and 76% were at least somewhat worried . A similar percentage of caregivers reported that their ALwO was very or extremely worried about their weight , but fewer reported that their ALwO was at least somewhat worried . More ALwO than caregivers worried a lot or a little about the ALwO's weight affecting their future health . --- Weight Loss Overall, 39% of caregivers believed that their child will naturally slim down as they grow older, compared with 29% of HCPs . A greater proportion of caregivers than ALwO agreed that the ALwO could lose weight if he/she set their mind to it (Figure --- Weight Loss Overall, 39% of caregivers believed that their child will naturally slim down as they grow older, compared with 29% of HCPs . A greater proportion of caregivers than ALwO agreed that the ALwO could lose weight if he/she set their mind to it . More ALwO than caregivers felt that weight loss was completely the responsibility of the ALwO , while 21% of HCPs thought that losing weight was completely their patient's responsibility . Similar proportions of ALwO and caregivers indicated that they or their child had attempted to lose weight in the past year, but more ALwO reported they were somewhat or very likely to try to lose weight within 6 months than caregivers reported for their ALwO . 3). More ALwO than caregivers felt that weight loss was completely the responsibility of the ALwO , while 21% of HCPs thought that losing weight was completely their patient's responsibility . Similar proportions of ALwO and caregivers indicated that they or their child had attempted to lose weight in the past year, but more ALwO reported they were somewhat or very likely to try to lose weight within 6 months than caregivers reported for their ALwO . For each question, response options were prespecified, and only one option could be selected. Panel A presents the proportion of participants who indicated that they "strongly agree" or "somewhat agree" with each statement. For panel C, the "not very likely" category includes the response options "not likely at all" and "not very likely", and the "very likely" category includes the response options "very likely" and "extremely likely". Percentages may not sum to 100% due to rounding. ALwO, adolescents living with obesity. Figure adapted from [14]. For ALwO and caregivers, the most frequently reported reason why ALwO wanted to lose weight was to be more fit/in better shape, while for HCPs, it was a desire to improve popularity and social life . andQ109, and caregiver data are based on responses to caregivers Q113, Q110a, and Q111. For each question, response options were prespecified, and only one option could be selected. Panel A presents the proportion of participants who indicated that they "strongly agree" or "somewhat agree" with each statement. For panel C, the "not very likely" category includes the response options "not likely at all" and "not very likely", and the "very likely" category includes the response options "very likely" and "extremely likely". Percentages may not sum to 100% due to rounding. ALwO, adolescents living with obesity. Figure adapted from [14]. For ALwO and caregivers, the most frequently reported reason why ALwO wanted to lose weight was to be more fit/in better shape, while for HCPs, it was a desire to improve popularity and social life . The barriers to ALwO weight loss reported by ALwO and caregivers were similar ; the most frequently reported barriers were an inability to control hunger, liking to eat unhealthy food, and lack of motivation. HCPs most commonly agreed that unhealthy eating habits, preference for unhealthy food, and insufficient exercise were barriers to ALwO losing weight, while inability to control hunger ranked 8th . The definitions of success for ALwO weight loss according to ALwO, caregivers, and HCPs are reported in Figure S5. The barriers to ALwO weight loss reported by ALwO and caregivers were similar ; the most frequently reported barriers were an inability to control hunger, liking to eat unhealthy food, and lack of motivation. HCPs most commonly agreed that unhealthy eating habits, preference for unhealthy food, and insufficient exercise were barriers to ALwO losing weight, while inability to control hunger ranked 8th . The definitions of success for ALwO weight loss according to ALwO, caregivers, and HCPs are reported in Figure S5. --- Conversations about Weight Over half of ALwO felt they could talk honestly about their weight with their mother or father, but only 23% felt this way about conversations with their HCP. The most common ALwO-and caregiver-reported barriers to conversations about weight with an HCP were not seeing their/their child's weight as a significant medical issue and the ALwO already knowing how to manage their weight . The most frequently reported barrier for ALwO to initiating conversations about their weight, as perceived by HCPs, was the ALwO not feeling comfortable bringing up the topic . Approximately half of caregivers had talked to their child's doctor about their child's weight in the course of the last year. Both ALwO and caregivers most frequently reported that the parent/caregiver usually started the conversation ; contrary to this, --- Conversations about Weight Over half of ALwO felt they could talk honestly about their weight with their mother or father, but only 23% felt this way about conversations with their HCP. The most common ALwO-and caregiver-reported barriers to conversations about weight with an HCP were not seeing their/their child's weight as a significant medical issue and the ALwO already knowing how to manage their weight . The most frequently reported barrier for ALwO to initiating conversations about their weight, as perceived by HCPs, was the ALwO not feeling comfortable bringing up the topic . Approximately half of caregivers had talked to their child's doctor about their child's weight in the course of the last year. Both ALwO and caregivers most frequently reported that the parent/caregiver usually started the conversation ; contrary to this, HCPs reported that they initiated weight discussions the majority of the time, but only 29% of HCPs felt they are responsible for starting the conversation . A greater proportion of ALwO than caregivers had been informed about their/their child's obesity diagnosis by their doctor. HCPs reported that they inform, on average, 87% of ALwO or their caregivers about the adolescent's obesity diagnosis. Furthermore, 50% of the HCPs reported they always recorded an obesity diagnosis in the medical record, and 37% recorded it most of the time. Among ALwO and caregivers who had talked with an HCP about their/their child's weight in the past year, 40% of ALwO and 71% of caregivers reported feeling comfortable doing so, while 26% and 9%, respectively, were not comfortable participating in these discussions. Most ALwO and caregivers reported at least one positive feeling following the latest discussion with an HCP, with the most common feelings reported to be "motivated", "supported", and "hopeful" . Among ALwO and caregivers who had not talked with an HCP about the ALwO's weight, 44% of ALwO and 55% of caregivers would feel comfortable discussing it. their/their child's obesity diagnosis by their doctor. HCPs reported that they inform average, 87% of ALwO or their caregivers about the adolescent's obesity diagnosis. thermore, 50% of the HCPs reported they always recorded an obesity diagnosis i medical record, and 37% recorded it most of the time. Among ALwO and caregivers who had talked with an HCP a their/their child's weight in the past year, 40% of ALwO and 71% of caregivers rep feeling comfortable doing so, while 26% and 9%, respectively, were not comfortable ticipating in these discussions. Most ALwO and caregivers reported at one positive feeling following the latest discussion with an HCP, with the most com feelings reported to be "motivated", "supported", and "hopeful" . Am ALwO and caregivers who had not talked with an HCP abou ALwO's weight, 44% of ALwO and 55% of caregivers would feel comfortable discu it. --- Figure 5. The feelings of ALwO and caregivers following their latest weight discussion with an Percentages represent the proportions of respondents who selected each of the prespecifie sponse options for ALwO/caregiver Q410, among the subset of ALwO who had discussed w with an HCP in the past year or the subset of caregivers who had discussed their child's weigh an HCP in the past year . The responses of caregivers represent own feelings rather than their perception of their child's feelings. The net positive category Percentages represent the proportions of respondents who selected each of the prespecified response options for ALwO/caregiver Q410, among the subset of ALwO who had discussed weight with an HCP in the past year or the subset of caregivers who had discussed their child's weight with an HCP in the past year . The responses of caregivers represent their own feelings rather than their perception of their child's feelings. The net positive category is the proportion of participants who selected at least one positive answer ; the net negative category is the proportion of participants who selected at least one negative answer . ALwO, adolescents living with obesity; HCP, healthcare professional. Figure adapted from [14]. --- Weight Management Most HCPs believed obesity to be a chronic disease . As reported by ALwO and caregivers, the methods most commonly used by ALwO for managing weight in the past year were becoming more physically active and improving eating habits . The most effective methods for managing weight according to HCPs were becoming more physically active, improving eating habits, and reducing screen time . --- Discussion This analysis of the Spanish cohort from the ACTION Teens survey study provides key insights regarding the attitudes, perceptions, and behaviors of ALwO, caregivers, and HCPs, as well as possible obstacles to effective obesity care and Spanish culture-specific factors. There was misalignment between ALwO, caregivers, and HCPs, notably around the perception and impact of obesity and regarding conversations about weight. The findings of this study may change the strategy for interacting with caregivers and adolescents with obesity. The study also highlights the importance of training HCPs not only in biological problems related to obesity, but also in the perceptions, attitudes, and motivations of the people involved. One-quarter of ALwO and almost half of caregivers considered their/their child's weight to be normal. The lack of recognition of the ALwO's obesity status may reflect the relatively small proportion of ALwO and caregivers who had been informed of their/their child's obesity diagnosis by their HCP. In contrast, HCPs stated that they informed the majority of ALwO/caregivers about the ALwO's diagnosis of obesity, with almost all reporting that they record obesity diagnoses in medical records most or all of the time. This aligns with the global ACTION Teens study, in which a greater proportion of HCPs reported informing ALwO/caregivers of the ALwO's obesity diagnosis compared with reports from ALwO/caregivers themselves [14]. This indicates a miscommunication, both in Spain and globally, between HCPs, ALwO, and caregivers with regard to obesity diagnoses, highlighting the importance of education and communicating weight status to ALwO and caregivers. Compared with the global cohort, more Spanish caregivers tended to misperceive their child's health/weight, with a greater proportion perceiving the overall health of their ALwO to be at least good and believing the weight of their ALwO to be normal [14]. This result aligns with a cross-sectional analysis of Spanish school children that reported parental underestimation of the weight category of children with obesity in 94% of cases [16]. The tendency for parents/caregivers to underestimate the overweight status of their children has been reported in other studies worldwide that included Spanish participants [17,18]. The greater propensity among Spanish caregivers, compared with those in the ACTION Teens global cohort, to believe their child's weight was normal may be related to greater indulgence towards pediatric obesity and failure to recognize the detrimental impact of obesity in Spain. The high proportion of caregivers with overweight or obesity may also have impacted how they perceive the weight of their child. Visual normalization of obesity has been shown to contribute to the perceptions of caregivers, whereby increased exposure to larger body sizes leads caregivers to shift their perception of normal weight in children, leading to an underestimation of the weight of their child [19,20]. Previous data indicate that the lack of perception of obesity by the family is a determining factor in the development or worsening of childhood obesity [21]. Hence, family involvement is considered fundamental to the prevention of childhood obesity [22]. Considering attitudes towards weight loss, many caregivers thought that their child would slim down with age. This could indicate a misunderstanding of obesity in adolescence and may be an important barrier to seeking support from their HCP for management of their child's weight. Delayed initiation of medical management may be compounded by views among ALwO and caregivers regarding personal responsibility for weight loss. Approximately one-half of ALwO and one-quarter of caregivers felt that losing weight was solely the ALwO's responsibility. Feelings of personal responsibility for weight loss among individuals with obesity are not uncommon, arising from a narrative that positions obesity as a choice and/or lack of willpower [23]. Furthermore, when analyzing barriers to weight loss, there was discordance between HCPs and ALwO/caregivers on the importance of the inability to control hunger, with ALwO/caregivers, but not HCPs, reporting this as one of their top three barriers. This may indicate a misunderstanding around the biology of obesity, even among HCPs, and calls for improved education on this topic that can be further communicated to patients. There were no remarkable differences between Spanish and global caregivers in terms of key motivators or barriers to weight loss. Key motivators in all respondent groups in Spain and globally included those related to appearance , improved social status , and improved self-confidence [14]. The barriers to weight loss most frequently reported by ALwO and caregivers, both in Spain and globally, were the inability to control hunger, lack of motivation, and a preference for unhealthy foods, while HCPs most often reported poor diet and insufficient physical activity [14]. Although 88% of Spanish ALwO reported barriers to weight loss in the present study, almost three-quarters of ALwO and over one-half of caregivers indicated that they/their child would be somewhat or very likely to try to lose weight within 6 months. Similar proportions of ALwO and caregivers were reported in the global study [14]. As such, HCPs should feel encouraged to initiate conversations about weight with ALwO and caregivers. Guidance from the American Academy of Pediatrics highlights the importance of communicating weight status to a child with obesity and their family in order to guide the subsequent evaluation and treatment [24]. However, this first step can be challenging for ALwO, caregivers, and HCPs alike, and can present an initial barrier to the advancement of obesity management. Only a small proportion of HCPs believed that they are responsible for initiating conversations about weight. Furthermore, approximately two-thirds of HCPs reported discomfort among ALwO in bringing up the topic of weight. In contrast, only 26% of ALwO who had discussed weight reported a lack of comfort with the conversation. Moreover, most ALwO and caregivers felt positive about previous weight-related discussions with an HCP. Across those who had not previously had a discussion about weight with an HCP, around half of ALwO and caregivers would feel comfortable with the conversation. As such, HCPs should feel confident in raising the topic of weight in order to facilitate effective treatment strategies. Interestingly, while 51% of Spanish caregivers reported having spoken with their child's doctor about their weight within the last year , less than 10% reported being informed of a diagnosis of obesity [14]. This indicates a miscommunication between HCPs and caregivers that is particularly prominent in Spain compared with other world regions. This may reflect a lack of understanding among Spanish caregivers or an unwillingness of HCPs to engage in challenging discussions. The recommendations for the prevention and treatment of childhood obesity in Spain are mainly focused on changes in diet and increased exercise, paying little to no attention to the attitudes, behaviors, and motivations of the family and the adolescent, and the importance of these in ensuring successful management of the child's obesity [25][26][27]. It will be important to integrate the results of this study at the primary care level to facilitate early detection of childhood obesity and provide information about obesity to the family. In Spain, children are checked by the primary pediatrician on numerous occasions during the first years of their lives. As contact with the family is frequent during this time, recommendations related to obesity may have a substantial impact. The strengths and limitations of this analysis of the Spanish cohort are comparable to those of the overall ACTION Teens study [14]. This was the first evaluation and comparison of the lived experiences of three parties involved in obesity management in Spain. This analysis provides valuable information that may be used to align and achieve the objectives of all participating groups. The main limitations of the study include the risk of selection bias and the potential for measurement error due to use of online surveys, the self-reported weight and height data that may have led to underestimation or overestimation of ALwOs' BMI, the cross-sectional study design, and lack of information on body composition and pubertal stage. Furthermore, only 36% of ALwO in the Spanish cohort were female, and so results may be biased towards male ALwO attitudes, perspectives, and behaviors; subgroup analyses based on sex are outside the scope of this manuscript. The majority of ALwO and caregivers were not matched. In addition, HCP respondents did not directly care for the surveyed ALwO; therefore, it was not possible to carry out paired studies. Overall, our findings should be generalizable to the wider population of ALwO, caregivers of ALwO, and HCPs treating ALwO in Spain. Although online surveys typically have poor response rates, broad inclusion and exclusion criteria were used to ensure that respondents were representative of ALwO and HCPs treating ALwO in Spain. Furthermore, stratified sampling and demographic weighting of caregiver data were used to ensure that the sample was as representative as possible. Additionally, while there is potential for selection bias with online surveys, this risk was mitigated by not specifying the topic of the survey in adolescent/caregiver email invitations and carefully designing screening questions to ensure that adolescents and caregivers did not know the purpose of the study until they had met the qualification criteria. --- Conclusions This analysis of the Spanish cohort of the cross-sectional ACTION Teens survey study highlights existing barriers to effective obesity management of ALwO in Spain, as well as Spanish culture-specific factors. It revealed misperceptions of weight status in a high proportion of ALwO and caregivers of ALwO, and suggested that ALwO are more likely to worry about their weight than their caregivers. Both ALwO and caregivers reported not being able to control hunger as an important barrier to weight loss, while HCPs were more likely to report a poor diet or a lack of physical activity. In addition, communication difficulties and misalignments exist between HCPs, ALwO, and caregivers. HCPs assume that ALwO will not feel comfortable starting conversations about their weight, and a minority of HCPs perceive themselves as being responsible for initiating weight-related discussions. Meanwhile, there is a willingness among both ALwO and caregivers to talk with an HCP about the ALwO's weight; they feel comfortable discussing it and feel positive after having the conversation. Improved communication between ALwO, caregivers, and HCPs is required, alongside a better understanding of the existing barriers to weight loss. --- Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/nu15133005/s1, Table S1: HCP primary medical specialty and professional experience. --- Data Availability Statement: Data will be shared with bona fide researchers submitting a research proposal approved by the independent review board. Individual participant data will be shared in data sets in a de-identified and anonymized format. Data will be made available after research completion. Information about data access request proposals can be found at novonordisk-trials.com. ---

Although the prevalence of pediatric obesity is rising, understanding of the perceptions, attitudes, behaviors, and barriers to effective obesity care among Spanish adolescents living with obesity (ALwO), their caregivers, and healthcare professionals (HCPs) is lacking. In 2021, the crosssectional ACTION Teens survey study was conducted in 10 countries; results from the Spanish cohort are presented herein. The survey was completed by 648 ALwO, 644 caregivers, and 251 HCPs in Spain. A total of 25% of ALwO and 43% of caregivers thought that their/their child's weight was normal, and more caregivers than ALwO perceived the ALwO's health to be at least good (95% vs. 59%, respectively). Only 53% of ALwO and 9% of caregivers reported receiving an obesity diagnosis, despite HCPs reporting they provide diagnoses to 87% of ALwO/caregivers. Although 65% of HCPs felt that ALwO may not be comfortable discussing weight, only 26% of ALwO who had discussed weight with an HCP (n = 488) reported not feeling comfortable. Inability to control hunger was a key barrier to ALwO losing weight identified by ALwO/caregivers, but not HCPs. Improved communication between the three groups, a better understanding of barriers to weight loss, and improved health education on obesity are needed in order to enhance obesity care in Spain.

Introduction While we have witnessed positive shifts in attitudes towards people with disabilities over recent decades, the label of 'intellectual disability' continues to be a dominant and stigmatising one. Labelling theory suggests that humans manage the world around them through categorisation and by applying labels to themselves and to individuals or groups around them. These labels can acquire negative valence where the labelled group is considered deviant from cultural norms. The labels, due to their stigmatising nature, can negatively influence individuals' self-identity and behaviour. As such, labelling has been criticised for instigating and maintaining stigma and activists with intellectual disabilities have called to "label jars, not people". Social identity theory focuses on the notion that group membership and identification with one's in-group can shape a person's sense of self through comparisons with out-groups. Thus, belonging to a stigmatised group has implications, likely negative ones, for how individuals feel about themselves. This can lead to rejection or denial of group membership in order to save face or avoid prejudice as discussed by Finlay and Lyons , or alternatively, some individuals may try to raise their self-esteem through engaging in collective action to bring about social change this would however require individuals to accept the label. Branscombe, Schmitt, and Harvey suggest that experiencing a sense of belonging in a stigmatised group can buffer against negative experiences such as prejudice and discrimination, through social support. The question of how people with intellectual disabilities should deal with their intellectual disability and the devalued status in society associated with this label has been the focus of debate for several decades. Social role valorisation proposed that by adopting socially valued roles, people with intellectual disabilities would be able to 'pass' and thus escape their deviant status. These ideas were highly influential in the intellectual disability field but much criticised by some for apparently failing to challenge the status quo, and implicitly accepting the lower value assigned to anyone with an intellectual disability . In response, Szivos and Griffiths proposed that the answer for dealing with a stigmatised identity is not to attempt to 'pass' but rather to enable those affected to examine the meaning of the disability and to 'own' it through a process of increasing self-acceptance. They also noted that dealing with a stigmatised identity by trying to assimilate into mainstream society can have potential negative consequences, not least derogation of members own group and concomitant social isolation. However, the suggestion of consciousness raising as a positive paradigm was rejected, for example, by Emerson who questioned the merits of boosting individuals' sense of self by focusing on their membership of a group defined by impairment, and suggested instead we should focus on the multiple roles and identities people with intellectual disabilities have, or should have, access to. Adopting an understanding of identity as socially constructed and fluid rather than fixed and moving away from group-based definitions of self can go some way to avoid the somewhat black and white tone of some such debates. This is in line with systemic ideas regarding human behaviour being embedded within relational contexts . Accordingly, individuals are described to have a relational sense of self; for example, someone may view themself as having an intellectual disability in some contexts, while in others this label purporting membership of a group of people with intellectual disabilities may seem irrelevant. Nonetheless, whether a person deemed by others to have an intellectual disability views themselves as having an intellectual disability, at least in some contexts and at some times, has implications, for example, for engagement with initiatives and support services targeting this population, including efforts to engage them in self-advocacy. If one does not identify with the intellectual disability label, one may well regard such initiatives as of little or no personal relevance and reject all support associated with the label, which may have benefits but also negative consequences. Furthermore, the centrality of intellectual disability to one's social identity is likely to affect how one makes sense of others' negative reactions. For example, one may view them as a sign of disablism and respond with righteous anger, or alternatively one may feel personally rejected and attribute this either to others' negative traits or states or to other aspects of one's own social identity . Self-reported stigma associated with having an intellectual disability has been found to be positively associated with psychological distress and negatively with quality of life . It stands to reason that a prerequisite for relating hostility and prejudicial actions to one's intellectual disability is that one views oneself as having such a disability, and/or an awareness that others view one in this vein. Beart, Hardy, and Buchan reviewed research that attempted to answer the question of how people with intellectual disabilities construct their social identity, and centrally how, or rather whether, they view 'intellectual disability' as part of their identity. Their much cited review concluded that many people with intellectual disabilities seem unaware of their 'intellectual disability' status or do not see it as applicable to them. They put forward various explanations for this, including a wish to distance oneself from belonging to a stigmatised group, or the use of denial as a defence mechanism. --- This review Given that Beart et al.'s review has been widely cited and that its conclusions have important implications, we felt it timely to review empirical work on this topic published over the intervening years. Our review set out to answer the following questions: 1) To what extent are people with intellectual disabilities aware that they have an intellectual disability? 2) How do they feel about the 'intellectual disability' label ascribed to them? 3) Are they aware that others view the label negatively? --- Method --- Search strategy Rapid review methodology Studies were included in the review if they were published in English in a peerreviewed journal since 2004, but not included in Beart et al.'s review, to account for time lag in publications. Articles were included if they reported relevant original empirical work, regardless of methodology. Articles that were concerned with the impact of involvement in self-advocacy or participatory research , even though relevant to our focus, were excluded from this review. --- Review process The database searches produced 483 articles. These were assessed for eligibility on the basis of the title, where indicated abstract, and full-text. Only five articles identified through the PsycINFO searches met the inclusion criteria. An additional six articles were identified via a citation search for Beart et al. . A search of the reference lists of all 11 articles thus identified yielded a further five relevant studies. The process of selecting studies for inclusion in the review and reasons for excluding studies is presented in Figure 1. --- Results In total, 16 articles reporting on 16 separate studies were included in the present review. A summary of these studies is provided in Table 1. --- Overview of studies The 16 studies identified were carried out in six different countries: most in the UK , and one each in the USA, South Africa, Australia, Norway, and Taiwan. Twelve studies employed qualitative methods, which included semi-structured interviews , photo voice and conversational interviews , focus groups , and participant observations . Three studies employed quantitative methods, which involved self-report measures. One study employed a mixed methods design, using both interviews and questionnaires. The sample sizes of the sixteen studies ranged from 2 to 229 participants. The total number of participants from all of the studies was 784, which included adults , young people aged between 15 and 22 years , and children aged between 10 and 14 years . Participants were recruited from various different settings including community services, agencies, day centres, schools, special educational programmes, self-advocacy groups, community housing and hospitals. -Insert Table 1 here- --- Main findings The main findings of the studies are summarised below in relation to the three questions guiding this review. While we have attempted to answer the review questions separately, many of the studies answered these questions simultaneously. --- Are people with intellectual disability aware that they have an intellectual disability? Across the studies identified, most people with intellectual disabilities appeared aware of their diagnosis, or acknowledged that they were in some way 'different'. This awareness was sometimes the result of receiving a formal diagnosis from a professional, and/or receiving support from or attending intellectual disability or other specialist services . Awareness was often present from a young age but appeared to increase as individuals got older and learned to negotiate their intellectual disability label . The level of individuals' awareness of their disability differed across studies, with some suggesting a high proportion of individuals were aware that they had an intellectual disability . Other studies concluded that people with intellectual disabilities had little awareness of their intellectual disability, or were uncertain or confused . A few studies suggested people appeared to have no awareness, e.g., rejecting the label altogether as of no personal relevance . These different responses suggest that whilst many may be aware that they have an intellectual disability , they may consciously or unconsciously reject the label, or may struggle to understand the concept and its implications. This is demonstrated by a marked lack of articulation and clarity in studies where individuals were asked to describe what an intellectual disability is, what it means to have this label, and whether it applies to them . Further to this, despite most people being aware of their intellectual disability , the results from many of the studies suggest that it may not be considered an important part of their identity. Instead, they often focus on other attributes, roles, and competencies when describing themselves, and rarely refer to their intellectual disability in self-presentations . Finlay and Lyons found that when individuals with intellectual disabilities were asked to describe themselves in an interview, only four out of 36 participants used the label when referring to themselves. Even when specifically asked about their intellectual disability , only 13 interviewees applied this label to themselves, whereas 13 others rejected it. As such, the literature suggests that regardless of the level of awareness of carrying the label, many do not see their intellectual disability as an important or meaningful part of their identity. This has implications for social interactions with others whose central focus may well be the intellectual disability. --- How do they feel about the intellectual disability label ascribed to them? The majority of the studies suggest that people with intellectual disabilities feel negatively about the intellectual disability label ascribed to them. Feelings such as shame, embarrassment and dejection were commonly reported, with individuals showing obvious discomfort when discussing this topic . People associated the label with inability and demonstrated a reluctance to reveal their intellectual disability to others . The label was also associated with feelings of anger, powerlessness and frustration .and conflict between accepting and rejecting this label was observed . In line with this, the results suggest that individuals felt that the label had a negative effect on many aspects of their lives, as it made them feel isolated and different to other people, restricted them from having relationships with others, limited their independence and decision-making, and increased the likelihood of being a target for insults . As such, some felt that it was unhelpful to label themselves and other people with intellectual disabilities . These negative feelings about the intellectual disability label in turn increased feelings of unhappiness and discontent, which led some individuals to dissociate or distance themselves from the label. This was achieved through seeking to create a new identity for themselves , resisting any inference that they have an intellectual disability , rarely using the label in reference to themselves, or even rejecting it completely . However, it is also important to note that not all people with intellectual disabilities felt negatively about the label they had been given. For example, Norwich & Kelly found that whilst the majority of participants felt sad, hurt, upset or frustrated about having an intellectual disability, 23% expressed neutral feelings . Furthermore, the label was sometimes viewed in a positive light; it can allow individuals with intellectual disabilities to have access to support and resources that are contingent on this label . However, the focus of the studies identified in this review was rarely on positive perceptions or implications of the label. This is important as the stance of these studies by their very nature may be seen as focusing on intellectual disabilities as problematic. --- Are individuals with intellectual disabilities aware that others view this label negatively? The findings from the identified studies suggest that many people with intellectual disabilities are aware that others view this label negatively. This awareness is often the result of negative treatment and interactions they have experienced, for example, people laughing at them, subjecting them to insults, treating them differently, ignoring or rejecting them . Kenyon et al. found that many participants expected the 'mainstream world' to treat them poorly and unfairly due to previous experiences. This awareness of others' negative views appears to be present from a young age; children and young people aged 15-17 from both mainstream and special schools showed awareness of the stigma attached to their intellectual disability label within society . Additionally, those attending a mainstream school reported stigmatising treatment at school, as well as by society in general, particularly from their non-disabled peers. Interestingly, Ali et al. found age and severity of intellectual disabilities to be associated with self-reported stigma. Older participants and those with moderate intellectual disabilities reported more negative experiences and discrimination, such as being mocked or treated differently. Further evidence that many people with intellectual disabilities are aware that this label is viewed negatively comes from other sources, including aforementioned findings that many want to dissociate themselves from the label and create a new and separate identity for themselves . It is also evident in findings that many resist any inference that might lead them being labelled as having an intellectual disability , rarely identify with or use the label in reference to themselves , or reject it altogether . Research also suggests that people with intellectual disabilities often engage in downward social comparisons to distance themselves from others who have an intellectual disability and to present themselves in a better light . Taken together these findings suggest that they are aware of the negative connotations and stigma attached to the label. In addition to these results, evidence cited above that many people with intellectual disabilities feel ashamed and embarrassed about their ascribed label, are reluctant to reveal their identity to others, and often experience tension and/or conflict in accepting that they may need support due to their disability, adds to the idea that they understand that the intellectual disability is viewed negatively within society. --- Limitations of the studies One significant limitation of the studies included in this review concerns their sample sizes. Five of the 16 studies had less than 10 participants, limiting the generalisability of the findings. For example, Jahoda et al. examined data from two case studies. Whilst case INTELLECTUAL DISABILITIES AND IDENTITY: A REVIEW 13 studies can offer in-depth insights, caution must be exercised when interpreting their findings. However, some studies had much larger sample sizes . The use of opportunity sampling in the majority of the studies is another limitation. Many participants were recruited through day centres or the equivalent and self-advocacy groups, and as such, the results may not be generalisable beyond these particular samples. Findings from studies that involved members of self-advocacy groups , for example, may give an overly positive picture as such participants are more likely to be aware of and to feel more positively about their intellectual disability label, having been part of a group that advocates for greater equality. Caution should also be exercised in generalising the findings of Chen and Shu's study to cultural contexts that are dissimilar to Taiwan, the study's setting. Whilst many young people in their study were aware of their intellectual disability, the authors note that Taiwan has a strong emphasis on academic success, and as such, intellectual disability may be viewed more negatively than in cultures that place less pressure on academic success. The definition of intellectual disability used across studies was often inconsistent and at least some studies included people with 'borderline' intellectual disabilities . Furthermore, in some studies researchers inferred that participants believed themselves to be treated negatively because of their disability without explicitly asking this; for example, Ali et al. assessed responses to statements such as "people look at me funny…" without checking whether participants felt others responded to them negatively because of their disability, which compromises the interpretation of the results. It is also important to point out that 'double' stigma may have been present in some studies. Many participants may have also experienced stigma for additional reasons, for example, emotional/mental health problems , other disabilities or conditions , or perhaps due to racism . Thus, caution should be exercised when interpreting the results purely in terms of carrying an intellectual disability label. Furthermore, the lack of longitudinal studies in this review is a prominent limitation; the static results reflect awareness and feelings about individuals' intellectual disabilities and the label at one point in time, rather than capturing views across different time points and thus are limited in what they tell us about the process of identity development. --- Discussion The results from the 16 studies included in this review indicate that most people with intellectual disabilities appear aware of the intellectual label ascribed to them, or that they view themselves as in some way 'different'. Some clearly reject any such suggestion though. Of note, they commonly present their intellectual disability to researchers as of little relevance to their self-image and instead focus on other attributes, roles, or characteristics when describing themselves, suggesting that intellectual disability is not a central or dominant aspect of their identity. This finding is in line with identity process theory , which suggests that we all have a 'constellation of identities' and go through a process of assimilation/accommodation and evaluation in forming our identities. The theory also proposes that a threatened identity may result in the use of coping mechanisms to manage threat, in line with the finding that some individuals distance themselves from the label. Most studies in the current review concluded that people with intellectual disabilities feel negatively about their ascribed label, and experience shame, discomfort, but also anger, powerlessness and frustration when discussing the label. However, this was not the case for all. Beyond the reasons cited in the literature for neutral or positive views of the label, such as entitlement to use specialist services and access to scarce resources, giving positive meaning to the label is also central to the self-advocacy movement and collective action by people with intellectual disabilities . The integrated social identity model of stress postulates that social groupings can foster a sense of belonging in a stigmatised group, which can buffer against negative experiences such as prejudice and discrimination . Crabtree, Mandy and Mustard draw similar conclusions, suggesting this involves complex mechanisms for people with intellectual disabilities which operate through in-group comparisons with those who are less able, moderating the value placed on various attributes salient in providing an alternative selfevaluation and a shared group identity with which to reject stigma. Finally, most people with intellectual disabilities seem well aware that others view the intellectual disability label negatively, often as a direct result of negative social interactions they have experienced across their life course. This emphasises the important role the social sphere has in the construction of social identity for people with intellectual disabilities and provides evidence for developmental systems theory models which emphasise that individuals are both a product and a producer in the development of their social identity through the bidirectional relationship between the individual and the environment or context . What needs to be considered next is how this model might inform effective interventions to enable the construction of positive social identities. --- Limitations of the Review Whilst this review has contributed to gaining an insight into how people with intellectual disabilities view the label they have been given, the limitations must be carefully considered before drawing strong conclusions from the results. Firstly, only the PsycINFO database was searched due to the rapid review methodology adopted. Therefore relevant papers published in journals not included in PsycINFO may have been missed. Secondly, we are limited to results yielded from the search terms and criteria utilised; for example we included only research published in English. Thirdly, no formal appraisal of the quality of the research was conducted but the limitations of the studies have been outlined. Finally, we focus on how social identity is constructed relating to intellectual disability; thus, broader influences on social identity construction were not synthesised in the present review. --- Conclusions and Implications Whilst people with intellectual disabilities generally appear aware of the label ascribed to them, some distance themselves from the label or reject it altogether. Overall, having an intellectual disability mostly appears not to be a dominant feature of their identity, although it may be in others' eyes and therefore the most widely researched. Whilst many have described negative feelings associated with the label, ascribing the label to oneself can also have positive consequences, such as enabling access to support, resources, and liberation via self-advocacy. The findings of this review have implications for both research and practice. There is a need for more robust evidence to inform interventions aimed at attitude change and stigma reduction. Future research should explore the role of group identification, social support and peer support in making people with intellectual disabilities more resilient to the negative consequences of stigma. Studies employing longitudinal designs would allow an understanding of identity development that goes beyond a snapshot at one point in time and a richer consideration of the many and varied factors that contribute to identity formation in people with intellectual disabilities. In terms of practice, the early cultivation of advocacy, citizenship and rights approaches in educational programmes with young people with intellectual disabilities seems vital if young people are to be resilient in the face of stigma associated with their disabilities. Fostering empowerment, confidence and positive schemas during adolescence could help sow the seeds of not only a positive identity, but also higher aspirations. Clinically, an understanding of social identity formation may serve to improve and refine both the formulation of psychosocial difficulties and the selection of appropriate interventions, for example, narrative approaches may be highly appropriate in deconstructing dominant narratives and the power relations that underpin them and may be helpful in promoting a positive sense of self . Indications that people with intellectual disabilities can be buffered against the stress of a stigmatised identity through coming together as a group, emphasise the value of self-advocacy groups and suggest a need for more group-based interventions for people with intellectual disabilities. --- Source of Funding This review was completed as part of a wider project funded by the Baily Thomas Charitable Fund. -Over 90% aware of their intellectual disability -44% expressed negative feelings about their intellectual disability; 23% were 'not bothered'; 33% had mixed feelings -Special school pupils had more positive views of own educational abilities than mainstream school pupils -15% of pupils minimised or denied their intellectual disability: secondary school pupils more likely to than primary school pupils -Common labels -Some identified themselves as having a disability -Feelings of difference experienced, despite lack of articulation -Some aware of stigma and others' negative viewsthey showed discomfort with topic of disability -Feelings of unfairness, blame and rejection expressed --- Conflict of Interest No conflict of interest has been declared.

Background: A 2005 review by Beart, Hardy and Buchan, asking how people with intellectual disabilities view their social identities, has been widely cited, indicating this important topic needs an updated review. This review covers research on how people with intellectual disabilities view their ascribed label; to what extent they ascribe it to themselves; and whether they recognise it as devalued in society. Method: Rapid review methodology using PsycINFO, citation-and hand-searching identified relevant studies.The 16 studies identified indicate that the majority are aware of their ascribed label, or acknowledge they are 'different'. Others reject it, focusing on alternative attributes or roles. Most recognise others view the label negatively and express feelings of shame, anger and powerlessness.The review advances our understanding of social identity formation in people with intellectual disabilities, with implications for future research and practice to support construction of positive social identities and stigma resistance. Beart, S., Hardy, G., & Buchan, L. (2004). Changing selves: a grounded theory account of belonging to a self-advocacy group for people with intellectual disabilities.

Introduction In almost every social setting, when individuals have trouble performing a task that they need to accomplish, there are people available to assist them. As so many activities migrate to computer environments, we need to think of assistants. Help for online activities may be available as information and as communication -in asynchronous and synchronous form. It may also come from user agents, an alternative that takes much more computational effort to design and develop. This paper discusses WNH, the Web Navigation Helper [7,8], which as its name suggests is a user agent for helping users with special needs to navigate the Web. The metaphor underlying WNH is that of an interpreter -a mediator that is able to understand different languages and cultures, and can therefore translate material from one into the other in different kinds of situations. Specifically, WNH can translate web site interfaces into specialized interaction dialogs . The translation is not automatic; it is the result of manual interface adaptations by well-informed Web users, who can use the WNH dialog editor to specify how a given web site's interface can be translated into an appropriate interface for the elderly, for example. This interface, which works truly as a Web Navigation Helper, is then made available for elderly users that need help to do tasks on the Web. The size and diversity of the user population in need of help are huge, and accessibility solutions are practical only if they can scale up to benefit the largest share of the targeted population. Hence, manually produced help usually fails to make a difference in the accessibility scenario. However, there is a limit to automation when it comes to translating web site's interfaces into more accessible dialogs. If even human translators are known for making regretful context interpretation mistakes, what can we say of automated software and its ability to produce contextually adequate translations? A key element in this context is that the targeted beneficiaries of the translation are individuals with special needs. One way to circumvent the limitations of automated solution, especially with the advent of Web 2.0, is crowd sourcing. Now that a large number of individuals can be mobilized, there have already been attempts to use the power of social participation to promote more accessibility on the Web [11]. Thus, WNH was initially conceived as an accessibility tool to be developed counting on large-scale social volunteering. However, as we found out later, the Brazilian culture, for which WNH is primarily targeted, values certain practices and structures that go in a different direction. Therefore, we report empirical research that led us to this realization and discuss what this may mean for promoting accessibility in Brazil and similar family-oriented cultures. In Section 2, we present a brief overview of WNH. In section 3, we describe three empirical studies carried out to inform the design and evolution of WNH. Then in Section 4, we present our current conclusions and outline the future research work. --- A Brief Overview of WNH WNH is a Firefox plug-in and currently works in conjunction with CoScripter, a macro recorder for the Web developed by IBM Research Almaden [9]. CoScripter macros automate tasks enabled by a particular Web application whose interaction with users WNH can mediate. Macros must be created by scripters, typically nonprofessional Web users with high levels of digital literacy, who will then develop mediating dialogs for each macro step that requires user input. When a macro is created, all interaction with the browser is captured and recorded. Thus, input values that go into textboxes, options that are selected in list boxes and checkboxes, links and buttons that are pressed, all of these are captured for further use in the mediating dialogs. For example, a simple movie search macro recorded on Google Movie Showtimes for Rio de Janeiro might look like this in CoScripter: 1. * go to "http://www.google.com/movies" 2. * enter "Toy Story 3" into the first textbox 3. * click the first "Search Movies" button The instructions above correspond to recorded input events and values. Note that values and events recorded in lines 1 and 3 can be reused if users want information about movies in Rio de Janeiro. Likewise, we do not know how the actual interface looks like; all we know is that there is a textbox where users can fill in movie titles and a button that launches search. The scripters' job is to adjust the macro and create mediating dialogs that may look like in Fig. 1 . --- Fig. 1. How WNH mediates Web navigation In order to allow each user to search for a different movie, scripters use an existing CoScripter mechanism to create variables whose values can be filled in during the execution of a CoScript. In the next step, scripters use the WNH specialized editor to design small screens like the one shown in Fig. 1 and then to associate them with instructions where user input is necessary . In Fig. 2 we show WNH editing screen shots that briefly sketch the flow that culminates in the creation of a mediating dialog, for a registration page. First , the scripter writes the screen title and the dialog text. Second , he sees which instructions will have their input captured in this particular screen. Third , the scripter chooses which labels and other interface elements will appear on the dialog screen . Fourth , he indicates the data types for input. Finally , the scripter can preview the dialog screen. --- Fig. 2. Creation of a dialog in WNH editor Scripters can add help information to dialog screens. A special button labeled "Dúvidas" , as shown in Fig. 1, appears on each screen. When the targeted user presses that button during interaction with WNH, she can resolve doubts about what she must do, the meaning of specific terms, the effects of the task, the overall goal of the web site, and so on. Notice that these doubts must be anticipated by scripters, or else the end user will have no help. With WNH installed in her Firefox, she can select one from a set of available mediated activities. Once she does it, she begins to interact with WNH. The user's interaction with WNH guides background navigation controlled by the underlying macro. In this way, the user completes the desired task using an alternative style of interaction. For further details on how WNH works, see [7]. --- Empirical Studies Informing the Design of WNH Our first approach to designing WNH was inspired by crowd sourcing. We hoped to count on volunteers with high levels of digital literacy to create mediating dialogs for users with special needs. They might even do this anonymously, if they wished. We planned to give them dialog templates, designed by HCI specialists, to guide their interface mediation. But in order to know what other requirements we would have to meet, we ran two exploratory experiments with participants that viewed themselves as potential WNH volunteers. We used a qualitative methodology, as is often the case with exploratory studies. In the first experiment, we worked with seven IT professionals and in the second with eight experienced Internet users. The groups were intentionally selected with complementary skills. Whereas IT professionals had no particular training in written communication, selected users in the second experiment were well-trained Language students, taking a course in digital narratives and related topics . In the first experiment, participants were quickly debriefed about WNH and its purpose. Then they were asked to go through the experiment scenario. The main character in it wants to develop mediating dialog screens for an elderly lady who wants to visit a Bus Travel Company web site and look for travel options . The CoScript for the task was previously prepared by our team, so all the participants only had to design the dialogs, and make the appropriate associations between them and the CoScript instructions. Because the WNH editor was not fully implemented at that time, we used printed form sheets, which participants filled out with their solutions. At the end of the experiment, each participant had a short interview with us, about the experiment, the technology, and the participant's perceptions and suggestions regarding WNH. The results of this experiment gave us much useful information. This group had no difficulty to understand how CoScripter combined with WNH. As a rule, they could indicate, easily and correctly, when their dialogs should appear. They could also easily name and use interaction elements in their dialogs. Interestingly, one participant used a different interface element in his dialogs, clearly rephrasing what the original interface was saying. Another participant, at a certain point along the mediated dialog, directed the user to the company's web site itself in order to select a particular bus service code. This achieved a kind of transition from mediated to nonmediated interaction, which was nice to see. Only three participants completed all of their dialogs within the time limit for the task, and the others fell short of completing them all by only a few screens. All participants thought of questions and answers to go in the "Doubts" section. The number of Q&A's elaborated by participants ranged from 3 to 7. Five participants created Q&A's for all dialogs where user input was required. The style of communication in all dialogs included various instances of technical jargon like "procedure", "this script is optimized", "valid date on dd/mm/yyyy format", and the like. In some cases, the participants even spoke for the bus travel company, saying things like: "Unfortunately, it's possible that your city isn't served by our company."; "We apologize, but apparently this city isn't served by Viação Cometa"; "Contact Customer Service at…"; "Bear in mind that we cover the largest number of routes…". The most relevant results of this experiment with IT experts were that the styles of mediating dialogs varied considerably from one participant to the other, even though their professional profile was homogeneous. Participants expressed their difficulty to communicate appropriately with the targeted user. Clearly, participants were unsure about how much explanation was necessary. None of them really knew how to deal with the stereotyped user depicted in the test scenario. So, in spite of having initially said they would be glad to act as volunteers to develop WNH dialogs, at the end of the experiment they were faced with unanticipated difficulties and were no longer sure that they had the necessary knowledge to achieve the task. Three participants explicitly suggested that companies and commercial businesses operating in the Web should be interested in creating the dialogs themselves. One of them explained that companies might use this to publicize their concern with users with special needs. Before we ran the second experiment, we interviewed the instructor of an Internet course for the elderly offered by the Rio de Janeiro State administration, and attended some of the classes to observe how they went, especially as the elderly learners practiced their lessons in the computer. Our goal was to put together some guidelines that the next participants would be able to use while developing their mediating dialogs for a similar test scenario. The second experiment was run for the same purpose as the first: to investigate what kind of support WNH volunteers needed in order to design good mediating dialogs. Participants were non-experts in IT, but had advanced skills in written communication. Although the experiment scenario was similar to the one used with the first group , the experiment procedure was different. Instead of running individual sessions with each participant, we used a class situation, with the whole group. They all saw the same introductory presentation, telling them about WNH, CoScripter, and our goal with the technology. We emphasized the case of elderly users. Then each participant received sheets of paper with their task description, the paper forms for the dialogs, and a summary of guidelines and facts about computer interfaces for elderly users. This step was done individually, by each participant. At the end of the experiment, we brought all participants together for a focus group session to discuss the experiment. The group as a whole needed more time than the first group to achieve the task. Five of the eight participants understood how CoScripter and WNH worked together. The other three were confused when associating dialogs with CoScript instructions. Only two of the eight participants had no problems with naming interactive elements. The other six experienced a variety of breakdowns. For example, they talked about 'icons' when none were involved in the interaction, or referred to radio buttons as circles. Three participants gave us evidence that they did not really understand how CoScripter automated script execution. For example, they created a dialog for commands where no user data was necessary. Unlike the first group, this group generated very few Q&A's for the "Doubts" section. To our surprise, most participants did not consult the list of guidelines and information about elderly users' needs. Regarding the content and style of dialogs, as a rule, this group's dialogs were very informative. All the participants embedded useful explanations in a number of dialogs, providing detailed explanations about how to use interface controls, for example . The focus group session raised a number of relevant points. One participant said he believed that the less technical the information, the closer the elderly would feel to the technology, and the more comfortable with it. Some of the participants manifested having difficulty to explain the basics of interaction . One participant said that he was worried that the language and style he used would not sound childish to the elderly. One participant said that as she listened to the oral presentation he thought everything was easy and obvious, but when it came to actually writing down the mediating dialogs, it became difficult . Interestingly, one participant said that he was rather pessimistic about Brazilians engaging in this sort of social volunteering online. He did not see "people helping other people at large", he said. This statement started a lively discussion about when and if Brazilians were likely to help others, rather than 'be selfish' . There was a consensus in the group that everybody in this culture would be happy to help their grandmother and grandfather, or some other elderly person in their family. These were people they knew, and notas one participant put it -"the whole wide world". It thus became clear that the point of the discussion was the reach of one's initiative online, which extended far beyond the limited context that necessarily informed what one was doing when trying to help someone with a special need. In other words, helping their grandmother was as far as they were willing to go, because they felt they needed to know the person they were trying to help. Helping unknown people, which was the context of WNH we presented to them, was something about which a number of participants said they were rather pessimistic. However, all agreed that once they developed a script for somebody they knew well, there would probably be no problem whatsoever with sharing it with other people that they knew. Notice that this keeps the decision about sharing or not sharing scripts would be then made on a case-by-case basis, not escaping human control . The two experiments gave us much valuable input for the design of WNH. However, one finding had much deeper consequences than all others did. The focus group discussion in the second experiment taught us that we had tacitly accepted the fact that crowd sourcing was the alternative for having contextually well-adapted dialogs for users with special needs. The pressure for scalable accessibility solutions led us to aim at users at large. The reaction of one of the participants in the second experiment showed us that we were unconsciously expecting WNH scripters to work for stereotyped beneficiaries, running the risk of missing precisely those special contextual elements that distinguish this user population from other users that can deal with standard interfaces. We then reframed WNH and began to think of it more clearly as an end-user development tool to create software agents that can achieve what Clayton Lewis refers to as "human companionship and support" [10]. This new design conception opened a new way of computer-mediated social participation, although it did not impose to scripters a networked sociability model [3] as our previous design did [7,8]. We ran a third experiment to explore the new design concept. It was structured in three subsequent phases, which shared the same scenario setting and involved two different participant roles: WNH scripters and WNH end users. We recruited four participants to play the scripter role . For the end user role, we recruited six users that had just started learning how to use computers and the Internet. In the first phase, scripters were asked to produce verbal descriptions of interaction involved in the experiment scenario to two different listeners. In one case, they would be talking over the phone to a colleague . In the second, they would be talking face-to-face to an elderly family member or friend . The participant's job was to explain orally to the other person how to use an online mortgage calculator made available by a Brazilian public bank. In the second phase, scripters listened to a brief description of our project and were introduced to the WNH Editor. Then they were asked to use the editor to create mediating dialogs for the web site they had just described and explained. Their targeted user should be the elderly person they were addressing themselves to in the first phase. In the last phase, participants matching most of the 'digitally illiterate elderly person' profile were asked to use the mediating dialogs created by scripters. The group was divided in two: three participants used dialogs created by one of the HCI researchers and the other three used those created by one of the teachers. At the end of experiment sessions, we had a short interview with the participants. This experiment was used for triangulation with previous experiments in the sense that the tasks and activities that scripters had to do were very similar to the ones we had before. The design vision and context this time was, however, substantially different: we moved from online volunteering to family help, and from paper forms to the real WNH editor. Moreover, we did an internal triangulation in the third experiment, by having end users actually using the dialogs produced by the scripters. We collected considerable evidence that the new design perspective we adopted is more naturally understood by participants, who can easily engage in the proposed scenarios. During the interviews, all scripter participants, without exception, told some story about relatives that had problems with computers and asked them to help . All agreed that WNH could be effectively used to create mediating dialogs, especially to facilitate frequent tasks done by relatives with lower levels of digital literacy. They also said that they would like to start using WNH as soon as it became publicly available. Some participants explicitly said that WNH gave them an opportunity to help their friends and family in a much more practical way. The teachers, in particular, said that WNH would be useful at home, with their relatives, and with the elderly students in the Digital Inclusion program. The more experienced of the two teachers even said that if he had more time, he could do "air-tight dialogs", with which it would be almost impossible for the end user to make mistakes during the navigation. When asked who would benefit from WNH, most participants answered that all lay users and the elderly, in particular, would. Some of them said they also thought that blind and deaf users, as well as users with literacy deficits, could also be helped with WNH. The end user group of participants showed real interest in WNH. All of them said that they would like have WNH installed in their personal computers to help them do most frequent tasks. We observed that successful navigation with WNH had mostly to do with breaking interaction into a series of small dialogs . This, in itself, circumvents a number of problems having to do with 'what to do next' confusion. We also observed that once users knew how to use an interface control element , they did not care to read short instructions and tips on how to use them that scripters had included in their dialogs. They jumped right into interaction, which was a small but noticeable gain in efficiency. In some cases, when skipping the reading led to mistakes, the end users were warned by WNH that something had gone wrong. This situation gave us the opportunity to see the importance of anticipating errors and doubts when creating WNH dialogs. In all cases where these were available, end users eventually corrected the problem and went on to next steps. We asked to the end user group if they thought that, after using WNH dialogs repeatedly to do the same task, they would stop using mediation and go directly to the web site original interface. All of them agreed that the dialogs actually help them learn to navigate the web site directly. However, half of them said that they would prefer the easier way, which would probably be to continue using WNH. The other half said that, after some time practicing with WNH, they would surely want to migrate to the web site original interface. We welcomed this result since it tells us that WNH is perceived by some as a scaffold in the path of digital literacy acquisition. --- Conclusions and Future Work Designing tools for accessibility is more than a technological matter. The research described in this paper illustrates some of the benefits of qualitative methods in research, since they allow us to go deep into explicit and implicit meanings manifested by participants in the course of empirical studies. Likewise, we, as designers, in the process, are naturally led to question our own explicit and implicit meanings regarding the technology we are proposing to users. Results in this kind of research setting can be very surprising and entail radical changes. Our study gave us a window onto Brazilian sociability, which is considerably influenced by a different sociability model than the one embedded in accessibility technologies proposed elsewhere . In sensitive issues like dealing with users with special needs, Brazilian sociability is closer to a community model [5] than a networked model [3]. In communities, individuals experience strong personal relations and intimacy. An individual's identity is built and sustained by such relations and collective values. "Others" are not unknown, faceless individuals; they can always be chartered in a space of inter-personal relations, because of their ties with people one personally knows. The third experiment showed us how deeply true this is, since all participants told us stories of elderly users that were "digitally illiterate" and how family members or friends were there to help them in various kinds of situations. We believe that the crowd-sourcing model may not work for technologies like WNH in the Brazil . Accessibility issues are sometimes related to a "universe of one" [2]. The needs are so special, so specific, that it is difficult solve problems by automating generic solutions. Having someone who is not personally involved with the user generating solutions may also lead to missing many relevant aspects of the user's special context. The new design vision for WNH brings together both universe of one and community-oriented sociability perspectives. We believe that this new vision somehow introduces a new type of technology in our country. We are talking about family-oriented IT, as an additional alternative to assistive technologies. This technology is strongly related to end user development alternatives, as is the case with technologies produced in the context of the CLever project, at Colorado University [4]. WNH is thus an alternative technology. It can promote and reap the benefits of inter-personal relationships in computer-mediated environments and activities. Besides, end users know whom they must talk to if they have further needs or if scripts and dialogs stop running because of a change in the website that they refer to. This kind of maintenance task can be costly for larger social volunteering initiatives. Another lesson we learned was that the mediated interaction supported by WNH could be used for other purposes. For example, it can be useful for teaching and learning activities, or for expressing one's version of the interactive potential supported by web sites and services , or even for exploring design or redesign activities . We still have a long way to go. Our next steps involve reviewing and fixing some usability problems we discovered with the experiments. We would also like to do more experiments to evaluate the further uses of WNH. On a more theoretical note, and in tune with previous research of ours in Semiotic Engineering [6], we would like to explore how WNH can be used to study different metacommunication strategies and styles, towards a rhetoric of HCI. In particular, when running new experiments to observe the use of WNH in family contexts, we might have a glimpse of affective dimensions of metacommunication that we have never explored before. Paying closer attention to cultural issues silently lying beneath technologies opens different avenues for research. Not because this will necessarily lead to inventing new technologies, but perhaps because it can show us how and why to combine existing technologies and then deploy them in different completely cultural settings.

This paper presents the Web Navigation Helper (WNH), an interface agent for users with special needs originally developed for Brazilian users. WNH mediates scripted interaction with web sites, by providing alternative dialogs with appropriate style, structure, etc. The paper reports the results of qualitative empirical studies done at the early design stages. In particular, it shows how our design vision changed when findings from initial studies revealed that the technology we were about to develop was implicitly guided by a sociability model that was not prevalent in the Brazilian culture. The main contributions of the paper are to expose the process by which we became aware of cultural factors affecting the design of accessibility agents, and to propose a kind of technology that may be adopted in cultures whose sociability models are based on personal relations with friends and family members.

Background The rapid increase in the urban population due to urbanization in the last one decade has contributed to the urban poverty in India [47]. It is manifested in the form of inadequate provision of housing and shelter, water, sanitation, health, education, social security, livelihoods, and food security along with the special needs of vulnerable groups like women, and children [13]. The health of urban children as measured with malnutrition in the existing context of urbanization, has become an important issue in India [26]. Malnutrition among children under age five is a paramount public health concern in many developing countries, and India is no exception [33]. Child under-nutrition is expressed in different forms, including stunting , wasting , and underweight [43]. In developing countries, an estimated 32% of children under five were stunted, 3.5% were severely wasted, and 20.2% were underweight [4]. The rate of stunting in India has declined from 48% to 38.4%, wasting has increased slightly from 19.8% to 21%, and underweight has decreased from 42.5% to 35.8% between the two time-periods of NFHS-3 and NFHS-4 [17]. Malnutrition in India is clustered in central, northern, and eastern regions [46]. In a study conducted in Indian settings, it was found that the spatial clustering of malnutrition was visible in those geographical pockets where poverty was high, women's education was low, and BMI level among women was below average [22]. Malnutrition is primarily caused due to the immediate causes , and the underlying causes [9]. Studies in countries other than India had also raised the issue of child growth significantly. In a study, it was found that there was a substantial poor/rich gap in the nutritional status in the urban residence of Kenya [11]. Moreover, the same study argued that poor nutritional status in urban areas is attributed to poor environment and housing, poor access to drinking water, food, and health services. One of the studies found that the children belonging to poor households were undernourished not only because of poverty, but also due to limited use of maternal health care services and poor care resulting from the lower educational status of parents and poor health of mothers [5]. Despite high levels of stunting, wasting, and underweight among under-five children in India, studies related to measuring growth faltering as measured with z score are very limited [14]. Studies have found that growth in weight appeared to start right from birth, and there appeared a fall in WAZ score for the first 3 months, and after that, the rate of fall appeared to slow down [25]. The study was done to compare NFHS-1 and NFHS-3 data to provide a comparison for all India level growth among children aged 0-3 years [25]. Growth in infancy was associated with poorer educational outcomes in later childhood. However, these outcomes, when adjusted for confounding factors, suggested that childhood educational attainment was more strongly influenced by other socio-demographic factors, especially maternal education and the quality of the home environment [15,42]. Similarly, it was found that early childhood stunting is associated with deficits in cognition and educational achievement in late adolescence [42]. India ranks 102 in the Global Hunger Index [12], shows the severity of under-nutrition existence in the country. Moreover, other studies too carved out the fact that the poverty nutrition trap does exist in India [19]. There exist lack of literature focusing on examining the socio-economic gap for malnutrition among children in urban India. Therefore, this paper aims to measure undernutrition among urban poor and non-poor children below age five in India. It also analyses the changes in socio-economic inequality in undernutrition among children in urban India from 2005 to 06 to 2015-16. --- Methods --- Data The data used in this paper have been taken from two rounds of National Family Health Survey namely NFHS-3 and NFHS-4 . The primary objective of the NFHS was to provide essential data on health and family welfare, as well as data on emerging issues including the nutritional status of children under age five at national, state, and district levels. Therefore, data from NFHS-4 was considered as a benchmark to examine the progress made by the country in various health indicators over time. Decisions about the overall sample size required for National Family Health Survey were guided by several considerations, paramount among which was the need to produce indicators at the national levels, as well as separate estimates for urban and rural areas. The data is a nationally representative data which provide essential data on health and family welfare, as well as data on emerging issues in these areas. Additionally, National Family Health Survey is a part of Demographic and Health Surveys which are nationally-representative household surveys that provide data for a wide range of monitoring and impact evaluation indicators in the areas of population, health, and nutrition. The findings of this study are based on a total of 15,241 urban children under 5 years in NFHS-3 and 53,483 urban children under five years in NFHS-4. The information related to child's anthropometric measures covered measurements of height, weight and hemoglobin levels for children; Details about the NFHS-4 designs, tools, and protocols are presented in the national report of NFHS-4 [17], and all relevant information is available in the public domain on http://rchiips. org/ NFHS. --- Categorization of variables --- Outcome variable The three anthropometric indicators were used for children under five years: stunting ), underweight , and wasting . The measure used was recoded as per WHO standards [45]. --- Predictor variable The study divided the predictors into five groups. 1) Maternal Factors 2) Socio-economic factors 3) Child-related factors 4) Environmental factors and 5) Health seeking factors. Maternal factors include age , Education , age at first birth , Parity and mother's anaemia status . Socioeconomic factors include caste , religion and Wealth status . Child factors consist age , sex of the child , anemia status , diarrhea, birth weight , and colostrum feeding . The environmental factors include the source of drinking water , open defecation , stool disposal , and geographical regions . Health seeking factors included the use of Skill Birth Attendants, Place of delivery , full immunization includes children who received BCG , measles, and three doses each of DPT , and Polio. Full ANC is defined as women who had four or more ANC visits, had at least two tetanus injection, and consumed 100 IFA tablets/syrup for the most recent birth. Further, Colostrum feeding, defined as a child received breastfeeding within 1 h after delivery. The variable of wealth status was created using the information given in the survey. Households were given scores based on the number and kinds of consumer goods they own, ranging from a television to a bicycle or car, and housing characteristics such as source of drinking water, toilet facilities, and flooring materials. These scores are derived using principal component analysis. National wealth quintiles are compiled by assigning the household score to each usual household member, ranking each person in the household population by their score, and then dividing the distribution into five equal categories, each with 20% of the population. Further, wealth status of the household was recoded as poor and non-poor. To depict the poor and non-poor differentials, we clubbed first two wealth quintiles as poor and next three quintiles as non-poor . --- Statistical analysis Concentration index The concentration curve is obtained by plotting the cumulative proportion of outcome variables on y-axis against the increasing percentage of the population ranked by the socio-economic indicator on x-axis. The curves shows that whether the socio-economic status related inequality in the outcome variable prevails or not. If the curve is above the line of equality that means the index value is negative hence it shows that the outcome variable is disproportionally concentrated among the poor and vice-versa. Income-related inequality in stunting, underweight, and wasting was measured by the concentration index and the concentration curve , using the wealth score as the socio-economic indicator and binary outcome as stunting, underweight, and wasting. The concentration index is defined as twice the area between the concentration curve and the line of equality. Concentration index measures the inequality of one variable over the distribution of another variable . The index ranges from -1 to + 1, where the index value of 0 shows no socio-economic inequality. However, positive value of index shows pro-rich inequality and vice-versa. Additional on the either scales higher the value, higher the extent of socio-economic inequality. The study used Wagstaff decomposition analysis to decompose the concentration index. Wagstaff's decomposition demonstrated that the concentration index could be decomposed into the contributions of each factor to the income-related inequalities [41]. For any linear regression model on health outcome , such as. y ¼ α þ X k β k x k þ εð1Þ The concentration index for y, C, can be written as follows, C ¼ X k β k x k =μ À Á C k þ GC ε =μð2Þ Where μ is the mean of y, x k is the mean of x k , C k is the concentration index for x k , and GC ε is the generalized concentration index for the error term . Equation shows that C is equal to a weighted sum of the concentration indices of the k regressors, where the weight for x k is the elasticity of y with respect to x k ðη k ¼ β k x k μ Þ. The residual component captured by the last term reflects the socio-economic status related inequality in health that is not explained by systematic variation in the regressors by income, which should approach zero for a well-specified model. Each contribution is the product of elasticity with the degree of economic inequality. Moreover, the percentage contribution is obtained by dividing each absolute contribution by total absolute contribution multiplied by 100 to obtain the estimates. --- Results Table 1 shows the distribution of children under age five by selected background characteristics among urban poor and non-poor in India. It was found that significant health problems like anaemia and diarrhoea among under-five children were more prevalent among those from urban poor as compared to those from urban nonpoor. On contrary, health-seeking behaviour was better practised by urban residents as compared to non-poor population. Figure 1 portrays the prevalence of child growth as measured through stunting, wasting, and underweight among urban poor and non-poor in India, during 2005-06 and 2015-16. It is evident from the results that three key indicators of growth among children under age five, namely stunting, underweight, and wasting were higher among urban poor children than among urban non-poor children for both the time-periods. Further, results show that the prevalence of stunting and underweight has declined among urban poor as well as non-poor children during the last one decade, i.e., from 2005 to 06 to 2015-16. However, the prevalence of wasting has increased, from 2005 to 06 to 2015-16, among urban poor as well as urban non-poor children, which may be primarily due to seasonal variation in the month of surveys. By and large, a similar result has been found in case of growth for poor and non-poor children in urban India during 2005-06 and 2015-16. The result found that the average growth is higher among urban poor than in urban non-poor for each classification of child anthropometry measures . For standard error, please refer supplementary Table S1 andS2. Table 2 presents the average growth among children as measured through z scores for HAZ, WHZ, and WAZ by segregating them into different age groups. Mean WAZ, HAZ and WHZ scores for children under 5 months were , and in 2005-06; and , and in 2015-16 for urban poor and non-poor respectively. Significant differences in z score in poor as well as non-poor children were visible from the first 5 months itself. Of the total growth differential as measured with z score for height-for-age at 59 months among urban poor and non-poor in 2005-06 , 38% and 22% of growth took place at first 5 months after birth; whereas, in 2015-16, the distribution increased to 41% and 25% respectively. Similarly, of the average growth in weight-for-age at 59 months among urban poor and non-poor in 2005-06 , 71% and 68% of growth took place at first 5 months after birth; whereas, in 2015-16, the distribution was 41% and 25% respectively. For weight-for-height at 59 months among urban poor and non-poor in 2005-06 , 112% and 113% of growth took place at first 5 months after birth; whereas in 2015-16, the distribution was 99% and 131% . In explaining economic inequality for underweight in 2005-06, maternal education was the most significant contributor in explaining the majority of the inequality followed by open defecation , and place of delivery . In 2015-16, the contribution of maternal education slightly decreased to 25.4%, but remained as a significant contributor in explaining the inequality followed by open defecation , and parity . In 2015-16, nearly half of the economic disparities in underweight among urban children was explained by maternal factors , followed by environmental factors , and health seeking factors . Table 5 --- Discussion Despite recent achievements in economic progress, India failed to ensure better outcomes in child growth. Results further suggested that there exists socio-economic clustering that is further widening the gap in securing development among children. Through this study, we have tried to explain the factors contributing to economic inequality in stunting, underweight, and wasting among children below five years in urban India. There is little data to understand the timing of growth faltering in under-five age children in India, and whatever limited study conducted to understand the growth faltering among children measured growth faltering through reference growth standards of NCHS methodology [25]. This study has the edge over previous studies in measuring growth faltering as this study used WHO standards to measure the growth faltering among children over NCHS methodology [34]. The result confirms that wasting among urban poor and non-poor children was more prone during the first year of birth, and stunting among urban poor and non-poor children is more prone during the second year of birth. Growth flattering in underweight is more during the second year of birth among urban poor children and the fifth year of birth among urban non-poor children. Critically, children aged 21-25 months were more prone to growth faltering than the children of any other age group. Studies found that children in low-middle-income countries experience rapid growth faltering during the first two years of life [40]. Maternal factors remained a prime contributor to inequality. Stunting in simple terms is defined as -> 2 SD height-for-age [38], and measures the past or chronic child undernutrition. The mother's related factors explained a little more than half of the inequality for stunting among children during 2015-16. Various studies in Indian set up are in coordination with this study, concluded that maternal factors such as mother's age mother's education, age at first birth, parity, and anaemia were highly associated with stunting [28,36]. Of all the mother's factors, mother's education is the most significant contributor to the economic inequality in defining stunting among children. Not only in India [10], but in developed [44] and developing countries [2,20] also, mother's education was found significantly associated with stunting among children. Among environmental factors, defecation and stool disposal explained the highest economic inequality in stunting among children. A study conducted in Rwanda for 10 years, using decomposition analysis reported that defecation has no association with stunting [30]. Another study conducted in rural Zimbabwe employing cluster-randomized trials found that WASH practices implemented in rural areas are unlikely to reduce stunting [16]. However, few studies found that sanitation and defecation significantly affect stunting among children [24]. A study conducted in Indian settings utilizing NFHS data concluded that the practice of open defecation influences the inequality in stunting [35]. Health seeking factors explained about 30% of the inequality in stunting in 2005-06, and in 2015-16, it could only explain about 10% of the inequality. From various works of literature, it can be summed up that full Antenatal Care and Skill attendants at birth had increased over time in India [1]. Underweight is defined as -> 2 SD weight-for-age, and it depicts the past as well as present undernutrition. During both the time, mother's factors remained the highest contributory factors that explained the inequality in underweight among children aged 0-5 years. Employing the decomposition approach, a study concluded that maternal factors improve WAZ among children aged 0-5 years in India [27]. In other studies also, the role of mother's education in explaining the inequalities in underweight was highlighted in Indian settings [29,35]. Of the environmental factors, defecation remained the highest contributor to the inequality in underweight with nearly 17%, during both the time-periods. Open defecation is a problem among masses in India, and recently the Indian Government has done well to tackle the issue [6]. On the base of the decomposition approach, a study conducted in India confirmed that open defecation positively influences the inequality in underweight [35]. A study conducted in five Indian states; Uttar Pradesh, Karnataka, Madhya Pradesh, Rajasthan, and West Bengal, found that children from open defecation free districts were less underweight than children in non-ODF districts [37], thus confirming the importance of open defecation in understanding the The decline in the contribution of the place of delivery in explaining the inequality signifies the equitable access of institutional deliveries in urban areas among poor and non-poor. A study found that in India, the utilization of institutional delivery care has increased from 43% in 2004 to 83% in 2014 [21]. This increase in institutional delivery had helped in shrinking the gap of inequality in underweight, as explained by place of delivery. Wasting is defined as -> 2 SD weight-for-height among children. It is a measure of current nutrition status, also known as acute undernutrition. The inequality in wasting among children has decreased, but the prevalence of wasting has increased. Around the world, wasting had reduced at a plodding pace over the last 40 years, but in countries like India and Sri Lanka, wasting has increased [39]. Maternal factors were the highest contributory factors in explaining the inequality in wasting among children during 2005-06, but the contribution of maternal factors in inequality in wasting had declined significantly to 22.2% in 2015-16. In previous studies, maternal factors were deemed as the contributory factors in explaining the inequalities in wasting among children aged below five years [23]. Nearly half of the inequality in wasting among children had been explained by open defecation in 2015-16, citing the importance of sanitation. Since the weight-for-height is a measure of current nutritional status and is likely to be affected by small illnesses episodes and infectious diseases [23], the contribution of defecation in inequality in wasting among children is highly visible. Unavailability of toilet facilities within households promotes open defecation, and in turn, it facilitates water-borne diseases, which negatively affect the current health and nutrition status among children. Open defecation reinforces undernutrition via increased susceptibility to infections and through reducing immunity [31]. --- Strengths and limitations One of the strengths of the study is the considerable sample size for both the survey periods. This study explores the inequalities in stunting, underweight, and stunting among urban poor by employing survey data, which is 10 years apart, thus giving a clear picture of decadal change in economic inequality in stunting, underweight, and wasting among urban children. Decomposition allows our understanding of various factors --- Conclusions The study made a reasonable attempt to describe the growth pattern among children below five along with the inequality in stunting, underweight, and wasting among children in urban India. Rich-poor inequality in malnutrition calls out for the expedite policy interventions aimed at poverty reduction in urban areas. There is also a need to provide mass education regarding nutrition and health, along with focussing on the correlates that aim at improving mother's education. Environmental factors also promoted the inequality in stunting, wasting, and underweight in urban children; there is a need to focus the enigma of open defecation in India. After the launch of Swachh Bharat Mission in 2014, the government of India has focused a lot on building individual household latrines [6]. Despite having access to latrines, people prefer to defecate in open [3,8,18,32]. There is a need to promote the use of latrines among Indian masses. There is a need to bring a change in behavioural aspects and preferences of people to promote latrine use. Information, Education, and Communication activities are encouraged to clamp down the economic inequality in stunting, Table 5 Estimates of decomposition analysis for contribution of various explanatory variables for wasting among urban children under five years in India, 2005-06 & 2015-16 underweight, and wasting among urban children. Since the cognitive and developmental deficiency resulting from malnutrition may significantly be irreversible after two years of age [7], it is being understood that the 'window of opportunity' for preventing malnutrition ends at age two years. It is therefore encouraged that policy-makers shall evenly target children for nutritional programs, with emphasis on children aged below two years. All the nutrition-specific and nutrition-sensitive interventions in urban areas should be prioritized, focusing on urban poor, who are often clustered in low-income slums. Rich-poor inequality in child growth faltering calls out for integration and convergence of nutrition interventions with policy intervention aimed at poverty reduction in urban areas. There is also a need to expand the scope of the ICDS program to provide mass education regarding nutrition and health by making provisions of home visits primarily focusing at pregnant and lactating mothers. --- --- Socio-economic factors --- Caste --- --- --- --- Competing interests The authors declare that they have no competing interests. ---

Background: With increasing urbanization in India, child growth among urban poor has emerged as a paramount public health concern amidst the continuously growing slum population and deteriorating quality of life. This study analyses child undernutrition among urban poor and non-poor and decomposes the contribution of various factors influencing socio-economic inequality. This paper uses data from two recent rounds of National Family Health Survey (NFHS-3&4) conducted during 2005-06 and 2015-16. Methods: The concentration index (CI) and the concentration curve (CC) measure socio-economic inequality in child growth in terms of stunting, wasting, and underweight. Wagstaff decomposition further analyses key contributors in CI by segregating significant covariates into five groups-mother's factor, health-seeking factors, environmental factors, child factors, and socio-economic factors. Results: The prevalence of child undernutrition was more pronounced among children from poor socio-economic strata. The concentration index decreased for stunting (-0.186 to -0.156), underweight (-0.213 to -0.162) and wasting (-0.116 to -0.045) from 2005 to 06 to 2015-16 respectively. The steepness in growth was more among urban poor than among urban non-poor in every age interval. Maternal education contributed about 19%, 29%, and 33% to the inequality in stunting, underweight and wasting, respectively during 2005-06. During 2005-06 as well as 2015-16, maternal factors (specifically mother's education) were the highest contributory factors in explaining rich-poor inequality in stunting as well as underweight. More than 85% of the economic inequality in stunting, underweight, and wasting among urban children were explained by maternal factors, environmental factors, and health-seeking factors. Conclusion: All the nutrition-specific and nutrition-sensitive interventions in urban areas should be prioritized, focusing on urban poor, who are often clustered in low-income slums. Rich-poor inequality in child growth calls out for integration and convergence of nutrition interventions with policy interventions aimed at poverty reduction. There is also a need to expand the scope of the Integrated Child Development Services (ICDS) program to provide mass education regarding nutrition and health by making provisions of home visits of workers primarily focusing on pregnant and lactating mothers.

Introduction Previously, an extensive amount of research have been conducted on, e.g. lease structure, rent determinants, and the effect of location, size, and anchor or non-anchor tenants as factors that affect rent levels . Despite previous research efforts, it is not well-known how landlords and retail tenants' beliefs differ in the negotiation phase of retail rent. Our study will add to the knowledge of the real estate research community and to practitioners, by investigating how retailers' and real estate landlords' views differ in Sweden during the coronavirus disease 2019 pandemic on factors they deem important in negotiating rent by investigating differences in their beliefs. At this stage, our results do not demonstrate causality. Causality as given by a formal structural equation model would be beyond the scope of this explorative research study. To inform us about critical variables in a rent negotiation, we have conducted a qualitative study using semi-structured interviews. Seven managers being responsible for negotiating retail rent as landlords and tenants participated in the interviews. In addition, interviews were also made with an advisory board, consisting of eight professionals with experience of rent negotiation. A web-based survey was then sent to respondents negotiating retail rent in Sweden to extract information on the variables that were found to be important in deciding retail rent in late spring 2021. We used principal components to discover which variables in the set that form independent factors. Principal component analysis allows us to reduce the number of variables seemingly uncorrelated to a fewer number of principal components explaining the variance in our dataset. PCA accomplishes this reduction by identifying directions where the variation in the data is maximal. We use PCA with communalities and varimax rotation, which allows us to represent and interpret complex data by a relatively small number of factors. The purpose of using PCA and statistical tests is to reveal differences that may exist between landlords' and retail tenants' beliefs about the process of negotiating retail rent. We hypothesize that landlords' and retail tenants' views on variables used to determine rent will differ, depending on past experiences and expectations of future changes in the economy. The survey was sent to 421 respondents; 156 respondents answered the survey, resulting in 106 fully complete answers. Responses were measured on a seven-point Likert scale. Results from the PCA and statistical tests show that real estate landlords are prone to believe that factors such as regional growth are more important while negotiating rent than do retail tenants. Moreover, we also find differences in beliefs on the importance of e-commerce, perceived trust and customer focus; also, our sample reveals differences in their views that depend on education. Thus, a landlord determines these factors as more important in the negotiation phase than a retail tenant. Our findings give new insights into the complex process of negotiating retail rent. Seven underlying dimensions in the rent negotiation process were labeled: regional growth, rent and vacancies, GDP growth, e-commerce, customer focus, external information and trust. The study is organized as follows. Section 2 provides a literature review of relevant research on retail rent. In Section 3, we discuss our methodology and how the data were generated. Section 4 provides results, and, finally, in Section 5, we conclude our results. --- Lease negotiations 2. Literature review Resent research by Guven et al. discusses the effect of competition and concentration on retail rent. Yang et al. use information asymmetry theory to explain rent and vacancy adjustments. Nanda et al. discuss the potential impact of e-commerce and digitalization post COVID-19 pandemic. Pretorius and Cloete discuss the complexity of online sales on a turnover-based retail rent in South Africa. However, research on retail rent during the negotiation process conducted using factor analysis from a rent negotiation perspective between landlords and tenants has, to our knowledge, not been published earlier. Thus, there exists a research gap in factors that affect landlords' and tenants' different views during rent negotiation. We hypothesize that landlords' and retail tenants' views on variables used to determine rent will differ, depending on past experiences and expectations of future changes in the economy. The following literature review will present previous research on rent determinations. From an agency perspective, Ross concluded that the optimal risk-sharing contract for a risk-averse tenant and a risk-neutral landlord is a pure turnover rent. However, in the presence of moral hazard, an optimal incentive contract is, instead, one including a fixed rent as well as a turnover rent. The level of turnover rent in relation to base rent has been thoroughly studied; it has been found that it is highly related to externalities, and lease structure can be considered helpful in reducing the conflict in interest between risk-averse tenants and risk-neutral landlords . Furthermore, according to agency theory, anchors and non-anchors affect sales, and lease contracts should be constructed to stimulate desired actions from all relevant parties. Thereby, contracts are written certain ways to internalize externalities, using both base and turnover rent, resulting in optimal allocation of space in shopping centers . Sirmans and Guidry pointed out four main determinants affecting rent negotiation and market rent: customer-drawing power, design/configuration of the property, locational characteristics and market conditions. Customer-drawing power is, in turn, substantiated by the size and is positively correlated to the level of the rent. Furthermore, customer-drawing power and rent levels are affected by age, implying that the age is negatively correlated to the market rent. A third factor affecting the customer-drawing power is the type of anchor tenants. Nationally well-known retailers are known as traffic attractors, having a positive effect on the drawing power and, hence, increasing the market rent. The design/ configuration, according to Sirmans and Guidry , has an impact on rental levels. By design or configuration, the authors do not mean design in architectural terms, but rather how the center is built and how the tenants are positioned. It is, according to the authors, important to focus on having anchor tenants in places with high visibility and, in that way, attracting traffic to the anchor tenants as well as to other tenants. An optimized disposition of the tenants has a positive effect on the rental level. When investigating what effect the locational attributes and market condition have on the rental level, some main subfactors, such as population density, growth potential and income in the area, are argued to be the main locational attributes that have a positive correlation with the rental level. The amount of vacant space and general economic conditions have a substantial effect on the outcome of rent negotiation. Research has shown that rent levels differ greatly depending on the type of tenant. Gould et al. carried out quantitative research to define the difference in rental levels between anchor tenants and non-anchor tenants. The first hypothesis was that anchor tenants received discounts on rents due to their ability to draw customer traffic, not only to their own stores but also to the other stores. Their second hypothesis was that non-anchor tenants paid rents at a premium. The regression model was based on approximately 2,500 contracts in the United States, with anchor tenants representing 58% of the leasable area, but accounting for only 10% of the rental revenue. --- JERER 16,2 The fact that anchor tenants, or customer-drawing tenants, generally enjoy what could be viewed as rent subsidies is often referred to as rent discrimination. Despite its discriminating nature toward smaller tenants, rent discrimination is widely recognized and considered fully rational from a property owners' perspective . Furthermore, Gatzlaff et al. have argued for the importance of anchor tenants. Their results showed that the loss of an anchor tenant lowered the customer-drawing power, increased the overall vacant space and, hence, decreased the rental level. The store location factor and its effect on retail sales and, in extension, retail rent have been widely studied using multiple approaches. Carter and Vandell applied bid rent theory, indicating that there is a relationship between rent and distance to the center of the shopping mall. At a district or regional level, the trade-off between location and retail agglomeration has been studied, with results showing that clustering is of greater importance for explaining consumer patronage than location . In a context of ground-floor commercial activities, the importance of centrality for retail concentration has been established, where centrality is defined as a function of closeness, betweenness and straightness . These determinants are still relevant and judged to have an impact on rental levels in the negotiation process. E-commerce and its effect on retail leases have been, at least partly, studied. As retailers shift focus from on-site to off-site sales, the role of the physical store is going through a great change, and, as part of that change, revenue is being transferred from physical stores to online stores. This transformation is also subject to changes in the parameters going into a rent negotiation. As the purpose and business models of physical retail change, lease agreements need to be altered, as turnover rent is a lease structure better fit for the past than the future . Gyllenberg and Koppfeldt focused on the negotiating gap between landlords and tenants in the retail sector and found that the gap between the two actors was significant, not least since the growth in e-commerce is complicating rent negotiations. --- Our objective The literature review has shown that previous research on rent and rent negotiations has focused on lease structure, rent determinants, the effect of location, size and anchor or nonanchor tenants, and the effects of e-commerce. Our objective is to increase knowledge of rent negotiations by investigating differences in beliefs held by property landlords and retailers on factors that they deem important in rent negotiation. Our first research question is thus, how do landlords' and tenants' beliefs on factors deemed to be important differ when they are negotiating rent? According to Huffman , the rate of adjustment in decision-making is positively correlated to the level of education. Decision-makers with higher education more quickly grasp changes and more quickly and accurately make adjustments. Brooks and Williams found that decisions of respondents facing an action to sell or keep their investments in a downturn in the economy are affected by the highest level of education of the respondents. Respondents having a higher level of education were positively correlated with keeping their investments in the downturn. Differences exist in susceptibility to behavioral biases due to education, with those having a higher level of education being more confident when making investment decisions . Positive relationships between an individual's educational level and their degree of risk tolerance have been observed . We will further investigate if differences in scores on the seven factors may depend on the respondents' educational levels. Our second research question is, do scores on the seven factors solution vary depending on educational levels? --- Lease negotiations 3. Method and data 3.1 Factor analysis Factor analysis using principal components, or PCA, is explorative in nature and is commonly used for investigating whether an underlying pattern exists in a set of data Berry . Factors are thought to reflect underlying processes that have created the correlations among the variables. Principal components, or eigenvalues, can be plotted, making it possible to determine whether samples can be grouped into a number of factors . However, factor analysis is not without problems. One of the drawbacks is the inability to test the solution as in regression and logistic analysis; another problem is the infinite number of available rotations. PCA is an effective tool in data reduction in the early stages of research to investigate relationships among variables . --- Interviews The aim of the seven semi-structured interviews was to gain the landlords' and tenants' point of view on determinants of rent for retail premises as an input to our quantitative survey . All of the interviewees are active in the leasing of retail premises in one way or another. The respondents were selected due to their expertise across multiple cities and regions in Sweden, as well as in managing lease agreements . The interviewees had a close focus on lease structure and how to construct an optimal contract from their different perspectives as landlords and tenants. In addition, interviews were also made with our advisory board, consisting of eight professionals with long experience of rent negotiation. The interviews showed that, although the rent determinants covered in previous research are still relevant in determining rent, landlords and tenants indicated that other factors might have considerable impact as well. Both parties could see, first, that the future calls for shorter lease terms and more The rotated factor structure and potential factors JERER 16,2 flexibility than has been appropriate historically. Second, expectations about the future have a large impact on the outcome of rent negotiations, and the existing discrepancy between retailers and real estate owners is mostly dependent on a gap in their expectations. It became obvious that real estate owners use the conventional rent determinants to determine rent levels. Simultaneously, retailers did not put much effort into recognizing these factors. Retailers expressed that the profitability of a store is the main factor in determining whether the rent level is reasonable for a particular store in a rent negotiation. Real estate owners and retailers agree that real estate values are limiting real estate owners' willingness and possibilities to reduce rents. Similarly, retailers are limited in their willingness and possibilities to pay rents above certain levels of the occupancy cost ratio . Landlords' and tenants' different expectations in rental negotiation widen the gap between retailers and real estate owners. In order to narrow this gap long term, there is an obvious need for greater transparency and joint efforts in predicting future developments in factors that have an impact on rent negotiation. To further widen our perspective on the negotiation process, an advisory board was added, consisting of eight economists and consultants with experience in rent negotiations. The board advised on variables that affect rent decisions in practice and gave their points of view on rent negotiation. The board also participated in validating survey questions, as well as in pre-testing the web-based survey before sending it out to respondents. --- The survey Our web-based survey consisted of 37 questions, of which 22 questions were measured on an ordinal seven-point Likert scale ranging from completely disagree to completely agree. One example is the following question: The negotiated rent is affected by how the regional economy has developed during the past year. We also included questions about socioeconomic factors, such as gender, age, the level of education and the type of occupation as landlord or retailer. Questions were asked about the respondents' experience in negotiating rent in years and the typical size of the citylarge/ medium or smallwhere the respondent negotiated rent to capture specific details of the respondent. The importance of anchor tenants and current rent levels as potential anchor values was included in the survey. Questions about the importance of gross domestic product growth and regional growth, the effect of e-commerce on rent levels and various sources of information used during the negotiation phase were also included. Furthermore, we asked whether market value and vacancies have an effect when rent is determined. Trust in the negotiating part and trust in legal institutions were also included in the questionnaire. The effect of COVID-19 on the turnover was included, as well as whether subsidies were received, in order to capture information on the effects of the ongoing pandemic on business. For summary statistics and questions used in the web-based survey, see Table 1. --- Data Contact information about the respondents were received from the following representative organizations for landlords and retailers in Sweden: HUI Research , City i Samverkan , F€ oreningen FFE , Public Housing Sweden and Unionen . Data to our study were collected from a data register managed and updated by these organizations. The respondents were selected from five different regions in Sweden. respondents. The western region, including the city of Gothenburg, contributed 17.2%. The remaining 8.0% of the respondents came from the northern part of Sweden. A total of 55.8% of the respondents categorized themselves as mostly negotiating rent in a large city; 38.8% negotiated rent in a mid-sized Swedish town, and 5.4% in a small town. When asking about the location of a typical store where the respondent is negotiating rent, 40.5% state that a typical store is located in a local street environment. A total of 44.5% of the respondents mostly negotiate rent in a retail center. Approximately 15.0% of the respondents were typically negotiating rent in an external trade area. The web-based questionnaire was sent to respondents during the late spring of 2021. A reminder was sent to those who did not answer after two weeks, and a second reminder after three weeks. In total, we reached 421 respondents who had experience in rent negotiation. In total, 156 respondents answered the questionnaire; however, only 106 were complete, leading to a response rate of 25.2%. Since our sample is limited in size, our results should be viewed only as a starting point in exploring differences in beliefs between landlords and tenants. The survey was answered by 36 females and 80 males. The mean age was 48 years . Of the respondents, 56% had between 7 and 25 years of experience in negotiating rent. Also, 64% classified themselves as tenants and 36% as real estate landlords. Education has four categories for highest finalized education . Respondents identified as landlords have a higher educational level than do those who classified themselves as tenants . An independent t-test showed that the difference between educational levels was statistically significant . We will further investigate differences in the importance of education on different factors, depending on educational level, using factor analysis and independent t-tests of mean differences. See Table 2 for an overview of differences in education levels. --- Results A factor analysis was made using SPSS Statistics 25.0. The cumulative variation by seven factors was 67.7% . The rotated factor solution and potential factors are found in Figure 1 and Table A2. The criteria for determining factors are an eigenvalue equal or higher than one. The factors can be thought of as representing regional growth, rent and vacancies, GDP growth, e-commerce, customer focus, external information and trust . An independent t-test and analysis of varience are used for statistical analysis. PCA with varimax rotation revealed a potential factor solution with eigenvalue . --- Lease negotiations During the interpretation of the rotated factor solution, an additional factor was added for conceptual reasons . Three variables from factor 1 were thus entered into a seventh factor named customer focus. The Kaiser-Mayer-Olkin test and Bartlett test of the suitability of the correlation matrix for factor analysis showed that the data set was factorable . Each factor was analyzed for reliability using Cronbach's alpha, and the internal consistency of all of the factors is above recommended levels, except the factor for external information . The following reliability values apply to the seven-factor solution: factor , regional growth 5 0.87; factor , rent and vacancies 5 0.69; factor , GDP growth 5 0.82; factor , e-commerce 5 0.93; factor , customer focus 5 0.76; factor , external information 5 0.51 and factor , trust 5 0.79. Factor , external information, has questionable reliability since it is below 0.7 . The correlation between the factors is found in Table A4. Correlation analysis shows that the factors have low inter factor correlation , which means that there are no multicollinearity problems between the factors. --- Analysis The literature review has presented factors that affect retail rent levels, such as lease structure, rent determinants and the effect of location, size, and anchor or non-anchor tenants. Our first research question concerned how landlords' and tenants' beliefs differ in factors deemed to be important when negotiating rent. We analyze differences in means using independent t-tests. Our results show that property landlords and tenants put different weights on the factors found in our study. Real estate landlords agree, to a greater extent, based on a seven-point Likert scale, that factors relating to regional growth, e-commerce, customer focus and trust affect the negotiated rent to a higher degree than do tenants. This means that landlords and tenants don't disagree on factors such as rent and vacancies, GDP growth and external information. Landlords and tenants put the same weight on the importance of these latter factors when negotiating rent. Statistical results in mean values and t-tests between landlords and tenants are found in Tables A5-A6. Our statistical results show that landlords have a significantly higher mean value on the factor regional growth than do tenants . Four questions in the survey capture how the regional economy developed over the past years and how it will develop in the current and next years, and how the industry developed over the past years and how it will develop in the current and next year. Since their mean values differ significantly, landlords are likely to value the factor regional growth higher than do tenants during a negotiation of rent. We find a significant difference in mean values also for the factor e-commerce . The questions concerning e-commerce in the questionnaire are framed to determine whether the negotiated rent is affected by the development of e-commerce during the past year, as well as for the current and next year. Differences in mean values show that landlords believe that this is the case to a greater extent than tenants do. There are also significant differences for the factor customer focus . The questions in the questionnaire are framed to determine whether the negotiated rent is affected by changes in consumer behavior and proximity to anchor tenants, as well as by the tenant mix in the past year, and for the current and next year. The differences indicate that landlords are likely to value the factor customer focus higher than tenants do during a negotiation of rent. The perceived trust between landlords and tenants turned out to be significantly different . The three questions in the questionnaire are framed to determine whether trust is given in confidence between the parties. Differences in mean values in trust between landlords and tenants show that landlords on average have greater trust in their counterpart during a negotiation than tenants do. Our second research question concerns whether differences exist in educational levels on factors that are deemed as important to landlord and tenants. From previous research on decision-making, we expect to find differences in beliefs depending on educational difference, see Table 4. We do find significant educational differences on the following factors: rent and --- Lease negotiations vacancies, e-commerce and customer focus. Testing for significant differences on the factor level, we find that those respondents with a five-year or higher university degree have a significantly higher mean value on the factor rent and vacancies than do those respondents having a gymnasium degree . We see the same result in the factor e-commerce: respondents having a five-year or higher university degree score higher than those respondents having a gymnasium degree . For the factor customer focus, we find a significant difference between a three-to four-year university degree and a gymnasium degree . Regarding respondents having a three-to four-year university degree significant differences exist in the factors e-commerce and customer focus . Depending on if the highest educational degree is on the gymnasium level or at a five-year or higher university degree, there are significant differences in beliefs on the importance of these three factors. We also find significant differences between respondents having a three-to four-year university degree and those having a university degree of five years or higher on the factors rent and vacancies and external information . Does higher education lead to different beliefs on factors deemed important in rent negotiation? As shown above, we discovered significant differences on the factors rent and vacancies, e-commerce and customer focus. In order to explore further differences in educational levels, we have added a table to show how landlords' and retailers' mean values differ on the seven factors, depending on their role and educational level. We specifically note that landlords and tenants with a 3-4 years' university education rate regional growth higher than retailers do . The factors rent and vacancies are rated higher by tenants having a 5-year or higher university degree than landlords do . We also find that landlords with a gymnasium education rate the factor e-commerce higher . The factor customer focus, which includes questions on the importance of anchor tenants and tenant mix, is rated higher by landlords having a gymnasium and 3-4 years university degree than retailers do . Landlords having a 5-year or higher university degree rate the factor trust much higher than landlords . We control for the respondents' number of years of experience in six different categories: 1-2 years, 3-6 years, 7-15 years, 16-25 years and >25 years of experience. We find a significant difference only in the factor GDP growth for respondents having 3-6 years' experience, compared with those with >25 years of experience , as well as for those with 7-15 years of experience versus those with >25 years of experience . No other significant differences based on experience are found. --- Conclusions Previous research on factors that affect rent levels have focused on the importance of e.g. lease structure, rent determinants, and the effect of location, size, and anchor or non-anchor tenants. Even though our sample is small, new insights have been found regarding the importance of industry and regional growth, e-commerce and customer focus, as well as trust between landlords and tenants in the negotiating process. Landlords were found to put more weight on the importance of these factors than do tenants. Such differences are likely to have an effect on their standpoints during the negotiation process. Our results were found in the upturn of the Swedish economy from late spring into the fall of 2021, during the COVID-19 epidemic. During this phase, landlords were likely to put more weight on the importance of industry and regional growth than did tenants, thus explaining their more positive score on the seven-point Likert scale. Landlords place greater importance on the actual and expected turnover from e-commerce during negotiations than do tenants. This might be because JERER 16,2 tenants have a more informed view of the actual turnover from e-commerce than do landlords. This information gap may disturb the negotiations of a fair rent because of lack of a mutual understanding of revenue from e-commerce. We also find significant differences in trust between the parties, where landlords have a higher trust for the counterpart. This difference was not expected, and this gap in trust indicates that trust is a factor that should be in focus to smooth the negotiation process. Previous research on the effect of educational levels on decision-making show that decision-makers with higher education more quickly grasp changes and adjust more quickly than those with lower education. Moreover, financial actions were found to be positively correlated with individuals who were more confident when making investment decisions, as well as having a higher level of risk tolerance. We also found differences in the importance of the categories of rent and vacancies, e-commerce and customer focus, depending on formal education; this led to significant differences between respondents having a gymnasium degree and those with a five-year or higher university degree. Such differences exist also between respondents having a three-four-year university degree and those having a five-year or higher university degree. This means that education levels matter, as they give different perspectives on various factors. For example, respondents having a five-year university education hold a different view on regional growth and rent and vacancies than do respondents having a gymnasium degree. In order to explore further differences in educational levels, we added a new table to show how landlords' and retailers' mean values differ on the seven factors, depending on their role and educational level. Meanwhile, controlling for the number of years of experience for respondents, only the factor GDP growth shows a significant difference in two out of the six age groups. Our results show that not only have objective factors, such as lease structure, the effect of location, size, and anchor or non-anchor tenants, an effect on rent levels, but subjective factors, such as the importance of regional growth, also play a role in rent decisions. However, using PCA, we cannot discover the causes and effects of these factors during an actual negotiation, since that would require an additional experiment that goes far beyond the scope of this paper. From a practical point of view, our findings provide new insights into the different views that exist between landlords and retail tenants in a rent negotiation. Trade organizations, landlords and tenants can increase transparency by putting forward information on these factors before a negotiation takes place in order to smooth differences in their beliefs based on educational differences. To use our findings, we suggest that retailers focus on those factors where landlords have a significant higher mean value to increase the possibility of reaching through with their arguments during a rental negotiation. Following this advice, we do not mean that one part is better informed, or that more information is the key to success; instead, information should be provided that is perceived to be important. Thus, it is more likely that landlords will listen to arguments that they themselves believe are important. From our results, landlords have a significantly higher mean value on, e.g. the factor regional growth than tenants do . The Swedish economy from late spring into the fall of 2021, during the COVID-19 epidemic, was in an upturn in the business cycle. During this phase, landlords would have been likely to put more weight on the importance of regional growth than would tenants, to motivate increasing rents due to an expected rise in sales. Retail tenants should, in their turn, put forward information on how regional growth is likely to have less impact on their sales. Exactly how negotiations would play out using our findings have not been in focus in our research at this stage. A10. --- Lease negotiations --- Appendix Landlord: Citycon is a listed real estate company traded on the Nasdaq Helsinki stock exchange. The company owns 40 shopping centers in 5 countries in the Nordics: Finland, Estonia, Sweden, Norway and Denmark Landlord: Vasakronan is the largest real estate owner in Sweden, with 174 properties, holding office and retail premises in the four biggest cities Landlord: Di€ os is a listed real estate company traded on the Nasdaq Stockholm stock exchange. The company owns commercial and residential properties in northern Sweden Tenant: KICKS is the largest cosmetics chain in the Nordics, with a total of 250 stores in Sweden, Finland and Norway. The company has 135 stores in Sweden Tenant: KappAhl is a large fashion chain, with a total of 350 stores in Sweden, Norway, Finland and Poland.

Purpose -The purpose is to increase knowledge of rent negotiations by investigating differences in beliefs held by property landlords and retailers on factors that they deem important in rent negotiation. Design/methodology/approach -This study investigates differences in subjective beliefs held by landlords and retail trade tenants on factors that affect rent levels during the rent negotiation process using a factor analysis approach. Semi-structured interviews were made with seven large real estate owners/landlords and retailers and eight experts in negotiating retail rent to elicit variables that have an impact on retail rent. Thereafter, a web-based survey was sent to 421 respondents who had experience in rent negotiation. Several factors were extracted using factor analysis. The data collection was made in Sweden during the coronavirus disease 2019 (COVID-19) pandemic in late spring 2021 Findings -Significant differences are found in beliefs held by landlords and retail trade tenants in four out of seven-factor: regional growth, e-commerce, customer focus and trust. Landlords rate these factors higher than retailers do. There are also systematic differences between landlords and retailers depending on their education levels on the following factors: rent and vacancies, e-commerce and customer focus. The number of years of experience did not prove to be significant instead differences are found to exist in factors Research limitations/implications -Not only do traditional factors of importance, such as lease structure, the effect of location, size and anchor or non-anchor tenants, have an effect on negotiated rent levels. Differences in other factors also exist, such as regional growth, e-commerce, customer focus and trust factors that may play an important role in the negation of retail rent. Practical implications -The findings provide new insights into the different views on factors that affect rent negotiations between landlords and retail tenants. Knowledge of such differences may increase the overall transparency in the negotiation process. Transparency may be increased by putting forward information on these factors before a negotiation takes place, in order to smooth differences in their beliefs. Social implications -If transparency in the negotiation process of retail rent increases, time to reach an agreement, stress and anxiety can be reduced by putting forward information on factors where differences exist between landlords and retailers JERER 16,2

Climate change is arguably one of the greatest public health threats of our time. Climate change is caused by rising greenhouse gas emissions from human activity, resulting in higher global average temperatures and changes to environmental and human systems. The American Academy of Pediatrics issued a policy statement in 2007 encouraging pediatricians to reduce carbon emissions and support families experiencing the effects of climate change. 2 We agree with this call to action. However, given the current national focus of policymakers, funders, and health care systems on addressing SDOHs, along with the ubiquitous impact of the changing climate on the social and environmental circumstances in which children live, we recommend climate change be defined as a critical and often overlooked social determinant. Viewing climate change as an SDOH could potentially help coalesce a large group of ideologically different stakeholders around the common goals of decreasing health care costs and improving child health. Similar to other SDOHs, climate change worsens health, increases health care costs, disproportionately impacts vulnerable communities, and exacerbates the effects of other SDOHs. Research has revealed that the changing climate, including longer and hotter summers, more frequent and intense storms, sea-level rise, more severe droughts, and poorer air quality, are inextricably linked to poorer health. The child health impacts are numerous and include worsening asthma and allergies; physical trauma from disasters; mental health symptoms, including posttraumatic stress disorder after disasters and anxiety about the future; increased Dr Ragavan conceptualized, drafted, and revised the manuscript; Drs Marcil and Garg helped conceptualize the manuscript and critically reviewed the manuscript for important intellectual content; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work. PEDIATRICS PERSPECTIVES exposure to infectious diseases; and lack of access to adequate food and clean water. 2,3 Projected costs of climate change to the health care system are staggering. The health costs of 6 climate-related events from 2000 to 2009 was 14 billion dollars. 4 Therefore, addressing climate change could help control health care expenditures while simultaneously promoting child health and well-being. Although climate change impacts everyone, children, communities of lower socioeconomic status, and communities of color will disproportionately experience negative health sequalae and have fewer available resources to adapt to the changing climate. 2,5,6 Poverty and climate change are interlinked because those experiencing poverty are more likely to be impacted by climate change, and climate change further exacerbates poverty. 6 However, comprehensive reports on addressing child poverty by both the National Academy of Medicine and the AAP do not mention climate change. Similarly, climate change threatens global food and water availability 5 ; yet, to our knowledge, food insecurity interventions do not currently incorporate climate change. Framing climate change as an SDOH provides an opportunity to incorporate approaches to address both the unique health effects of climate change and its impact on other SDOHs. --- STRATEGIES TO ADDRESS CLIMATE CHANGE AS AN SDOH Like we have done with other SDOHs, the pediatric community must consider ways to thoughtfully address climate change through medical education, clinical practice, community and scientific partnerships, and professional responsibility. --- Medical Education Trainees and practicing pediatricians should learn about climate change, including the impact of the changing climate on child health and strategies to address patient and parent concerns about climate change. We argue that climate change education should be incorporated into existing SDOH curricula, which have been shown to be effective in improving medical trainees' knowledge about SDOHs, ability to identify patients experiencing social needs, and awareness of appropriate resources. 7 --- Addressing Climate Change in Clinical Practice Discussing climate change during pediatric visits is within our scope of practice and recommended by the AAP. 2 The Yale Program on Climate Change Communication revealed that 77% of US adults believe schools should teach climate change; thus, it is likely parents feel similarly about pediatricians. However, no data exist on addressing climate change during pediatric visits. Further research should explore the perspectives of parents and pediatric providers on the importance of discussing climate change during visits, which topics would be the most meaningful to address, and barriers and facilitators to climate change discussions. Screening for SDOHs is less meaningful without tangible resources for families. Although climate change has profound impacts on child health, providing resources in clinical settings may be challenging. In this regard, climate change is different from other SDOHs for which a referral to a clinic or communitybased resource may have an immediate impact. Additionally, some components of addressing climate change may be best accomplished on a policy level. That being said, pediatricians can help educate about disaster preparedness and provide referrals to families impacted by climate change. The pediatric community should also familiarize itself with actions individuals, families, and communities can take to both decrease carbon emissions and reduce the health impacts of climate change and incorporate these recommendations during routine anticipatory guidance. Pediatric providers may also consider having developmentally appropriate materials available to support parents who are trying to talk with their children about climate change. Finally, providers must be aware of how climate change impacts other SDOHs to effectively address a range of unmet social needs. --- Community and Scientific Partnerships To comprehensively and effectively address other SDOHs, pediatric providers have begun to work in multidisciplinary teams. For example, to support families experiencing violence in the home, health care providers have partnered with hospital and community-based intimate-partner violence advocates. Similarly, the pediatric community needs to collaborate with climate scientists, clean energy leaders, community-based organizations, and behavioral health specialists with expertise in environmental-related mental health. Pediatricians should also work with policymakers to advocate for legislative policies aimed at reducing carbon emissions, just as we have done for policies about other SDOHs, immigration, and the social safety net. --- Professional Responsibility The US health care system is responsible for 10% of the US greenhouse gas emissions, and hospitals are the second-most energy intensive commercial buildings. 8 The patient-centered medical home may itself lead by example in adopting climate-friendly practices, an action similar to the ownership clinics have taken over other SDOHs. The pediatric community should also consider its own carbon footprint, both individually and as a profession. For example, traveling by air creates a heavy carbon footprint; it may be worth considering a professional commitment to choosing alternative means of transportation for attending conferences or using Web-based platforms. Our profession must take responsibility for our actions and work together as leaders to reduce carbon emissions. --- CONCLUSIONS We call on the pediatric health care community to view the changing climate as a critical SDOH that both increases costs of care and negatively impacts child health locally and globally. From that vantage point, we urgently need to integrate climate change work into all facets of our professional lives to further health equity and promote child well-being.

Social determinants of health (SDOHs), defined as the social circumstances in which people are born, grow, live, work, and play, profoundly affect children's health and drive health disparities. 1 SDOHs are shaped by the distribution of money, power, and resources at global, national, and local levels. The list of potential SDOHs is expansive and includes food insecurity, housing instability, violence exposure, structural racism, poverty, and immigration-related stressors. Addressing SDOHs within the pediatric medical home has garnered considerable interest lately as a potential solution to both improving health across the life course and mitigating health care costs. 1 However, one critical determinant is missing from the pediatric community's definition of SDOHs: climate change.

Introduction In the current era of globalization, easy access to information, news, and interaction through social networking sites has become more convenient with the use of gadgets . The development of communication technology, from postal services to public telephones and smartphones, has significantly transformed our way of communication . Communication has become an integral part of our daily lives. Social networking sites, as a form of internet-based communication, play a crucial role in facilitating international interactions and friendships . The positive impact of social networking sites is evident from the increasing number of new users each year. These platforms not only enable us to connect with people from around the world but also make it easy to add new friends . Social networking sites offer various benefits, such as the ability to communicate with idols without direct meetings and assisting students in finding the information they need . Instagram, as one of the most popular social networking sites, allows users from different countries to interact and connect with each other .With the presence of social networking sites, communication and friendship have become easier in the era of globalization, allowing people to obtain information about other countries without physical travel and establish friendships with individuals from diverse nations . --- Literature Review --- A. Social networking A type of social contact that takes place online is known as social networking. Users may connect with individuals from all over the world through social networking, and they can quickly and simply exchange information, photographs, videos, and opinions. Online platforms like websites, social apps, or social media can be used for social networking. --- B. Cross-national interactions and friendships facilitated by social networking Social networking has made international connections and interactions possible more quickly and easily. People may quickly locate and communicate with people from various nations through social networking. Here are a few strategies for using social media to promote connections between nations: --- 55 • Discover and Join International Groups or Communities: On social networking sites, users may join a variety of groups or communities based on particular interests or pastimes. Users may look for and join groups that are relevant to their hobbies or interests, where they can communicate with others from other countries who have similar interests. • Use location-based search features: A few social networking networks provide this function. These qualities can be used by users to locate individuals from particular nations or areas. Users may start talks and make friends by looking for people who live in the target nation. • Follow Pages or Accounts from Other Countries: There are several social networking accounts or pages that represent other nations. Users may follow these accounts or pages to learn more about the history, customs, and other intriguing details of that nation. Users may make new acquaintances by interacting with other followers and account owners. • Join Global Discussions or Forums: Social networking sites also include discussion areas or forums that address particular issues. Users can take part in worldwide debates or forums covering current events, culture, and other interesting subjects. Users may communicate with people from other nations and widen their buddy network through these chats. The advantages of cross-national connections and friendships made possible through social networking include: • Increased Cultural Tolerance and Understanding: Interacting with individuals from other nations enables users to better comprehend various cultures, customs, and values. This can improve cross-cultural tolerance, appreciation, and respect. • Opportunities to acquire Languages and Cultures: Interacting with people who speak a language natively can assist learners of that language acquire it more quickly. Users can also develop a greater grasp of that nation's culture and way of life. • Opportunities for Global Networking: Through social networking, people may create large global networks. These networks may be helpful in a variety of areas, including international project partnerships, jobs, and education. Instagram is a well-known social media site that people use to connect with others globally and make friends. Users may submit, view, and engage with other users' content on Instagram, a picture and video sharing app. Users may quickly locate and communicate with people from various countries using Instagram. Here are some tools and techniques you may use to locate international friends on Instagram: • Instagram has a location-based search tool that enables users to look for posts or other users in a particular nation or area. people may use this function to look for people from the nation of their choice and start conversations with them. • Instagram users can use hashtags that are distinctive to certain nations or cultures. Users can find posts and other users who have connections to or interests in that nation by utilizing hashtags that are appropriate. • Users can follow Instagram pages or accounts that are related to a specific nation. These articles or pages often provide information about the people, places, food, and daily life of that nation. people may communicate with other people who have similar interests by following these accounts or pages and learn more detailed insights and information about the nation. • Through the Direct Message function: Instagram users may send private messages to other users using the Direct Message function. Users may start talks with people from different nations using Direct Message and develop connections with them . Users may broaden their social network, exchange stories, and learn about various cultures using Instagram. Instagram is a social networking tool that promotes international connections and useful and pleasurable interactions. --- Research Methotology The research methodology is the way how research is carried out. In general, there are two According to the Knopf literature review is the activity of summarizing the body of writing for a topic. The literature review summarizes various sources related to the topic being discussed. Literature review can be done at the beginning of a study or it can be the ultimate goal of a study. Literature review can answer problems theoretically and practically. Synder submitted that literature review has a very important role in research. Literature review can be the basis for developing knowledge, creating directions and policies, and if done well can create new ideas and directions for a particular field. The method used in this research is descriptive method. This research is a qualitative type research by applying literature study. Literature study is used to build concepts or theories related to the role of social networks in friendship between countries. In conducting research, a problem formulation will be made as previously stated. Then there will be a collection of literary sources. The sources that can be used are books, scientific journals, proceedings, and other relevant sources. After that, interpretation is carried out from the study sources that have been collected to produce information or knowledge. --- Result and Discussion According to research conducted by Collin, Rahilly , Richardson, & Third the use of social networking services is very important in strengthening and developing relationships or relationships, especially as a result of the merging of the online and offline worlds. Social networking services can strengthen relationships that have been formed before. Some forms of online social networking such as instant messaging are generally used by smaller groups of participants and are used to maintain existing friendship networks. In addition, social networking services also have an important role in the development and exploration of new relationships. For some people, especially those who are socially isolated, online relationships formed from social networking services are close relationships because online interactions are the chosen way or the only way to carry out social interactions. Even some of these online relationships can lead to offline relationships. This shows that social networks can form close relationships for friends between countries. There are positive aspects of having social networks for learning foreign languages. They also mentioned that social networking platforms are able to improve foreign language learning skills, especially writing skills. The existence of interaction through social networks is a person's motivation to communicate with others. Piantari also stated that technological developments, especially for social networks, led to code-switching. Code-switching or code-switching is the transition from using one language to another. Code-switching behavior is a phenomenon that occurs in bilingual or multilingual communities. One of the social networks is the Instagram platform, users can engage in interactions with individuals from various countries. Instagram gives users the opportunity to make friends internationally without being limited by distance. Users can communicate with people from other countries through comments, private messages and shared materials . Strong emotional attachments and better relationships between people of different cultural origins may result from this type of engagement. The number of active users on Instagram has increased dramatically since the platform's introduction in 2010. The site has been able to draw in a wide audience because to its userfriendly layout, appealing features, and emphasis on visual material. Instagram is always improving and enhancing the user experience with features like feeds, stories, IGTV, and Reels, which helps to drive up the number of its active users . This fosters possibilities for individuals to deepen their connections and get a respect for various cultures . Instagram has made it possible for people from other nations to collaborate and work on collaborative projects. Users may communicate directly and establish conversations with tools like Direct Messages and comments. For instance, a team of artists from several nations can collaborate to produce original pieces of art. It promotes the sharing of knowledge, experiences, and creativity across cultural boundaries, which eventually strengthens global ties and creates robust social networks. Social networking platforms also play an important role in social activation and global awareness. Through online campaigns, users can mobilize support for social and humanitarian issues at the international level. They can participate in movements, share information, and raise funds to help people in need around the world. Platforms such as Change.org and GoFundMe provide an effective means of rallying support and building awareness of global issues. Social networking platforms facilitate cross-cultural knowledge exchange and learning. Users can follow accounts or groups that focus on a specific topic, such as science, art, music or languages. They can interact with individuals who share their interests and knowledge, exchange information with each other, and broaden their understanding of various topics. This generates diverse knowledge and enables cross-country collaboration in learning contexts . --- Conclusion Social media platforms, particularly Instagram, have a big influence on international relationships and interactions. Users of this site are able to communicate with people from other nations. Users may deepen their awareness of the customs, food, and way of life of various nations by following user accounts from other cultural backgrounds. Users now have more chances to get insights that take into account cultural diversity and to increase their general knowledge of the globe. Additionally, Instagram offers users the chance to meet people from all around the world without being constrained by physical distance. Through comments, private messaging, and content sharing, users may interact with users in different countries. Strong emotional attachments may be formed via this type of engagement, which also strengthens friendships between people from various cultural backgrounds.

In today's digital world and with advances in technology, social networking platforms are an important tool for finding relationships or relationships between individuals on a national to international scale. This study aims to examine the role of social networks in providing faceto-face interactions between individuals on an international scale. We conducted this research by collecting literature that discussed this before. The results of this study provide an in-depth understanding of how social networking platforms help expand social networks beyond national boundaries and how international friendships are formed from interactions that occur in social networks. With this research, it can make an important contribution to our understanding to create new opportunities in exchanging information about culture and friendship across countries through social networks that facilitate social connections that cross geographical boundaries.

Introduction Contraceptive discontinuation is defined as starting contraceptive use and then stopping for any reason while still at risk of an unintended pregnancy [1]. It has been reported to be higher for short-acting methods such as condoms, injectables, pills and traditional methods as they can be discontinued by the user herself compared to longacting reversible methods such as implants and the Intrauterine device which require a visit to facility to discontinue [1][2][3]. The method-related reasons and contraceptive failure have been reported as the predominant causes for contraceptive discontinuation [4]. Postpartum period in the first 12 months following childbirth has been associated with high unmet need for contraceptives coupled with unintended pregnancies [5]. Breastfeeding practices and beliefs about return of menses as a marker of fertility resumption during the postpartum period, makes it difficult for women to determine their fertility risk. Thus, women are less motivated to start contraceptive use while breastfeeding [6,7]. Previous investigators have demonstrated that women who discontinue contraception use during the postpartum period may opt not to use any method or switch to different modern method which is less effective than the previous method at preventing pregnancy, and thereby exposes women to risk of unintended pregnancy, abortions and mistimed pregnancies/births [7,8]. Sustaining postpartum contraception use is important for woman's fertility because it ensures optimal birth spacing, prevents unintended pregnancies, abortion and has an impact on infant and child survival [9]. Examining postpartum contraceptive discontinuation will shed a light on the knowledge gaps in contraceptive use such as trends and determinants for contraceptive discontinuation, and will help in reducing unmet need for family planning [10]. It will also provide evidence for areas that require coordinated efforts between different stakeholders involved in family planning programs and the government. This will help to improvement quality of services for family planning and reduces the discontinuation postpartum contraception. Previous studies from developing countries showed that, on average, 19-64% of women discontinued using reversible contraceptive methods by the 12th month of use [4,11]; [12][13][14][15]. The discontinuation rate for condoms within the first 12 months is higher than intrauterine devices , and up to 40% higher for other methods such as pill, injectable, periodic abstinence and withdrawal [3,4,16]. Contraceptive discontinuation among women with no desire to get pregnant increases the risk for unwanted pregnancies [8]. This reflects a failure of family planning programs and services [3,4,[17][18][19]. According to Tanzania Demographic and Health Survey, only 15.5 and 22.4% of women reported using modern contraceptive methods at 3 months and 12 months post-delivery respectively, especially during postnatal visit probably due to contraceptive counselling [20]. A cohort study conducted in Northern Tanzania among 5284 pregnant women who were followed from 6 to 15 months postpartum, reported that 34% of women initiated contraceptive use during the postpartum period and 25% of the participants started at 7 months postpartum [21]. Authors in this study noted that 18.8% of contraceptive users discontinued at 15 months postpartum. The reasons for contraceptive discontinuation include partner disapproval , side-effects , wanting a child and other reasons [21]. The most recent study in Tanzania reported that short contraceptive methods were associated with high rate of discontinuation compared to long term acting contraceptives [13]. There is scant information on contraceptive discontinuation rates, patterns and associated factors postdelivery. This study aimed to determine predictors of contraception discontinuation at 3, 6, 12 months postpartum among women of reproductive age in Arusha city and Meru district, Tanzania. --- Methods --- Study design and setting This study was conducted in two districts of Arusha region from December 2017-June 2018. --- Study population, sample size and sampling method The study sample included women of reproductive age aged 16-44 years who started to use family planning methods after a delivery that occurred at least 12 months prior to the survey. Multistage cluster sampling with probability proportional to size was used to draw respondents from 13 streets in Arusha City and 2 wards in Meru District. The details of the sampling procedures have been described elsewhere [15,22]. A contraceptive calendar is a contraceptive history collected for each woman who, or whose husband, was not sterilized at the calendar's start. The data were recorded in a calendar matrix, consisting of rows and columns, with each row of the calendar representing a particular month. The sample size was estimated based on number of events included in the contraceptive calendar, which included retrospectively reported month-by-month during 31 months before the interview. A total of 12,203 contraceptive use events were recorded during the study period. We excluded 3216 episodes which started prior to the calendar period of 31 months before the survey; 2883 episode because no birth within 31 month; 2211 episode because the episode started before the latest birth; 2494 episode because no method used, birth, termination, or pregnancy; 184 events during 0-3 months before the survey [methods unknown & sterilization/other ]. The remaining 1215 episodes of use within the last 3 to 28 months prior to interview constituted the final sample size . --- Study variables The main outcome was contraceptive discontinuation after starting to use of contraceptive at three time points; 3 months, 6 months, and 12 months post-delivery. The independent variables were selected based on the previous studies. These include sociodemographic characteristics: maternal age, maternal education and wealth index. Women were asked of the number of antenatal care visits during the last pregnancy, number of months since birth to first family planning use, received contraceptive counseling during ANC or PNC visits and attendance of postnatal care after last delivery. The type of contraceptive methods used were IUD, pills, male condoms, injectables, implant, rhythm and withdrawal. Women were also asked for the reasons related to discontinuation of modern contraceptive. --- Data collection methods Data collection was conducted through face-to-face interviews using questionnaire administered using tablets. A team of trained research assistants including medical doctors, statisticians, demographers and social scientist were used to collect information from the participants. A standardized questionnaire in Kiswahili language which was adapted from the Tanzania Demographic and Health Survey 2015-2016 [12] was used to collect information from the study participants for household survey to meet the WIE objectives. These variables includes women age, education level, marital status, area of residence, wealth index, utilisation of contraceptive methods, cost to access the family planning services, type of facility, distance to health facility, type of contraceptive used and availability of family planning commodities to mention a few. --- Data analysis Data analysis was performed using STATA version 15. Descriptive statistics were summarized using frequency and proportions for categorical variables. The percentage of women who had discontinued their method of contraception at 3, 6 and 12 months were reported. Adjusted Odds ratios with 95% confidence interval for the factors associated with contraceptive discontinuation were estimated in a multivariable logistic regression model. The correlation between discontinuation at certain time with sociodemographic characteristics were separately evaluated. We used the logistic regression because the dependent variable is a dummy variable . Independent variables we used for the regression analysis were reported in Table 1. Only variables which were significant in the bivariate analysis were included in the final model. --- Results --- Characteristics of study participants We have a total of 1215 episodes from the calendar data . About one third of all episodes were aged 25 to 29 years old. More than half of all episodes had primary education. Majority of all episodes were reported to have received family planning counselling during ANC visit and postnatal care visit . --- Contraceptive discontinuation at 3, 6 and 12 months postpartum Overall, 11.5% of all episodes discontinued at 3 months. Of these, injectables, pills and male condoms were discontinued by 16, 16 and 9% of the sample respectively, while implants and IUD discontinued by 1.55 and 3.23% at 3 months. On the other hand, of the 11.5% episodes discontinued at 3 months, 27% were LAM users . Likewise, there was 19.4% episodes discontinued at 6 months, where injectables, pills and male condoms were discontinued by 24.64, 25.44 and 20.37% of the sample respectively while implants and IUD accounted for 3.84 and 8.34% during the same period. At 6 months, 49% of those 19.4% episodes discontinued were LAM users . Furthermore, about 29.4% episodes discontinued at 12 months, injectables, pills and male condoms accounted for 38.15, 36.48% and 50. 49% of the sample while implants and IUD contributed 5.11 and 9.06% at 12 months while 76.12% of 29.4% episodes discontinued at 12 months were LAM . --- Multivariable analysis for determinants of contraceptive discontinuation at 3, 6 and 12 months postpartum The determinants for contraceptive discontinuation across time points are shown in Table 5 . In 5). --- Discussion This study examined the determinants of postpartum contraceptive discontinuation at 3, 6 and 12 months postpartum. Overall, 11.5% of all episodes discontinued at 3 months, 19.4% discontinued at 6 months and 29.4% discontinued at 12 months. Women aged 40 to 44 years had significantly lower odds of contraceptive discontinuation at 3 months postpartum compared to their counterparts aged 16 to 19 years of age. Furthermore, women who reported using implants and pills had significantly lower odds of contraceptive discontinuation at 3 months, 6 months and 12 months compared to injectable users. The average number of months since child birth to first family planning use was 3.8 months, this was similar to what was observed in rural Ghana [23] where the average time of first family planning use following child birth was 3.5 months and in contrast to what was observed in Nairobi slums where initiation of contraceptives following child birth occurred 7 months after delivery [11]. The fact that women initiate contraceptive use early in our setting may be encouraging, however the type of methods used must be borne in mind because short contraceptive methods have been associated with high rate of discontinuation compared to long term acting contraceptives [13]. Thus, short-acting contraceptive methods do not guarantee adequate birth spacing and the prevention of unwanted or mistimed pregnancies. The most common methods discontinued after postpartum initiation in this study were LAM, pill and injectable. This is consistent with the previous reports from Tanzania Demographic and Health Survey and Health and Demographic Surveillance System in Magu district in Tanzania [13,20]. Our finding was different from Malawian study, where women who reported long-acting methods and injectable use at 3 months post-delivery were more likely to continue compared to those using pills, condoms, traditional methods [2]. It also differs from the South Africa that reported high rate of implant continuation at 12 months at a rate of 86% [7]. The difference in findings could be explained by the differences in characteristics between the studied population especially the cultural barriers. Furthermore, the high contraceptives discontinuation rate in our population calls for a need to provide women with education on contraception, increase access to contraceptives as these will facilitate women to have informed choice and decision towards contraceptive use. However, there may be a possibility of some women experiencing forms of coercion to adopt a LARC the immediate postpartum period and choosing later to have the method removed. The lower odds of postpartum discontinuation among women aged 40-44 years in our study is consistent with previous report in Kenya by [11] where higher odds of contraceptive discontinuation among adolescent women at 3 months, 6 months and 12 months compared with the adults counterparts. Previous studies in Tanzania and Nepal also reported the effect of age and type of contraceptive methods used on post-partum contraception discontinuation [13,24]. The high contraceptive discontinuation among young women could be explained by their fertility desire to have more children. Unlike the short-term contraceptives, the lower discontinuation rate for long-acting methods could be explained by its difficulty to remove which requires health care professionals with cost implications [25,26]. --- Strengths and limitations of the study The strength of this study is that, being a communitybased study may reflect the representation of what is happening at the ground in the general population. We also used a rigorous data collection methods to enhance validity for observed findings. Despite its strength, some limitations which need to be taken into account while interpreting our finding. First, being a cross-sectional in nature, the study cannot establish a causal effect. Secondly, we did not collect information on sexual resumption, menstrual resumption, partners and service related factors which would further explain the methods discontinuation during the course of time. --- Conclusions Lactational amenorrhea, male condoms, injectables and pills users had the highest rates of discontinuation compared with implants and IUD users. Women aged 40 to 44 years had lower odds of postpartum contraceptive discontinuation at 3 months while implants and pills had lower odds of contraceptive discontinuation at 3 months, 6 months and 12 months. Addressing barriers to prolonged contraceptive use amongst younger women and knowledge on attributes of contraceptive methods and their potential side-effects is warranted. In addition, the programs should assist women to timely switch to a method of their preference when they discontinue the method that fail, do not meet their expectations or cause side-effects. --- --- Authors' contributions SEM, BE and IS designed the study. All authors were involved in data collection. RS and MJM conducted data analysis. MJM, RS and BE drafted the manuscript. MJM RS RM CA BE SEM and IS contributed to the final draft of the manuscript. All authors reviewed and gave their final approval to the manuscript. --- Funding There was no fund allocated for this project. This was part of the student project. --- --- --- Competing interests The authors declare that they have no competing interests. All authors read and approved the final manuscript. ---

Background: Postpartum contraceptive discontinuation refers to cessation of use following initiation after delivery within 1 year postpartum. Discontinuation of use has been associated with an increased unmet need for family planning that leads to high numbers of unwanted pregnancies, unsafe abortion or mistimed births. There is scant information about contraceptive discontinuation and its predictors among postpartum women in Tanzania. This study aimed to determine predictors of contraception discontinuation at 3, 6, 12 months postpartum among women of reproductive age in Arusha city and Meru district, Tanzania. Methods: This was an analytical cross-sectional study which was conducted in two district of Arusha region (Arusha city and Meru district respectively). A multistage sampling technique was used to select 13 streets of the 3 wards in Arusha City and 2 wards in Meru District. A total of 474 women of reproductive age (WRAs) aged 16-44 years residing in the study areas were included in this analysis. Data analysis was performed using STATA version 15. Odds ratios (ORs) with 95% confidence interval (CI) for the factors associated with contraceptives discontinuation (at 3, 6 and 12 moths) were estimated in a multivariable logistic regression model. Results: Overall, discontinuation rate for all methods at 3, 6, and 12 months postpartum was 11, 19 and 29% respectively. It was higher at 12 months for Lactational amenorrhea, male condoms and injectables (76, 50.5 and 36%, respectively). Women aged 40-44 years had lower odds of contraceptive discontinuation at 3 months as compare to those aged 16 to 19 years. Implants and pills users had also lower odds of contraceptive discontinuation compared to injectable users at 3, 6 and 12 months respectively. Conclusion: Lactational amenorrhea, male condoms and injectables users had the highest rates of discontinuation. Women's age and type of method discontinued were independently associated with postpartum contraceptive discontinuation. Addressing barriers to continue contraceptive use amongst younger women and knowledge on method attributes, including possible side-effects and how to manage complications is warranted.

INTRODUCTION Empowerment enables women to realize their potential and participate fully in every aspect of life . It also improves their capacity for household decisionmaking with implications for their personal welfare and that of their children . The need to empower women is a right in itself and a means of attaining economic growth , poverty reduction , and good governance . The inclusion of women's empowerment as part of the 2000-2015 Millennium Development Goals and the 2015-2030 Sustainable Development Goals shows the priority accorded. While the concept of women's empowerment is a universal notion, its association with poverty reduction indicates that more attention needs to be paid to it in developing countries . Institutions like the World Bank, United Nations agencies and local non-governmental organizations have implemented multiple programs aimed at improving the socioeconomic status of women. These programs create educational opportunities and access to micro-credit, among others , with the aim of building financial independence capacity of women as a means for empowerment. Empowerment increases the ability of women to participate in decision-making . As women's access to wage employment improves, their contributions to the household and their negotiations improve, and their bargaining and eventual participation in household decision-making also improve . Given the centrality of decision-making to the conceptualizations of power, decisionmaking therefore becomes the most frequently used measure of agency measure for empowerment . The authors of this current study therefore examined the implications of women's labor market participation for their involvement in household decision-making in Nigeria, bearing in mind the predominant attention paid to the economic/instrumental dimensions of women's empowerment relative to its agency dimensions. This research examined the association between women's labor market participation and household decision-making in 2003, 2008, and 2013 and looked for consistencies in the pattern of association between women's labor market participation as well as participation in household decision-making over this period. The study explored the direct and indirect route through which women's labor market participation was associated with participation in household decision-making. This study is based on the extant micro-level phenomenon that women's access to education and employment opportunities result in a reduction in household poverty and increased women's empowerment . The authors acknowledge that choice and control are also indicators of women's empowerment and that there is a nexus between education and the empowerment of women . The study focused on women's access to paid employment as a path for empowerment, measured by their participation in household decision-making. --- METHOD The study area, Nigeria, is a middle-income country and the most populous in Africa, ranked seventh in the world with a total population of about 190 million people. The country is diversified in ethnicity and religion. It is divided into six geo-political zones, namely the South-West, South-South, South-East, North-East, North-West, and North-Central. The data used for the study was extracted from the Nigerian Demographic and Health Survey , a nationally representative survey. The NDHS is often conducted every 5 years, hence we extrapolated data from the three most recent surveys to show a change in outcomes over an 11-year period . There is currently no newer NDHS data. Although the 2018 NDHS survey has recently been concluded, the dataset is yet to be made available and accessible, hence this study analyzes NDHS 2003NDHS , 2008NDHS , and 2013. . The NDHS is a nationally representative survey that provides population and health indicators. The 2003 NDHS used twostage cluster design sampling to select 365 clusters and chose 50 households systematically from each cluster. A total of 7,620 eligible respondents were successfully interviewed. The NDHS 2008 had a total of 888 clusters selected from a complete list of households, with an average of 41 households taken from each cluster through equal probability systematic sampling and a total 33,385 women were successfully interviewed. Lastly, NDHS 2013 used a three-stage stratified sampling design to select a total of 904 clusters with a fixed representative sample of 45 households per cluster and completed interviews of 38,948 respondents. For the purpose of this study, only women aged 15-49 years who were "currently in union or living with a man" at the time of survey were included in the analysis. A total sample of N = 56,620 was analyzed. All variables used in this study were adopted from the women recode file. Some variables were measured as defined in the NDHS while others were recoded for the purpose of analysis. Explanatory variables such as "level of education, " "residence" and "household wealth index" were all adopted from the NDHS. The recoded variables were employment within past 12 months [none = 0, professional = 1, sales/services = 2, manual = 3], religion [Christianity = 1, Islam = 2, Traditional/Others = 3], and age group [15-24 = 1, 25-34 = 2, 35-44 = 3, 45+ = 4]. The independent variable was "Employment within the past 12 months" and the dependent variable "women empowerment" was proxied by three variables. They are "decision on own healthcare, " "decisions on large household purchases, " and "decision on visits to her family or relatives." Each of the outcome variables is binary "0 and 1" and thus binomial logistic regression was adopted to assess the relationship between variables. Other covariates such as age group, religion, employment status, educational level, household wealth index, and region were controlled for in the analysis. Age group and religion were redefined and recoded. Age group was recoded into four groups: 15-24 years, 25-34 years, 35-44 years, and >45 years. Religion was recoded into Christian, Muslims, and others. Explanatory variables adopted were age group, religion, employment status, educational level, household wealth index, household headship, place of residence, and region. Age group and religion were redefined and recoded. Age group was recoded into four groups: 15-24 years, 25-34 years, 35-44 years, and >45 years. Data analysis was conducted using STATA 15.0. Univariate analysis was conducted to show the frequency distribution of each respondent's characteristics. Bivariate analysis was conducted to highlight the associations between the dependent and independent variables using Pearson's chitest with the level of significance set at p = 0.01. We used binomial logistic regression to predict the proxy dichotomous variables of women empowerment from the exposure variables. Appropriate weighting was applied throughout the analysis to ensure representativeness. --- RESULTS Table 1 shows the background characteristics of ever-married women in Nigeria by employment status in 2003, 2008, and 2013. There was a decrease in the proportion of ever-married women who were not employed who had these characteristics from 2003 through 2008 to 2013: those 35 years and older; those with primary and secondary level of education; those from the middle, richer, and richest household wealth quantiles; Christians; male head households; urban residents; and those residing in Northcentral, Northwest, Southeast and South-South Nigeria. Over the study period, unemployment increased among ever-married women from the poorest household wealth quantile , with female household heads , and from Southwest Nigeria. Table 2 shows the socioeconomic profile of ever-married women in Nigeria by their participation in different forms of decision-making-health, household purchase, and visit to family. More women in the 15-24 age group, with no education, with the poorest wealth index and not employed did not participate in any decision-making process in 2003, 2008, and 2013, respectively. It was only in South-West Nigeria that the proportion of women who did not engage in any of the decisionmaking processes decreased from 27.9% in 2003 to 16.4% in 2008 and 10.1% in 2013. Also, the proportion of women who did not engage in any of the decision-making process increased from 55.0% in 2003 to 57.3% in 2008 and 60.0% in 2013 in North-West Nigeria. Figure 1 shows the proportion of ever-married women who could take decisions by education and employment status. Overall, a higher proportion of women with higher educational status could take decisions relating to health, major household purchase, and visit to family. Also, the proportion of evermarried women who could take decisions relating to health increased with the level of education. The proportion of women who could participate in decision-making regarding large household purchases increased over the years among women with employment status. Notably, the gap between women that had no employment and household decision-making widened increasingly over the years. The gap between those that had some level of education and those that had no education also widened and was most apparent when it came to decision-making on major household purchase. Table 3 is a cross-tabulation of respondents' work status with involvement in household decision-making. Significantly more ever-married women with professional employment status were able to take part in decision-making regarding health, household purchases, and visits to family in years 2003, 2008, and 2013. For the same period, significantly less ever-married women were about to take part in household decision-making. Figure 2 shows that the odds for women's participation in decision-making on health increased with education status when compared to those without education from 2003 to 2013. Also, women with higher household wealth status compared with women with the poorest household status, residents in rural areas compared with residents in urban area and residents in North-east Nigeria compared with residents in North-central Nigeria had better participation in decision-making regarding health from 2003 to 2013. The odds decreased for Muslim women when compared with Christian women and for female-headed households over the same period. The difference within the age groups also decreased over the same period. Figure 3 shows that the odds for women's participation in decision-making on major household purchases increased with education status when compared with women without education and with higher household wealth status compared with women with the poorest household status from 2003 to 2013. The odds decreased for women who were resident in Northeast and Northwest Nigeria over the same period. The odds also decreased for women who were resident in all regions compared with Northcentral Nigeria; and for gainfully employed women when compared with those who were not employed from 2003 to 2008. Figure 4 shows that the odds for women's participation in decision-making regarding visits to family increased with household wealth status compared with women with the poorest household status from 2003 to 2013. The odds decreased over the same period for women who were resident in Northwest Nigeria and for Muslim women. --- DISCUSSION The results of the study highlight the dynamics of household decision-making in Nigeria over an 11-year period. The study showed that that women's access to paid employment, educational status, and household wealth index improved their participation in household decision-making. However, the single factor that consistently increased the likelihood of an ever-married woman to be involved in all forms of household decision-making from 2008 to 2013 was household wealth. The study also found that religion affected the ability of ever-married women to participate in household decision-making. The ability of Muslim women to participate in all forms of decision-making decreased from 2008 to 2013. What the authors seemed to observe is a phenomenon where the balance of power in household decision-making improved as the household wealth index improved. This balance of power also seemed to improve as the educational and employment status of women improved. A recent study in Spain shows that when demographics, family, and labor market characteristics were controlled for, the educational status of both men and women were the major influencers of an egalitarian household decision-making process . In part, the findings of Albert and Escardíbul's study in Spain is consistent with the findings of this current study and with Mejiuni's study in Nigeria which reported that educational status influenced the household decision-making process for women . Other sociocultural factors, such as religion and geographical factors like residential location, also influenced household decision-making for women. This study shows that the Islamic religion reduced the decision-making ability of ever-married women over the study period. Low decision-making power is pertinent among women who reside in Northern Nigeria where Islam is dominantly practiced. These women were reported to be less able to take part in household decision-making compared to women in Southern Nigeria where other religion are dominant. Mercy affirmed this finding in her study that, more than any other geographical settings in Nigeria, women in northern Nigeria have very limited educational opportunities and very few of their children who were girls completed their primary education, mostly due to early marriage, poverty, and oftentimes religious reasons. The ability of women to take part in household decisionmaking reflects their autonomy. Largely, this study suggests that the autonomy of married women may have improved over the last 11 years in Nigeria. Although the educational disparity between males and females persists and is worse at the tertiary level of education, despite the increase in school enrolment figures over the years , the findings of this current study show that educational enrolment for women has improved over the study period. Other studies have shown that women who are educated and especially those who have secondary or more education will likely have more autonomy to make unilateral or joint decisions regarding their health and other family members . Consistent with other studies , the findings of this current study have implications for maternal and child health because improved autonomy of women would result in improved household decision-making regarding major household purchases and improved nutritional status of members of the household. For a country like Nigeria where there has been no substantial improvement in the under-5 mortality index, improving the ability of women to take decisions on health and improving her social capital through her ability to take decisions to visit family members can increase maternal health service utilization and maternal under-5 mortality . Summarily, the authors of this current study noticed that women's levels of participation in household decision-making varied significantly across the six geopolitical zones in Nigeria. Women residing in geopolitical zones located in northern Nigeria had lower household decision-making autonomy when compared with women in Southern geopolitical zones. Some studies attributed this to lower level of education and lower employment status . This has further implications for health conditions and welfare status, given the report that maternal mortality and under-5 mortality rates are worse in Northern Nigeria . --- CONCLUSION This study showed that over the last 11 years, there have been changes in the status of women in ways that affect their decision-making power. Education of women has proved to be a significant contribution to enhancing a balance in equity between men and women in household decision-making. The attenuating effect of Islam on this prospect needs further investigation and interventions for married women living in Northern Nigeria. With respect to variation on the influence of Islamic religion on women decision-making between the North and South, we suggest that adherence to stricter practices of Islam in the North relative to the South could have accounted for the variation. Such adherence to religious practice increases male dominance and male dependency. Islamic extremism has also increased over the years. It could also be partly a function of the relatively higher level of exposure to western education by Muslim women in the South. However, this opens a frontier for further research. Like all cross-sectional studies, what this study has been able to establish is an association between variables rather than a cause-effect relationship. The findings can, at best, help establish a hypothesis for a study that determines a causeeffect relationship between household decision-making and employment status of ever-married women in Nigeria. --- ETHICS STATEMENT The study employed a secondary dataset of the Nigerian Demography and Health Surveys. This is a nationally representative survey conducted by the Measure DHS/ICF International in partnership with Nigerian National Population Commission. All protocols and survey instrument were approved by the Institutional Review Board of Nigeria and IRB of the ICF International. The corresponding author obtained approval from Measure DHS to access, use and analyze the dataset of the study area. --- --- Conflict of Interest Statement: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Prior studies have assessed economic/instrumental dimensions of women's empowerment relative to its agency dimensions. This study assessed ever-married women's participation in the labor market as a form of agency for empowerment and household decision-making in Nigeria. The study utilizes secondary data from three national surveys of Nigeria 's 2003, 2008, and 2013 Demographic and Health Surveys (DHS) to assess the differences and relationship between indices and the status of women's labor market participation and their empowerment or household decision-making over an 11-year period. Explanatory variables adopted were age group, religion, employment status, educational level, household wealth index, and region. Binomial logistic regression was used to predict the proxy variables of women's empowerment from the explanatory variables. Findings showed that women's access to paid employment, educational status, and the household wealth index improved their participation in household decision-making. However, the single factor that consistently increased the likelihood of an ever-married woman to be involved in all forms of household decision-making from 2008 to 2013 was household wealth. Also, religion affected the ability of ever-married women to participate in household decision-making. The ability of Muslim women to participate in all forms of decision-making decreased from 2008 to 2013. The study concluded that the factors that enhance household wealth will enhance married women's ability to be involved in household decision-making. Education of women is, however, a significant contribution to enhancing the balance of inequity between men and women in household decision-making. The attenuating effect of Islam on this prospect needs further investigation and interventions for married women living in Northern Nigeria.

The policies in many respects represent a poised effort on the part of the learning ministries to improve educators' awareness of culturally-responsive pedagogy, as well as to bring Aboriginal peoples' socio-historical traditions to the fore . The trilogy of policy documents refer specifically to the critical role of Aboriginal communities, as described in the literature , in fostering positive social relationships with learning institutions , improved academic outcomes for Aboriginal learners , and healthy environments . In addition, each of the documents recognize the pivotal role of schools and training institutes in terms of endorsing life-long learning and community and civic responsibility . The policies may also be a response to the fact that "the experience of Indigenous youth remains principally ignored in scholarly, as to an even greater extent, policy discourse" . --- Pending Research While the policy initiatives by Ontario's learning ministries undoubtedly represent a collective attention to Aboriginal learners, my current line of inquiry is revealing some noteworthy initial findings that are deserving of further analysis. Interestingly, the policy documents reflect a dominant discourse of an assumed normative educative stance; more precisely, the jargon of data-driven outcomes and evaluative statements throughout all three policies seem to implicate upon normative educative paradigms that, in turn, creates a conceptual tension with the selfdeclared intentions of the policies themselves. Even with the distinguished political profile of these educational policies, the substantial investments of taxpayers' dollars, the government's goodwill to reach out to Aboriginal communities, and the perceptible moral action to redeem Aboriginal learners' social dependency and prepare them for the labour markets, one is still left to wonder about the potential consequences that may be in store for Aboriginal peoples if they choose not to self-identify. --- Overview The jargon of data-drive outcomes is symbolic of normative values based largely upon Eurocentric measures. The 2007 Ontario First Nation, Métis, and Inuit Education Policy Framework , relevant for K to 12 provincial schools, emphasizes the need for "reliable and valid data" to measure Aboriginal student learning. The 2009 Solid Foundations for the Road Ahead document cites the importance of using "reliable First Nation, Métis and Inuit student data" to measure the percentage of students achieving provincial standards on large-scale externally-delivered tests. --- The Aboriginal Postsecondary Education and Training Policy Framework (Ministry of Training, Colleges and Universities, 2011) identifies the imperative nature of analyzing "reliable and relevant quantitative and qualitative data" to evaluate programs and services. The following issues must be considered: 1. I and others have discussed in previous studies the problematic nature of using culturallyinappropriate standardized instruments to evaluate and measure Aboriginal student achievement and progress, particularly given the fact that the same policies cite the holistic and traditional epistemic values of Aboriginal students and communities . 2. The respective policies encourage Aboriginal peoples to self-identify their Native ancestry. It remains to be determined if such endorsements to self-identify positions the onus directly upon Aboriginal peoples to abide by the ministry initiatives. Each policy goes to great lengths to describe the substantial amounts of public monies that have been invested in order to improve Aboriginal learners' achievement and thereby close the educational and socio-economic gaps that have resulted in their dependency upon various social assistance agencies. The implication exists that in order for Aboriginal learners to benefit from the self-described and oft-referred "support" of the tax-paying public, and hence break the cycles of dependency, they need to self-identify so that the necessary data can be collected and the government can report on Aboriginal learner progress in light of these educational and training investments. Similarly, woven throughout the discourse of all three policies are evaluative statements that imply directly that the government's "commitment" to Aboriginal learners includes consultation with Aboriginal community leaders . By citing these collaborative partnerships between government and local First Nation communities, the learning ministries position themselves as collegial and inclusive bodies. The general public, thus, can attest to the fact that these relationships are built on representation as presumably so too are the policy initiatives. The government's dedication to improving the apparent plight of Aboriginal learners seems to embrace not only a political but also moral cause, especially given the "great progress . . . in developing stronger working relationships with school boards and Aboriginal organizations" , as one example of an evaluative statement among many others. Yet, the policy discourse positions Aboriginal learners in a potentially dubious predicament. The attention invested in these high profile educational policies, combined with the various sources of funding and the government's efforts to consult with Aboriginal communities might seem convincing enough for Aboriginal learners to self-identify. However, the political, cultural and epistemic realities and respective complexities associated with self-identification may not necessarily be fully explained in the policy documents. By choosing not to self-identify, Aboriginal learners may be perceived by the general public as rejecting the goodwill of policymakers and educators and as being resistant to the offers of assistance. Such a perception may in fact contribute to the conceptual tension already inherent in the policy discourse. --- Further Analysis The aforementioned policy documents produce a dominant discourse of a normative educative space that through the jargon of data outcomes and evaluative statements may create a profound conceptual tension for Aboriginal learners. Subsequently, and still to be determined, is how the Aboriginal communities and First Nations representatives are receiving these documents. Said differently, how have the community consultations materialized first from supposed embedded understandings to policy, and currently in practices related to selfidentification? I suspect that the educative view of these policies will not be uncontested for too long.

The learning ministries in Ontario have made a concerted effort to underscore Aboriginal learners' needs and preferences in publicly-funded and assisted schools and training services throughout the province. Through a trilogy of policy documents, the Ontario Ministry of Education (OME) and the Ministry of Training, Colleges and Universities (MTCU) have addressed expanded definitions of learning and sought to unfold the socio-cultural and epistemic values related to Aboriginal student and community worldviews: 1. The Ontario First Nation, Métis and Inuit Education Policy Framework (2007) commissions the province's boards of education, school administrators and teachers to create culturally-sensitive schools and classrooms. 2. The OME's Sound Foundations for the Road Ahead ( 2009) is a progress report on the outcomes for the aforementioned Policy Framework (2007) and aims to assure taxpayers that progress is being made in regards to policy implementation. 3. The MTCU's Aboriginal Postsecondary Education and Training Policy Framework (2011) addresses the educational gap between Aboriginal and non-Aboriginal peoples and focuses upon training outcomes and skill-development programs for Aboriginal learners.

Plain English summary Social norms are shared values, beliefs, and attitudes within a group, and they may impact individuals' behaviors, including their health behaviors. We studied the relationship between norms and women's decisions to deliver their child in a facility in rural Ghana. We surveyed 508 women and interviewed 79 participants including mothers, their husbands, mothers-in-law, community leaders and health care providers. We learned that women's beliefs that their families approved of facility delivery and that other women in their community delivered in a facility were important in their decision making. Community teams made up of community leaders and health care professionals had a positive impact on women's beliefs and their decisions to deliver in a facility. Community-based efforts to increase facility delivery should help women's support systems, made up of family members such as husbands and mothers-in-law, understand their important role in encouraging women's access and use of health facilities. --- Background Sub-Saharan Africa accounts for approximately twothirds of the global maternal deaths [1]. Specifically, in Ghana the maternal mortality ratio remains high at 310 deaths per 100,00 live births [2]. As many maternal deaths can be prevented through pregnant women's use of health facility birth with the assistance of a skilled birth attendant [3,4], efforts have been made in the past decade to increase facility-based childbirth in many sub-Saharan African countries [2]. In Ghana, health facility births increased from 55% in 2008 to 79% in 2017. This increase can be attributed to nationwide initiatives to implement intervention strategies to improve maternal health services use, as well as the Ghana health ministry's implementation of free national health insurance and maternal delivery services [5]. Facility births in rural areas of Ghana, where more than half of the nation's population reside, are significantly lower than the national average [2]. Reasons for this include limited access to health facilities, cost and time associated with utilizing the facilities, and social factors such as social norms related to childbirth practices [6]. Research to improve maternal health services utilization has in part focused on social norms as important determinants of place of childbirth. Social norms refer to shared values, beliefs, or attitudes of a group that impact perceptions and behaviors of the group members [7]. These norms may be sanctioned by influential individuals within a community or community members [8]. Community norms favoring facility birth have been positively associated with an increased rate of childbirth in health facilities [9][10][11]. Normative beliefs about the importance and quality of health facility birth within a community may be shaped by health outreach and intervention strategies and, in turn, increase the likelihood of health facility births among women [11,12]. In Ghana, studies have pointed to a shifting trend toward community acceptance of health facility birth, due to changing perceptions about the benefits and quality of care in these facilities [13]. Social norms may also be sanctioned by members of one's own personal network such as relatives and friends. There is limited evidence related to the relationship between network norms and health facility birth [14]. Our recent examination of the role of personal network characteristics in women's place of childbirth revealed that women with greater numbers of close relatives that approve of facility birth, and knowledge of more women in their network that received pregnancy-related care in a facility, are more likely to give birth in a health facility [15]. Less is known about the interactive impact of network and community norms on facility birth. Social network theories indicate the importance of connections and norms to health behaviors [7,16,17]. For example, network and community norms have each been individually examined as influential determinants of health decisions about women's pregnancy-related care continuum [10][11][12]18]. Previous work has linked community perceptions about the quality of formal health systems with facility delivery [12], and across six African countries, community-level norms -e.g., percentage of women in the community that delivered in a health facility, husbands in the community that approve of family planning -were associated with women's decision to deliver at a health facility [19]. Specifically in Ghana, Speizer and colleagues found a significant association between norms about facility births and women's use of facility delivery, as well as a significant interaction between these norms and women's decision-making autonomy. Whereas research is limited on network norms and facility births, recent evidence from Ghana suggests that normative support for facility birth from network members was associated with facility delivery [18]. Both norms can inform health intervention programs and strategies to further improve uptake in health facility births. However, it is unknown whether network or community norms are more salient than the other, or if both norms have a cumulative effect on health facility birth. The purpose of the present study is to examine the association of network and community norms with facility birth, following a quality improvement intervention to improve facility births in the Northern and Central regions of Ghana; both regions are representative of the populations impacted by their birth delivery preference. We hypothesized that network and community norms will each be positively associated with, and have an interactive effect on, facility birth. --- Methods --- Study design A Quality Improvement Maternal and Newborn Referral project in Ghana was implemented to improve the referral process for pregnant women and sick newborns in need of comprehensive medical care [20]. We have previously described the study setting [15]. In this study, we examined data from the project's end-line assessment that was conducted in 2015. We used a mixed methods convergent design to separately collect and analyze quantitative and qualitative data within the same period; and the qualitative findings enabled the interpretation of the quantitative findings [21]. As women's perceptions of their social networks are well correlated with measured attributes of their network members [22], we used egocentric network measures to assess rural Ghanaian women's social networks [23]. --- Study sampling and participants The quantitative study was a cross-sectional community-based study using a 3 by N cluster sample design, as detailed elsewhere [18]. Within each of 3 designated districts in the Northern Region , 30 communities were randomly selected to be either an intervention or a control community; the same approach was used in the Central Region . A total of 1,260 women including those with recent birth and those of reproductive age were interviewed. The present study focused on women with pregnancy in the past 3 years , and the analytic sample was 508 after excluding women who had missing information on all key variables of interest. Using a purposive sampling approach, the qualitative study as detailed elsewhere included both in-depth interviews with women and focus-group discussions [18]. We recruited a subsample of the survey participants for the IDIs. A sample of these women's husbands was interviewed. FGDs were conducted with mothers-in-law selected from a community in the district where the IDIs were conducted. Additional IDIs were conducted with leaders of healthcare improvement collaboratives made up of health facility heads. FGDs were also conducted with members of healthcare improvement collaboratives consisting of both health facility workers and community leaders involved in the process of improving the referral process. --- Data collection For the quantitative study, we developed and administered household surveys to the women. Interview guides and focus guides were also developed to collect qualitative data from women, husbands, MILs, and members of the healthcare improvement collaboratives. The IDIs and FGDs were audio recorded in locations selected to protect confidentiality that could be indoors or outdoors or in a private community space and were later transcribed into English. Male and female research assistants, who received training and were experienced in data collection in the local languages of the study communities , conducted household surveys, IDIs, and FGDs in the local languages. This study was exempted from ethics review by University of North Carolina -Chapel Hill's Internal Review Board, as it was considered a program evaluation. All participants verbally consented before participating in the study. --- Quantitative measures The outcome measure, health facility birth, was based on whether women participants delivered at a health facility during their most recent pregnancy. The response was dichotomized, yes or no. Key independent measures included network and community norms adopted from existing measures and review of the network literature [8,[24][25][26][27]. Network norms, family and friend approval of facility birth, were measured as: "how much do your close relatives, and friends, you described in the previous section approve of or encourage the use of health facilities for care during pregnancy and childbirth?" Response options were recoded into higher approval and lower approval . Additionally, network sought facility care, was derived from the question: "How many of the women you know of have gone to the health facility for their pregnancy related care?" The response categories were recoded as greater number , some, and fewer number . Community norm measures were perceived number of women that deliver in facility, men that support facility birth, and MILs' attitudes toward facility delivery. The variables were based on the following successive questions: 1. "How many women do you think in your community deliver their baby in a health facility?" 2. "In your opinion, what percentage of men in your community is supportive of facility delivery?" 3. "In your opinion, what percentage of mothers-inlaw in your community is supportive of facility delivery?" Response options for each question were recoded into majority and minority . Control measures, associated with use of health facility delivery, included maternal age, education, employment, household wealth, religion, marital status, ethnicity, parity, region, and decision-making autonomy [28]. We created the wealth variable based on a similar approach used in previous studies [11,20]. We used three household characteristics -type of fuel used, type of toilet, and location of kitchen-to assess wealth. The poorest households were those that use wood for fuel, have a non-improved toilet and have a kitchen outside the house. We coded households with two out of three of these characteristics as medium wealth, and households with one or none of these characteristics as wealthiest. The autonomy variable was based on the item: "who usually makes decisions about health care for you?" Response options were as follows: respondent alone, partner alone, or others . --- Qualitative instruments We developed interview guides for the IDIs and FGDs. These included questions about women's and their husband's experiences and perceptions of the role of network and community norms in their pregnancy and childbirthrelated care. Husbands were also asked to describe their role as well as their network's involvement in health decisions and support for their wives' pregnancy and delivery. MILs were asked to describe their role as well as their network's involvement in health decisions and support for women's pregnancy, and to describe community perceptions about place of childbirth. Leaders and members of healthcare improvement collaboratives also provided information on their work with community members and women's families to promote facility-based pregnancy care. --- Data analysis Quantitative analyses were conducted in SAS version 9.34 . We examined whether network and community norms and control variables were associated with facility birth using chi-square test and t-test. Multivariable logistic regression models were used to test the association between network and community norms with facility births, adjusting for control measures. We also assessed whether the interaction between community and network norms had an effect on facility birth [29]. The regression analyses were twotailed and adjusted for clustered survey design. Qualitative analysis began with a close reading of all transcripts in Atlas.ti software . Subsequently, the research team, including LEC, KC and CEL, discussed emerging themes from the transcripts, memos, and fieldnotes, which resulted in development of preliminary codebook for thematic analysis. LEC and KC applied the codes to the transcripts and refined the codebook with input from the team [30]. LEC conducted coding checks to ensure that the coded data reflected codes defined in the codebook. Discrepancies identified were discussed and resolved by the research team to ensure coding accuracy and consistency. The research team analyzed and interpreted the data by reviewing the code outputs and developed code summaries and analytic matrices [31,32]. This yielded emerging themes capturing study participants' perceptions of how network members impact women's pregnancy and birth experiences, as well as community perceptions about norms surrounding pregnancy. The matrices enabled comparison of community and network influences. Also, we identified common themes shared by all participants by comparing the experiences and perceptions of women, husbands, MILs, and members and leaders of the health care improvement collaboratives. --- Results --- Quantitative findings -descriptive characteristics Over half of all women were 25-34 years of age . More women who had homebirth had no formal education than women who had facility birth . A greater proportion of women who had homebirth than those who had facility birth did unpaid work or were unemployed , were in the poorest household wealth category , were married or living with a partner , had more children , and indicated that their husband alone made decisions about their health care . In terms of network norms, there was a significant difference between women who had facility and homebirth in their perceptions that close family members and close friends had higher approval of facility-based pregnancy and delivery care . More women who had facility delivery than homebirth perceived that most women they know have had facility-based pregnancy care . Related to community norms, more women who had facility birth compared with homebirth indicated that most women in their community have facilitybased childbirth , and men and MILs in their community were supportive of facility birth. --- Qualitative findings -network and community norms about place of childbirth Overall, 17 out of the 36 women gave birth at a health facility. Interview participants highlighted ways in which both network and community norms played a role in women's pregnancy and delivery experiences. They provided insights into how each type of norm impacted women's place of childbirth. In terms of network norms, various study participants described ways in which network members uniquely informed or directly influenced the birth location experiences of mothers. The role of network norms in women's pregnancy and delivery experiences was mostly expressed by mothers that experienced, or intended to experience, health facility delivery. A notable exception was one mother who maintained that her sister influenced her decision to have homebirth. Most mothers remarked that their network members had successfully experienced facility births, and thus advised them to utilize facility delivery. For example, a NR mother mentioned that her husband and sister influenced her decision to give birth in a health facility. In describing her sister's role, she explained, "We were both pregnant, but she gave birth before me at the hospital, so she advised me to deliver at the hospital. " Several women received this type of advice from various network members. For instance, a CR mother mentioned that her sister-in-law, grandmother, and friend influenced her decision to give birth in a health facility through the advice she received. --- Interviewer: Why was [your sister-in-law] in support to you giving birth at the hospital? Respondent: She said giving birth at the hospital is better than home birth so I should go to the hospital. Interviewer: How about [your grandmother]?… [Friend]? Respondent: She too the same; she said giving birth at the hospital is better than home birth. Another CR mother who received similar advice from her sisters offered insight into the magnitude of influence her family had on her place of birth. She stressed, "In this house, everybody goes to deliver at the hospital. " This quote highlights a distinction between network and community expectations. Network members enforced their normative expectations through the various types of support they offered to women during their pregnancy continuum. In addition to advice from their network, women received resources including money, transportation, food, and help with housework, which enabled them to access and use facility-based antenatal, labor, and delivery care. Related to community norms, many women and husbands perceived that facility or homebirth utilization among women was based on community expectations or perceptions of where women should give birth. Several women echoed the sentiment of a 23-year-old mother : "In this place if it is not critical, you do not go to the hospital. " She explained that although women may sometimes go to a health facility for management of a complicated childbirth, it was customary for women to have homebirth. Therefore, she fully anticipated giving birth at home. Women's adherence to community norms also extended to health facility delivery. Many women and husbands described their interactions with health providers as part of the pregnancy care experience. They considered facility birth an essential part of ensuring the health of both mother and infant. A 37-year-old farmer offered the following clarification. --- Interviewer: Do you know of any woman who gave birth at home and not in the hospital --- The reason [for allowing my wife to go to the health facility] is that I have seen a lot of women in this community delivering there, so if there is problem, they [health providers] will manage it. They are the best people to handle it, if my wife has any complication during delivery. --- Quantitative findings -relationship between norms and health facility delivery The relationship between each network and community norm and health facility birth among women was examined controlling for sociodemographic factors including age, education, employment, household wealth, parity, marital status, religion, ethnicity, region, and decisionmaking autonomy . In terms of network norms, women who perceived that their close family had a higher approval of facility-based pregnancy and delivery care had higher odds of having facility birth than those who perceived their family had a lower approval. Respondents perceiving that most women they know have had facility-based pregnancy care had higher odds of having facility birth than those who perceived few women they know had facility care. For community norms, facility birth was associated with women's perception that most women in their community have had facility birth , and that most MILs in their community were supportive of facility birth . Network and community norms had no interaction effect on location of birth, and thus a final model adjusted for these measures, in addition to the sociodemographic controls . Women's perception that their close family had a higher approval of facility-based pregnancy and delivery care , and that most women in their community have had facility birth , remained significantly associated with facility birth; the strength of associations were attenuated. --- Table 2 Association between network and community norms with health facility birth among women Note: Each cell in the unadjusted models column is a separate model. The Adjusted model column adjusted for network and community norms measures that are significantly associated with facility birth. The models controlled for age, education, employment, household wealth, parity, marital status, religion ethnicity, region, and who usually make decision about your healthcare. *p < 0.05; **p < 0.01; p < 0.001 --- Norms --- Qualitative findings -impact of healthcare improvement collaboratives on norms Women, husbands, and MILs, as well as members of healthcare improvement collaboratives, indicated that health outreach efforts were important in swaying network and community norms toward facility-based delivery. Most perceived that network norms were more resonant in the normative shift toward facility-based care. For example, collaborative members engaged in promoting health facility care for pregnant women indicated that health outreach efforts were designed to cater to both women and their networks as well as the community at large. Through the collaboratives, health providers offered women pregnancy-related health information, antenatal care, and support for facility delivery. As explained by a health care improvement collaborative member , [we accomplish our goals] through health education and effective home visiting. Sometimes during the antenatal meetings, we do health talks on how to prevent anemia, and when we go on home visits too, we do the same thing. Home visits and women's facility care visits provided an avenue for health providers to directly encourage facility birth among women and their network members. In targeting various communities, they also visited churches, mosques, and community gatherings such as durbars or festivals. One collaborative member explained that education to promote women's utilization of facility delivery was accomplished through going to churches and giving them talks . Another participant explained: That they [collaboratives] also have community volunteers; they also gave advice on the food they should be eating, the way they should be living for healthy life and safe delivery. In addition to using large gatherings as avenues to educate community members, health providers worked with trusted and prominent community leaders to facilitate community acceptance of facility delivery. Notably, the head of the collaborative described a community leader as an ambassador to his community as he encouraged women to utilize facility-based pregnancy delivery. --- Qualitative findings -normative shift towards facility birth Participants including women, MILs, and collaborative members contextualized women's experiences by expounding that across many communities there has been a shift from homebirths to facility births. In their view, community norms have begun transitioning from home births to facility delivery due to the increase in education outreach that has enabled women's networks to facilitate pregnant women's use of facility-based care, by advising women and providing them with needed resources . As explained by the head of the collaborative , although initially women and their network facilitated homebirth, Their perception about the hospital or formal health care is now good. We have a number of drugs they will give to pregnant women. But when women give birth at home, they cannot be certain of the drugs the birth attendant will give them. They see the medication in the hospital as something better. Whereas facility or homebirth for some mothers was based on observing customs in their communities, for most women, network members that are traditionally involved in their pregnancy experiences had a prominent role in decisions about their place of childbirth. Most participants acknowledged that many women's interactions with their networks resulted in use of facility delivery. A collaborative member identified the shift in husbands' support from home to facility birth: Most men preferred home deliveries to hospital deliveries before Project Five Alive . But now that we are educating them, they know that hospital deliveries are safer than home deliveries. Many men are now allowing their wives to deliver at the hospital. This quote demonstrates the change in perceptions about facility delivery, and the effectiveness of outreach efforts to reorient network norms toward facility birth. Although it highlights husbands' roles, participants also mentioned that other network members including inlaws were a critical part of the effort to improve women's use of facility-based delivery. --- Discussion We found that network and community norms were independently associated with facility-based delivery among women in rural Ghana. In terms of community norms, women who perceived that other woman in the community have facility birth had a higher odds of facility birth than those who perceived other women did not. The odds ratio for network norms was, however, higher than community norms; women who perceived that their families approved of health facility care had five times higher odds of having facility birth than those who perceived their families did not. The qualitative findings suggest that both network and community norms impacted women's use of health facility care. Health outreach efforts involving healthcare improvement collaboratives influenced the behaviors of women in accessing and using pregnancy-related care including facility-based delivery. However, in alignment with the quantitative findings, study participants suggested the network norms played a larger role in their decisions about facility delivery than community norms. Our finding that community norms are associated with facility delivery is supported by previous research. Specifically, Speizer and colleagues found that Ghanaian women's perceptions that other husbands or MILs in their community supported facility delivery, and their perceptions of a higher number of women in their community who delivered in health facilities, were significantly related to facility delivery [11]. In the present study, the only community norm that remained significant in the adjusted model were women's perceptions that women in their community delivered in a facility. Support from other men and other MILs in their community did not emerge as significant in the final model. While the support of MILs in influencing women's place of childbirth has been well documented [28,33], the present study suggests that perceptions related to where women within the community give birth were more powerful than perceptions of other MILs' , or other husbands' , support for facility delivery. Study findings suggest that communitylevel outreach that highlights the increasing rates of institutional delivery among rural Ghanaian women will be important to further encourage facility-based delivery. Our examination of the relationship between personal network norms and facility delivery contributes to the sparce literature on the role of network norms on facility births. In prior qualitative research, network members, which include husbands, MILs, and mothers, seem to play prominent roles in influencing women's pregnancy and delivery care experiences [28,33,34]. For example, women in rural Burkina Faso and rural Ghana relied on their husband and inlaws to make decisions about their use of facility-based pregnancy and delivery care [28,34]. Resonating with the findings in the present study, the authors of these studies suggested that the influential role of network members included provision of needed resources. Our study, which uniquely quantifies the relationship between network norms and facility delivery, also revealed that network members provided women with critical support to enable their use of facility-based care. Our findings related to the importance of network norms and support from women's personal network specifically underscores the need for continued efforts to promote support for facility-based pregnancy-related care among women's personal networks. For example, health education during home visits, community gatherings, and prenatal care visits could further strengthen the role of network members in facilitating the use of facility delivery among rural Ghanaian women. --- Limitations There are a few limitations worth noting in this study. The quantitative study was based on cross-sectional data; thus, we were unable to establish temporal relationships between network characteristics and use of facility delivery. As our analyses focused on women who had given birth three years prior to the survey administration, there is the potential for recall bias in women's responses and the opinions of network members may have changed after women gave birth successfully at a health care facility. In cases where the gender of research assistants who administered surveys and conducted interviews did not match with study participants, response bias may have occurred. Participants were asked probing questions to address this potential bias. Also, the network data may be limited in that we did not directly survey or interview women's personal networks. The qualitative interviews were conducted in native Ghanaian languages and transcribed and translated into English; errors in translations and nuances in meaning may have been missed. In anticipation of this concern, the research team maintained interview summaries and memos, and met regularly during the data collection process to debrief. --- Conclusion Network and community norms are critical influences on women's use of facility delivery. While we found the perception that most women in the community have facility delivery was associated with facility delivery, network norms were even more pronounced in women's access and use of facility delivery during pregnancy. Our findings indicate that quality improvement initiatives are well positioned to impact both community and network norms in under-resourced communities like those in rural Ghana. These initiatives should focus on highlighting the shifting trend toward facility delivery in rural communities and the role of women's personal networks in supporting facility-based pregnancy-related care. --- --- Abbreviations --- IDIs In --- --- Competing interests The authors declare that they have no competing interests" in this section. ---

Background Reducing pregnancy-related deaths in Sub-Saharan Africa through increases in health facility births may be achieved by promoting community norms and network norms favoring health facility births. However, the process of how both norms shift attitudes and actions towards facility delivery is little studied. We examined the association of network and community norms with facility birth, following a quality improvement intervention to improve facility births in Ghana. Methods A 2015 mixed methods evaluation of a Maternal and Newborn Health Referral (MNHR) project in Ghana included a cross-sectional survey of women (N = 508), aged 15-49 years; in-depth interviews (IDIs) with mothers (n = 40), husbands (n = 20) and healthcare improvement collaborative leaders (n = 8); and focus group discussions (FGDs) with mothers-in-law (n = 4) and collaborative members (n = 7). Multivariable logistic regression was used to examine the association of network and community norms with facility birth. Thematic analysis of the qualitative data was conducted to explain this relationship.The network norm of perceived family approval of facility delivery (AOR: 5.54, CI: 1.65-18.57) and the community norm of perceived number of women in the community that deliver in a facility (AOR: 3.00, CI: 1.66-5.43) were independently associated with facility delivery. In qualitative IDIs and FGDs both norms were also collectively perceived as influencing facility delivery. However, network norms were more influential in women's utilization of facilitybased pregnancy-related care. Healthcare improvement collaboratives were important in swaying both network and community norms toward facility-based delivery by offering pregnancy-related health information, antenatal care, and support for facility delivery.Quality improvement initiatives impact both community and network norms. To be most impactful in advancing facility-based pregnancy-related care, these initiatives should focus on highlighting the shifting trend toward facility delivery in rural communicates and promoting support for facility delivery among women's personal networks.

Introduction It is much easier for a medical professional to build a relationship with a patient if both come from the same cultural and socio-economic background. Patients report greater satisfaction with treatment and improved communication experiences [1]. Research on health inequalities also indicates reduced access to healthcare and discriminatory experiences of cultural minorities: when treating patients from a different cultural and lower socioeconomic background, providers demonstrate implicit bias and prejudice most likely due to a lack of cultural competence or unconscious bias [2], which can have implications on health outcomes [3,4]. Patients often complain in such situations about a lack of empathy and a failure to understand the doctor's communication style, which is also reflected in lower adherence and inadequate use of preventive and health services [5,6]. This compelling evidence suggests that it can be beneficial for healthcare quality if medical professionals represent the communities they serve, particularly because of their ability to understand the social circumstances of health and illness and communicate in a patient-centered way. Continuous societal changes and the increasing diversity of the population require the diversification of the medical profession, whose benefits are often emphasised as a strategy to reduce health disparities and improve access to high-quality care for minority or underserved populations [7]. Diversifying the healthcare force also increases the chances of a doctor-patient cultural and/or socio-economic match as studies show that physicians from minority groups are more likely to practice in underserved communities [8,9]. One way this can be attained is through widening participation strategies aimed at recruiting students from disadvantaged or minority backgrounds. Such strategies include contextualizing admission criteria for health-related undergraduate programs that may be otherwise challenging for disadvantaged students [10]. Studies in the USA report on a lack of ethnical diversity, with black and Hispanic people being particularly underrepresented in the group of physicians [11][12][13][14][15][16]. Recent research suggests that inequalities in access to the profession have not changed in the past decade: concerning students' socio-economic background, an overrepresentation of high-income households has remained stable [17]. This situation is also known in the UK, where 80% of medical school applicants come from a low number of schools and colleges, more likely to be expensive or private [18]. Moreover, up to 34.5% of UK applicants live in the most affluent tenth of postcodes and have better chances of being admitted into medical schools than applicants from less wealthy backgrounds [19]. This evidence suggests that current education systems may be perpetuating issues around inequalities between different population groups. This deserves attention, as it is a societal responsibility. In Germany, 80% of children coming from academic families go into higher education compared to only 27% of children from non-academic families [20]. About 70% of medical students have at least one parent with a university degree [21,22] although in the general population, this level of education is achieved by only 20% of 40-to 60-year-old people. While the population is becoming increasingly diverse [23], the population of medical students remains homogeneous. Thus, medical students do not seem to represent the population whose health problems, beliefs, and needs they are supposed to understand. Whereas in other countries, medical education stakeholders try to address the issue of the unequal representation of population groups in medicine [10,24], the discourse around unequal access to the study of medicine is missing in Germany. At the same time, German studies on medical students' and physicians' socio-economic and cultural background are lacking. In this study, therefore, we explore the diversity of physicians in the city of Hamburg and German medical applicants and students from the socio-economic and cultural point of view. --- Materials and Methods --- Study Design, Participants, and Data Collection This is an observational study using an online questionnaire to explore the socioeconomic and cultural characteristics of participants from two different groups: physicians in Hamburg and medical applicants in Germany. All 15,195 physicians registered with an email address at the Hamburg Chamber of Physicians were invited to take part in the online SEKHHA survey . In the stav survey, we invited 11,287 medical applicants and students who had agreed to participate in research projects upon taking admission tests, of which 2013 were included. In both cases, the invitations were sent by email and the data were collected between June and August 2022. Inclusion criteria consisted of being a member of HCP , applying for medicine or dentistry , and participants answering a sufficient number of questions to determine either their cultural or socio-economic background. --- Instruments We developed an online questionnaire that contained items related to participants' gender, type of secondary school attended, final secondary school grades , and native language. We used previously validated instruments from the national KiGGs study [25] for the measurement of the main outcome variables. The objective socio-economic background was calculated taking into account income, parents' education, and occupation at the time of application for medical school. For the computation of the SEB, point values were calculated for each individual dimension. With regard to education and occupational status, the higher score of the parents was taken into account. For each dimension, a minimum of one and a maximum of seven points were assigned. The individual dimensions were thus included equally in the SEB score. Quintiles were estimated based on the distribution of the score in the general population. For the subjective SEB participants reported on their perceived socio-economic status. As for the description of migratory background, we collected data on participants' citizenship and participants' and their parents' places of birth [25]. Having a migratory background is defined as participants or one of their parents being born in a country other than Germany. We used migratory background as a proxy for cultural background. In order to improve the comprehensiveness and readability of items, the questionnaire was pretested with a group of 10 physicians. --- Data Analysis Assuming small effect sizes, we estimated in a power analysis that a minimum sample size of 600 participants in each survey would be sufficient to detect subgroup differences. Descriptive statistics were used to illustrate the socio-economic and cultural characteristics of participants. We carried out binomial tests to investigate significant differences in the proportion of migratory background distribution when comparing the physicians in Hamburg and the medical applicants and students in Hamburg and Germany to the general population. Whenever the availability of census data allowed it, we age-matched the subgroups of medical applicants and students to the corresponding general population. We focused in our analysis on the four most frequent cultural groups in Hamburg and Germany. To address issues with multiple testing we implemented Bonferroni-Holm corrections [26]. We used predictive mean matching [27] for missing values, particularly for the income variable which had between 35 and 39% missing data. A logistic regression was calculated to explore if the probability of a medical applicant being admitted into a German medical school can be determined by socio-economic and migratory background by adjusting for high school GPA and gender [28]. Although not the main purpose of the study, the logistic regression is relevant to the field of study, particularly with regard to practical implications and recommendations regarding widening participation policies. The four independent variables considered as predictors generally constitute the main criteria based on which international universities recruit widening participation students. In order to more easily interpret odds ratios for numeric independent variables, we calculated the relative likelihood as follows: ORˆ [24]. All calculations were computed with SPSS 27, R 4.1.2. [29], and the R package mice [30]. --- Results In total, 1812 physicians and 3336 medical applicants and students took part in the SEKHHA and stav surveys, respectively. Excluded from the study were physicians who were not HCP members , physicians and medical applicants who answered an insufficient number of survey questions, and candidates who applied for veterinary medicine or pharmacy . Our final sample included 1457 physicians in Hamburg and 2013 medical applicants in Germany . The majority of Hamburg physicians were female, an even higher proportion of women apply for and are admitted into Hamburg medical school . Across all study subsamples, participants were in their majority German citizens, German native speakers, and graduated from grammar school . The high school GPA of applicants and students was on average better than that of Hamburg physicians. The proportion of Hamburg medical applicants who had at least one parent working as a physician was similar to that of Hamburg physicians . Compared to Hamburg physicians, an even higher proportion of medical applicants in Germany relied on their parents' support to finance their studies. The average objective and subjective SEB at the entrance into medical school was similar across all subgroups in the study, with Hamburg physicians having a slightly lower mean objective SEB compared to medical applicants and students . The objective and subjective SEB correlated at ρ = 0.6 . Figure 1 illustrates that the lowest three quintiles of objective SEB are vastly underrepresented in all subsamples of the study and in particular amongst applicants and students admitted in Hamburg. Physicians, applicants, and students with migratory background are underrepresented compared to the general population and these differences are statistically significant . People with Turkish roots are underrepresented amongst Hamburg physicians and German medical applicants and students . The proportion of applicants with Polish roots who apply for medicine in Hamburg is also lower than the one in the general Hamburg population with migratory background . The Afghan community is underrepresented in the population of physicians in Hamburg . As shown in Table 2, compared to the general population in Germany aged 20 to 25, a much lower proportion of young people with Syrian roots in Germany apply for and are admitted into medical schools . Overrepresented in the SEKHHA study were Hamburg physicians with Iranian and Austrian roots . In the subgroup of applicants in Germany, we similarly found an overrepresentation of people with Iranian and Austrian roots; these proportions remained almost the same after admission into medical school: Iranian roots , Austrian roots . High school GPA, objective SEB, and gender contributed significantly to the probability of being admitted into medical school according to the logistic regression model which correctly classified 64.8% of cases overall . For one unit change in the SEB, there is an Exp change of 1.07 in the outcome, which means that students with the highest SEB score are 3.1 times more likely ) to be admitted into medical school than students with the lowest SEB score, after adjusting for high school GPA and gender. The effect of the migratory background was not statistically significant in our sample. We additionally conducted a sensitivity analysis to test how the probability of being admitted into medical school can be predicted by migratory status, gender, and SEB alone; in this regression model, the Exp for SEB was 1.085 while migratory background and gender became statistically insignificant. --- Discussion In this study, we present evidence on the diversity of physicians, medical students, and applicants, which has not been available so far for Germany. --- Socio-Economic Diversity The great majority of medical applicants, students, and physicians came from the top two quintiles of the SEB score. The overrepresentation of the highest quintile was particularly manifest amongst medical students indicating that inequalities in accessing medical studies in Germany have at least maintained if not worsened, a finding consistent with existing international evidence [31]. Moreover, as similarly found in a Swedish study [32], medicine appears to stay in the family, as one in five study participants had at least one parent that was a doctor. As in many other countries [33], high school GPA is an important selection criterion for medical students in Germany and can predict study success [34]. However, young people with low socio-economic status are strongly underrepresented among school leavers with good grades that are needed for admission into medical school. The link between social class and education is particularly pronounced in Germany where PISA assessment results depend on individual socio-economic status to a greater extent than in the other OECD countries on average [35]. Therefore, the underrepresentation of the lower SEB among medical applicants could also result from self-selection, as pupils from underrepresented groups may not even consider studying medicine or may not have the confidence to do so. --- Cultural Diversity In line with international studies, our research indicates that Hamburg physicians and medical applicants in Germany form a culturally homogenous group [14,19]. The underrepresentation of participants with migratory background in German higher education has been previously reported on in education reports: persons of the same age with a migratory background are less likely to have a university degree and more likely to have no vocational qualification [36]. One explanation for this type of underrepresentation is that migratory background is closely linked to socio-economic status. This fits our observation that the underrepresentation is especially dominant for cultural groups with a low SEB in Germany [17,37]. It is for example well known that people with Iranian roots living in Germany are amongst the most educated and generally have a very good income [38]. Our logistic regression model similarly suggests a higher effect of socio-economic status compared to migratory background. --- Practical and Political Implications The underrepresentation of certain groups in the medical workforce has two main implications: questionable fairness in the selection of medical applicants and worse quality of healthcare. Regarding fairness in student selection, each applicant should have a fair chance to enter medical school, regardless of socio-economic or cultural background. German politics can benefit from experiences in other countries where strategies to increase the chances of students from disadvantaged backgrounds to study medicine have been implemented as briefly mentioned in the introduction under the name of "Widening Access" or "Widening Participation" [10,39,40]. In Canada, it is ensured that members of underrepresented groups are involved in the development and assessment of admission procedures [41,42]. In the Netherlands, the selection by lottery will be reintroduced after years of using admission tests [43]. In the USA, quotas and bonuses for minority groups are used [44,45]. This "affirmative action" approach is controversial and subject to legal action. In the UK, adjusted criteria for applicants with low SEB are implemented for contextualized admissions; in one medical school, every applicant's high school GPA is considered in comparison to the average school attainment, which proved to have a significant effect on widening participation [46]. Moreover, recruitment campaigns are organized in disadvantaged areas to encourage pupils from underrepresented groups to inform, empower, and raise the aspirations of these groups to apply to medical school. A combination of mentoring programs and summer schools has proved to be successful [10]. Student selection in Germany is highly regulated by law. A sentence of the constitutional court in 2017 has led to a reform of selection rules, which medical schools have to apply. The question remains whether there is a political will to create a legal framework that could accommodate the implementation of widening participation approaches in Germany. Following the argument of healthcare quality, the need for a diverse healthcare has been emphasized in German health policy documents which underline the importance of designing "diversity sensitive healthcare" [47]. A diverse healthcare force has the potential to improve the cultural competencies of healthcare staff and the health outcomes of patients. Although there is no guarantee that physicians from a minority group will serve in their own community, international studies show that this can be achieved [8,9]. Human resources for health migration is a phenomenon that partially led to increasing diversity; however, decision-makers need to consider the disastrous consequences for the health personnel's country of origin such as insufficient healthcare delivery and increased mortality [48,49]. The current evidence suggests that strategies to obtain a more diverse healthcare force should rather focus on the recruitment of local talent. Medical education alone cannot solve the issue of underrepresentation of population groups in higher education and widening participation is surely an issue that needs to be addressed by all sectors of society, particularly the school system. Widening participation interventions need to be initiated at an early educational stage so that each child can optimally develop their abilities. However, in order for their graduates to be able to address the diverse preferences, communication needs, and demands of all patients, medical schools must consider criteria other than merit and academic achievement. --- Limitations With participation rates below 20%, we cannot state with confidence that our findings are generalizable. Our sample size calculation indicated that the sample is sufficient to generate solid findings; however, some comparisons for certain cultural groups were not meaningful because of the very small number of cases in each subgroup. Future research should be conducted at the national level to gain a better insight into the diversity of the medical workforce and studentship across Germany. We have not performed any parity adjustment for income in the case of people whose parents may have been employed in a different country; therefore, we may have underestimated some participants' socio-economic background scores. For the income variable, we had a high number of missing data ; given the good correlation between the objective and subjective socio-economic background, researchers should consider in the future the use of subjective measures. Alternatively, the school social index could also constitute a good measure for socio-economic disadvantage. Our study could not establish a direct causal link between the socio-economic status and access to the medical profession. Further research is needed to investigate what determines students' ambition and motivation to study medicine. We used migratory background as a proxy for the cultural background as it is generally the common way of reporting on ethnicity used in Germany. Although an additional variable on participants' sense of belonging to a particular culture has been collected for the SEKHHA study, it did not appear to be a reliable variable and therefore it was not used in the stav study, nor reported in the current publication. Although we report on participants' gender, the current study is mainly focused on the migratory background and SEB; subsequent studies should investigate the role of gender, particularly in the context of an increasingly female-dominated medical profession. Finally, there are different dimensions of diversity that deserve researchers' attention such as, for example, personality traits and risk or time preferences, which can influence treatment choices and adherence [50,51]. --- Conclusions People from lower socio-economic backgrounds and minority groups are underrepresented amongst physicians, medical applicants, and students in Germany. This may be a result of an unjust education system, which mostly favors affluent households and potentially discriminates against migratory backgrounds. The results of the current study could guide the development of widening participation strategies aimed at increasing the diversity of the healthcare force. Such strategies could include actively recruiting and sustainably supporting disadvantaged students to guarantee their admission and success in medical schools. --- Data Availability Statement: The data presented in this study is not publicly available and cannot be shared due to privacy and ethical restrictions. ---

Medical professionals who represent the communities they serve are in a better position to understand patients' social circumstances and communicate in a more patient-centered way. International studies show limited diversity and underrepresentation of certain social groups in the population of physicians and medical students. We designed an observational study to investigate the cultural and socio-economic diversity of physicians and medical applicants in comparison to the general population in Germany. We invited 15,195 physicians in Hamburg and 11,287 medical applicants in Germany to participate in an online survey between June and August 2022. The lower three quintiles of objective socio-economic background (SEB) were vastly underrepresented in all subsamples of the study and in particular amongst applicants and students admitted in Hamburg: 57.9% of physicians and 73.8% of medical students in Hamburg originate from the top quintile of SEB. The Turkish and Polish communities were particularly underrepresented in the group of physicians from Hamburg and medical applicants and students in Germany (p = 0.02; p < 0.001). In line with existing evidence, the vast majority of physicians and medical students come from the most affluent households when entering medical school. Widening participation strategies are needed to facilitate fairer access to the study of medicine in Germany.

INTRODUCTION Family mealtime is a salient context for promoting child and family health. Shared family mealtimes have been linked to better child health outcomes, including higher consumption of healthy foods, lower consumption of unhealthy foods, and healthier eating patterns. 1 Recent longitudinal research has demonstrated that frequency of family meals during childhood predicts reduced risk for obesity in adulthood. 2 Indeed, recommending that families share meals together each week is a key child health promotion message. Despite the well-documented importance of family mealtimes, television and other electronic media use during mealtime may compromise its potential benefits. For example, mealtime TV and other electronic media use is associated with poorer overall dietary quality of foods served at meals 3 and overall dietary intake. 4 To prevent unfavorable mealtime practices in childhood, it is essential that pediatricians identify, early in child development, which families may be at greater risk for TV use during mealtime. Identifying these early risk factors will enable tailoring of obesity prevention messages provided during well child visits. Characteristics of both children and their parents may play a role in family mealtime practices. Children with more challenging dispositions, for example those prone to negative emotions and lability, may be more likely to be pacified with media during meals. Likewise, parents lacking effective discipline strategies may use TV to reduce conflict, or simply be more permissive with TV use during mealtime. Although prior studies have examined child temperament and/or maternal factors as predictors of screen time in infants and young children, [5][6][7] it is not yet known whether child negative emotionality or poorer parenting practices predict later mealtime TV use in middle childhood. Another gap in the literature is that prior studies have used parent reports of screen time, whereas observational methods may provide a more nuanced picture of children's meal time TV exposure. The current study seeks to fill these gaps by examining whether early childhood factors predict observed child TV exposure and TV engagement during mealtime in middle childhood. --- METHODS --- --- Procedure Three typical dinnertime meals in the home were video-recorded by mothers over a oneweek time span. Mothers were provided with a camera and instructions on how to do these recordings. Instructions were to record the entire meal and the child's upper body, and that the child's plate and drink should be in view throughout the video. --- Observational Coding The second dinnertime meal recorded was selected for coding. Trained research assistants coded each video in 10 second intervals indicating whether the TV was audible and whether the child attended to the TV screen or not . The research assistants were trained to reach a reliability criterion of a least Cohen's κ >.70 . After achieving reliability , the coders independently coded the remaining videos and double coded 20% of videos to protect against coder drift. These interval codes were summed across the meal and collapsed into three categories: no TV exposure, background TV exposure , and TV engagement . --- Measures Suboptimal discipline practices were measured using the Parenting Scale. 9 This scale measures lax and harsh parenting practices in the context of managing child behaviors. Items include, "When my child does something I don't like… I often let it go," "When I'm upset or under stress…I am picky and on my child's back," and "When my child misbehaves…I get so frustrated or angry that my child can see I'm upset." To get an overall score of dysfunctional parenting practices, we took the mean of all items . Response options ranged from 1 to 7, with higher scores reflecting poorer parenting practices. Child negative emotionality was measured using the Emotion Regulation Checklist Lability/ Negativity subscale, 10 which was also completed by the mother. Items on this scale include: "is prone to angry outbursts, tantrums easily" and "responds angrily to limit setting by adults." Response options ranged from 1 to 4 . The mean of the items on the Lability/Negativity subscale was calculated, with higher scores indicating greater negative emotionality. This scale evidenced good reliability in the current study and correlated with suboptimal discipline practices at r = .13 . --- Data Analysis To test whether early childhood negative emotionality and discipline practices predicted relative risk for background TV exposure and child TV engagement during mealtime compared to no TV exposure , we conducted a multinomial logistic regression, adjusting for child sex , child age, child race/ ethnicity, and maternal education level . Negative emotionality and suboptimal discipline practices were entered in the same model as the independent variables. We chose the reference category of no TV exposure given that research has indicated that both background TV and TV engagement may have negative effects on child health. 11 As such, we were interested in whether these two separate TV use variables were predicted by early childhood factors. --- RESULTS --- Sample Characteristics See Table 1 for demographic characteristics of sample. Half of the children in the sample did not have TV on during mealtime. Approximately 21% of children had background TV exposure. The remaining proportion of children had TV engagement ; on average, this proportion of children engaged with TV during 43% of their coded mealtime intervals . No significant differences in demographic variables or baseline measures of interest were found between participants with baseline only data versus participants with family mealtime observations at the second wave. --- Predicting TV Exposure and Engagement The multinomial logistic regression examined whether negative emotionality and suboptimal discipline practices were associated with increased risk for TV exposure during mealtime, relative to the reference category of no TV exposure. Neither factor predicted an increased risk for background TV exposure. However, both child negative emotionality and suboptimal discipline practices independently predicted a greater risk for child to engage with TV during mealtime compared to no TV exposure . Children higher in negative emotionality at preschool were more likely to engage with TV than to have no TV exposure during mealtime later in childhood. A one unit increase in negative emotionality resulted in 2.11 times increase in risk of engaging with TV versus no TV during mealtime later in childhood. Children of mothers who utilized suboptimal discipline practices when their child was younger , had a greater risk of engaging with TV during mealtime relative to no TV exposure during mealtime ). There was no interaction between suboptimal parenting and child negative emotionality on risk of TV engagement versus no TV exposure. --- DISCUSSION The objective of this study was to identify early childhood predictors of TV exposure during mealtime-an important context for pediatric obesity prevention. We found that both child and parent characteristics predicted child TV engagement during mealtime. Children with greater negative emotionality in early childhood were more likely to engage with TV during mealtime than to have no TV. Similarly, early parenting disciplinary practices characterized by over-reactivity and laxness increased the risk for child TV engagement versus no TV during mealtime approximately two years later. Children who have mood instability and are prone to negative emotions , may be challenging to parent in certain contexts, such as mealtime. Organizing and managing family mealtimes can be stressful; 12 attempting to feed or engage with a child with mood instability could amplify the stress of mealtime. Indeed, one challenge reported by parents in carrying out family mealtimes is parent-child conflict and power struggles between the parent and child. 13 Thus, it is plausible that families of children with poorer emotion regulation may attempt strategies such as allowing television viewing during mealtime in order to prevent conflict and achieve mealtime goals. Suboptimal discipline practices during early childhood also predicted increase risk for child TV engagement during mealtime two years later. There are several potential mechanisms by which an overreactive and lax parenting style could promote TV engagement during mealtime. First, parents who over-react to child behaviors may experience mealtimes as particularly overwhelming and stressful. For families already experiencing multiple stressors, such as the low-income families in our study, TV during mealtime may be a means by which to create a quiet/calm atmosphere . It is also possible that with a lax or permissive parenting style, rules about TV exposure during mealtime may not be enforced or implemented. Indeed, this has been found in another study wherein lax maternal parenting style was associated with greater amounts of child TV time. 14 It may also be that harsh parenting early in childhood may engender a stressful emotional climate in the home and/or foster less communicative parent-child interactions over time. As has been suggested by others, 15 harsh/over-reactive parenting style may be stressful for some children. Thus, in the context of mealtime, children may attend to TV instead of engaging in potentially stressful or unrewarding interactions with their caregivers. This may also explain why we found that dysfunctional parenting practices only predicted child TV engagement and not simply whether the TV was on and the child did not watch it. It is important to distinguish that these early childhood characteristics did not predict background TV exposure, relative to no TV, during mealtime-only whether the child attended to TV during mealtime. This could be explained by the fact that background television or "constant television households" are more common among low-income families. 16 Half of the families in the current study had audible television programming during the family meal. However, our results show that of the families with TV on, approximately 43% of children did not attend to the screen at all during their meal. It is possible for these families, TV noise is a normative background sound in their environments. Although the observational methods are a strength of this study, there is a possibility for the Hawthorne effect, wherein some parents may have not allowed TV to be on during the mealtime because of the video recording. These results are limited to TV exposure during mealtime but it is possible these early predictors of mealtime TV exposure may also apply to new mobile technology. With the recent trends in children's mobile device ownership, 17 it will be important for future research to also assess children's use of other screen media, as it is possible that these parent and child risk factors may also be precursors to mobile device use during mealtime. It should also be acknowledged that other caregivers responsible for child feeding were not assessed in the current study. Results are specific to the primary caregiver but future research should assess whether parenting practices of the other caregivers may also associate with mealtime TV use. It is important to acknowledge that these results are based on low-income families, the majority of whom are non-Hispanic White. Thus, our findings may not generalize to families from higher income brackets or of different racial/ethnic backgrounds. However, given that obesity disproportionally affects children from low-income backgrounds, our findings still have important implications for those most in need of obesity prevention strategies. When treating children with emotion regulation difficulties, pediatricians could elicit from parents how they manage mealtimes and provide alternatives to TV use to manage child difficult behavior. Likewise, messages about no TV use during mealtime may be especially important to parents who have over-reactive and lax parenting styles. In terms of obesity prevention interventions, improving early childhood emotion regulation and the emotional climate of family mealtime may mitigate the risk of unfavorable mealtime practices, such as child TV engagement. Future research should examine whether improving parenting practices and child emotion regulation reduces children's TV use during mealtime and lowers obesity risk. --- CONCLUSIONS Both child emotion regulation and dysfunctional parenting practices in early childhood predicted an increase in risk for child TV engagement during dinner time two years later. Low-income families with these risk factors could be identified early to prevent obesitypromoting mealtime practices. Interventions may target general parenting style and child emotion regulation as one means by which to reduce mealtime TV use. --- WHAT'S NEW? Both child negative emotionality and dysfunctional parenting practices in early childhood predicted an increased risk for child TV engagement during mealtime two years later. Families with these risk factors could be identified early to prevent unfavorable mealtime practices. Note. Adjusted for child race/ethnicity. Child sex was coded as male = 1; female = 2; maternal education level was coded as 0 ≤ high school degree/GED and 1 > high school degree/GED. --- Miller. The funders of this work did not contribute to, participate in or influence the study design, collection, analysis or interpretation of data, writing of the manuscript or the decision to submit this manuscript for publication.

OBJECTIVE-To identify whether child and mother characteristics in early childhood predict TV exposure and engagement during mealtime in middle childhood. METHODS-A total of 220 low-income mother-child dyads participated. Children were 4.26 years old (SD = 0.51) at baseline and 5.94 years (SD = 0.68) at two-year follow-up. Mothers completed baseline measures of child negative emotionality and parenting practices. Family mealtimes were video-recorded and coded for background TV exposure and child TV engagement. Multinomial logistic regression tested whether child emotionality and parenting

Among im/migrants, defined as both legal immigrants as well as migrants who move from one country to another but lack legal status , women often face greater difficulty accessing health care than men . Reasons for this difference may include overlapping social determinants of health such as gender roles, social class, employment options, and legal status that differentially influence access to services for women . In Canada, women outnumber men in 'dependent' categories of immigration , and have been shown to face pervasive challenges accessing conventional labour markets in Canada resulting in overrepresentation in informal sectors of work such as sex work . Despite facing acute health inequities and pervasive barriers to health access, patterns of health access and unmet health needs among im/migrant sex workers remain poorly understood . Globally, researchers have documented that SWs frequently experience unmet health needs including suboptimal preventive care testing and cervical screening) and inadequate treatment following abuse, rape, or assault . Canadian researchers have also suggested that SWs have inadequate access to sexual and reproductive health services and cervical screening . As many of these unmet health needs are persistent in settings where high-quality health care is available and contact with general practitioners is frequently reported in the general population , health care barriers faced by SWs can be attributed to negative attitudes of health care providers, fear of being judged within health care settings, as well as the broader stigma and criminalization that shape SWs' everyday lives . Researchers in Vancouver, Canada have found that im/migrant SWs are more likely to face reduced access to certain health-related and preventive services, including HIV and Hepatitis C Virus testing, as compared to their Canadian-born counterparts . In addition to sex work-related barriers to health access, im/ migrants in sex work may face additional barriers related to language differences, unfamiliarity with or lack of information regarding local health systems, concerns regarding legal immigration status, lack of insurance coverage for recent arrivals, financial barriers, isolation from health services, and differing cultural perceptions of health . Currently, limited attention has been directed to the type or duration of migration, gender-, or occupation-specific experiences in im/migrant health research -researchers have historically regarded im/migrants as fairly homogeneous . However, im/migrant populations are diverse and evidence suggests that dynamic changes in legal status , access to housing, social networks, and social integration experienced over the course of migration are highly linked to changes in health status . For example, the "healthy migrant effect" proposes that some types of im/migrants are often healthier than non-migrants at the time of migration and increasingly adopt behaviours and health outcomes that are more similar to non-migrants over time . At the same time, however, theories on the disruptive effects of migration posit that disruptions can take place immediately following arrival in a new destination, and can attenuate for some im/migrant groups over time . These hypotheses have rarely been examined among im/migrant SWs internationally, with a particular paucity of evidence within North American settings; moreover, the majority of studies to date have been cross-sectional and qualitative in nature. Investigating patterns of health access and unmet health needs among SWs, particularly in relation to im/migration experiences, remains critically needed in order to inform sex worker and im/migrant-tailored approaches to health services delivery in immigrant-receiving countries internationally. Findings have important policy and programming implications for health care providers, public health specialists, and policy makers in immigrant-receiving countries given trends towards increasing patterns of immigration and globalization worldwide . As such, we seek to investigate the relationship between im/migration experience and other socio-structural factors on unmet health needs among SWs in Metro Vancouver over time. --- METHODS --- --- Data Collection SWs completed interview-administered questionnaires by a trained interviewer and HIV/STI/HCV serology testing by a project nurse at enrolment and biannually. The main questionnaire elicited responses related to socio-demographics, sex work patterns, physical work environment factors, and social/interpersonal and structural environment factors. Geographic data were also collected to understand spatial trends in health patterns. Biolytical INSTI rapid tests were used for HIV screening, urine samples were collected for gonorrhea and chlamydia, and blood samples were tested for syphilis, herpes simplex virus-2 antibody, and HCV. All participants received an honorarium of $40 CAD at each biannual visit for their time, expertise, and travel. The study received ethical approval through Providence Health Care/University of British Columbia Research Ethics Board and continues to be monitored by a Community Advisory Board comprised of more than 15 community agencies . --- Dependent Variable Unmet health need was assessed based on the question: "How often can you get health care services when you need it, in the last 6 months?" Answers included: always , usually , sometimes , occasionally , never, and non-applicable . Responses were grouped into binary categories of having experienced unmet health needs: "yes" versus "no" . --- Independent Variables Time-fixed demographic and individual-level variables measured at baseline included: age , gender/sexual minority , Indigenous , im/migration experience , high school completion or greater , and duration of sex work . Time-updated individual and biological factors included injection and non-injection drug use, and HIV seropositivity. Interpersonal factors considered were: physical and sexual intimate partner violence and lifetime abuse/trauma . Time-updated socio-structural variables with occurrences in the past 6 months included: homelessness; current unstable/transitional housing status; police harassment including arrest; poor treatment by health care professional; and community threats/assaults of sex work). Work environment factors considered were social cohesion and primary place of servicing clients . --- Statistical Analyses Baseline descriptive statistics including frequencies and proportions for categorical variables or measures of central tendency and variability were calculated for all variables and stratified by whether participants reported unmet health need in the past 6 months. Differences between SWs who reported unmet health need and those who did not at baseline were assessed using Pearson's Chi-square test for categorical variables and the Mann-Whitney U test for continuous variables. Bivariate and multivariable generalized estimating equations with a logit link function and exchangeable correlation structure, that account for repeated measures by the same respondents, was used to prospectively examine independent correlates of events of unmet health need over the 4-year observation period. Potential confounders as described in previous literature, factors hypothesized a priori to be related to unmet health needs , and variables with a significance level of less than 5% in bivariate analyses were considered for inclusion in the multivariable model. Model selection was done using a backward process, with the final model being selected as the one with the lowest quasi-likelihood under the independence model criterion value, as previously described by our group . Analyses were performed using the SAS software version 9.4 . All p-values were two-sided. --- RESULTS Among 742 street and indoor SWs, a quarter reported unmet health needs at least once over the 4-year study period, contributing to 255 reports of unmet health needs out of 2602 observations included. Of the 742 participants, 559 returned for at least one follow-up visit, with a median of 3 follow-up visits and median of 21.2 months under follow-up. The median age of participants at baseline was 35 years . Approximately one-quarter were im/migrants to Canada, with 10.5% of the cohort being recent im/migrants and 13.8% being long-term im/migrants at baseline. In bivariate GEE analyses over the 4-year period, elevated odds of unmet health needs was significantly associated with recent im/migration , long-term im/migration , police harassment including arrest , and lifetime abuse/trauma . Participants with a shorter duration in sex work , who used non-injection drugs , were living with HIV , and experiencing unstable/transitional housing were less likely to experience unmet health needs. In the final multivariable GEE model , recent im/migration , long-term im/migration , police harassment including arrest , and lifetime abuse/ trauma remained significantly and independently associated with elevated odds of unmet health needs in the last 6 months. --- DISCUSSION Our results reveal that recent and long-term im/migration, historical violence and trauma, and policing were linked to enhanced unmet health needs among SWs in Vancouver, Canada. Recent im/migration to Canada had the strongest independent effect on unmet health needs among SWs, with a three-fold increased odds of unmet health needs as compared to non-migrants. Long-term im/migration was also associated with an almost two-fold greater odds of unmet health needs. Our findings are consistent with conclusions drawn by Vancouver-based researchers who show connections between policing and workplace violence with reduced access to health services for SWs . We are one of the first groups to examine unmet health needs within the context of im/migration experience among SWs. The influence of recent and long-term im/migration on unmet health needs suggest the need for increased attention to barriers to health access faced by im/migrant SWs, which may arise from both im/migration-related barriers as well as the barriers faced by SWs more generally . While some researchers have documented that immigrants typically report no difference or a decreased risk for unmet health needs in comparison with non-migrant populations , our findings are in line with international qualitative and epidemiological research elucidating the disruptive impacts of im/migration on health care access for marginalized women . The effect of im/migration experience on unmet health needs of SWs is likely related to changes in the social and structural determinants of health over time. For example, im/migrants may adopt health behaviours of the local-born population over time -with health profiles and health care access of im/ migrants becoming more similar to non-migrants as duration of residence in the destination country increases . Greater unmet health needs experienced by recent im/migrants may be explained by a combination of concerns regarding low socioeconomic status, precarious legal status and working conditions, loss of social networks, unstable housing, and language barriers . Among im/migrant SWs, these inequities may be magnified by macrostructural determinants related to sex work, including stigma, discrimination, and criminalization of sex work . Specifically, institutional barriers that may affect obtainment of health services for im/migrant SWs include fear of disclosing sex work to health providers, foregoing care in order to continue earning income, denial or delay of public health insurance, high cost of private health insurance, and ineligibility for social assistance and subsidized housing . Not having a provincial health insurance card was strongly correlated with reporting institutional-level barriers to health care among SWs, as determined by a researcher in Vancouver . This is especially relevant among recent im/migrants awaiting the approval of provincial health insurance and those with precarious status . Additionally, our findings may relate to shifting gendered power dynamics in relation to immigration status. Women who move to Canada are twice as likely to be classified as 'dependent' immigrants . Classification as a 'dependent' immigrant may relate to specific health-related social determinants that have been linked to increased barriers to accessing services and adverse effects on physical and mental health . We recommend further research that investigates the intersections between legal status over the course of migration, unmet health needs, and health outcomes in order to examine how these dynamics change over time -for example, as im/migrant women improve their language abilities, build social networks and potentially gain enhanced access to health care services. Finally, our results illustrate the importance of the broader determinants of im/migrant and non-migrant SWs' unmet health needs, including lifetime violence/trauma and policing. Researchers studying SWs' health have documented that law enforcement approaches to sex work exacerbate HIV risk and barriers to health services by isolating and displacing SWs, effectively pushing SWs away from accessible health and support services . For example, due to police activity, street-based SWs who inject drugs avoid health facilities and syringe exchange services, thereby interrupting treatment and prevention efforts . Police presence has also been linked to SWs' mistrust of authorities and fear of arrest under criminalized sex work contexts that may then discourage obtainment of health services . Findings of our study reveal links between partner, client, and communitylevel violence to barriers and reduced access to health services . --- Recommendations We found that the disruptive effects of im/migration may affect health care access, thus contributing to unmet health needs of im/migrant SWs. As such, we find it important to promote integrated health models that include community and social support servicesstrategies that enhance culturally appropriate, community-based delivery of health services for im/migrant SWs. Examples include SW-only drop-in centres and other low-threshold services tailored towards im/migrant SWs , as well as outreach to hard-to-reach women . Language-specific counselling and support services that address sexual, physical and emotional abuse as a result of violence and trauma should be made available. We also recommend policy reform in the area of immigrant health to ensure that im/migrant women are connected to care upon arrival to Canada and receive continued care during the transitions of im/migration. Additionally, attention to broader social determinants that shape health access for all SWs, at the macrostructural and work environment levels , is needed . There is a continued call to remove punitive measures against SWs and clients in order to increase support and implementation of health and social interventions for all SWs, as well as to improve police relations . --- Limitations While researchers have previously established that health-seeking behaviours are different among individuals with different immigration statuses , our analysis did not include information on legal immigration status due to limitations on currently available data. To achieve a more nuanced understanding of im/migrants' experiences in sex work, we recommend that future studies of unmet health needs and health access among im/ migrant SWs should focus on changes in legal status over the course of migration. A major strength of our study is the prospective longitudinal design and GEE analysis as it accounts for repeated measures by the same respondent. While our study findings may not be fully generalizable to other sex work settings , our sample included SWs from a wide range of sex work environments, including streetbased, indoor, and online spaces. Since our analysis included a series of sensitive topics , cases of underreporting may have occurred as a result of social desirability bias. However, questionnaires were conducted in safe spaces by experiential and non-experiential outreach staff where a strong community rapport exists as a result of weekly outreach to outdoor and indoor sex work environments. --- CONCLUSION We found that im/migration experience has differential effects on unmet health needs of SWs and suggest that accessible and consistent health services during the early years of migration are crucial to ensure that SWs' health needs are met. Our findings underscore the importance of comprehensive structural interventions for both im/migrant and non-migrant SWs, including changes across immigration, health and sex work policy . At the community and institutional levels, cultural and language-specific health and support services should be made available for and developed in partnership with im/migrant SWs. While our study is situated in the Canadian context, findings have the potential to inform immigration policy that promote equitable and adequate access to health care in Canada and other immigrant-receiving countries. ---

Despite universal health care in Canada, sex workers (SW) and im/migrants experience suboptimal health care access. In this analysis, we examined the correlates of unmet health needs among SWs in Metro Vancouver over time. Data from a longitudinal cohort of women SWs (AESHA) was used. Of 742 SWs, 25.5% reported unmet health needs at least once over the 4-year study period. In multivariable logistic regression using generalized estimating equations, recent im/migration had the strongest impact on unmet health needs; long-term im/migration, policing, and trauma were also important determinants. Legal and social supports to promote im/migrant SWs' access to health care are recommended. Globally, it is estimated that there are over 240 million international migrants worldwide ('Migration | UNFPA -United Nations Population Fund', n.d.), with women representing almost half of all migrants internationally ('Migration Facts and Figures', 2015). Women are migrating for purposes of seeking new economic and social opportunities, family reunification, and improved health and security (Goldenberg, Liu, Nguyen, Chettiar, & Shannon, 2014;Platt et al., 2013;Zimmerman, Kiss, & Hossain, 2011). However, resettlement in destination countries often results in a number of barriers to health for women including language barriers, discrimination, and limited access to services; and migrant women are consequently disproportionately represented within precarious forms of

INTRODUCTION Human interaction lies at the heart of social life: a central feature of the human condition is that our daily life is spent in the presence of others . Over the last two decades, social encounters in everyday life have diversified significantly, taking on new and different forms due to social media. The Internet has introduced new possibilities for communication, and to be co-present , and the social web i now plays a significant role in the social relationships of many individuals. With the popularity of the social web and mobile devices, people have at their disposal a broad array of technologies for communicating with others and being social. Today, in the north part of the globe, for some of us it is difficult to imagine social relationships without access to the social web. Emotions are key to the dynamics experienced in social relationships, and this is also valid for those social encounters that take place online . In this study, emotions are understood as relational, i.e. relationships and practices are involved in the production of emotions . Daily acts of emotion constitute one important thread in the fabric of civilization , given that all social relations-from the simplest face-to-face encounter to complex social interactions-involve emotional responses. People might express emotions online and are able to perceive the emotions of others online too . In this way, emotions could be understood as 'patterns of relationships' between individuals or groups of people . Consequently, the ways in which emotions are communicated online surface in 'patterns of relationships […] and these also result in patterns of activity that can become dispositions-ways of acting in particular situations […]' . These patterns of relationships can vary for particular social groups and settings. The present study is interested in beginning to explore the patterns of activity in expressing emotions during social interactions online, as well as the factors that affect them. Recent research suggest that interaction rituals are possible online to a micro-level. Now, this study tries to extend those results to the mesolevel of analysis and offer potential for further macro-level analysis. Individuals consider each other's attitudes, and rehearse actions and social acts online. Groups of individuals generate interaction rituals that are common to them . Considerable variation exists in the experience and display of emotions in F2F interactions . This variability suggests that, to an important degree, subjective experiences and emotional beliefs are both socially acquired and socially structured. The social web could be expected, then, to have a role in shaping emotional expression online. Collins ' theoretical approach suggests that assembling human bodies in F2F interaction allows an interaction ritual to be constructed. However, recent theoretical developments suggest that co-presence is possible online too, entailing a wholly different and complementary set of strategies-and interaction rituals-to, for instance, signal immersion on the online encounter . Moreover, in online encounters, emotional expressions can emerge and can become particulars ways of acting, affected by social factors and socio-cultural characteristics . This article is particularly concerned with exploring the extent to which emotional expression during online interactions is affected by two social factors-age and gender-and by individuals' level of engagement with the social web, in a particular socio-cultural context. The article is structured as follows: the Literature Review discusses existing research on age, gender, social web engagement and emotion, and provides the basis for developing a conceptual framework for the analysis of emotional expression online; Research Questions, Data and Method introduces the data and methods used to contrast the framework empirically; the following two sections outline and discuss the results; the final section presents conclusions and future directions for research. --- LITERATURE REVIEW In earlier Internet studies, the dominant view was that emotions are very difficult to express online . More recently, the study of the relationship between emotions and the Internet has become a central theme of analysis: a trend that has been aided by the proliferation of public data-mainly textual-that social web applications make available, and by the advances in techniques that can be used to obtain and analyse such data, like text mining, natural language processing and sentiment analysis . Several studies argue that emotions can be expressed just as well online as in F2F situations . Some studies go even further, suggesting that emotions can be expressed more openly online, given the lack of social constraints and the 'disinhibition factor' that the Internet may offer . There is currently a limited but growing body of research that deals with the role of emotions in online social relationships , and how emotions matter in online communication . Today, few would question that emotions can be conveyed online, thus the focus of research has shifted from whether emotional communication is possible online towards the analysis of the differences between emotional expression online and F2F . However, the analysis of differences in how social groups experience and express emotions on the Internet is scarcer. --- Differences in emotional expression online There is a large body of literature dealing with age and gender differences in Internet use . Following an initial focus on inequalities in Internet access, the interest has shifted to the study of the background characteristics that impact how the Internet is used . This section reviews literature that studies how age and gender affect Internet use, focusing on their impact on emotional expression online. It also introduces the concept of social web engagement as an additional variable that could potentially affect emotional expression online. Young people are a particularly active group on the social web . Although the term 'digital native' has been rightly criticised , it is acknowledged that young people hold a privileged position regarding individual capacity for technology use, as they grew up with digital technologies as an integral part of their everyday lives. This view implies that year of birth is a key determinant of Internet use, in terms of quantity and quality and in terms of digital skills . Consequently, there is a considerable amount of literature about age differences regarding Internet use and online behaviours . However, we do not know much about age differences in the expression of emotions online. Previous literature on F2F emotions suggests that people become less emotional and that they seem to have less emotional expressivity as they get older . Thus, age could be expected to be a relevant factor affecting the self-report of emotional expression on the social web too. Gender differences in online emotional expression are a more established area of research. A number of studies agree that female Internet users are more expressive of their emotions , and are more prone to use the social web to communicate with others, than males . The results pointing at gender differences are broadly consistent with those on F2F communication research . Yet, the evidence is not unequivocal regarding the gendered use of the Internet to communicate. Helsper finds no gender differences in the use of the Internet as a communication tool. In addition, Preece and Ghozati do not find a relationship between empathy expression in an online group and gender, although the presence of women in the groups correlates with a higher number of empathic messages. To further develop this research, we consider gender differences in emotional expression on the social web. Internet usage can no longer be described only in terms of frequency or Internet connection versus non-connection. Some people go online to interact with others only occasionally; many others make it an integral part of their lives, connecting regularly or constantly through mobile devices. This variation can be expected to have consequences on the ways in which people use the Internet . Whilst some research has been conducted on social media usage as an online venue where emotional communication takes place , there is much scarcer research on the effects of social web level of use on emotional expression online. A study about online gaming suggests that greater Internet experience can reduce emotional competence. Yet, other studies suggest that more skilled users are likely to engage in more types of online activities-including communicating with others-and with greater frequency . Differences in emotional expression for users with different levels of social web engagement could also be expected, but they have been largely ignored by previous research. For this reason, this study controls for levels of access, by focusing on frequent Internet use, and investigates which kinds of web applications have a larger effect on emotional expression. --- Emotions online in social relationships People use a wide variety of cues to express emotions online. We take into account this variety, and explore the use of three types of cues: non-verbal, action and verbal . These three categories are used to conceptualise emotional expression on the social web and to develop a conceptual framework . The definition of these three cues on the social web is based on a previous ethnographic study about online social interaction as discussed below. We expand previous ethnographic findings and use this conceptual framework to explore whether emotional expression on the social web is affected by gender, age or levels of social web engagement. --- FIGURE 1 Non-verbal cues: Non-verbal cues are the observable actions that accompany social interactions and do not imply language and words: facial expressions, blinks, speech pauses, gestures etc. In F2F research, non-verbal cues, which are typically produced outside conscious awareness, are considered as signals conveying reliable information about people's emotions and a measure of affect and socialisation . The same points do not necessarily hold for online communication, as non-verbal cues are, primarily, conscious decisions and the only way of conveying non-verbal cues unconsciously is through the webcam or Voice over Internet Protocol . Previous research argues that the equivalent to facial expressions online are the emoticons . Hence, emoticons could be used as a strategy for non-verbal emotional expression on the social web. Furthermore, existing literature suggests that the use of multimedia elements to express emotions is a common practice among Internet users . Action cues: Action cues occur when people purposively act in response to something. They are different from body movement, because they are goal-directed or voluntary acts . In addition, they are also perceivable stimuli that people produce and sense in their everyday interactions with others. Action cues can take place in online communication too. People may shut off their computers when they get angry, for instance, or may wish a hug or a kiss during online encounters. In relation to action cues, we also explored individuals' tendency to increase communication depending on the emotions they experience, in order to explore the concept of openness to communication . Verbal cues: It is commonly acknowledged that non-verbal communication is the prime medium for emotion. Despite this, the verbal expression of emotion is important in F2F interactions and even more so in online settings. The verbal/written cues of emotion online are those expressed in words. This expression can be direct, by coming outright and saying 'I am feeling happy today' or more subtle, such as through word choice or language intensity . According to Baym, 'text based new media afford many ways to express emotions' . As illustrated in this section, few studies have so far looked at how different social groups vary in their expression of emotions online. Questions about the relationship between age, gender and the way people engage with the social web and emotional expression remain under-theorised. Moreover, the bulk of studies on social media draw on college-aged samples and focus on specific applications. This does not allow for an examination of the use of the social web across generations and platforms. In order to address this gap, this study introduced a conceptual framework based on previous research, which is tested in the reminder of this article. --- RESEARCH QUESTIONS, DATA AND METHOD --- Research questions This analysis explores the individual foundations of variations in self-reported emotional expression on the social web, an issue that has not been previously explored in the literature. In order to address this gap, we ask how age, gender and social web engagement relate to self-reported volumes of emotional expression online, based on cues. In other words, do older people report more emotional expression on the social web than younger people? Do women report more emotional expression on the social web? Do people who report higher social web engagement also report higher levels of emotional expression? --- Sampling and survey administration procedures We use data collected from a non-probability sample of Spanish frequent Internet users . The dataset captured information on people's self-reports on emotional expression, as well as details of their social web engagement and social factors allowing us to analyse whether systematic differences exist between different ages, genders and social web engagement groups in the sample. The use of emotional expression self-reports on the social web is relevant for two reasons. First, it provides a quantitative research design that complements content analysis studies , which are necessarily focused on a single application and generally analyse sentiment. Second, it is ethically responsible, unlike the recent, controversial Facebook experiment . However, it also raises questions regarding the capacity of surveys to capture the complexity of emotions in relationships. In this sense, it is worth mentioning that the survey was carried out after an ethnographic project that focused on understanding those complexities . The study also recognises the socio-cultural dimension of emotions and does not pretend to generalise results beyond the sample of Spanish frequent Internet users. The questionnaire was distributed online using multi-modal administration methods. The social web is a viable means of obtaining a sample of individuals who frequently use the Internet , i.e. the focus of this research. The researcher initially used their own Social Network Sites and email contacts to distribute the survey, asking those contacts to further circulate it. The survey included a set of questions to identify Spanish nationals who are frequent Internet users-our target group-regardless of their place of residency. In total, 301 individuals answered the survey. As is generally the case with online surveys, caution in interpreting the results is required, due to the convenient nature of the sample and the biases of self-reported surveys . --- Measures The survey was prepared in Spanish and included 21 questions about emotional expression online , background characteristics of respondents and social web engagement measures . Emotional expression on the social web: Following previous ethnographic work on emotional communication online , we developed a measure based on the conceptual framework presented in Figure 1, using a five-point Likert scale that asked respondents to self-report on a set of statements about their ways of expressing emotions on the social web . Responses were aggregated into non-verbal cues , action cues and verbal cues , and computed into a summary measure of emotional expression . The emotional expression values in the survey ranged from 19 to 63. Background variables: Age was operationalised in five age groups and 5 the oldest . As was to be expected, the sample was made up, predominantly, of young and middle-aged individuals: the largest group of respondents was those aged between 25 and 34 , followed by those between 35 and 44 . The least represented group was those aged between 55 and 64 . There were no respondents over the age of 64. Gender was operationalized as a binary dummy variable, indicating whether the individual reported to be male or female . There were approximately the same number of women as men in the sample . --- Social web engagement: Digital inequalities literature has demonstrated that Internet usage should no longer be operationalized in terms of time and frequency of connection only, but rather should take the type of use during connections into account too. We used five variables to measure social web engagement: frequency of Internet usage; number of email accounts used; number of Social Network Sites used; number of instant message systems used; and the types of interpersonal relationships that individuals maintained online. We defined frequency of Internet use as an ordinal variable based on three levels of usage: 'I am always connected or at one click away from being connected' ; 'I connect daily' ; 'I connect at least once a week' . Survey respondents who reported to connect less frequently than once a week were excluded from the analysis, given our interest in frequent Internet users and in those who make online communication an essential part of their social relationships. To provide a more nuanced account of IU for social communication, we also asked respondents to report on their use in different platforms for communication through the Internet: the number of email and instant message accounts , and the number of Social Network Sites in which they were active in the six months prior to survey. The EMAIL measure ranged from 0 to 20; the IM measure ranged from 0 to 7; and the SNS measure ranged from 0 to 10. Finally, we asked respondents about the different kinds of interpersonal relationships that they keep online. It seems to be an important factor in understanding self-disclosure in online social interactions, so we would expect it to affect emotional expression as well. Participants were asked to report whether they communicated with their partner, friends, colleagues, family, acquaintances and/or strangers. The IR measure ranged from 1 to 6. --- Data analysis Data were analysed through descriptive statistics and hierarchical linear regression. The analysis first studied the IU of subjects in our sample , then generated descriptive statistics on group differences based on IU alongside other forms of social web engagement. Gender and age differences in IU were examined using chi-square tests. T-tests and ANOVA were used to compare gender and age differences against IM, EMAIL and SNS. Hierarchical linear regression was used to model the relationship between emotional expression online, individual characteristics and social web engagement: IU, IM, EMAIL and SNS, and IR. A hierarchical model is particularly appropriate to analyse the different levels of social web engagement. In order to deal with missing data in the emotional expression variable, we used multiple imputation to create five complete datasets using the Amelia statistical package for R . The regression coefficients from each dataset were combined using the rules described by Little and Rubin . --- RESULTS --- Individual characteristics and social web engagement Table 1 provides information on gender and age differences in IU in our sample. --- TABLE 1 The percentage of respondents reporting daily connection or to be alwaysconnected is considerably higher than for weekly connections . Men in our sample are more likely to be in the 'always connected' category than women , but the chi-square tests show that gender differences in IU are not significant . Young respondents were more likely to be 'always connected' than other age groups. This was the highest figure for all age groups, followed by the 35 to 44 age range . Differences in IU by age, however, are not significant . The lack of statistically significant differences between groups shows that gender and age differences are blurred among frequent Internet users. --- FIGURE 2 Respondents had, on average, 3.06 email accounts, 1.97 instant message accounts and were active in 2.65 Social Network Sites . Participants reported that they communicate chiefly with friends and colleagues . The least popular kind of online communication was with strangers, but even so, around a third of respondents reported to interact within this category of social relationship . On average, respondents communicated within 4.00 types of interpersonal relationships, meaning that they maintain a broad range of kinds of relationships, including with strangers. Table 2 presents summary statistics on IM, SNS, EMAIL and IR. This provides a more nuanced description of the social web engagement of our sample. Again, we find no statistically significant gender differences for either EMAIL, IM and SNS, or in the range of relationships kept online. Given the focus on social relationships, it worth emphasising that there are not significant differences in the range of relationships by age or gender. All the participants interact mainly with friends, colleagues, acquaintances and family, but they also communicate with their partner and strangers to a lesser degree. Understanding social web use for emotional communication requires attention to both, the applications used and the relationships themselves. --- TABLE 2 Regarding age, we found statistically significant differences for IM and EMAIL. To further explore the differences, we conducted a Tukey multiple comparison of means. The results show that the significant differences for IM and EMAIL are based on variances between those in the 55-64 and the 16-24 age groups. --- Emotional expression on the social web Figure 3 details respondents' self-reported experiences of emotional expression on the social web. Regarding the use of non-verbal cues to communicate, 50% of participants agreed or completely agreed that they use emoticons to express emotions online. However, previous research has referred to the difficulties involved in recognising, interpreting and conveying emotions, due to the lack of non-verbal cues on the social web. Thus, participants were also asked about VoIP and webcam usage, as potential vehicles to express emotions better. As Figure 3 shows, surprisingly, the majority of respondents reported that they did not use webcams or VoIP tools to express their emotions better . Thus, according to our respondents, these tools are not as effective as could be expected in offering non-verbal cues. Alternatively, it could be that individuals in our sample 'prefer' the limitations associated with the lack of provision for non-verbal cues. Indeed, video clips and music are used more often than webcams to express emotions . By contrast, verbal cues are essential in online communication. Our findings, in line with those of Baym , strongly suggest that written text assists in emotional expression: 42% of respondents completely agreed or agreed that they are more explicit in what they say when trying to convey emotions online. On the other hand, it is intriguing that frequent Internet users consistently reported that they use exclamation marks when they are surprised, but they do not use other strategies to signal negative emotions, such as capital letters for shouting or using strong language. This pattern would suggest that negative emotional cues could be more suppressed online than cues associated with positive emotions or ambivalence. --- FIGURE 3 This finding is consistent with the results for action cues, where respondents reported to use the web often to communicate a lot when they feel happiness , or always . By contrast, communication in relation to sadness is much more constrained: only 16.4% of respondents agreed or completely agreed that they communicate a lot with others online when they are sad. This difference supports the idea that the openness to communication is fundamental for encounters on the social web , with this openness being mediated by the mood of the individual . Consistent with Quan-Haase and Collins , we also found high levels of usage of the 'no replying' action , which is an option that is much easier to implement online that F2F. --- Multivariate analysis Table 3 shows the results of the hierarchical multiple regression modelling. The analysis investigates the relationship between gender, age, levels of social web engagement and the self-reported emotional expression on the social web . Seven models are presented, to explore which model best fits the data. Emotional expression online is regressed, first, on respondents' gender and age . The third model adds IU. Models 1 and 2 show that gender and age alone are not good predictors of emotional expression, as only 8% of the variance is explained. When IU is added , the three independent variables are better predictors of emotional online: the model explains 17% of the variance. In Models 5 to 7, we include the independent variables EMAIL, SNS, IM and IR in the analysis. The model fit improves, with Models 6 and 7 explaining 26% of the variance. --- TABLE 3 The modelling suggests one main finding regarding gender and emotion: women report expressing more emotion on the social web than men. This result is broadly consistent in all seven models. Alone, gender does not explain much variance in our sample, but the gender effect size increases when we include social web engagement variables in the models, which suggests that a model with a more nuanced description of social web engagement considering IU, IM, SNS and EMAIL provides a better fit and is able to explain differences in emotional expression on the social web to a greater extent. We also find differences in the reporting of the independent variable between age groups. The results show a negative relationship between emotional expression online and age: older Internet users tend to report lower levels of emotional expression, although the effect size of age decreases when we include social web engagement variables in the model. This means that to understand why age is important better and explain the variable, we need to consider the interactions between age and social web engagement. The effect size of IU decreases substantially as we introduce other social web engagement variables, suggesting that is also relevant to consider social web engagement variables such as IM, SNS and EMAIL to predict self-reporting of emotional expression. That is coherent with our theoretical point of departure and implies that we can achieve a more accurate prediction of people's emotional expression on the social web if we introduce complementary measures of engagement in the analysis, as illustrated by Models 4 to 7. The study of the interaction of the regressors suggests that gender, age, IU, IM, SNS and EMAIL are relevant for the study of differences in self-reported emotional expression. Both Model 6 and 7 provide equal model fit , but as the inclusion of the IR regressor does not improve the fit of the model, we select Model 6. This model shows a positive relationship between gender, IU, IM, SNS, EMAIL and self-reported emotional expression: being female, younger or reporting greater levels of social web engagement are associated with higher levels of emotional expression. The effects of age and gender are larger than the effects of social web engagement. So, ascribed characteristics are, on the whole, better predictors of self-reported emotional expression than other aspects covered in this study. --- DISCUSSION To answer our research questions, we explored the ways in which the reporting of emotional expression online relates to gender, age, IU and social web engagement among Spanish frequent Internet users. Regarding whether women who are frequent Internet users report more emotional expression on the social web than men, our results concur with previous research that has studied gender differences in online emotions . First, females in our sample reported higher levels of emotional expression than males. These findings are also in line with those of Kapidzic and Herring , which prove that woman tend to be more expressive of their emotions than men in their online communications. These results also suggest that self-reported emotional expression follows similar trends along gender lines to reported F2F emotional expression . Second, older people reported lower levels of emotional expression on the social web than younger users. This finding also aligns with F2F studies on emotions expression, which have found that younger people exhibit greater levels of emotional expression . Finally, regarding the relationship between social web engagement and the independent variable, our results indicate that all the explored forms of engagement affect it. A covariate analysis allows a more detailed description of the effect of each Internet application-IM, SNS and EMAIL-in reported emotional expression. This shows that the impact of Internet usage decreases when we control for other engagement variables, such as IM. Among the covariates introduced in the model, the most substantial effect size is for IM, which is consistent with previous research that reports that much emotional communication on the social web happens in IM applications . As already mentioned, the nuanced description of social web engagement offered by the inclusion of variables beyond connectivity suggests that taking into account the varied uses and platforms helps in predicting patterns of emotional expression online. Our results are limited but important in this respect, as they suggest that greater use of the social web is associated with greater levels of reported emotional expression online, and that being younger and female is also related with higher levels of reported emotional expression, which is consistent with the results obtained in F2F research. Moreover, the findings support the formulated theory that differences in the way people use the social web shape their emotional communication online. By studying these social factors and characteristics, this study suggests that in these particular social interactions-i.e. online encounters-different ways of acting emerge that are affected by age, gender and social web engagement. It suggests that being female, young, with higher levels of Internet usage and social web engagement correlates with reporting higher levels of emotional expression during online social interactions. This study, following previous ethnographic findings , proposes that these practices can be considered interaction rituals, constructed in assembling communication acts in online interaction. Different ways of acting are being created and normalised online. Internet users have developed shared strategies to express emotions online and to build meaningful encounters, even without human bodies and F2F interaction -as theorised by Collins . Caution is warranted when interpreting these results, given the non-representative nature of our sample. One limitation of this study relates to the self-reported nature of the survey . Another limitation is the restricted background of the sample-i.e. only Spanish people-but this narrow scope also set the grounds for further studies comparing different socio-cultural realms. Our analysis underlines several points. First, it highlights a need for online research to focus on aspects that go beyond specific web applications or connectivity when studying interaction rituals online. The results of this study suggest that different levels of engagement affect the self-report of emotional expression. This pattern indicates, in turn, that expressing emotions on the social web is a complex matter and should also be studied beyond the confines of a single web application and considering different kinds of interpersonal relationships. Although useful, research on specific web services can lead to oversimplification. Second, our results suggest that measuring social web engagement only by frequency of connection could lead to an incomplete overview of group differences amongst highfrequency Internet users. The people in our sample do not interact on the social web by means of a single application, nor do they interact within one type of interpersonal relationship . Thus, a conceptualisation of engagement for IU is very limited. An alternative to a technology-driven approach is a user-centred approach, focused on the Internet user and her/his social interactions . The proposed conceptual framework aims to offer a quantitative way to study self-reported emotional expression on the social web that goes beyond sentiment analysis: it is able to analyse ritual interactions online once they have been described, and takes into account the different ways in which people engage with the social web to communicate with others. --- CONCLUSIONS This study aims to contribute to the emerging body of research that has been seeking to enhance our understanding of online interaction and, especially, of online emotions. The study introduces a novel theoretical and methodological approach to the literature of emotions in social relationships. In doing so, the approach follows recent calls for more diverse ways to study emotional experiences in everyday life, using a variety of methodological perspectives. The framework introduced arises from the idea that greater attention should be devoted to the investigation of an emerging emotional terrain online, and aims to provide a framework to explore it. Through communication processes, societies develop interaction rituals, rules and strategies in different settings. A contemporary analysis of interaction rituals needs to consider the online realm as an integral part of socialisation processes and needs to explore differences in ways of acting online. The novelty of this approach lies in the fact that it does not focus on the content of the message and does not describe what is being said only. Rather, our proposed approach underlines the importance of looking at emotional cues and whether they might become a general way of expression -a social norm. It is crucial to note that online social interactions are affecting structural factors that influence socialisation processes by themselves. For example, the structure of social web applications and services affects the way in which we experience and convey emotions. Moreover, as the highly controversial Facebook emotion experiment showed, emotional contagion is related to emotional expression, and it can happen online even without any verbal cues intervening in the communication process. This could be explored in the future, using the conceptual framework proposed in this article to research to how we are brought up or trained to express and manage emotions online. Future research could also use probabilistic and larger samples, as well as different types of social web users to further test our findings. Additionally, it could extend our study by exploring cultural differences in emotional expression online or the interaction rituals that are emerging in different social-cultural contexts, amongst different groups of users and geographies. This framework we have put forward has the potential to deal with the question of the commodification of emotions and the homogenisation of emotional expression, as well as the extent to which both take place on the social web. There are many unexplored issues in this area, such as the detailed study of specific emotions and their manifestation online. Finally, future research could seek to understand better the impact that online emotional expression may have in F2F emotional communication. i Here, 'the social web' is defined as a set of relationships that link people on the Internet, also known as 'social machines' . The social web concept is useful because, unlike terms such as Social Network Sites or Social Media, it stresses the social aspects beyond any software, platform or application, and beyond any device. --- TABLES AND FIGURES 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

B e n eit o-M o n t a g u t, Ro s e r 2 0 1 7. E m o tio n s, ev e ry d ay life a n d t h e s o ci al w e b: a g e , g e n d e r a n d s o ci al w e b e n g a g e m e n t eff e c t s o n o nli n e e m o tio n al e x p r e s sio n. S o ciolo gic al R e s e a r c h O nli n e 1 0.

Introduction Playing and physical activity are essential for child development as they enable cognitive, physical, emotional, social, and motor learning, as well as well-being. The UN Convention on the Rights of the Child states that every child has a right to play. However, many children worldwide are not meeting physical activity guidelines [1] and have decreased opportunities for active outdoor play, which are both vital for a child's healthy development [2]. In Germany, only 13% of girls and 17% of the boys aged between 4-17 years meet the WHO physical activity guidelines [3] recommending that children engage in at least one hour of moderate-to-vigorous physical activity a day [4]. Additionally, due to parental concerns about road safety and stranger danger, children nowadays have limited independent mobility and therefore less opportunity to play outside [5][6][7][8]. Physical and social environmental factors determine children's physical activity and outdoor play [9]. With regard to the physical environment, play facilities such as playgrounds are important requirements for being physically active [10,11]. Therefore, to provide opportunities for children to be physically active, play, and interact with other children outdoors, accessible, usable, and safe physical activity and play spaces are needed [12,13]. Public playgrounds constitute important settings for children to play, experience and interact with their social and physical environment, recognize and test their own abilities, and develop social, physical, and motoric skills. Furthermore, playgrounds facilitate positive experiences such as fun, creation, self-efficacy, social interaction, creativeness, and physical ability [14] and may contribute to increased levels of energy expenditure in children [15]. Boys and girls often behave differently in regard to their physical activity and play behavior, as postulated in various theories on gender [16]. Empirical research studies have shown that boys and girls choose to play at and with different things in different ways and they differ in terms of their physical activity levels, types, and preferences [17]: For example, girls tend to be less physically active than boys overall [1,18] in school playgrounds [19][20][21], as well as out of school playgrounds [22]. In playgrounds, the following differences with regards to preferred play behaviors of girls and boys have been observed: For example, boys are more likely to play space-consuming and popular ball games such as football [14]. Krajicek playgrounds, which are newly created public playgrounds of the Richard Krajicek Foundation and which are "primarily aimed at offering play areas for team sports like soccer and basketball" [23], have been found to be more attractive for boys than girls. Additionally, Anthamatten et al. [24], as well as Floys et al. [25], found that boys utilize play fields, courts, or basketball areas more often than girls and exhibit higher levels of MVPA on hard-surfaced play areas and play equipment areas than girls in a schoolyard setting. Furthermore, boys and girls seem to be responsive to different playground features and qualities: in school playgrounds, girls experience a higher enjoyment level than boys for a number of activities like playing tag games, being active during play activities, walking, creating/making things, using imagination, climbing, sliding, hiding, sitting, and resting/relaxing [26]. The social environment is an important predictor of children's physical activity. As stated in Social Cognitive Theory [27], children's behavior and behavioral choices are strongly affected by social models. Several modeling phenomena, each with different operating mechanisms, could be distinguished [28]. One phenomena of social modeling, which could be assumed as a predictor of children's physical activity behavior, occurs when "others' actions serve as social prompts, inducements or response-cuing effects" [28]. The same-sex imitation hypothesis assumes that children develop tendencies to prefer imitation of same-sex models [29]. More general, imitation of social models occurs as a function of credibility and relevance of the social models. For example, parents could be significant models for children: Parental modeling of active behaviors has been shown to positively influence children's participation in physical activities such as outdoor play, sport, and walking [30,31]. Besides parental modeling, the presence of peer models being physically active is associated with an increase in physical activity [32,33]. In particular, in a playground setting, peer modeling and companionship/playing together may be especially relevant for the encouragement of physical activity and play [34]. Nevertheless, the presence of other persons can also lead to a decrease in physical activity or inhibit physical activity and how children play. For example, parental restrictions can lead to lower levels of children's physical activity [35] and the presence of groups of teenagers can evoke feelings of fear, limiting physical activity behavior [36]. Furthermore, playground density is inversely related to physical activity in school playgrounds [20,37,38], which indicates that besides supportive effects of companionship and modeling, a lack of space per child prevents physical activity. While companionship is the precondition for many play activities like tag games, ball games, or seesaw, and while peer modeling and support could foster physical activity, a high number of children in the playground may hinder physical activity. A large amount of studies exist that examined girls' and boys' physical activity and outdoor play in school playgrounds [39][40][41][42][43], but not in public playgrounds. Usage, physical activity, and play in public and in school playgrounds differ from one another. Children use school playgrounds obligatorily, while public playgrounds are visited during their leisure time voluntarily or with the encouragement of their parents/supervisors. In school playgrounds, children are surrounded with friends and familiar children attending the same school, whereas in public playgrounds, there is a greater variety of children and persons of different age groups unknown to the children. Additionally, the number of persons in public playgrounds varies more widely than in school playgrounds. Consequently, physical activity and outdoor play behavior in public playgrounds could differ from those found in schools. The effects of the social environment on physical activity and play behavior in public playgrounds have rarely been examined by gender [10], especially in a non-school playground setting. Therefore, the aim of this study is to examine: gender differences in physical activity, social play behavior, and preferred play equipment of children in public playgrounds; the relationship between social environmental factors and the physical activity of children in public playgrounds; and whether this relationship is moderated by gender. --- Method --- Study Design A quantitative, observational study was designed to examine weekday physical activity and outdoor play in children in playgrounds. The study was conducted in a district of Constance , a middle-sized town in the south of Germany near the Swiss border. With approximately 3000 children and adolescents aged between 0-17 years [44] out of 22,000 total inhabitants of the district , it represents a district with a high proportion of young people. The district includes an area of 181.89 ha with a dense population in which a high proportion of the inhabitants are German and a low unemployment rate . All ten playgrounds in the district of Petershausen-West were analyzed in this study. Detailed information on quality aspects of the playgrounds and playground users is provided elsewhere [45]. All procedures performed in this study were in accordance with the ethical standards of the Ethics Committee of the University of Konstanz and with the 1964 Declaration of Helsinki and its later amendments. No informed consent was required from the playground users because only de-identified data was collected in this observational study. --- Data Collection A direct observation tool was used to assess the number of users in the playgrounds , children's physical activity levels, activity types, group sizes, social interactions, and usage of playground equipment. The observation tool comprised items of the System for Observing Play and Leisure Activity in Youth [46], the System for Observing Children's Activity and Relationships during Play [47], and a newly developed item on children's usage of playground equipment. This item captured whether and which playground equipment the observed child played with. Data collection took place from April to September 2015. The observers were trained by studying the SOCARP and SOPLAY protocols and by using the observer training videos [46,47]. The trained observers visited the playgrounds for one hour each between 10 a.m. and 8 p.m. to conduct direct observations based on a momentary time sampling procedure [48]. Thus, the observers visited the playground on a defined date and conducted the scans and individual observations. Scans were conducted to observe all playground users in the playgrounds and for individual observations, one child was randomly selected. During school days, the observations took place between 12 p.m. and 8 p.m., during the summer holiday between 10 a.m. and 8 p.m. The direct observation included scans of all playground users and observations of individual children. Up to five scans were conducted within one hour of observation . During the scans, the observers let their eyes wander from one side to the other of the playground. To enable counting of the number of playground users and their physical activity levels, a mechanical counter was used [46]. Each scan encompassed six of these views over the playground. In the first view, the number of playground users within an age group was counted . The following views were conducted to count the number of female children in relation to their physical activity level and the female adolescents in relation to their physical activity level. Following this, the procedure was repeated for male users. In the present study, children were assigned to the children's group when the estimated age was ≤12 years. Detailed information on physical activity levels of children in the playgrounds and playground usage is reported elsewhere [45]. For individual observations, up to four different children were observed within one hour of observation. For this, the observers randomly selected one child playing in the playground. Each child was observed for five minutes . Each interval included 10 s of observation following 20 s of recording. In order to keep the timeframes, a timer was used. During each hour of observation, the observation procedure was repeated up to four times. To control for time and observer effects, all playgrounds were equally examined by all observers and the hours were evenly distributed over the day . During the observations, the observers had a fixed position in the playground from where they could observe the entire playground. To avoid reactivity by children, the observers behaved as unobtrusively as possible. In case of questions by parents or supervisors, the observers offered an informative letter containing details of the aims and procedure of the study, the responsible researchers and institutions, and data security measures. Observers also provided verbal information about their data collection in the playground and about the study, if requested by parents/supervisors or children. --- Measures --- Children's Physical Activity Behavior A modified version of the System for Observing Children's Activity and Relationships during Play tool developed by Ridgers et al. [47] was used to assess a targeted child's physical activity level, activity type, and group size in the playgrounds. It was modified to also assess usage of playground equipment. The physical activity level of the target child was recorded as sedentary , walking , or vigorous [47]. A child who was walking or engaging in vigorous physical activities was defined as being moderately-to-vigorously physically active [47]. Thus, every child was rated as being sedentary, moderately active, or vigorously active. Social group size was defined as the total number of persons in the group in which the target child was located and encompassed four categories: alone , small , medium , and large . Activity type reflected the nature of the activity the target child engaged in the playground. To capture activity type, five categories were given based on SOCARP [47]: sports game , active games , sedentary activities , locomotion , and activities with playground equipment . The observer designated the type of activity with the best fit to the child. Furthermore, in the observation protocol, it was recorded whether the target child was using the following playground equipment: things to hang from, things to slide down, things to climb up, things to climb through things to stand/walk on, swings, carousels, see-saws, playing fields, or sandpits. --- Playground Density and Presence of Children and Adults in the Playgrounds To capture the playground density and the number of children in the playground within a defined time period by direct observation, a modified version of the SOPLAY observation tool [49] was applied. Using the direct observation tool, the number of users per age group and the number of children featuring a physical activity level were examined separately for males and females. Gender and age group were rated from the observers through visual assessment. In line with the original SOPLAY instrument, sedentary behavior was defined and captured as lying, sitting, or standing. Walking was categorized as moderate physical activity and activities such as running, climbing, and jumping as vigorous activities. Children who were walking or engaging in vigorous physical activities were assigned to the moderate-to-vigorous physical activity group . Variables of the number of present children, boys, girls, and adults were calculated. Furthermore, the number of active boys and girls, respectively, were calculated. Based on these variables, a variable of the number of same-sex children in the playground and a variable of the number of active same-sex children were calculated. Playground density was calculated by the quotient of the number of persons in the playground and playground size. Playground sizes were captured from the playground manual of the City of Constance [50]. --- Confounding Factors Daily weather variables were recorded as potential confounding factors. Daily weather was captured using weather protocols provided by two German weather service webpages . --- Statistical Analysis Continuous variables are presented as the median with minimum and maximum values, as well as mean plus standard deviation . For categorical variables, we report the absolute number and percentage. A comparison of categorical variables between girls and boys was assessed by a chi 2 test or Fisher's exact test. The main outcome variable is the percentage of being in MVPA during the up to 10 time intervals a single child was observed. We used a beta random variable to model the distribution of the percent values and used a multivariate regression analysis [51] to examine the effects of the independent variables on the outcome variable . These independent variables cover relevant factors of the social environment. All beta regression models were conducted by adjusting for temperature and rainfall. Interaction effects between some independent variables and gender are analyzed and based on statistical significance included in the final overall model. Stratified models for boys and girls were also considered. The multivariate regression can only handle values in the open interval . Thus, we transformed the value 0 to 0.001 and the value 1 to 0.999. Consequently, on the logit scale, the response variable takes on values between-7 and 7. Two-sided p-values < 0.05 were considered as significantly different. No adjustment for multiple testing was done because the p-values should be interpreted in a descriptive manner. All analyses were conducted with SAS Software 9.4 . --- Results A total of 578 scans and 338 individual observations of children and adolescents in the playgrounds were conducted. Of these individual observations, 266 observations for children between four and 12 years are used in this analysis. Each playground was observed 18-21 times. The size of the playgrounds ranged from 280 m 2 to 3800 m 2 . On average, 6.4 children were observed during a scan, and on average, 3.9 children were moderately-to-vigorously physically active. Similar numbers of boys and girls were counted during the scans. Further descriptive information on social environmental measures is presented in Table 1. Further information on user characteristics of the playgrounds is presented in Table S1. Children's physical activity and play behavior observed in the playgrounds are presented in Table 2. Within our study, most children were at least moderately active , with more boys having higher proportions of MVPA than girls . Boys and girls preferred similar playground equipment to play with: the only statistical significant difference was observed in regard to things to stand or walk on . However, the numbers of children using a specific type of playground equipment were small, leading to three observations of children in the dimension "things to climb through" and up to 44 observations in the dimension "swings". Approximately, one quarter of the observed children played alone , and most observed children played in groups of two to four persons with no significant gender differences . Only one girl and one boy were observed playing in a group of ten or more persons. Significant gender differences were found concerning the activity types: boys were more likely to engage in sports and active games than girls , while girls were more likely to be walking/running , to play on playground equipment , or be sedentary than boys. In the multivariate regression models , the interaction of gender with number of active children in the playground, as well as the number of opposite-sex children, predicted the MVPA of the observed children. The models stratified by gender revealed that the number of active children was positively associated with MVPA and the number of opposite-sex children was negatively associated with MVPA only in girls . However, narrowly missing statistical significance, the number of active children was also positively associated with MVPA and the number of same-sex adults was negatively associated with MVPA in boys. Playground density and group size did not predict MVPA in any of the models. Besides the presented regression models which were adjusted for weather , we also ran these models without adjustment. No substantial differences were found in these models, but model fit was better in the adjusted models. = reference; * p < 0.05; models were adjusted for rainfall and temperature; Data was weighted using the complex sample procedure. --- Discussion The present study contributes to a better understanding of children's physical activity and play in public playgrounds. It examines social environmental influences on these behaviors in detail and takes gender differences into account. Previous studies in this area primarily focused on physical activity behavior of children in school playgrounds [41,42] or often did not analyze gender influences in depth [52,53]. The results of this study showed that most children were moderately or vigorously active when playing in the playgrounds. Thus, the proportion of active children was considerably higher than in other studies conducted in public playgrounds or parks in the Netherlands, the US, and Australia [23,25,54]. Additionally, the present study revealed gender differences in the preferred type of activity and the usage of equipment. Overall, no significant gender differences in physical activity levels were found, which is surprising because this finding is in contrast to many other studies [19][20][21][22][23]25]. However, preferred activity types differed between boys and girls, which is in line with the findings from other studies [19][20][21]53]: girls engaged more often in sedentary activities, locomotion, or activities on playground equipment, and boys were more likely to play sports or active games. Consequently, it seemed that the differences in types of activities did not lead to different levels of physical activity and that both boys and girls are actively playing in public playgrounds for most of the time. Moreover, another study revealed that boys were more likely to play with materials associated with boys like cars and girls tended to occupy play spaces with activities related to girls like the doll corner in a kindergarten setting [15]. Furthermore, in a playground of a primary school from South Africa, girls were excluded from playing soccer with the boys [55]. The authors of these studies assume that there are stereotypical gender boundaries in play spaces that foster children to play according to the stereotypes. However, children crossing these gender boundaries have also been observed in these studies, but in the primary school study, the children not conforming to gender norms were often bullied or disapproved [55]. Thus, there might also be stereotypical gender boundaries preventing children from conducting activities not conforming to gender norms in public playgrounds. Overall, none of the social environmental variables predicted MVPA in boys: Only in girls did the number of active children in the playground predict participation in MVPA and the number of boys being present was inversely related to MVPA. Playground density, group size, and presence of adults did not predict MVPA-neither in girls or boys. Thus, in the present study, the presence of active children who might serve as models or as active companions , led to increased levels of MVPA of girls in public playgrounds. This finding is similar to the findings of an experimental study from Horne et al. [32] that was conducted in a primary school setting. In this study, physical activity levels increased during an intervention, in which fictional peer models acted as role models for physical activity [32]. However, in this study, the effects of the fictional role models were found in both genders. Additionally, in a study of children's park-based physical activity in Durham NC, which did not account for interactions with gender, the presence of other active children had the strongest association with physical activity [25]. Consequently, the presence of other children being physically active could be a relevant factor fostering physical activity in children, especially in girls. Furthermore, in the present study, the presence of same-sex children did not predict MVPA in boys or girls, respectively, indicating that the gender of models or play companions was irrelevant. This contradicts the same-sex hypothesis published by Bussey and Perry [29], which postulates that children tend to imitate models of their own sex. However, girls were less likely to be active when more boys were present at the playgrounds. The presence of boys could probably hinder physical activity in girls playing in playgrounds, given that boys tend to use larger spaces when playing [56]. Furthermore, boys may exclude girls while playing in the playgrounds [55,57,58] and therefore, reduce girls' opportunities to take part in active games like ball games or playing tag. Supporting this presumption, in the present study, it has also been shown that girls were less often playing active or sports games than boys. For example, as shown by McGuffey and Rich [59], boys could have used hegemonic masculinity to regulate girls' boundaries in outdoor play and to keep them out of the boys' domains like sport games. However, dismissive behavior and social interactions have not been analyzed in the present study. Hence, we do not know the mechanisms that prevented girls from being physically active when more boys were present. Previous research showed that playground density was inversely associated with physical activity levels of children in school playgrounds [20,[37][38][39]. However, these studies took place during recess and playground density seems to be much higher during recess compared to public playgrounds: The mean density in the present study was 0.008 children per m 2 , while in the study of Cardon et al. [20], it was 0.15. In the study of D'Haese et al. [37] the play space per child was 12.18 m 2 , by Ridgers et al. [38] it was 6.2 m 2 , and by Van Cauwenberghe et al. [39] it was 7.4 m 2 , resulting in playground densities of 0.082, 0.161, and 0.135 children per m 2 , respectively. Consequently, the lower playground density in public playgrounds could offer better opportunities for being physically active than in school playgrounds during recess. Boys and girls were organized in similar group sizes. Neither boys nor girls had more children around them to play with. Nonetheless, we did not analyze the types of interaction in the playgrounds. Consequently, we do not know if the present children or adults were interacting with the observed child in a supportive or restrictive way. For example, besides motivating or inviting the child to play with each other, the adults could have restricted the child's activities by prohibition on the grounds of fear, as shown in other studies summarized in a meta-study by Lee et al. [60]. Moreover, other children could have bullied the observed child physically or verbally so that it retired and was less active: For example, an ethnographic study of children aged 6-10 in South Africa showed that boys were bullying girls and boys who did not confirm to gender norms and excluded them from their soccer play [55]. Another observational study showed that bullying was a very common behavior in playgrounds [61]. --- Strengths and Limitations The strength of the current study is the analysis of physical activity and social play behavior of children in neighborhood playgrounds and time away from school. This study also takes into account gender differences and provides real-time data by observation of physical activity and social environmental aspects in public playgrounds. A comprehensive data set of observations of play and physical activity behavior of individual children and their social environmental surroundings has been collected. This study also has some limitations. Firstly, only 10 playgrounds in one district of a middle-sized German town were assessed, only on weekdays and in the summertime of a single year. Furthermore, the number of boys and girls using playground equipment was somewhat low . Thus, these results are limited in their generality. Secondly, misjudgments could have occurred due to difficulties in estimating the age and gender of the playground users by subjective assessment using human observers. And as we did not capture the personal data of the observed children, we were not able to analyze the relevance of socio-economic status or migration backgrounds. Thirdly, the observations might have affected the children and playground users and their behavior. Finally, as we conducted an observational study with a quantitative observation protocol capturing presence and activity levels of playground users, we did not gather information on the interaction behavior of the social environment. Hence, we do not know if the social environment supported or hindered the physical activity and play behavior of the observed children. --- Conclusions The current study showed that the presence of active children contributed to girls' physical activity levels in playgrounds. Additionally, girls' physical activity was suppressed when boys were present in the playground, indicating that girls could have been distracted or bullied by boys simultaneously playing in the playgrounds [55]. Potentially, public playgrounds should provide more spaces for being physically active-especially for girls. The results of the present study indicated gender stereotypical play behavior in public playgrounds. It seemed to be necessary to break down these stereotypes to achieve more multifaceted play opportunities for boys and girls, respectively. However, physical activity levels in public playgrounds seem to converge between boys and girls and differences decrease. Hence, public playgrounds have the potential to foster physical activities in boys and girls and thereby contribute to child health. During weekdays, frequently visited public playgrounds like the playgrounds in the study district can offer opportunities to play with other children and to engage in moderate or vigorous physical activities together with other children. In summary, playgrounds seem to be places for boys and girls to be physically active and to interact with other children whereof they could benefit with respect to their physical activities and resulting health outcomes. --- Supplementary Materials: The following are available online at http://www.mdpi.com/1660-4601/15/7/1356/ s1. Table S1: Information on user characteristics of the playgrounds. Author Contributions: A.K.R. designed and conducted the data collection, planned the data analysis, and conceptualized and drafted the paper; S.S. contributed to the discussion and provided edits to the paper; Y.D. contributed to the discussion and provided edits to the paper; and G.K. conducted the statistical analysis, wrote the statistical analysis section, and provided edits to the paper. --- Funding: The publication costs of this article were funded by the German Research Foundation/DFG and the Chemnitz University of Technology in the funding programme Open Access Publishing. ---

Background: Few studies have delved into the relationship of the social environment with children's physical activity and outdoor play in public playgrounds by considering gender differences. The aim of the present study was to examine gender differences and the relationship of the social environment with children's physical activity and outdoor play in public playgrounds. Methods: A quantitative, observational study was conducted at ten playgrounds in one district of a middle-sized town in Germany. The social environment, physical activity levels, and outdoor play were measured using a modified version of the System for Observing Play and Leisure Activity in Youth. Results: In total, 266 observations of children (117 girls/149 boys) between four and 12 years old were used in this analysis. Significant gender differences were found in relation to activity types, but not in moderate-to-vigorous physical activity (MVPA). The presence of active children was the main explanatory variable for MVPA. In the models stratified by gender, the presence of opposite-sex children was a significant negative predictor of MVPA in girls but not in boys. Conclusions: The presence of active children contributes to children's physical activity levels in public playgrounds. Girls' physical activity seems to be suppressed in the presence of boys.

Introduction Increasing attention is being paid to the identification and prevention of unintended pregnancies globally and in sub-Saharan Africa [1,2]. An estimation of the prevalence of unintended pregnancy in sub-Saharan Africa showed that 39% of the 49 million pregnancies in 2008 were unintended, that is, they came earlier than desired or were not wanted at all [1]. There is ample evidence on the negative effects of unplanned pregnancy and fertility on infant, child and mother's health [3][4][5], household economic conditions, population growth, and the attainment of the Millennium Development Goals [6]. Unintended pregnancies have been shown to adversely influence maternal and child health seeking behaviors, birth outcomes, and women's quality of life [7][8][9][10]. It has been shown that women who experience an unintended pregnancy are more likely to seek an abortion, which in many cases will be illegal and unsafe in sub-Saharan Africa [3,11]. About a third of unintended pregnancies in sub-Saharan Africa are estimated to end in an abortion [1]. Recent studies in sub-Saharan Africa have been undertaken to examine the extent and determinants of unintended pregnancies. These studies have demonstrated that women experiencing unintended pregnancies are older, more likely to be unmarried, of higher parity, and poorer than women who have not experienced an unintended pregnancy [8,12]. A number of studies on unintended pregnancy prevalence and consequences in sub-Saharan Africa are quantitative [8,[12][13][14][15][16], while others include qualitative data collection [13,16,17]. Many of these studies are from rural areas [12,16], from Nigeria [13,[15][16][17] or from Eastern Africa [8,12,14]. The 2010-2011 Senegal Demographic and Health Survey indicates that 24.4% of pregnancies in the last five years among women ages 15-49 were considered to be unintended, including 20.4% that came too soon and 4% that was unwanted [18]. Secondary analyses of the 2010-2011 Senegal Demographic and Health Survey indicate that a greater percentage of women in urban areas reported their pregnancy as coming too soon or being unwanted , compared to women in rural areas [19]. In the Senegal context, where procreation is only considered socially acceptable within marital unions [20,21], it is not surprising that pregnancies that are experienced outside of union are often considered to be unintended. In particular, estimates from the 2010-2011 Senegal Demographic and Health Survey indicate that 81% of pregnancies in the last five years among unmarried urban women were reported as unintended. With increased urbanization in sub-Saharan Africa and in Senegal [22], understanding the extent and the determinants of unintended pregnancies is important for ensuring that all women have access to the most effective methods of family planning in order to reduce the occurrence of unintended pregnancies and lower the risks associated with unsafe abortion. With urbanization often come changes in social and sexual norms. In Senegal, a recent study demonstrated that about a third of female urban youth ages 15-24 reported being sexually experienced and among sexually experienced female youth, a third had premarital first sex; these youth are at risk of an unplanned pregnancy [23,24]. Other recent studies on youth in Senegal indicate that unintended pregnancies are often the consequence of a lack of knowledge about reproductive and sexual health as well as a lack of communication between young people and their parents [25]. Given that marriage and childbearing are closely tied in the Senegalese context, rarely has unintended pregnancy been examined among married women; this may be related to fatalistic and pro-natal attitudes in this mostly Muslim population [26]. In the urban context, where it is becoming more expensive to have large families and fertility desires are rapidly declining, studying the magnitude of unintended pregnancy is important for informing future family planning program strategies seeking to target urban Senegalese women most in need. This paper seeks to fill these gaps in our understanding about unintended pregnancy using recently collected data from women from urban Senegal. The objectives of this paper are to: a) examine the extent of unintended pregnancy among urban women who have ever been pregnant; b) identify key determinants of unintended pregnancy experience among those women; and c) make programmatic recommendations for improving urban women's access to and use of family planning to meet current and future fertility desires. --- Methodology The data from the baseline household survey of the Initiative Sénégalaise de Santé Urbaine were used for this study. This survey was implemented by the Measurement, Learning & Evaluation project in 2011. MLE is the evaluation component of the Urban Reproductive Health Initiative established by the Bill and Melinda Gates Foundation in three African countries and in the State of Uttar Pradesh in India. The data were collected from a representative sample of 9614 women ages 15-49 in six urban areas: Dakar, Pikine, Guédiawaye, Mbao, Kaolack and Mbour. Multistage sampling was used to obtain a representative sample of women from each site. In the first stage, enumeration areas were selected in each city with probability proportional to their size; a total of 268 enumeration areas were selected across the six sites. Following a detailed listing of households, twenty-one households were drawn randomly from each selected enumeration area, in the second stage. Finally, all eligible women ages 15-49 years in each household were approached and asked for consent to participate in the survey following completion of a household questionnaire. Prior to the interviews, the household head agreed that the interviewer can approach eligible teenagers to request their participation in the study. Surveyed women who had a pregnancy in the last two years prior to data collection, were asked if their last pregnancy was desired at that time, wanted later, or was not wanted at all. When the woman was pregnant at the time of the survey, the current pregnancy was considered in the analysis as the last pregnancy. A pregnancy was considered unintended if the women responded: "wanted later" or "not wanted". The main outcome variable for this study is the intentionality of the last pregnancy during the two years prior to the survey; this variable was coded as: 'not pregnant in the last two years' , 'intended last pregnancy' and 'unintended last pregnancy'. For this analysis of intentionality of the last pregnancy, the study population is women who have ever been pregnant in their life . The total number of women in the study sample is 5769 in the six study sites. The main independent variables considered include: education level , religion , marital status , age ,whether the woman worked in the last 12 months , number of living children and ever use of family planning . Other independent variables specific to women in union, such as type of marriage , who makes decision on household finances ,who makes decisions on the number of children to have within the couple and discussion between spouses on family planning within the couple are included in the model focused on the married sample. These couple-level variables, reported by women, allow the assessment of a possible influence of partner relations on the pregnancy outcome. Details of these variables are in Table 1. Also included in the analysis is an indicator of economic well-being, the household wealth, calculated as in DHS [27] from households' assets using principal components analysis. The variable is further recoded into a three equal category variable coded as poor, middle and rich. Final models also control for the urban site using Dakar as the reference group. Descriptive analyses were used to assess the level and trends of unintended pregnancies among the study population. Then, multinomial logistic regression analyses were undertaken to identify factors associated with the occurrence of unintended pregnancy, with women reporting unintended pregnancy as the reference group. Three different models were performed to examine a) all women in the sample ; b) only women in union ; and c) only women in union, including the couple-level variables . All descriptive and multivariate analyses were performed using weights and adjusting for the clustered nature of the data using the svy commands in Stata statistical software version 12. Ethical approval for the study protocol and the informed consent process was obtained from the University of North Carolina at Chapel Hill Institutional Review Board and from the Senegalese Ministry of Health's National Ethics Committee. --- Results --- Characteristics of the sample Table 1 shows the sample of women who have ever been pregnant in the six urban sites included. A large proportion of these women are from the Dakar site , reflecting the overall distribution of the population across the six sites. The sites of Mbour and Kaolack include low numbers as these are smaller cities . Women included are generally in their prime reproductive years, with about 60% under the age of 35. As expected, about a third of women are in each of the wealth groups. There is also a low level of education with nearly 8 women out of 10 not reaching the secondary level. As is found throughout Senegal, the overwhelming majority of the sample is Muslim. More than 80% of the women who have ever had a birth are in union. This is not surprising in the Senegalese context where marriage and childbearing remain strongly connected [20]. About seventy percent of the women in union are in a monogamous union and 29.9% are in a polygamous union. The sample is mostly composed of unemployed women with 85.6% who have not worked in the last 12 months preceding the survey. The percentage of women who report that they have control over their financial resources is 42.1%. Notably, 48.3% of the women have already had more than 3 children and 60.3% ever used modern family planning including sterilization, intrauterine device, injections, implant, pills, male condom, female condom, emergency contraception, lactational amenorrhea or spermicides. About 6 women in union out of 10 women in union have ever discussed family planning with their spouses and more than half report that the decision on the number of children to have is made by the woman or jointly with her husband/partner. --- Experience of unintended pregnancy by socio-demographic characteristics As shown in Table 2, half of the ever pregnant women did not have a pregnancy during the past two years prior to the survey. Thirty six percent of the sample reported their last pregnancy in the last two years as intended and the remaining women reported their pregnancy in the last two years as unintended. Across the cities, women from Dakar and Guédiawaye were the least likely to have had a pregnancy in the last two years. The other groups without experience of pregnancy during the last two years, include rich women , women over 40 years old , Christian women , and unmarried women . Therefore, more than half of the sample has birth intervals that appear reasonably long . Looking at the sub-groups with higher prevalence of unintended pregnancy in the last two years, it is noted that 20.0% of women in Pikine reported their last pregnancy in the last two years as unintended. Likewise, 20.1% of poor women reported an unintended pregnancy in the last two years. Finally, a quarter of the youngest women reported that their last pregnancy in the last two years was unintended. --- Multivariate analyses The multinomial logistic regression analysis to examine the factors associated with the occurrence of an unintended pregnancy is presented in Table 3 and Table 4 --- . As shown in Table 3, Model 1 is carried out for all women in the sample. The results show that richer women are less likely to have an unintended pregnancy than to have had no birth in the last two years as compared to poorer women; conversely, they are more likely to be non-pregnant in the last two years than to have experienced an unintended pregnancy. Women aged 25 or older are less likely to have had an unintended pregnancy than to have not gotten pregnant as compared to women under age 25. The opposite pattern is found by number of living children; women with two or more living children are more likely to have had an unintended pregnancy than to have had no pregnancy as compared to women with less than 2 children. Model 1 also includes the comparison between women who had an unintended pregnancy versus women who had an intended pregnancy. As a main result, richer women are less likely to have had unintended pregnancies than intended pregnancies than poorer women. Likewise, women in union and women age 25 and older are also less likely to have had unintended pregnancies than intended pregnancies as compared respectively to unmarried women and women under age 25. Having two or more children is associated with a greater likelihood of having an unintended pregnancy than an intended pregnancy as compared to having fewer than two children. Also, women who ever used a family planning method are more likely to have had an intended pregnancy than an unintended pregnancy compared to women who never used a FP method. Finally, women living in Guédiawaye and Pikine are more likely to have had an unintended pregnancy than an intended pregnancy. Model 2 is the same as Model 1 but it is developed only for women in union. Overall, the same pattern for all women is observed; this is a consequence of the fact that the overwhelming majority of the women in the sample are married or in union. The only difference is that the use of a contraceptive method in the past is only associated with the distinction between having an unintended vs. an intended pregnancy; women who ever used contraception are significantly less likely to have had an unintended than an intended pregnancy. Table 4 which includes Model 3 is for the same sample as in Model 2 but includes the couple-level variables only asked to women in union. Model 3 shows the same results for the demographic factors as found in Model 2. The examination of the couple-level variables indicates that decision-making about income and discussion of FP between spouses are associated with pregnancy experience and intentionality of the pregnancy. In particular, women who work and are involved in decision-making regarding the management of their own financial income, as compared to women who work and are not involved in decision-making, are less likely to have had an unintended pregnancy. In particular, these women are more likely to have been non-pregnant or to have had an intended pregnancy in the last two years. Further, women who did not work in the last 12 months as compared to women who work and are not involved in financial decision-making are less likely to have had an unintended than an intended pregnancy in the last two years. Finally, women who have discussed family planning with their partner are significantly less likely to have had an unintended pregnancy in the last two years and more likely to have had no pregnancy in the last two years, as compared to women who did not discuss family planning with their partner. --- Discussion Although previous studies have shown that the risk of unintended pregnancy is higher among unmarried women [21,25], this study demonstrates that married women also experience unintended pregnancies. This result is indicative of unmet needs for family planning among urban women in union and the need to pay greater attention to groups traditionally thought to have lower need for family planning. This study also demonstrates important distinctions between urban women who have intended pregnancies and those who have unintended pregnancies. Indeed, women with unintended pregnancies are more likely to be poor, from a young age group and multiparous . Moreover, it appears that a low involvement of married women in decision making within the couple and a lack of discussion on FP with the partner are associated with higher experience of unintended pregnancies. Our findings are similar to those from other studies in sub-Saharan Africa on the extent of unintended pregnancies and factors associated with the occurrence of unintended pregnancies. However, it should be noted that most of these studies are at a national level [28,29] or include only rural samples [13]. The results of our analyses suggest the need to focus on improving the targeting of family planning programs to urban women, particularly urban poor women, as a way to ensure that they can meet their changing fertility desires. Some socio-demographic factors are not significantly associated with the occurrence of unintended pregnancy. That result was expected given the homogeneity of the sample relatively to the religious group . Our findings indicate differences in unintended pregnancy experience across the urban sites included. Women from both Guédiawaye and Pikine experience more unintended pregnancies than women in Dakar, even after controlling for the wealth groups. Given that these sites are both part of the region of Dakar, future studies are needed to better understand ethnic, religion, and behavioral differences in these sites; we also need more information on access to contraception and other health services in these sites. Case studies that include qualitative data collection may be needed to obtain a clearer picture of why these sites are higher risk for unintended pregnancies than the Dakar site. This study is not without limitations. First, the fact that pregnancies not resulting in a live birth were not taken into account in the study constitutes a source of bias. Those pregnancies that end in abortion are likely to be unintended and thus the true prevalence of unintended pregnancy is likely higher than shown here. In addition, the retrospective question on intentionality of the pregnancy could lead some women to reconsider their responses now that the birth took place. Similarly, Table 3 Multinomial logistic regression relative risk ratios and 95% CI from analysis of whether ever pregnant women had an intentional or an unintentional pregnancy/birth or no pregnancy/birth since 2009 among all women and then among women in union by socio-demographic characteristics, six urban sites in Senegal, 2011 because pregnancy intentions are self-reported, women may under or over report unintended pregnancies and there is no way to know the direction of this effect. Finally, the data are cross-sectional and thus it is not possible to know the direction of causality between the variables of interest. For example, while we hypothesize that married women who speak to their spouse are less likely to have unintended pregnancies, the association may be the other way, the experience of an intended may lead women to discuss family planning and future fertility desires. With the data available, it is not possible to know the true direction of causality and thus we discuss associations between these interpersonal variables and experience of an unintended pregnancy. Notably, these limitations should have a minor impact on the scope of the study in view of the large size of the sample and that the main variables of interest are demographic factors associated with experience of an unintended pregnancy. --- Conclusion and recommendations This study demonstrates that unintended pregnancies in urban Senegal affect both unmarried and married women and the main correlates are parity, age and economic status. In urban areas where non-marital sex is becoming more common, and in a setting like Senegal that is predominately Muslim, programs need to consider strategies to get information and counseling to high risk women. This may mean undertaking outreach in poorer urban sites and providing community-based distribution of family planning methods or counseling and referral for women who want methods not available through outreach approaches. In addition, programs can be undertaken to target youth through youth corners in existing health facilities or training providers in offering youth friendly services. Implementation of targeted programmes will guarantee access to family planning for all categories of women in need. In urban areas characterized by economic insecurity, as in Senegal, it is essential to also consider strategies for promoting communication within couples on fertility issues. It is these types of targeted approaches that can help urban women to meet their fertility desires and reduce unintended pregnancies with the overall objective of reducing maternal mortality and morbidity in urban Senegal. --- Competing interests The authors declare that they have no competing interests. --- --- Authors' information

Background: In Senegal, unintended pregnancy has become a growing concern in public health circles. It has often been described through the press as a sensational subject with emphasis on the multiple infanticide cases as a main consequence, especially among young unmarried girls. Less scientific evidence is known on this topic, as fertility issues are rarely discussed within couples. In a context where urbanization is strong, economic insecurity is persistent and the population is globalizing, it is important to assess the magnitude of unintended pregnancy among urban women and to identify its main determinants. Methods: Data were collected in 2011 from a representative sample of 9614 women aged 15-49 years in six urban sites in Senegal. For this analysis, we include 5769 women who have ever been pregnant or were pregnant at the time of the survey. These women were asked if their last pregnancy in the last two years was 'wanted 'then' , 'wanted later' or 'not wanted'. Pregnancy was considered as unintended if the woman responded 'wanted later' or 'not wanted'. Descriptive analyses were performed to measure the magnitude of unintended pregnancies, while multinomial logistic regression models were used to identify factors associated with the occurrence of unintended pregnancy. The analyses were performed using Stata version 12. All results were weighted. Results: The results show that 14.3% of ever pregnant women reported having a recent unintended pregnancy. The study demonstrates important distinctions between women whose last pregnancy was intended and those whose last pregnancy was unintended. Indeed, this last group is more likely to be poor, from a young age (< 25 years) and multiparous. In addition, it appears that low participation of married women in decision-making within the couple (management of financial resources) and the lack of discussion on family planning issues are associated with greater experience of unintended pregnancy.This study suggests a need to implement more targeted programs that guarantee access to family planning for all women in need. In urban areas that are characterized by economic insecurity, as in Senegal, it is important to consider strategies for promoting communication within couples on fertility issues.

Introduction Ovarian cancer is one of the most challenging gynecological cancers in the world, with a profound negative effect on the quality and length of life affecting more than 70% of the women diag-nosed with advanced disease. 1 The absence of specific symptoms and useful screening tests make it difficult to establish an early diagnosis of ovarian cancer. In addition, most women have little or no knowledge about the risks, possible symptoms, and the dangers ovarian cancer poses to them. 1 Worldwide, ovarian cancer is the seventh most frequent cancer in women, accounting for 3.6% of various cancers diagnosed in women. It is the third most prevalent genital tract malignancy after cancers of the cervix and uterus. 1 According to the UAE cancer registry report in 2014, ovarian cancer is the fifth most commonly diagnosed cancer affecting women with a prevalence of 4%. 2 In the past, ovarian cancer was believed to be a silent killer due to a lack of symptoms in the early stage leading to late presentation of the disease with its related high mortality rates. 3 In the absence of effective screening, "appropriate clinical suspicion" of the signs and symptoms of the disease is currently the best strategy to detect ovarian cancer. 4 The hypothesis that ovarian cancer outcomes could be improved through better symptom recognition has been primarily advanced by Goff et al. who proposed a "symptom index" as a tool to screen women for further evaluation for ovarian cancer. The index is considered positive if pelvic or abdominal pain, increased abdominal size, bloating, difficulty eating, or early satiety occur more than 12 times per month but for less than one year. 5 Unfortunately, because these symptoms are non-specific, they may be attributed to other causes either by the women themselves 6 or by an inexperienced physician. 7 When these symptoms are detected promptly and investigated appropriately, they may aid early diagnosis and management, resulting in better prognosis and outcome for the patients. 8 This is because the survival rate of ovarian cancer depends significantly on the stage of the disease with the early-stage disease having more than a 90% five-year survival rate compared with less than 30% in advanced stage disease. 9 The --- Significance for public health In the absence of effective screening, "appropriate clinical suspicion" of the signs and symptoms of the ovarian cancer is currently the best strategy to detect the disease. To reduce the diagnostic interval, several countries have implemented referral guidelines and organizational changes. Most of these guidelines suggest that individuals presenting with symptoms indicative of cancer should be urgently referred to specialized investigative trajectories. However, many studies worldwide have shown that public awareness levels of ovarian cancer are poor, therefore women are often diagnosed at late stages when treatment is difficult. Our findings demonstrate that the overall level of recognition of risk factors and symptoms of ovarian cancer among women in the UAE is indeed low and within a critical level. Thus, strengthening education among women with regards to risk factors, early symptoms, and the need to seek early medical will be helpful in increasing ovarian cancer awareness and reducing the delay in diagnosis. --- Article risk of developing ovarian cancer is increased with advanced age; family history of ovarian, breast, uterine, and colon cancer; use of fertility drugs and nulliparity. 9 However, the awareness of these risk factors and symptoms of ovarian cancer is generally poor among women. 8,9 Several studies have shown that late-stage cancer diagnosis is associated with reduced survival. 10 This is especially the case for gynecological cancer, and timely diagnosis and treatment are thus considered essential for prognosis. For most patients with ovarian cancer, the diagnostic process is still initiated based on symptom presentation, although some patients are diagnosed through screening programs. 11 To reduce both the patient interval and the diagnostic interval, several countries have implemented referral guidelines and organizational changes. 12 Most of these guidelines suggest that individuals presenting with symptoms indicative of cancer should be urgently referred to specialized investigative trajectories. Many studies worldwide have shown that public awareness levels of ovarian cancer are poor, and women are often diagnosed at late stages when treatment is difficult. 13,14 Poor public knowledge of early symptoms of cancer has been considered to be the predominant reason for late-stage presentation or delay in the diagnosis. 15 The purpose of this study was to determine the level of knowledge and awareness among women regarding ovarian cancer risk factors, symptoms, and barriers to seek early medical care in the UAE. --- Design and Methods --- Study design This descriptive cross-sectional survey was conducted among women of all nationalities; from September 2019 until April 2020. --- Study participants and procedure The participants living in the UAE and willing to provide written informed consent were selected for the study. The participants were assured of the confidentiality of the information provided and the protection of their rights to privacy, mandated by the research ethics guidelines of the human research ethics committees. The inclusion criteria mandated participants to be women aged 20 and above, who are not health care providers, willing to participate in the survey and able to sign informed consent, having appropriate cognitive skills, ability to read and understand in English or Arabic . Both UAE citizens and expatriates were included in the study. Participants that had poor mental health and were unable to meet any of the above criteria were excluded from the study. The sample size was calculated using epidemiological information for a population of 682,000 , with an alpha of 5% and a 95% confidence level. We needed a minimum required sample size of 384. 16 --- Survey design A structured questionnaire was designed and adapted from the validated cancer awareness measure , 1 ovarian cancer awareness measure , 17 and a questionnaire developed by the National Ovarian Cancer Coalition. 18 The evaluation tool was then pre-tested among 20 women to assess the ease of understanding and time required for completion. The bilingual questionnaire in Arabic and English is represented in APPX I. The survey instrument consisted of 44 items. The first part of the questionnaire collected sociodemographic data, including Age, Nationality, Ethnicity, Marital Status, Education Level, Employment Status, Monthly Family Income , history of cancer, Smoking status, and Body Mass Index . The remaining sections of the questionnaire focused on knowledge and attitude toward ovarian cancer. Selected questions under ovarian cancer knowledge , and practices were used to develop a scaled scoring system to categorize these aspects as good, fair, or poor. Each correct or favorable answer chosen by participants was given one or two points; then, the points were summed for each participant. No points were deducted for wrong or unfavorable answers. The participants' scoring system was as follows: --- Article -Knowledge: Good , Fair , Poor -Practice: Good , Fair , Poor . --- Data analysis and statistics All collected data was entered into STATA version 15 for statistical analysis. Descriptive statistics were computed for the socio-demographic variables. The overall responses to each item of the survey were recorded as a percentage of the total. The percentage differences in the total responses were determined using the Chi-square test and statistical significance recorded for non-parametric data. For all analyses, alpha was set at 0.05. We selected only those variables that had a moderate or higher correlation that was also statistically significant . A Pearson correlation of 0.00-0.19 is considered to be very weak, 0.20-0.39 weak, 0.40-0.59 moderate, 0.60-0.79 strong, and 0.80-1.0 very strong. We also calculated Φ as an effect size for all significant correlations . --- Results Out of 602 individuals approached, 532 agreed to participate in the study . The majority of our respondents were between 25-39 years, UAE national, with middle eastern ethnicity, retired or homemakers, had college or higher degrees certificate, married, had low income, never smokers, and having an average body mass index . Most of our respon- dents have read about ovarian cancer, had regular check-ups with physicians, did not have a family history of ovarian or breast cancer. --- Article It was encouraging that the majority of individuals surveyed mentioned that they had heard about ovarian cancer; however, when we assessed their knowledge, the majority had fair or poor knowledge . When evaluating the attitude of participants, the majority mentioned that if they notice a symptom that might be a sign of ovarian cancer, they will visit a doctor within a week. Yet, most of the participants mentioned that they are not that confident on noticing an ovarian cancer symptom and most of the participants did not relate age with the incidence of cancer. We calculated Pearson's correlation coefficient between variables; a multiple regression model was also fitted to explore the role of possible predictors of the knowledge score. There were no statistically significant predictors of the Knowledge score using a multiple regression model. There was a significant difference between UAE nationals and non-UAE nationals on the regard of hearing about ovarian cancer, having a family history of ovarian cancer , and having a family history of breast cancer . There was a significant difference between educated and non-educated women participants in confidence level regarding noticing ovarian cancer symptoms . Having a history of cancer, or having a family member or a friend with history of cancer was significantly associated with higher ovarian cancer knowledge score. Reading about ovarian cancer was significantly associated with BMI and having a regular checkup with physicians . There were no statistically significant predictors of the Knowledge score using a multiple regression model. Knowledge score was only significantly associated with educational level . --- Discussion Ovarian cancer has been regarded as a "silent killer" due to the majority of patients being diagnosed at later stages of the disease with low 5-year survival rates . 19 To our knowledge, this is the first study conducted in the UAE to evaluate the level of awareness among women of ovarian cancer risk factors, symptoms, and anticipated time to seek early medical help should they develop suspicious symptoms. The recognition level of ovarian cancer symptoms among women in the UAE in this study was varied from 12-51%, and the most alarming ones, back Table 2. Knowledge on ovarian cancer among the women participants . The answers were on Likert scale as a five-point scale; however, they were re-grouped to 3 scales for knowledge scoring. --- [page 496] [Journal of Public Health Research 2021; 10:2073] pain, eating difficulties, persistent abdominal pain and bloating, were missed by most of the surveyed population. The most recognized ovarian cancer risk factors among UAE women were never having been pregnant, having a history of infertility, and going through menopause, while the least recognized were having a personal or family history of breast cancer, having ovarian cysts, and having a genetic predisposition for cancer. Previous studies conducted in several countries, such as USA, Malaysia, Oman, and Jordan, have shown that knowledge of risk factors of ovarian cancer among women was also low as found in our study. 1,3,20,21 The observed positive association between education level and awareness of ovarian cancer risk factors and symptoms indicate that educated women are in a better position to access and understand information regarding ovarian cancer. It is not surprising that women with a tertiary level of education had significantly better knowledge as compared with their counterparts. This is similar to the findings from other countries, where women with higher levels of education were more knowledgeable of the symptoms and risk factors of ovarian cancer. 6,22 Thus, education is an essential factor in increasing awareness of cancer symptoms and predisposing cancer risk factors, such as smoking, physical inactivity, obesity, and an unhealthy diet. 23 The role of education in increasing awareness --- Article Table 3. The relationship between knowledge and practice on ovarian cancer with demographic characteristics among participant women are presented. among women of risk factors and symptoms of ovarian cancer in this study confirms findings from earlier studies conducted worldwide, which showed that people with higher levels of education recognized cancer symptoms better than those who were literate alone or less educated. 20 Indeed, previous studies have shown that a low level of education is associated with a delayed presentation among breast and colon cancer patients, whereas those with a higher level of education showed a greater knowledge of early symptoms. 15,24 Our study also revealed that women were more knowledgeable about the risk factors of ovarian cancer compared to its symptoms, and this is consistent with the findings of another investigation. 21 Although participants did better in recognizing ovarian cancer's risk factors than its symptoms, it is still evident that they have only a low knowledge level. Knowledge of ovarian cancer risk factors is fundamental; it may facilitate participation in screening programs and early detection. This emphasizes the potential importance of raising women's awareness of the symptoms and risk factors, particularly in a country like UAE, where low awareness and knowledge levels seem to be dominant. This could be achieved by using announcements in the general media and social media channels. Leaflets could be distributed in local health centers, and hospitals in different regions in the UAE that might help to increase ovarian cancer awareness. Investigating potential interventions to increase awareness would seem to be the next step in our research. Besides, ovarian cancer awareness programs comprising regular lectures and workshops for women in general and women in secondary schools and universities should also be considered. When developing such programs, specific considerations for a UAE context should be borne in mind. First, an educational program should target women with low incomes and low levels of education. Second, these programs should be freely developed and easily accessible to the former group at schools, and primary health care centers close to their homes. To improve women's awareness of ovarian cancer symptoms and change their attitudes toward early detection practice and programs, using a change or implementation model is highly recommended. --- Conclusions Our findings demonstrate that the overall level of recognition of risk factors and symptoms of ovarian cancer among women in the UAE was low and within a critical level. Thus, strengthening education among women with regards to risk factors, early symptoms, and the need to seek early medical help should help to increase ovarian cancer awareness and reduce a delay in diagnosis. --- Limitation of study The results of this study are limited due to the cross-sectional study design; hence, any association found from the results cannot infer causal relationships. The findings of the study cannot be generalized for a larger population as the study respondents were recruited from the Dubai health authority, but could be considered indicative of the context measured. Nevertheless, a more extensive national study with recruitment of women from all emirates in the UAE is needed for better representative sampling. However, despite these limitations, the present study findings may be helpful to the health policymakers, health educators, clinicians, and nursing professionals in designing the interventions to increase the knowledge and awareness of women in the UAE on ovarian cancer symptoms and risk factors. --- Availability of data and material: All data and materials are available up on request. --- Ethics approval: The study was approved by the institutional review boards. All participants gave written informed consent before participation. Aggregate reporting of data assured to enhance confidentiality and accurate reporting by the respondents. The return of the completed survey also guaranteed the anonymity of participation constructs to an administrator, independent, and blinded to the study hypothesis. A code linking respondents to their surveys was kept isolated from the investigators. --- Funding: The authors have not received any funding or benefits from industry or elsewhere to conduct this study. Contributions: All authors participated in conducting the survey and writing the manuscript. SS, NA, SSS, ZSA, FNA, conceived the idea and study design; FRC and MA, data analysis, interpretation of results, study revision; SS, NA, SSS, ZSA, FNA participated in development of the data collection tool. FRC and MA critically revised, and approved the final version of the manuscript. --- Conflicts of interest: The authors reported no potential conflict of interest

Background: Studies have shown that public awareness of ovarian cancer is weak, and women are often diagnosed at late stages when treatment is difficult. This study aimed to determine the barriers for seeking early medical care and level of knowledge and awareness among women regarding ovarian cancer. Design and methods: A structured questionnaire was designed and adapted from the validated cancer awareness measure. Results: The recognition level of ovarian cancer symptoms among women in this study varied from 12-51%, and the most alarming ones, back pain, eating difficulties, and persistent abdominal pain, were missed by most of the surveyed women. The most recognized ovarian cancer risk factors were never being pregnant, having a history of infertility, and going through menopause, while the least recognized was having personal or family history of breast cancer. Non-UAE nationals women with higher educational levels had better knowledge of ovarian cancer compared to their peers. Conclusions: A change or implementation model is highly recommended in order to improve women's awareness of ovarian cancer symptoms. Strengthening education among women with regards to risk factors, early symptoms, and the need to seek early medical help should help to increase ovarian cancer awareness and reduce a delay in the diagnosis.

Background At the close of 2019, SARS-CoV-2, the virus that causes COVID-19, emerged in Wuhan, China, and rapidly spread throughout the world, being labeled a pandemic in only a few months after its emergence [1,2]. Nearly 2 years after the initial outbreak, as of April 8, 2022, there have been 496,355,574 confirmed cases of COVID-19, 6,170,720 cumulative deaths, and 11,085,254,518 vaccine doses administered globally [3]. The Omicron variant is also circulating globally, quickly driving up cases in regions that had previously seen reprieve owing in part to high rates of vaccination [4,5]. The only way to prevent COVID-19 from occurring is through both reducing exposure to SARS-CoV-2 and reducing susceptibility. Since COVID-19 is primarily transmitted through air droplets and aerosols [6,7], the main strategies for reducing exposure are social distancing [8], mask use [9,10], and limiting time spent in poorly ventilated spaces where air is shared with other people [11,12]. The main strategy for reducing susceptibility is vaccination [13,14]. With respect to COVID-19 prevention, individuals' personal decision-making about behaviors to reduce their susceptibility and/or exposure influences risk at both the individual and community-wide levels [15,16]. Individuals' health and illness-related decision-making is affected by many different sources of influence such as family members and friends, health care providers, political leaders, and the media. A plethora of behavioral science theories demonstrate how individuals' levels of knowledge and beliefs, skills, availability and accessibility to resources, and social support shape health choices [17][18][19][20][21]. Over the past decade, examination of the role played by social media in personal decisions that affect individual, family, and population health, including those related to infectious disease, has expanded rapidly. This viewpoint highlights the challenges, risks, and potential benefits that social media present in mitigating the COVID-19 pandemic. --- Challenges Airborne transmission of droplets and aerosols is the main way SARS-CoV-2 is transmitted [6,7]. Secondarily, although far less frequent, the virus can also be transmitted through droplets and aerosols settling on fomites and having hand/nose or hand/mouth contact [22]. In the early stage of the pandemic, emphasis was placed on transmission through fomites, and on handwashing and surface disinfection, as well as social distancing. Yet, little to no emphasis was placed on masking. Messaging about fomite transmission was eventually dialed back [23,24], while messaging on mask recommendations was altered entirely. In times of global and national crisis, the public clamors for the guidance of trusted experts to navigate rapidly evolving and potentially life-threatening situations. Early on, the experts at Centers for Disease Control and Prevention , the World Health Organization , and the US Surgeon General blasted messages that subsequently required modification. False messages of conspiracy rapidly surfaced, and charlatans promoted unsubstantiated claims to prevent disease. To some degree, future expert messages were discredited. For example, on February 29, 2020, US Surgeon General Dr Jerome Adams used Twitter to send the following message to the public: "Seriously people-STOP BUYING MASKS! They are NOT effective in preventing general public from catching #Coronavirus, but if healthcare providers can't get them to care for sick patients, it puts them and our communities at risk!" [25]. This sentiment was in concert with the early mentality of the CDC, WHO, US Office of the Surgeon General, and other constituents at the time. All of these entities would later reverse their views as further scientific evidence emerged regarding the ambient transmission of SARS CoV-2 [26][27][28][29][30]. With spread of the Delta variant during the summer of 2021, CDC messaging encouraged unvaccinated individuals to wear masks in crowded public places and for vaccinated individuals to wear masks indoors in areas of high transmission [31]. When the Omicron variant began to spread in late 2021, news reports began appearing with quotes from health experts advising the public to upgrade their masks [32][33][34]. However, the official CDC guidance on masking was not clarified until mid-January 2022 [31]. In a context of rapidly evolving scientific discovery, a key challenge for public health education is to adapt messaging in an information environment containing older content that continues to live and circulate online despite being obsolete or inaccurate from the outset. A second major challenge is that public health experts may disagree about recommendations. Communicating the nuances of scientific progress may be especially hampered by the fact that nearly 60% of the population has medium or low scientific literacy [35]. Given the uncertainty involved in an emerging global pandemic, the incremental and often contentious way in which scientific knowledge typically advances, and shifting and diverging recommendations, it is not surprising that public confusion is common and trust in official governmental agencies is hindered [35]. --- Risks Large numbers of individuals search for health information online [36,37]. As with previous infectious diseases [38], the public turned to the internet for information on the novel coronavirus [39][40][41]. Unfortunately, false information about the COVID-19 pandemic on social media has been extensive [42][43][44] and viewed by millions of people [45]. False information involves both misinformation, or inaccurate conclusions about a phenomenon drawn from incomplete or incorrect facts, as well as disinformation, which pertains to the purposeful spread of false information in line with a specific agenda [46]. The early days of the pandemic saw such a proliferation of false information online, leading the WHO Director General to declare that the fight was not only against the pandemic but against an "infodemic" as well [47]. An "infodemic" can be described as "an overabundance of information -some accurate and some not -that makes it hard for people to find trustworthy sources and reliable guidance when they need it" [47]. Preventing COVID-19 involves both reducing exposure to SARS-CoV-2 through masking, social distancing, and the avoidance of crowded, poorly ventilated spaces, as well as through reducing susceptibility via vaccination. In the United States, there are two types of vaccines available: mRNA vaccines and an adenovirus-based vaccine [48,49]. It has become clear that one of the concerns that the public has about the vaccine pertains to the comparatively short time spent on development, testing, and approval . Officials sought to assure the public that neither safety nor scientific integrity was compromised in the process [50][51][52]. Nevertheless, public concern that mRNA would alter one's genes and fear surrounding an untested technology persisted [53,54]. Surveys in the United States about attitudes toward vaccination continue to reveal persistent levels of hesitancy and refusal [55], with notable demographic and political differences [56][57][58][59][60]. The trustworthiness of online information about COVID-19 vaccination is endangered by those whose aim is to spread disinformation. Disinformation can be spread by anyone. One estimate was that 12 individuals were responsible for 65% of the mis/disinformation spread on Twitter, Instagram, and Facebook [61]. These individuals were identified as "anti-vaccine activists, alternative health entrepreneurs and physicians" [61] who spread a wealth of antivaccination content over the course of the pandemic [62]. The danger of mis/disinformation is epitomized by its use "to incite violence and crime targeted at ethnic minorities -which has resulted in deaths and displacement of children, led to lower child COVID vaccination rates, undermined trust in journalism and science, and drowned out marginalized voices" [63]. Mis/disinformation in times of health emergencies threatens national security and there have been calls for immediate intervention [64]; however, this is not easily accomplished in a societal context with freedom of speech and access to digital communication channels capable of reaching millions of people in minutes. A small number of social media accounts spreading disinformation have had widespread reach and have gained many followers during the pandemic [62]. To the extent that this small number of antivaccination messengers attract many users who are skeptical or are "on the fence" about vaccination, they can undermine efforts to mitigate community transmission and increase the risk of more pathogenic and virulent variants. The influence of misinformation and disinformation on acceptance of inaccurate COVID-19 information, behavioral intentions, and actual behavior remain equivocal [65][66][67]. Some evidence suggests that such communications are associated with acceptance of COVID-19 conspiracy theories [65,68] and other inaccurate information [69]. A belief in conspiracy theories may reduce the uptake of at least some preventative behaviors [70,71]. In sum, while the behavioral impact of misinformation and disinformation on social media is debated, what is clear is the role that social media has played in extending the reach of COVID-19-related falsehoods. Of course, even if low-quality or blatantly untrue COVID-19 online information has a minimal impact on COVID-19 exposure and susceptibility reduction, in the case of a highly contagious respiratory virus, the impact on community spread may nevertheless be of considerable concern. --- Benefits of a Multifaceted Public Health Presence on Social Media It has been well-established that social media can spread accurate, useful information. During the pandemic, social media companies have claimed to make efforts to remove harmful COVID-19 content and promote accurate information [72][73][74], although the degree to which they have been successful in doing so has been questioned [75]. Previous investigations have demonstrated examples of how social media have promoted positive behaviors such as the Global Handwashing Partnership [76]. Some research suggests that although misleading COVID-19 information has been widely shared on social media, evidence-based content may have been shared to a greater extent. Fortunately, public health agencies such as the WHO and CDC have increased their presence on key social media platforms [77]. Beyond creating their own content, public health agencies have begun to look to social media personalities to help disseminate prevention messages. For example, in 2020, the state of Texas moved to pay influencers to promote mask use and social distancing on TikTok [78]. We previously identified a shift in popular YouTube videos regarding the source of messages on COVID-19 from news media [79] to entertainers [80]. From January to March 2020, there was a large increase in the number and proportion of cumulative views garnered by widely viewed videos on hand hygiene, from 33,268,243 to 182,331,135 [80], and there was an increase coverage of face masks in the most popular videos [80]. Our work examining mechanisms of disease transmission across three successive samples indicated an increase from ~63 million views to more than 273 million views for videos specifically mentioning disease transmission. This increase coincided with a rise in the worldwide number of cases and the occurrence of COVID-19 transmission [81]. A Medline search performed at the time of writing this viewpoint indicated that these are among the first successive samples of YouTube videos published to date. Our studies suggest that tracking changes over time is useful and that social media may help to amplify scientific recommendations. Notably, successive sampling captures evolving situations rather than labeling something as misinformation simply because it was posted at a time when science about the topic was evolving. Similar to YouTube, there is active conversation about COVID-19 on TikTok. Earlier research suggested that, despite 1 billion views with the #Coronavirus hashtag, there was minimal discussion about transmission [82]. However, with more directed searching, we identified messaging about prevention [83]. Notably, a search of the hashtag #WearAMask indicated that the number of views of TikTok videos mentioning masks uploaded by the WHO was far less than those in the "Wear a Mask" campaign, despite the fact that the messages from the WHO were more aligned with scientific recommendations [6]. Humor, music, and dance were large components of the #WearAMask posts, but were not common in those from the WHO [83]. Consistent with the study on YouTube, these findings suggest that popular entertainers and the use of nonconventional approaches to messaging can help draw attention to public health messages. While conversations on TikTok could be health-promoting [83][84][85], they may also be counterproductive as was noted in our study of misinformation related to vaccination [86]. Thus, while there are challenges with and cause for concern regarding COVID-19 content on social media, social media clearly play a vital role in disseminating accurate information. --- Moving Forward Public health communication best practices suggest that in cases where science is not definitive, messages should be calibrated to make it clear that there is an evolving context [87]. Our work indicates that the content of official public health agency social media accounts does not receive as many views as communications posted by popular entertainers, influencers, or, in some cases, even consumers. Partnerships with carefully vetted content creators may help to extend the reach of accurate health information on social media, particularly to young people who tend to use these platforms in higher numbers than older adults [88]. Because there is a propensity for research highlighting misinformation as a means to gain attention [89,90], the dialog should be clear about what constitutes misinformation. Researchers should consider the date that material was posted and how scientific understanding changed over time, rather than aim to produce "clickbait" titles. There are degrees of misinformation and calibrating messages with a disclaimer will also help in this circumstance. Studies claiming they have identified misinformation should be forthcoming about how this was coded and what factors were taken into consideration, with concrete examples of what did and did not constitute misinformation and disinformation. Effective communication and education should be geared toward a very heterogenous public in terms of age, beliefs, culture, sociodemographics, literacy, and information-seeking sources. Over the last decade, social media have become a dominant source of information that consumers turn to for health and COVID-19 information. This situation prompted our early work to determine what is and is not being communicated. The findings to date suggest there is considerable room for improvement. A great deal of misinformation and disinformation has reached large numbers of social media users, pointing to a need for the agencies of the US Public Health Service to create communications to convey accurate and current information and appeals that will actually be viewed. --- Conflicts of Interest CHB serves as an Editorial Board Member for JMIR; she did not have a role in the review or editorial process for this article. This is an open-access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included. --- Abbreviations

Shortly after the first case reports in 2019, COVID-19 was declared a pandemic. Early messages from trusted experts, which later proved to be inadequate or incorrect, highlight the need for continual adjustment of messages to the public as scientific knowledge evolves. During this time, social media exploded with greatly sought-after information, some of which was misinformation based on incomplete or incorrect facts or disinformation purposefully spread to advance a specific agenda. Because of the nature of social media, information, whether accurate or not at the time posted, lives on and remains accessible to the public even when its usefulness has been discredited. While the impact of mis/disinformation on COVID-19 risk-reducing behaviors is debatable, it is clear that social media has played a significant role in both extending the reach of COVID-19-related falsehoods and promoting evidence-based content. Over the last decade, social media has become a dominant source of information that consumers turn to for health information. A great deal of misinformation and disinformation has reached large numbers of social media users, which points to a need for the agencies of the US Public Health Service to create communications to convey accurate and current information and appeals that will actually be viewed. This viewpoint highlights the challenges, risks, and potential benefits that social media present in mitigating the COVID-19 pandemic.

Indexing Key Words Manuscript format: research; Research purpose: modeling/relationship testing; Study design: nonexperimental; Outcome measure: behavioral; Setting: local community; state; Health focus: nutrition; Strategy: built environment; Target population: adults; Target population circumstances: education/income level; race/ethnicity; geographic location --- PURPOSE The Survey of Health of Wisconsin estimates that 37.8% of Wisconsin's adult population is obese 1 as calculated from the height and weight of participants measured by the field staff . These obesity rates are even higher than the national obesity prevalence calculated from self-reported height and weight data by Centers for Disease Control and Prevention ; and thus highlight the importance of understanding underlying factors contributing to excess weight gain in Wisconsin. Research shows that one of the factors that can lead to excess weight is frequent consumption of foods that are high in calories, fat, and processed carbohydrates and are purchased at retail food outlets. [2][3][4][5][6] Most of the data for this research, however, has been obtained in areas along the East and West coasts [7][8][9] . Hence, it is not yet known whether the link between frequent eating-away-from-home and high obesity rates also holds true in Midwestern Wisconsin communities. Furthermore, studies in coastal areas focused mainly on urban regions and limited data are available on associations of body mass index with proximity to restaurants 10 and frequency of meal consumption 11 in non-large metro areas. Considering the obesity prevalence in non-large metro areas of Wisconsin is similar to the large metro cities , an evaluation of frequent restaurant eating on excess weight gain in State's non-large metro areas is warranted. Frequency of visits to fast-food restaurants and its association with obesity has been extensively studied, [12][13][14] but only a few US based studies have investigated this relation in other retail food outlets such as, takeout outlets, 15 cuisine specific restaurants , 16 and total out of home eating. [17][18][19] Moreover, information is lacking on how other restaurant types, such as fast casual, all-you-can-eat or sit-down restaurants may be contributing to excess body weight, especially in non-large metropolitan settings in a Midwestern state such as Wisconsin in which two-thirds of the population lives in rural and urban communities with population of less than 250,000. 20 While we speculate that the fastfood chain restaurants in urban regions may be similar to fast-food restaurants in non-large metro areas, other restaurant types may differ and also have different associations between frequency of restaurant eating and BMI. It is our expectation that frequent eating-awayfrom-home at all restaurant types will be positively associated with a greater BMI in the non-urban and small-urban population. We also expect a dose response relationship between frequency of eating out at the various restaurant types and BMI in these Wisconsin regions. To inform interventions aimed at improving health, investigators have studied the influence of socioeconomic/environmental variables like income, lack of time, lack of resources, proximity to the restaurant, etc. for restaurant types where recurrent visits are associated with high BMI. [21][22][23] These evaluations have resulted in targeted interventions that have influenced the restaurant environment by changing foods on the menu, promoting healthy menu items, and/or implementing calorie labeling for menu items in chain restaurants, which is also mandated by the recent US Food and Drug Administration statute. 24 In several large cities, communities and restaurant businesses have already demonstrated positive outcomes by implementing consumer-focused strategies that improve quality of foods offered at the retail food outlets and enable patrons to make healthier choices. 25,26 For instance, one outlet in California reported an increase in the purchase of lower calorie menu items, after 2-years of creating healthy menu items and posting calorie information on the menu. 27 Similarly, 9 food outlets in San Diego, California created and promoted healthy menu items using table tents, posters, community events, ads in magazines, newspaper, and television. This intervention resulted in a 3.7% greater likelihood to purchase the healthy menu items than the control group after 1 year. 28 To identify strategies and initiate effective interventions that will work for the rural and small metro areas of Wisconsin that are studied here, additional information on the socioeconomic/environmental factors of the residents need to be collected. 29 Nutritional attitude and behavior related data of individuals are rarely included in predictive models of eating out frequency. Based on previous findings on the positive role of diet and health related knowledge on the overall diet quality, 30,31 it is our assumption that nutritionoriented consumers will avoid frequent restaurant visits due to advertised negative health consequences of retail foods. If our assumption is valid, especially for the non-urban and small urban Wisconsin areas, implementing community programs designed to impart nutrition knowledge may have a large impact on the frequent consumption of unhealthy restaurant meals and consequently, obesity. Hence, understanding how nutrition knowledge and attitudes in residents in the study area impact restaurant visits is critical to designing and implementing effective interventions. The primary aims of this analysis were: 1) to explore the association between frequency of eating-away-from-home and BMI in non-large metropolitan areas of Wisconsin; and 2) to identify the socioeconomic/environmental and nutrition attitude/behavior variables that may have influenced frequent restaurant visits. Evaluating this information will give us a novel insight into the eating-away-from-home behavior in non-urban Wisconsin communities. We worked with six, mostly small metro Wisconsin communities to collect population level data in order to inform evidence-based strategies that will improve healthy eating habits in these communities and aid efforts to reduce obesity. --- METHODS --- Design This cross-sectional survey was part of the Centers for Disease Control and Prevention's Community Transformation Grant effort to develop local interventions and inform policy to address issues related to an unhealthy lifestyle. Transform Wisconsin made an open call to counties in Wisconsin to apply for the CTG grant. Selection of counties was based on organization's capacity and readiness for change. We conducted a secondary data analysis of the information collected from six Wisconsin counties: Kenosha, Rock, Marathon, Winnebago, LaCrosse and Manitowoc between Oct. 2012 and Feb. 2013. Based on the CDC's classification, one of these counties can be categorized as rural, one as large fringe metro and remaining four as small metro. 32 Our data may not fully represent the State because 66.6% of the counties in our dataset classify as small metro, 16% as non-metro and 16% as large fringe metro, while of the 72 counties in Wisconsin, 64% are nonmetro , 15% small metro, 0.1% large fringe metro . These counties all received transformation grants to implement healthy eating, active living, and smoke-free environment interventions after our survey data were collected. --- Sample The CTG survey had two aims, assessment of smoking with an emphasis on multi-unit housing and assessment of active living and healthy eating. For sample selection, we stratified residences by county and multi-housing units and performed simple random sampling within strata. Because individuals in multi-unit homes are a small fraction of each community, we slightly oversampled multi-unit housing structures for adequate power. Mailing addresses of all households were purchased from Marketing Systems Group -GENESYS and a random sample of addresses were chosen. An advance letter describing the study was mailed to households 2-weeks prior to the in-person visit. Field interviewers visited selected households up to six times before eliminating that address from the role. If the residents were found at home, participation was discussed and a household screener was completed. All civilian non-institutionalized adults ≥18 years from each selected household were invited to participate in the study. After providing consent, each participant completed a 45-60 minute interviewer-administered survey, which was deidentified before data entry. Local field staff was trained to collect data in randomly selected homes in these six counties. Staff also measured height, weight, waist and hip circumference, blood pressure and BMI was calculated. Participants were given an incentive of up to $50 for participation. This study was determined to be exempt from IRB by the University of Madison Health Sciences IRB. --- Measures Frequency of eating out-To aid local obesity prevention efforts, respondents were asked to report the frequency of eating at different restaurant types including fast-food restaurants, fast casual restaurants, all you can eat restaurants and sit-down restaurants. Fastfood restaurants were defined as those similar to chains like McDonalds, Pizza Hut, Burger King etc.; fast casual restaurants were defined as somewhat quieter and slower paced than fast-food restaurants e.g. Noodles and Company, Panera Bread, or cafeterias; All-you-caneat restaurants were places where unlimited meals are served at one price e.g. Old Country Buffet, Ponderosa etc.; and a sit-down restaurants are places where people sit and a staff person takes an order. The response scale for eating out at each restaurant type was Never, Rarely, Sometimes , 1-2 times/week 3-4 times/week and >5 times/week. Based on the distribution of responses, these categories were collapsed to Never/Rarely, Sometimes , 1-2 times/week, >=3 times/ week. These questions were adopted from NHANES 2005-2006, modified, and incorporated in the Survey of Health of Wisconsin questionnaire. Same questions have been used yearly since 2008 to collect eating out information from Wisconsin residents. Socioeconomic/environmental and nutrition attitude/behavior variables-An important objective of this evaluation was to enumerate whether certain factors influence eating out at a certain restaurant type. For this analysis, participants were asked a set of validated questions on factors that may have influenced their dietary behaviors . 33 Responses were coded as "1) Applies to me" or "2) Does not apply to me". These questions covered socioeconomic/environmental variables, including lack of time, storage space, equipment, affordability and lack of healthy food choices. Additionally, questions including self-perception of body weight, self-perception of diet, family encouragement to eat healthy, lack of knowledge and lack of motivation were categorized as nutrition attitude/behavior variables. Since one of the aims of this data collection was to assess smoking in multi-unit housing, self reported data on smoking was collected. Participants were asked whether it is allowed to smoke inside their house. Reponses were coded as: 1) Not allowed, 2) Allowed sometimes, 3) Allowed anywhere in the house, 4) No rules about smoking inside the house. Based on the distribution of responses, these categories were collapsed to: 1) Smoke inside the house, 2) No smoking inside the house. --- Statistical analysis To account for the oversampling of households in the sampling design of the evaluation, all analyses used sampling weights. For analysis of the first aim, multiple linear regression models were created using BMI as a dependent variable for each of the four restaurant categories. All models were adjusted for common confounders for BMI including age, sex, education, income, smoking, and marital status. Smoking was included in the model due to its previously established associations with obesity. Because both the fast-food and sit-down restaurants were associated with BMI in our population group, for our second aim we combined the frequencies of visits to these restaurant types. A multivariate linear regression model was created with combined frequencies of restaurant visits as dependent variable and socioeconomic/environmental and nutrition attitude/behavior related variables as independent predictors. A total of 20 socioeconomic/environmental and nutrition attitude/behavior variables were investigated and the backward elimination procedure was used to remove the non-significant variables. The model was adjusted for age, sex, education level, marital status, income and family members per household. Frequencies of eating at fast casual and all-you-can-eat restaurant were also accounted for in these models. A P value of < 0.05 was chosen for statistical significance. Analyses were performed using SAS statistical package, version 9.4 . --- RESULTS Selected baseline characteristics of the study participants are presented in Table 1. Briefly, data was collected from 1418 individuals in six non-large metropolitan counties of Wisconsin with an average age of 48 years . Thirty-nine percent of the participants were obese, with an average BMI of 29 kg/m 2 . Frequency of eating out at different restaurant types is presented in Table 2. On average participants reported eating out 1.86 times per week. Overall, 21% of individuals reported going out to eat more than three times a week. Eating out at a fast-food restaurant was most frequently reported, followed by a sit-down restaurant. When we estimated the association between BMI and the frequency of eating out for each of the four types of restaurants in a single model, our analysis showed a significant positive association between frequent eating out at both fast-food and sit-down restaurants with BMI . A post-hoc analysis of BMI in each restaurant category with the frequency of eating out response indicated a dose-response effect in fast-food restaurants. No such dose response effects were observed in sit-down restaurant category . Fast-casual and all-you-can-eat restaurants did not reach statistical significance presumably due to their lower reported frequencies. Since both fast-food and sit-down restaurant visits were associated with greater BMI, we combined the frequencies of eating at these restaurant types for further analysis. Table 4 presents the model showing association of combined frequency of eating out at these two restaurants with socioeconomic/environmental and nutrition attitude/behavior variables. In our population group, individuals who described their diet as healthy and believed that their diet was keeping their weight controlled ate less frequently at fast-food and sit-down restaurants. Of particular note, 20% of the participants who considered their diet as healthy avoided both fast-food and sit-down restaurants . Of the total participants who considered their diet as healthy, 40% were overweight or obese; and 56% of participants who reported that their diet is keeping their weight healthy were overweight or obese. Individuals avoiding high fat food also never/rarely frequented the two restaurant types. Participants who reported lack of availability of healthy choices at shopping and eating venues and lack of cooking skills were more likely to frequent the two restaurant types. Frequency of eating out at restaurants was associated with BMI; however, research shows that already overweight/obese individuals also tend to eat away from home frequently. 34 To test whether it was true for our population group, we added BMI as an independent variable to the same model. No effect of BMI was observed, except lack of cooking skills was no longer associated with frequency of eating out . --- DISCUSSION This secondary data analysis adds to the growing literature on complex associations between socioeconomic variables and frequency of away from home food consumption. Participants who reported eating frequently at either fast-food and sit-down restaurants were more likely to have higher BMIs. We also found that individuals concerned with their diet and weight reported visiting these establishments less frequently compared to those lacked cooking skills and healthier food choices. This study differs from the previous work 8,21,22 by including nutrition attitude/behavior, indicative of barriers to healthy eating, in addition to the socioeconomic/environmental variables, as predictors of frequent restaurant eating and consequent obesity. This analysis is also novel because there is very limited data on frequency of restaurant eating and these influencing factors in non-large metropolitan Midwestern communities. We hope to use these results to aid non-urban Wisconsin communities develop targeted obesity prevention efforts such as, making healthy options available in restaurants and interventions on promoting healthy menu items. We evaluated the frequency of eating-away-from-home in our dataset because of its previously established positive associations with BMI. [35][36][37] Consistent with findings from those scientific analyses, we also found a positive association of frequent fast-food consumption with greater BMI. Moreover, our estimated increment in BMI of 0.8 kg/m 2 with every one-meal/week increase in fast-food consumption agrees with the 0.13 kg/m 2 reported in the CARDIA study. 13 This association may reflect the high content of energy, [38][39][40] total fat and saturated fat, processed carbohydrates, 39 sugar and lower content of fruits, vegetables and micronutrient density 12,39,41 in fast-food. In our analysis, we also found frequent sit-down restaurant visits to be positively associated with the BMI, which was in contrast to the findings of other researchers. 42,43 A decrease in body weight with frequent sit-down restaurant visits reported by Mehta et al. and Bezerra et al. 42,43 may be explained by the availability of healthier menu options in the restaurants their study population frequented. 43 Unlike these other reports, positive associations between frequent sit-down restaurant visits and BMI in our dataset indicate that sit-down restaurants in our communities include many bar-and-grill establishments and their menu options may not be conducive to healthy eating . It is important to note, however, that information on the menus of this restaurant type is not available and these are speculations. Another possible explanation for our positive association is that menu items vary broadly in calorie per serving, serving sizes and calorie density among sit-down restaurants, 44 and positive association with BMI in our population probably indicates personal preference for obesogenic menu items such as, high sugar, high fat energy dense foods. Further analyses are warranted to identify sit-down restaurants in which healthy choices are missing or limited so that stakeholders can create programs for healthier Wisconsin communities. Although our analysis was underpowered with respect to food outlets such as buffet restaurants and cafeterias, other investigators have indicated their strong role in overeating and obesity. 11,45 For example, Casey et al. in a cross-sectional survey data set identified that 33% of the participants visiting buffet restaurants frequently were obese. 11 Data on menu items and the kinds of foods consumed by Wisconsin residents at these different restaurant types is warranted. Since both fast-food and sit-down restaurant visits were positively associated with BMI, we combined their frequencies in order to identify the socioeconomic/environmental and nutrition attitude/behaviors that may influence frequent eating-away-from-home. Our analysis found that less cooking at home due to perceived lack of cooking skills is associated with frequent fast-food and sit-down meal consumption. Dave et al. reported a similar outcome in their cross-sectional analyses, where dislike for cooking was associated with higher frequency of fast food intake. 46 Larson et al. also supported these results by showing that frequent food preparations at home lead to less frequent fast food intake in young adults. 47 A novel finding of our analysis is that individuals who considered their diet to be healthy and believed that their diet is helping them maintain a healthy body weight avoided frequent restaurant visits. These results indicate that individuals who are aware of the benefits of a healthy diet also understand the poor diet quality of restaurant food and therefore avoid eating out frequently. Our findings suggest the need to develop effective programs to improve nutrition knowledge in Wisconsin communities. There are some limitations to our evaluation. As noted earlier, our data may not represent the state due to the inclusion of large number of non-large metro areas in the analyses. Additionally, this cross-sectional study design does not provide us with an opportunity to find a causal relationship between socioeconomic/environmental and nutrition attitude/ behavior variables and frequency of eating out at different restaurant settings. These data were collected to identify problem areas that communities could address. Moreover, the information collected on the frequency of eating out at different restaurant types is selfreported and may be inaccurate due to memory lapses on the part of the respondents and/or social desirability bias. We also lack information on medical advice that may have influenced the choice of restaurant or reduced the frequency of consumption and physical activity levels of the participants. Information about any systematic differences between those who participated in the study and those who declined to participate is also missing. Furthermore, Wisconsin population is predominantly Caucasian and therefore we did not include ethnicity as a confounder. In addition, our data collected for this evaluation did not include other eating out venues or take out businesses such as cafeterias, supermarkets, street vendors etc., nor information on the density of food outlets, which may also contribute to obesity. Our data set also lacked objective information on local healthier choices available outside the home. Further analysis of restaurant options around these communities will give us a clearer picture. Overall, the knowledge gained can be useful in many ways for developing effective interventions and policies to create healthier communities. In a literature review of community-based interventions to promote health eating in restaurants, authors concluded that point of purchase information with promotion, and increased availability of healthy choices were most effective in improving dietary intake outside the home in urban communities. 29 Recently, the same research group implemented a pilot intervention "Waupaca Eating Smart" focused on labeling, promoting and increasing availability of healthy foods in seven restaurants in two Midwestern rural communities. 48 Restaurant food environment scores improved significantly in the intervention group suggesting that this intervention may be successfully implemented in our communities. In a similar study called Baltimore healthy carryout trial, 49 researchers improved labeling on the menu boards, promoted healthy sides and beverages and introduced healthy combo meals in 8 carry-out locations in low income Baltimore communities. Results indicated an improvement in types of foods purchased and the intervention was immediately adopted as a citywide intervention. Since the communities we studied have similar structure, these restaurant-focused initiatives may be successfully implemented and may influence individuals to choose healthier items at restaurants. It is important to understand that people will be exposed to fast-food and sitdown restaurants every day; however, educating individuals to improve cooking skills or to rely on others they trust to cook for them may avoid dependence on these restaurant meals. Communities may also increase motivational and education programs focusing on the importance of a healthy diet and teaching tools to prepare healthy meals. In summary, the present study confirms previous research findings that frequent fast-food consumption is associated with higher BMI. In contrast, however, it weakens the cumulative data relationship between patronizing sit-down restaurants and obesity, by showing a positive association between frequent sit-down restaurant visits and BMI. These findings may be critical to strategically plan targeted interventions for non-large metropolitan and rural Wisconsin communities. Our findings also indicate that understanding the socioeconomic/environmental factors and nutritional attitude/behaviors variables that we speculate drive Wisconsin residents to eat at a restaurant frequently is critical to the success of community based obesity prevention strategies. --- SO WHAT? Implications for Health Promotion Practitioners and Researchers What is already known on this topic? Frequent restaurant visits are associated with BMI and obesity in large metropolitan communities; however the information on frequent eating at different restaurant types in non-large metropolitan and small metropolitan communities is lacking. Additionally, the influence of socioeconomic variables on frequent restaurant food consumption is well documented. However, the predictive models rarely included nutrition attitude factors that may influence frequent restaurant eating in non-urban settings. --- What does this article add? Frequent eating out at both fast-food and sit-down restaurants was associated with BMI in non-large metropolitan Wisconsin communities, with stronger association found for fast-food visits. Nutrition conscious individuals are less likely to visit restaurants frequently, while consumers lacking cooking skills and lacking food choices are more likely to visit restaurants frequently. --- What are the implications for health promotion practice or research? Our findings support that understanding variables that may influence frequent eating at obesogenic restaurants is critical to developing community-based healthy restaurant eating interventions. Considering that the majority of these interventions are concentrated towards large metropolitan populations and less attention has been given to less populated regions of the Midwest, these outcomes are especially important for development of effective healthy restaurant eating interventions. --- Author Manuscript Author Manuscript Bhutani et al. Page 17 --- TABLE 4 The influence of socioeconomic/environmental and nutrition behavior/attitude variables on predicting combined frequency of visits to fast-food and sitdown restaurant in six non-urban Wisconsin counties Variables

Purpose-We investigated associations between frequency of eating at fast-food, fast-casual, allyou-can-eat, and sit-down restaurants and body mass index (BMI) in non-large metro Wisconsin communities. To inform prevention efforts, we also analyzed socioeconomic/environmental and nutrition attitudes/behavior variables that may drive frequent eating-away-from-home.

Introduction Research on the effectiveness of digital technologies shows diverging outcomes. Some conclude that there is potential, but that school digital maturity and teaching practices do not align with the use of digital tools in society at large , and that digitalisation does not lead to increased academic achievement . While the literature is replete with research on educational technology and design interventions, approaches have been criticised that overlook axiology, that is; questions of principle and value , research on engagement in Learning Management Systems or evaluating teachers' IT skills , with calls for informative examples on how to intervene in the learning environment and use existing digital technologies effectively in situ. As education has shifted towards becoming more digital, particularly after the COVID-19 outbreak , it is important to understand how teachers and the digital technologies used for learning influence engagement and disengagement. While engagement is often described as the visible and measurable outcome of motivation , many teachers report that student disengagement is the biggest challenge they face in their classrooms . Where engagement is strongly correlated with proactive behaviours for learning, general school success and retention , disengagement is related to disruptive behaviours, negative attitude, withdrawal, absenteeism and school dropout . However, engagement and disengagement are malleable , and thus teachers, learning environments and digital technologies may influence engagement . Due to the strong relationship between engagement and disengagement to either school success or school failure , insights into how teachers are utilising digital technologies within blended learning environments and how these, in turn, influence engagement and disengagement are critical to schools . It remains important to realise that behind school success or failure lies an individual's success or tragedy, which further accentuates the need for a deeper understanding of how engagement and disengagement manifest, are altered or redeemed, alongside teacher considerations or agile didactic decisions in Blended Learning . A growing body of research has found that today's students are becoming increasingly disengaged in school; displaying increased levels of boredom , taking the opportunity to escape the classroom via digital devices when feeling bored , along with general passivity, zoning out, and even occurrences of sleeping , all of which may increase during a pandemic, but then without a teacher present to support the individual student. Recent reports have called for research that can inform the transformation of Emergency Remote Teaching into high-quality distance learning and concluded that there is a lack of research "... with a specific focus on how the use of collaborative learning with the support of digital tools affects socio-economically disadvantaged students" . In addition, a systematic review on global responses to the pandemic in secondary education also concluded that disadvantaged students have received very little focus . This study, therefore, explores the engagement compound in BL in a socially disadvantaged school to meet this gap, and contributes to the field as it brings together aspects of teacher leadership, digital technologies and management of engagement and disengagement in a real classroom setting, to explore meaningful facets. More specifically, this paper adds an exploration and further refinement of the engagement compound, which has been called for , where the digital context, work pace, learning design, and teacher self-efficacy are explored in relation to student engagement. Informed by the above, this study seeks to answer the following research questions: 1) How do the uses of digital technologies influence how students engage in a disadvantaged upper secondary school? 2) How does classroom leadership influence engagement in a disadvantaged upper secondary school? --- Background --- The blended learning context and the engagement compound BL combines online and physical elements, such as instruction, material, resources, and learning activities . For the purpose of this article, digital technologies refer to the devices , digital resources with learning content, or to support learning activities , digital infrastructure, , but also other hardware . The BL context is thus infused with varying kinds of digital technologies and resources. When entering a classroom , the teacher needs to establish an agreement -a teacher-student contract -which serves to remind and consolidate the structure, expectations, agreements and positions between the teacher and the students. The teachers communicate norms and expectations, explicitly and implicitly. Even if teachers would not establish a teacher-student contract, they cannot separate themselves from the school context and [Blended] learning environment . Kuh refers to teachers' negotiation and facilitation of needs as an engagement compound that establishes the roles, structure, expectations, agreements and positions between the teacher and the student. However, while the communication can be coloured by personality, structural and contextual factors, it may also be influenced by the BL context, the digital technology and teacher ITliteracy . While non-pedagogical digital tools and resources may be explored separately from pedagogical digital tools and resources , both may influence the engagement compound . The school, the teaching profession, the role of students, and even digital technologies, are not value-free; and individuals and digital technologies are heavily intertwined, which is why teaching cannot be separated from the context in which it happens . --- Teacher leadership Teacher leadership entails quite a few perspectives. For the purpose of this article, we explore the blended learning context, teacher self-efficacy, and management of student engagement in learning. --- The blended learning context and work pace Importantly, digital technologies have the potential to disrupt learning, and the maturity of BL can be viewed as moving from 'enabling and enhancing' to transforming learning . Leadership qualities commonly refer to an individual's traits and characteristics, even though the context may trigger and shape leadership qualities . In BL and online learning, teacher leadership demands the ability to lead with digital and physical tools and resources in both physical and digital learning environments. Digital technologies challenge the spatiotemporal aspects such as pace, place and time in relation to teaching and learning , and studies have proposed that teachers' workload could decrease as a result of 'working smarter not harder' . At the same time, leadership research has proposed that passive destructive leadership or laissez-faire type of leadership can be triggered by contextual factors such as lack of time, pressure and stress, which then impact teachers ability to exert the leadership they otherwise would . --- Teacher self-efficacy and the fostering of engagement A critical perspective relating to engagement and disengagement is self-efficacy. Established as a socio-cognitive theory, Bandura emphasised that perceived self-efficacy links one's own ability to manage situations. The self-efficacy theory is the motivation theory used to study teachers since it was first applied in 1977 . Teachers' views on their own ability to influence situations thus govern if they 'can' and 'want' to get involved. Regarding the disengagement compound, the negotiation of engagement and disengagement is strongly related to teachers' self-efficacy, as it places the teachers' perceived ability to influence students in focus. A teacher's self-efficacy affects their leadership in the classroom and determines if the solution will be implemented . Teachers' self-efficacy refers to their beliefs and to what extent they can influence or affect student learning . In addition, Hatlevik concluded that teachers' self-efficacy is related to their digital competence and uses of digital tools and resources in and for teaching and learning. Thus, what teachers do in the classroom is related to their efficacy, but their actions also directly affect student engagement and school outcomes . Perera et al explored teachers' personality profiles and how those related to self-efficacy, work engagement, and job satisfaction. They found that job satisfaction was the lowest among excitable teachers, while well-adjusted teachers was found to report significantly higher self-efficacy in relation to classroom management than teachers in the ordinary and rigid subgroups . As seen above, research have pointed the importance of exploring contextual influences in relation to leadership. Researchers have addressed a similar need when fostering engagement . Engle and Conant, highlighted that students need support, relevant resources and authority to engage productively, and that shared norms are needed to be able to hold students accountable for their learning engagement and that such guiding principles can inform teaching practices. Understandably, they did not view these in a BL setting. Some twenty years later, Shi et al, proposed that the blended learning setting in particular that needs to be taken into account when trying to engage students . --- Engagement and disengagement in blended learning Research has indicated that context affects engagement both sequentially and reciprocally and emphasised that how digital technologies are used, along with the considerations aimed at promoting engagement and redeeming or circumventing disengagement, is critical for learning . Together with how and when digital technologies are used, a digital context is shaped, which subsequently influences engagement . Building on previous engagement and disengagement research engagement and disengagement in BL can be understood as a multi-dimensional construct, consisting of four dimensions: a behavioural, a cognitive, an emotional and a social dimension, with engagement encompassing pro-learning behaviours, emotions, focus and interaction, and disengagement encompassing negative emotions, maladaptive behaviours and responses. --- Student self-beliefs It has been suggested that contemporary theories of learning generally include a section about student beliefs about their competence . Schmid and Petko explored students' beliefs about their capability of using digital technologies and the perceived usefulness of the same. They found that these aspects are often overlooked, even when digital technologies have a significant impact on the learning context. Schmid and Petko suggest that personalized learning using digital technologies has positive effects on students IT-realted beliefs in learning. However, they also found that "freedom of choice" of learning activities with digital technologies had no significant effect on student beliefs, which may indicate the need for instruction, guidance and leadership. While one aspect of self-belief relates to their competence, another part relates to a student's sense of belonging, and thoughts about their relationship with teachers and peers. Functional relationships are critical for students to sense that they are a part of a learning community , and identify as a learner together with other learners . Research has also pointed out that students' self-beliefs also influence how they experience factors related to their learning. A recent study showed that students who had a high perception of their competence to a much greater extent than students who perceived that they have low competence, nuanced their answers and pointed to several influencing factors: results, learning process and context, while students with lower self-confidence briefly stated that nothing was good. Similarly to Schmid andPetko , andPelikan et al. , Bergdahl et al. shadowed students across their school week, and concluded that student engagement varied, but the patterns seemed to be more related to how the teacher orchestrated the digital technologies than student interest in specific subjects. In fact, only one student compensated for poor orchestration with a devoted interest in a subject . Students coming together from several cultures may carry with them varying selfbeliefs that influence their learning . Research exploring the digital divide has focused on digital inequality and concluded that different groups might have limited access to digital tools, may have limited IT-literacy, and also indicated that, even when access and literacy exist, some groups do not benefit from the time they invest online , which during the pandemic has also hindered, for example, immigrant groups from benefiting from the shift to online services in society at large . In schools, studies have found that teachers may reduce the use of digital technologies and resources for immigrant students , which may be due to teacher consideration of student wellbeing, as second language students may have difficulty interpreting social cues, or experience their culture as devalued . Moreover, if left unsupported, students' negative self-beliefs may cause their disengagement to spiral, particularly in online learning . --- Methodology In order to explore the engagement concept in depth, a qualitative case study was conducted across five months . Case studies allow researchers to explore a phenomenon from multiple angles within their "natural setting" , which enables data triangulation and validation . --- Research context This case study was conducted in an upper secondary school, in a socially disadvantaged area, in one of the larger cities in Sweden. This is an example of purposeful sampling, which is an appropriate method for selecting sites for deep investigation in qualitative research . The upper secondary school welcomes students who have poor primary school results, and while some students study to gain eligibility to enrol with a national program, others are enrolled in a more practically oriented apprenticeship program. The school welcomes students all year round, which means that students, who have just arrived in Sweden, could enrol at any time. In this school, the work is structured around teams that work around their dedicated student groups. Each team of teachers includes a dedicated student health team: a school nurse, a counsellor and a special needs teacher. All teachers and students have their own laptops, use GSuite for Education, and have classrooms fitted with projectors. --- --- Ethical considerations and researcher bias The school principal and all participants provided written informed consent to participate. All respondents were informed of their right to withdraw from the study at any time without questions asked, and that data would be treated in line with current legislation and analysed and reported anonymously. While, the department had prior established connections with the school the observing researcher had not. The researcher remained an impartial observer during all observations, and did not interfere in any of the lessons. During the five months of observations, the school dedicated a room to the researcher, which further enabled the researcher to spend additional time at the school and with the teachers and students. This familiarity may have helped participants to feel more comfortable in the classroom whilst being observed. --- Data collection Multiple data sources were collected across the five months . Classroom observations were undertaken and documented using field notes and photos. Then the teachers assisted in suggesting students that could be observed, close to where the first author would sit to enable observation and dialogue . --- Identifying work pace During the first classroom observation, an emerging indication was that teachers worked to influence student work pace . This spurred the interest in observing, and making subjective notes on student and teacher work pace respectively. A schema was developed . The schema uses the letters A-E to reflect distinct characteristics of work pace. The work pace schema was added to the classroom observation schema and subsequently used to identify student and teacher work pace in relation to uses of digital technologies during class. Even though there is no equal distance between A -E, the categories were arranged so that A reflects higher engagement and E lower. With the purpose to compare teacher and student work pace, in terms of high and low, the observed pace was re-calculated into numeric values . --- Data analysis Data were analysed using thematic analysis and descriptive statistics, using actions and processes as units of analysis . After conducting classroom observations, field notes and photos were coded with memos. Thematic analysis focuses on meaning across a data-set. Data were coded using post-it notes, and codes were subsequently arranged into themes, reflecting instances of actions and processes . Codes were discussed between the researchers, explored for emergent, unexpected angles, and re-checked against the data. This reflected a combination of two styles of thematic analysis: 1) descriptive and 2) interpretative . Thematic analysis was used to identify how the themes could be visualised in a network display . By arranging and rearranging the themes in thematic maps, the visualising can reveal patterns, support conclusions of the analysis, and provide insights into the relationship between the themes . To ensure 'authentic triangulation', data collected and analysed were verified by the participating teachers . --- Results Figure 1 displays a network of themes related to the engagement compound. In exploring the engagement compound in BL, four main themes were identified that represent perspectives of influence: 1) The blended learning context, 2) Teacher leadership, 3) Blended Learning activity, and 4) The student as a learner . --- The blended learning context Under the category digital context, two themes were identified: I. Blended learning work pace and II. Unintended consequences. --- Blended learning work pace While students could choose which digital technologies to use to work toward their learning objectives, they were also observed to become passive due to using those technologies. Some students did not have a password and did not have access to the digital learning material -and they continued to be passive throughout the lesson. On the other hand, teachers' work pace was high, trying to assist students and encourage them to complete their work. The teacher was energetic and worked hard to try and meet student needs, while the students were passive. The teacher draws on the text for topics and tries to start a dialogue. Students are passive, waiting. The teacher encourages the students further by relating the content to the students' world. An emerging indication was that the student pace was observed to be related to and influenced by the teachers' teaching practices and lesson design. Because of this emerging indication, student and teacher work pace were observed throughout the lessons . All observations included exploring student and teacher work pace in relation to uses of digital technologies during class. In Fig. 2, where category E reflects the lowest work pace and category A the highest. Figure 2 reflects the occurrences of lessons in a certain combination of . Figure 2 reveals that some teachers work hard to sustain a high level of activity , but half of the teachers in category B do not succeed in reaching the same level of activity for their students. The teacher pace matched categories C or B , with a slight inclination for the higher pace. For students, the most common work pace was identified as categories D or C , with the lower work pace being slightly more common. In three observed lessons, teacher and student work pace were the same . Figure 2 also shows that three teachers employed a slower work pace than their students: Teacher: C; Students: B, or Teacher: D and Students C. Neither extreme was observed. The mean of teachers' observed work pace was 3. 30, which was higher than the students' mean of 2.69. Category A was not observed for teachers or students. Most often, teachers had a higher working pace than did their students. An emerging indication was that the student pace was observed to be related to and influenced by the teachers' teaching practices and lesson design. While category E does not include teacher-student interaction focused on learning, designing learning as described in category E may be deliberate for a specific learning goal. However, if the teacher always designs for category E, that decision can reflect values of the surrounding culture and attitudes and may also be a reaction to high work pressure, frustration, and a sense of giving up. --- Unintended consequences However, the effect of using educational technology in the classroom meant that students occasionally had to move positions so that they could charge their devices as needed when their batteries ran out, a design consideration that is now essential in modern classrooms. Another effect of using technology was the need for teachers to be cognisant of flexible lesson design, adapting their lesson plans when the Internet, in particular, was not working. When faced with this obstacle, one Music teacher asked the students to turn their computers off and sing instead. Teacher: "OK -you can turn off your computer now -and we'll sing". "Teacher: "Oh, the Internet is up again ... [The teacher can end the lesson by showing streamed media; a snippet of a band who performed the verse that the students had practised]. On several occasions, it was observed that the digital technologies the school provided could lead to unintended consequences, such as shared information unintended for students or only being able to listen or use the mobile phone for learning, only if they had access to one. The teacher logs in to [a digital learning resource] and shows the teacher view... when he does this, all the students can see everything on the teacher's screen as the teacher searches for the information needed for the lesson. Teacher: Now we are going to India. This is going to be funny. Other unintended consequences could mean teachesrs had to support students in overcoming challenges, for which there was not always time after class, leading to instruction on how to use IT competing with subject content during lessons. --- Teacher leadership Three sub-themes of teacher leadership were identified: I. Management of education, II. Teacher self-efficacy and III. Managing engagement and disengagement. --- Management of education In some classes, the characteristics of digital technology use related to managing education, such as distribution of learning materials and resources, including directing students to other resources for use outside of class, e.g. students needing to download an app to practice the bass guitar at home. When learning was online, attendance was an automatic feature in another application , where timestamps reflected student logins . --- Teacher self-efficacy Teacher self-efficacy includes the motivation to act but may also be influenced by external factors, such as information, organisational support and school culture. The teacher says he does not know why only nine students are present ... he says that he is very frustrated. There used to be classroom rules posted on the walls in all classrooms. He points to a poster saying, among other things, that mobile phones should be switched off. He sighs and comments that it is not a priority. It is hard not to be heard when you point out that we need to make efforts, such as locking the doors if the students arrive late. Instead, the teacher reports being met with a negative jargon that applies "a certain type of students". Teacher self-efficacy is only possible when students are present. On the other hand, views of having 'a certain type of student' may reflect a school culture of collective efficacy, impacting individual teachers. --- Managing engagement and disengagement Efficacy and knowledge may influence teachers to increase engagement or manage disengagement. The observations, however, included instances when engagement shifted into disengagement and vice versa. In three observed lessons , engagement was observed to shift into disengagement. In the observed classes, such instances revealed that instruction to actively work triggered engagement , but when the teacher continued to talk, instead of allowing work as promised, the students returned to their mobile phone games. In the two other classes, there was little consideration on active learning for all, especially when digital technologies allowed for simultaneous activity, and when the teacher took over from digital technologies, s/he would engage with one or a few students. The majority of students would then be passive. In five of the observed lessons , there were instances when student disengagement shifted into engagement. A low threshold invited students to engage in some lessons , even after missing classes. Often, little effort was observed that challenged students' cognitive level. Instead, there was a clear focus on raising the low achievers above the pass threshold. However, if lessons are always un-demanding to be inclusive, this risks lowering or omitting the cognitive challenge that spurs progression for the present students. Nonetheless, the way teachers respond to disengaged students signals teacher expectations and what will be accepted. In one of the lessons , it was observed that the teacher ignored the disengaged student and refrained from communicating any fostering norms. This reflects a delicate balance on how and when it may be suitable to approach disengagement, but it may also remain unmanaged. However, there was one teacher that succeeded with students that had failed in other classes. Teacher: he does not function well in the other classes. And when [another student's name] first came to me, she was very aggressive. But I talked to her, met her and talked about what she wanted in life and that I empowered. I supported and encouraged her. And now, now she's not like that. --- She functions very well [in class]." When asked about managing disengagement, the teacher forwarded that "it is not about offering treatment -it is about being human, engaging in dialogue and showing you care, that may help turn schooling around fully for students, then the challenges pale in comparison". In other instances, when the teacher-initiated interaction, students were observed to display a range of responses; some students took the initiative to learn, others would rest or even yawn. There was no instruction on what to do when the teacher was interacting with other students. --- Blended learning Two sub-themes were identified: I. Uses of digital technologies, II. Avoidance of digital technologies. --- Uses of digital technologies It was observed that digital and physical resources were often combined and that students were accustomed to bringing both digital and physical tools with them to class; some had pen, paper and books, and others also had laptops and mobile phones. After the film, the students work on the [digital learning resource] material... They can see the exercise on the whiteboard and their screens. A text is projected onto the whiteboard. The students can access the text through Google Classroom. The teacher reads: "Ebba has an exciting book with her and a chocolate bar to munch on. She sinks into the comfortable blue seat on the train. Lisa's mother will meet Ebba. But no Ebba gets off. "Why doesn't Ebba get off the tram?". The teachers remind the students of the built-in audio support: "You can listen to the text again if you have headphones." When the question is raised, students use translation apps, including image search in Google, to visualise what a tram is. Digital tools seem to be intertwined with the social and culture, that it is expected to function and serve as a foundation from which to manage learning , distribute tasks , using media . Digital media has become a standard in classrooms of today. --- Avoidance of digital technologies There were also situations when digital technologies were not used. Reasons included that writing by hand was needed or preferred, that digital applications were no longer supported by the developer, and a gap between digital uses during class and school structures enabling student access to digital technologies and digital literacy training. The digital context appears to frame the conditions under which the digital technologies can be used to both manage education and support learning. There were several instances when digital technologies were the cause of problems: such as teachers having only one device or outdated learning resources, which may indirectly affect student engagement negatively. --- The student as a learner Three sub-themes were identified: I. Belonging, II. Self-beliefs, and III. Individual challenges. --- Photo 1 Student using Google image search Photo 2 Student using a translation application --- Belonging When a student signals they dare to be passive during an active learning activity, they challenge the teacher and norms. The lesson derives from students making their own choices to engage in learning, with the student having to take his/her initiative to learn. When a student chooses not to make that choice, it can be interpreted as a rebellious protest, a signal that something is not right within him/her. Moreover, it may create an unsettled atmosphere, especially if the student holds a status position in the class. [Sound from phone]. The teacher does not react. Two students who had arrived late both used their mobile phones. One hides under the desk and spends the lesson time scrolling instead. This behaviour signals shared norms between the latecomers. However, in the class, other norms existed. Another student then takes the opportunity to signal belonging to another set of norms "I have already written everything". This student raises his hand every time the teacher asks a question. The disengagement is infectious, spreading to nearby peers. Moreover, if other students have unrest, struggle to concentrate or self-regulate, they may become distracted or take the opportunity to disengage actively. When several outbursts of disengagement happen simultaneously in class, the teachers' stress levels were observed to increase. The teacher could not oversee all behaviour and hence did not set boundaries directed to specific individuals, as they would act out behind the teacher's back. --- Self-belief and withdrawal One teacher described that the school caters for socially disadvantaged students; that some are even accustomed to physical abuse from their teachers. When arriving in Sweden, they are often unsure of the rules and norms in the Swedish classroom. As such, one critical aspect for these students is to re-evaluate their self-beliefs in the new context, where diverse cultures and norms co-exist. Many students were observed to display silence and withdrawal. While self-beliefs may refer to one's ability, it may also reflect insecurities. When the students withdrew into passivity, it appeared as if the passivity had different levels, as if one layer was a temporary idle mode, waiting for the teacher to activate them. The other level, observed when students were left alone for a longer time, was interpreted as if students retracted into a state of isolation and loss of agency, even in class. When the teacher is with the students, works to engage them, and helps them get started, the students begin to work, but without the teacher's constant prompts, awareness, and leadership, they tend to return to the passive state. When the teacher turned to manage disengaged students, the other students in the class received no teaching or instruction. The teachers were observed to balance continuing their teaching or managing student disengagement at the expense of educating the class. --- Individual challenges There were instances when teachers tried to engage students and then encountered other challenges, such as student knowledge gaps, or that latecomers could remain invisible in online classes as there was less disruption, which in turn might not trigger a teacher reaction. A third observed challenge was that different students might respond differently to the same situation. Students were also observed to respond differently to learning activities. In the same situation, one student proactively showed that he was participating in a situation that enabled passive presence. In a similar situation, another student had his eyes open, looked ahead, and observed to work but did not engage in classroom interaction. In a classroom, students take on different roles. There might not be room for every individual student to be proactive, who takes the initiative to talk. On the other hand, in classes with no student interaction, little dialogue or momentum, a verbal exchange can challenge the cognitive level. --- Discussion In answering the first research question: "How are the uses of digital technologies influencing how students engage in a disadvantaged upper secondary school?", we found that in BL, there were both uses and deliberate avoidance of digital technologies. In the observed BL classrooms, digital technologies were used to enhance learning, often combined with traditional practices. For example, classroom observations revealed that teachers frequently use their laptops to project content onto the whiteboard , which resembles the blackboard, but with digital equipment. While projecting content saves time and helps avoid problems that may arise when teachers have to turn their back against students, it remains a traditional approach . Teachers would then alternate between engaging in dialogues with students and directing questions to students, referring to the content on display. Expanding on previous research , we found that a teacher is physically managing the classroom by initiating and shifting interactions , tone of voice, signals, prompts, and by providing or withholding information. Our results also show the impact of digital technologies on the learning context e.g., when teachers initiated uses or avoided uses of digital technologies, experienced technology breakdowns or that students lacked necessary digital skills, equipment, or login details. Such contextual occurrences directly affected students' actual possibilities to engage proactively in BL activities and demanded the teacher to shift the order of learning activities promptly. Several teachers also displayed an accumulated understanding of students' vocabulary, anticipating new and potentially problematic words, and were attentive for cues, particularly related to students not understanding the language. With meticulous perfection, teachers identified students' level of knowledge, or expanded on students' insights, using gestures, tone of voice, visualisations and peer translations to communicate their message. However, in a digital learning context, relying on non-verbal communication, gestures and enthusiasm, is far from the design-thinking needed in a BL context. Merely offering a traditional class online may then cause some unexpected disturbances. For example, whilst turn-taking was an accepted method a couple of decades ago, it may be interpreted as un-demanding, slow and boring for the students of today, who may look to be simultaneously active using digital resources. Indeed, when relying on traditional ways of teaching, seven out of ten students were passive, and four out of these seven did not accept sitting passively but either initiated private conversations or turned to their mobile phones. That is, the very interaction that used to be effective, or at least accepted in the classroom some decades ago, was promoting passivity and reduced interaction . In regards to question two: "How does classroom leadership influence engagement in a disadvantaged upper secondary school?", classroom leadership was found to influence engagement directly. Results indicate that teacher leadership in BL entails both self-efficacy and the knowledge of how to design lessons that positively influence engagement and work pace and manage disengagement. As expected, ICT tools and related activities were used in almost all lessons. Still, the tools were almost always treated as something that should be handed over to the students to choose and utilise, without moderation or prior consideration by the teacher. In line with Valckx et al. , we agree that teachers' beliefs of what they can do to influence student learning relate to their self-efficacy. However, apart from leadership efficacy , we argue that digital self-efficacy is needed. There were differences in how ICT tools and related activities were used amongst the teachers, and there was no indication that access to technology, or students lacking digital literacy, would hinder implementing BL. On the contrary, students used many technologies as an integrated part when switching between their educational and privacy spheres . ICT tools were standard when it came to teachers managing the class and the content. However, they were rarely used in a deliberate way to shape learning or engagement. Warschauer et al. argued that how teachers integrate ICT tools and pedagogical thinking, aiming to promote engagement and learning, is critical, with teachers in this school seemingly making the same considerations throughout their lesson planning. On the one hand, it is understandable that ICT knowledge differs. On the other, developing an overall school strategy could lead more teachers to develop their use of ICT for pedagogical purposes . Developers of educational technologies should incorporate axioms, considering what teachers need and how those can be met, and considering that poor designs may trigger unwanted behaviours, considering whether learning will be increased, alongside the actual benefits of the software will be. The conditions to exert leadership is influenced by critical contextual factors . Thus, the digital context, work pace, designs for learning and teacher self-efficacy are viewed as critical aspects of the engagement compound. Bergdahl et al. concluded that engagement can be designed for if teachers are supported in becoming more aware of their potential to influence student engagement. Extending these findings, the results found that teachers' planning seemed to be teacher-centred and that active learning is not active learning for all students. When all students were expected to engage in learning and supported by lesson design actively, students were not observed to disengage. A lack of digital devices or outdated applications was also observed to impede teacher excellence. However, structural access needs to be combined with digital competence and awareness of implications that occur when shifting between the teacher and the digital tools and resources as agents for learning. When using digital technologies, a teacher could serve an unlimited number of students, as digital technologies are used to mediate one-to-many communication. In this case, though, when the teacher withdrew the digital technologies and turned to just one or a few students, the majority of students were left passive, and few students knew what to do. This resulted in most students being left passive for up to three times longer than the duration of observed technology breakdowns. Moreover, when comparing student and teacher work pace, teachers often worked at a more intense pace than their students . This implies that digital competence is more than subject didactics, or how to teach one's subject using digital resources and IT skills. We suggest that IT competencies should include aware considerations of consequences relating to digital technologies and digital leadership. Indeed, passive or disruptive presence negatively affected the learning climate for students and the working climate for teachers. The teachers were struggling with managing and redeeming disengagement, and often, individual or even collective denial was observed. Coming to class and being actively disengaged may be a students' way of repeating negative selfbeliefs; that no one cares. In the online classes, there was considerably less disengagement and managing of maladaptive behaviours. However, considering the uneven digital competencies and teachers struggling to offer support , not only wanting to give support but ensuring school structures adapt to the needs of support appears critical. In BL learning classes, it also became important how the teacher worked to establish a positive learning climate or react to disengagement signals. However, working with negative self-beliefs remains a challenge . --- Conclusion Building on the analysis and previous work, the results indicate that negotiation of engagement in the BL setting is a legitimate problem as the engagement compound is affected by both individual traits and context. Thus, a complex network of factors seem to influence learners engagement. • First the school context, which includes teacher workload with factors like time, pressure and stress that may negatively impact the conditions to realise positive leadership. • Second, digital technologies were found to influence leadership conditions. Thus, we propose that teachers' classroom leadership should include digital selfefficacy, and teachers' IT competencies should include digital leadership. • Third, results indicate that teachers' work pace was related to awareness of the impact of digital technologies, alongside a teacher's digital awareness of how to orchestrate digital technologies and resources to reduce work pace. • Moreover, results show that teachers manage engagement quite differently: reveal shifts both to engagement from disengagement, and vice versa. Therefore, the engagement compound can be understood as interactions within the BL context, the conditions for teaching and learning, and leadership execution, the learning activities and students' beliefs, sense of belonging and individual challenges. Thus, the negotiation of student engagement, in a BL context, relates to teacher self-efficacy; namely, their beliefs about their ability to influence students' engagement, teacher work pace, whether affected by digital competence or other stressful factors, influencing conditions to exert leadership, knowledge on how to design for engagement, and manage disengagement. --- Limitations and future work Generalisations from this study are limited. First, the sample size and number of schools are insufficient to generalise the conclusions. Second, the study was conducted in one school in a socially disadvantaged area. More students, schools and diversity of the same should be selected for increased generalisability. Moreover, despite observing several classrooms and lesson, any such observation does at best provide a snippet of that teacher's practice, and those students' engagement. Future research could explore negotiations in other BL settings or use a longitudinal design. The findings do however contribute to the research field, in terms of proposing instances and occurrences when digital technologies hinder and promote engagement, describe how engagement may shift into disengagement and vice versa, and point to the need to include managing in digital context as a skill for teachers. Future research should explore teachers' digital leadership and digital self-efficacy, learning designs, teacher considerations in relation to online and blended learning. --- Data availability Anonymised results are available from the authors by request. --- Code availability n/a --- Ethics approval n/a Consent to participate Written consent on voluntary and anonymous participation was obtained from all participants. --- --- Appendix --- Teacher and student work pace Table 1 describes characteristics of categories A-E reflecting teaching practices in BL classrooms and student practices from a perspective of pace. --- Authors contributions The first author planned and designed the study, conducted the observations and made the initial coding. Findings were discussed and the manuscript was co-authored by both authors. Funding Open access funding provided by Stockholm University. This research is a part of the project Malmö IT i Skolan , which is partly funded by the Department of upper secondary and adult education, Malmo, Sweden. --- ---

It is well-recognised that engagement is critical for learning and school success. Engagement (and disengagement) are, however, also influenced by context. Thus, as digital technologies add complexity to the educational context, they influence classroom leadership, lesson designs and related practices, and thereby engagement. Despite being critical, engagement and disengagement are not well explored concerning these influences, with a lack of research undertaken within socially disadvantaged schools. In this qualitative study, 14 classroom observations were conducted, during five months, in twelve classes in an upper secondary school in Sweden, along with dialogues with teachers (n=12) and students (n=32). The data were analysed using thematic analysis and descriptive statistics. Identified themes include digital context, teacher leadership, engagement and disengagement. A network of relations between the (dis-)engagement compound and themes is presented. The results identified processes in which engagement shifted into disengagement and vice versa; in particular, that the intention of active learning does not automatically translate to active learning for all students, although teachers employed a higher work pace than did their students. Teacher self-efficacy and awareness of how to manage digital technologies in and outside the classroom was found to play a vital role in facilitating engagement. Understanding the (dis-)engagement compound in blended learning environments is key to inform active and visible learning for future research and supportive organisational structures.

INTRODUCTION In high-income western countries such as Switzerland, cardiovascular disease is the leading cause of death. CV disease has its origin early in life due to genetic predisposition, environmental influence as well as socioeconomic and lifestyleassociated status. Over the last decades, the assessment of retinal vessel diameters and arterial stiffness by aortic pulse wave velocity have become well-established and recommended vascular biomarkers as a surrogate end point for CV risk assessment . We recently demonstrated that childhood obesity, high blood pressure and low physical fitness are associated with retinal micro-and macrovascular alterations in young children . Children with obesity and elevated BP had higher arterial stiffness and narrower retinal arteriolar diameters. Cardiorespiratory fitness was associated with a favorable higher retinal arteriolar-to-venular ratio as well as lower arterial stiffness . Environmental conditions such as socioeconomic status , migration background and parental lifestyle seem to play a major role in the development of CV risk factors in children and adolescence. Evidence suggests that the prevalence of CV risk factors is higher in children with low SES. For example, an inverse correlation has previously been described between household income and childhood overweight and obesity . The International Study of Childhood Obesity, Lifestyle and the Environment in 12 countries around the world found that parental educational level is also inversely associated with childhood overweight and obesity . Parental educational level seems to be contributed to a higher prevalence of childhood obesity , elevated BP and sedentary behavior . Passive smoking seems to affect development of childhood obesity and vascular health . Migration background has also been associated with higher prevalence of children with obesity and physical inactivity . Nonetheless, evidence for associations between socioeconomic status, parental smoking and physical activity behavior with childhood macroand microvascular health are scarce. This study, for the first time, aimed to investigate the association of SES, migration background and parental lifestyle with large artery stiffness and retinal microvascular health in young school children. --- METHOD Study Design and Participants This study is part of the large-scale, cross-sectional EXAMIN YOUTH study . Children were included if they were between 6 and 8 years old and had an informed consent from their parents to participate. Briefly, children with written informed consent were screened for large artery stiffness and retinal vessel diameters to assess macro-and microvascular health. Anthropometric parameters and blood pressure were determined according to standardized procedures for children. In addition, physical fitness was assessed on a separate day. The medical screening was performed in the morning on-site the regular school setting and children had to remain fastened before anthropometric and CV assessments. Parents were requested to complete a questionnaire on lifestyle behavior, level of education and SES . The questionnaires were handed out and recollected by the teachers. If the questionnaire was not returned to the teachers by the parents, an additional reminder was sent to the parents asking them to return the completed questionnaire. Approval from the Ethics Committee of the University of Basel was conferred. The study was designed according to the Guidelines for Good Clinical Practice of the Declaration of Helsinki and the manuscript conforms to The Strengthening the Reporting of Observational Studies in Epidemiology Guidelines . --- Measurements --- Retinal Microcirculation Retinal vessels were assessed using a fundus camera and an advanced image processing unit . This technique allows a non-invasive and semi-automated measurement of retinal vessel diameters. The method has been described elsewhere . Briefly, two valid images from the retina of the left and the right eye were taken at an angle of 45 • with the optic disc in the center. Retinal vessel diameters were estimated to central retinal arteriolar and central retinal venular diameter equivalents applying the Parr-Hubbard formula . The AVR was calculated from CRAE and CRVE. CRAE and CRVE were presented in µm. One measuring unit assigns one µm in the Gullstrand's normal eye. Reproducibility for retinal vessel analysis is high with an interclass coefficient for CRAE of r = 0.94 and a coefficient of variation of about 2% in young children . --- Large Artery Stiffness PWV serves as a well-established and validated indicator of arterial stiffness . The assessment of PWV was performed using the non-invasive and validated oscillometric Mobil-O-Graph Monitor with integrated ARCSolver software . PWV was achieved in a sitting position and appropriate small-sized cuffs were settled on the left upper arm. After a resting period of 5 min, a blood pressure measurement was performed, which ensured a calibration with systolic BP. Two measurements of pulse wave analysis followed. After checking every measurement for erroneous values, the mean of two valid measurements was used for further analysis. Data on the validity for use of the Mobil-O-Graph in children has previously been published . --- Anthropometric Parameters and Physical Fitness Anthropometric measurements and physical fitness assessments have been explained in detail elsewhere . Briefly, body height and weight were measured in light sport clothes without wearing shoes, and BMI was calculated. The InBody device has been validated in school-aged children and correlates strongly with the measurement of dual-energy X-ray absorptiometry . BP was assessed five times using an automated oscillograph after 5 min of rest and based on the recommendations of the American Heart Association. Physical fitness was objectively measured by the validated 20-m shuttle run test . --- Questionnaire Parents were asked to fill in a questionnaire regarding their SES, migration background and parental lifestyle. Questionnaire items were included from our previous study and translated into the seven most spoken languages in Switzerland. --- Socioeconomic Status SES included household income and parental educational level. Household income was determined as low , medium and high income. Parental educational level referred to the highest school level completed by at least one parent. A low educational level was defined as both parents having/ not having completed compulsory school education, whereas neither of them absolved a vocational training. A medium educational level was determined as at least one parent having completed compulsory school education and vocational training. A high educational level was defined as at least one parent having completed high-school education with or without tertiary education. --- Migration Background Children were categorized into children with two-sided non-European migration background , migrants with one-sided European migration background or without migration background . --- Parental Lifestyle Parental lifestyle consisted of parental PA level and smoking status. Parental PA was categorized into three groups. Low PA was determined as both parents never being physically active or no more than once a week. Medium PA was defined as at least one parent being physically active twice a week. High PA was defined as at least one parent being physically active several times per week or on a daily basis. Additionally, parents were asked if they were smoker, or non-smoker. --- Statistical Analysis Univariate analysis of covariate was used to analyse the association of household income, educational level, parental PA level and smoking status. Different models were fitted to adjust for age and sex as well as household income, educational level and parental lifestyle. Bonferroni post hoc testing was conducted to reveal the direction of the results. Additionally, we applied sensitivity analysis to adjust for further potential confounders. For quantitative analysis of parental PA categories and smoking status, bivariate analysis was performed. To indicate the amount of uncertainty the measurement of effect presents 95% confidence intervals and a two-sided level of significance of p = 0.05 denotes statistical significance. All analyses were performed using an up-to-date version of Stata 15 . The sample size of the study was given by the expected large number of children and parents giving their consent and was based on calculations of our previous smaller scale study . In this study conducted by Imhof et al., we assumed a moderate effect size for the influence of a physical fitness performance on retinal vessel diameters in children. The sample size calculated with the software G-Power using F-tests was estimated to be ∼290 children in total. --- RESULTS --- Population characteristics are presented in --- Analysis of Covariance The results for differences in group-categories are shown in Table 2. --- Socioeconomic Status Thirty percentage of children were in the group with low, 33% with medium, and 37% with high household income. Eleven percentage were children with low, 22% with medium, and 67% with FIGURE 1 | Flow diagram. highly educated parents. Children with low household income parents had a higher PWV compared to children with high household income parents, also after adjustment for educational level . Retinal vessel diameters were not associated with household income. Lower parental educational level was associated with a lower AVR and higher PWV , but not independent of household income. --- Migration Background In our cohort, 9% of children had two-sided non-European parents, 13% had one-sided non-European parents and in 78% of children both parents were European. Children with a European background showed narrower CRVE and higher AVR compared to non-European children, independent of household income and parental education. European migration background was not associated with arterial stiffness. --- Parental Lifestyle Based on parental PA level, 25% were classified as parents with a low PA level, 21% as medium and 54% as parents with a high PA level. Based on parental smoking status, 67% were non-smoking parents. Higher parental PA level was related to higher AVR and lower PWV , but not after adjustment for parental smoking status. Parental smoking status was associated with a lower AVR , but not independent of parental PA level. Children with smoking parents had a higher PWV compared to children with non-smoking parents , also after adjustment for potential confounders. Bivariate analysis illustrated that microvascular AVR increased according to increasing parental PA level and a non-smoking status . --- Association With Parental Gender Mother's and Father's educational level, migration background, smoking and PA behavior are shown separately in Supplementary Table 2. As a main result, children with a mother of high education had a higher AVR due to wider arterioles, independent of household income. The educational level of the father was not associated with childhood vascular health. Children with a European mother had a higher AVR due to narrower venules compared to children with a two-or one-sided non-European mother independent of household income and educational level. The migration background of the father was not independently associated with childhood vascular health. In children with high physically active mothers, retinal microvascular AVR was higher independent of parental smoking status. The PA level of the father was not associated with childhood vascular health. However, smoking status of the father was associated with an unfavorable lower AVR and a higher PWV independent of parental PA levels. --- DISCUSSION Our study demonstrated several detrimental associations of socioeconomic and migration status as well as parental lifestyle with macro-and microvascular health in young children. The influence of parental lifestyle on vascular phenotype was evident examining parental PA and smoking status with both, retinal vessel diameters and PWV. Children with smoking parents had higher arterial stiffness, independent of parental PA level. The analyses of the interrelation between parental PA categories and smoking status with microvascular AVR demonstrate the additive deleterious effects of parental physical inactivity and smoking on the microvascular phenotype of the children . Children with smoking and low physically active parents showed the lowest AVR in retinal microcirculation. Parental PA levels and smoking status affect the vascular phenotype in young children. It has previously been shown that parental PA and perceptions of competence are associated with parental support for children's PA and therefore, for children's PA at home . This could be one explanation for the association of parental PA level with vascular health in our study. In addition, there is evidence that childhood exposure of environmental tobacco smoke is associated with vascular functional impairments in adulthood . In line with our results, a previous study demonstrated that passive smoking is associated to attenuated endothelial dysfunction in 11-year old children . Vascular alterations are related to impaired arterial vasodilatation. Arterial dilatation is mediated in large part by the release and bioavailability of nitric oxide. Our findings suggest that the lower AVR and higher arterial stiffness in smokingexposed children may be due to the interaction of smoke particles with the nitric oxide pathway. In our cohort of young children, higher household income was independently associated with lower arterial stiffness. The Young Finns study previously reported that childhood SES, defined as annual household income and years of parental education, is inversely associated with PWV in adulthood . Previous studies have shown that parental educational level seems to contribute to a higher prevalence of childhood obesity , elevated BP and sedentary behavior . Lower maternal education has been associated with higher blood pressure in 6 year old children . However, no relationship between maternal education and arterial stiffness was found . In line with previous findings, parental education was not independently associated with PWV in our cohort. Our results show that high maternal education is related to favorable microvascular health . Interestingly, maternal SES and lifestyle behavior was primarily associated with vascular health of children. An exception was seen for smoking status of the father which was independently association with micro-and macrovascular alterations of children. In adults, it has been shown that narrower CRAE and wider CRVE as well as higher arterial stiffness are predictors for increased risk of CV disease . Therefore, disadvantaged SES may be an important influencing factor for the development of CV disease later in life. With respect to migration background, European children had narrower CRVE and a higher AVR compared to non-European children. Similar results were found in a previous smaller-sized study on migration status and retinal microcirculation . In our cohort, no association of migration background and PWV was observed. There is evidence that migration background is associated with CV risk factors such as childhood obesity and sedentary behavior . We recently showed that childhood obesity and elevated blood pressure are related to retinal vessel alterations and arterial stiffness in this cohort of children . Based on our findings, we assumed that both micro-and macrovascular beds give separate clinically relevant information on health disadvantaged of children with migration background. Some limitations have to be discussed. Our cross-sectional study is associative in nature and thus no differentiation in terms of causality can be made. Causal associations of vascular alterations with the development of CV disease in adulthood have to be verified by a longer-term follow up study. Our study was designed in a school setting and 45% of parents did not give consent for participation. Oftentimes parents find it difficult to follow instructions and to fully comprehend the content of the questionnaire and may therefore have decided not to participate . The low participation rates in questionnairebased surveys are common limitations . We were not able to compare participants with non-participants with respect to parental lifestyle and SES. However, children who did not participate in the medical screening and did not have a completed questionnaire had a similar BMI, but a lower physical fitness compared to those participating. The risk of a selection bias was therefore considered to be moderate. The oscillometric measurement with the Mobil-O-Graph device to estimate PWV is based on a calibration with systolic blood pressure. The assessment is thus associated with level of current systolic blood pressure. The assessment of PWV in large cohort of school children has been shown to be a validated approach . We refrained from additionally adjusting for systolic blood pressure in our statistical models to avoid over-adjustment. Our findings are related to a predominant Caucasian population with a small percentage of other ethnical groups. We have therefore chosen to characterize our population by looking at non-European migration background. Future studies have to investigate the association of SES, migration background and parental lifestyle with vascular health in other ethnic groups and populations from different countries. One strength of our study is the large sample size and the limited age range of young children. During childhood, mirco-and macrovascular function and structure continuously develop, and especially during puberty, age adaptations occur rapidly. Investigating different vascular beds in a large sample of children at the same age therefore reduces a developmental impact of our findings. Future studies may still investigate the association of SES, migration background and parental lifestyle with vascular health in a wider age range from infancy to adolescence to account for the dynamic pattern of small and large vessel development during childhood. --- CONCLUSION Parental PA has been found to be associated with better microand macrovascular health but not independent of parental smoking status. It appears that parental smoking may mitigate the positive association of parental PA on childhood vascular health. Parental smoking was itself independently associated with impaired large artery stiffness in children. Cessation of parental smoking therefore seems key to improve childhood vascular health. Prospective follow-up studies will have to prove if cessation of parental smoking can improve CV outcome in offspring long-term. SES and migration background also affect micro-and macrovascular impairments early in life. Primary treatment strategies will have to address reduction of socioeconomic barriers. Family-based interventions targeting cessation of parental smoking, increase of parental PA, and reduction of socioeconomic barriers may be effective means to counteract the development of CV risk and disease in offspring and during lifespan. --- DATA AVAILABILITY STATEMENT The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. --- ETHICS STATEMENT The studies involving human participants were reviewed and approved by Ethics Committee of the University of Basel . Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin. --- --- SUPPLEMENTARY MATERIAL The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpubh. 2021.610268/full#supplementary-material --- Conflict of Interest: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Background/Aims: Socioeconomic barriers and lifestyle conditions affect development of cardiovascular disease in adults, but little is known about the association of parental lifestyle and education with childhood health. We aimed to investigate the association of socioeconomic status (SES), migration background, parental physical activity (PA) and smoking status with micro-and macrovascular health in children.In 2016/2017, 833 school children (aged 7.2 ± 0.4 years) in Basel (Switzerland) were screened for retinal arteriolar-to-venular ratio (AVR), pulse wave velocity (PWV), SES, migration background and parental PA as well as smoking status. Results: High parental PA levels were associated with a favorable higher AVR (p = 0.020) and lower PWV (p = 0.035), but not independent of parental smoking status. Children with parents who smoked had a higher m/s] compared to children with non-smoking parents [4.32 (4.29-4.34) m/s, p = 0.001]. Children of parents with a low household income had a higher PWV [4.36 (4.32-4.41) m/s] compared to children of parents with a high household income [4.30 (4.26-4.34) m/s, p = 0.033]. Low parental educational level was associated with a lower AVR [0.86 (0.85-0.88)] compared to children with highly educated parents [AVR:0.88 (0.87-0.88), p = 0.007; PWV: 4.33 (4.30-4.35) m/s, p = 0.041]. Children with a European background showed a higher AVR [0.88 (0.87-0.88)] compared to non-European children [AVR: (0.86 (0.85-0.87), p = 0.034].Parental PA is associated with better macro-and microvascular childhood health. However, the positive association is lost when parental smoking is considered in the analysis. Socioeconomic factors seem to associate with subclinical vascular alterations in children. Primary prevention programs should focus on including parental lifestyle interventions and educational programs to reduce the burden of lifestyle-associated barriers in order to improve cardiovascular health during lifespan. Clinical Trial Registration: ClinicalTrials.gov Exercise and Arterial Modulation in Youth, https://clinicaltrials.gov/ct2/show/NCT02853747, NCT02853747.

INTRODUCTION This paper is about citizen participatory activities through information communication technology that promote collaboration and partnership between citizens and law enforcement agencies in community policing. Participatory policing practices refer to initiatives involving citizens acting in identifying and addressing safety and security issues in their neighborhood. They qualify as participatory practicing, where citizens are assisting law enforcement . The concept of citizen participation dates back to 1980' in the United States of America . Citizens as social actors may report or give clues about threats in their neighborhood for police action when interacting with police officers during police regular patrols. Police regular patrols may be in form of foot or vehicles patrol aiming to enhance visibility in the community . Mobile police service are designated days on a weekly, fortnight, or monthly basis depending on available resources. Are programmed to serve remote communities that otherwise may have to travel long distances to reach police stations. Beneficial for the people in hard-to-reach villages and hilly areas with sparse transport access and road connectivity. Citizens may also inform lawlessness activities to officers during Mobile police service . Citizens may also practice neighborhood patrols by volunteering to create a team that patrols within the neighborhood to look for any suspicious persons who may commit crimes when chances permit and report them to police. The existence of community forums provides a common platform for various stakeholders to interact to improve security in neighborhoods, such as government, police, and community . The primary objective of these forums is to highlight and address persistent community concerns before these escalate to bigger problems . Citizens are active in neighborhood watch program where a group of people living in the same area get actively involved to make their neighborhood safer by working together and in conjunction with local law enforcement to reduce crime and improve their quality of life . Obeagu in Nigeria, recommends regular and continuing training of neighborhood watch members through seminars and workshops on lawful operational tactics and recognition of citizens' fundamental rights. Some citizens depending on the need have become members of a group of volunteers who decide on their own to stop crime and punish criminals. It is an unofficial way to prevent crime or to catch and punish perpetrators of crime, where there is a conspicuous absence of official police presence to control. In Nigeria, vigilante groups can patrol in cities to deal with rebels reducing fear of crime . In Tanzania, known as Sungusungu deal with cattle raiders because the police are far from them . Somehow vigilantism is prone to opportunism and can generate violence, corruption, and social othering. In Indonesia citizens volunteer to rotate patrols by residents in the form of a system for maintaining community security in homes and make them safest place known as the Siskamling system. In Afghanistan volunteer forces known as the Arbakai are military forces composed solely of volunteers, without resorting to a military draft. They aimed at reducing kidnaps of women and children because the police could not help the communities in Afghanistan . Citizens benefit from Community awareness-raising by police . These activities are planned by police for the community in the form of campaigns and are meant to provide community members with the necessary information required to stay safe and avoid common dangers. All the above may be very convenient through virtual interactions amplifying interactions between citizens and police to manage security. Finally, complaint management to police is a system where any members of the community can raise their concerns related to the safety of their neighborhood or about the behaviors of police officials. Through digital platform the activity may improve submission of individual and community complaints and providing feedback. The community members can also provide their feedback related to various problems highlighted in the forum . The willingness of members of the public to share information with the police is central to the operation of the criminal justice system . --- Police Involvement in Communities Society needs police for the maintenance of law and order, preservation of public peace and protection of human rights . Police Act provides for community policing fora to influence police citizen partnership . Police are divided into national, regional and local levels . At the local level, the police coordinator consults, coordinates and collaborates with citizens to organize policecitizen partnerships. Community Policing Forum members are identified and coordinated at Traditional Authority, Group Village Headmen, and Village Headmen levels. The police Coordinator is responsible for awareness campaigns, advisory and directive roles towards crime management . In the study the partnership must be guided by Arnstein's rung no 6 of the ladder of participation Table 1 below, for cooperation, reciprocity and trust for effective social cohesion, collective efficacy and community guardianship. Police involvement in the neighborhood can improve the social organization of neighborhoods and can decrease feelings of insecurity . Source: However, community policing is negatively affected by lack of resources . In Malawi, community policing is challenged by lack of knowledge by police and lack of recognition of community policing; corrupt practices; lack of confidentiality; poor working relationships; and lack of resources . However, spontaneous and unorganized proliferation of ICT applications in community policing has both supplemented and supplanted conventional policing . Maqsood et al. claim that ICTs may augment for lack of resources in community policing in Southeast Asia. --- Application of Icts On Citizens' Participatory Practices The digital platform has accorded space for the interaction of citizens through forming social networks to solve society's problems. Statistically, it shows, that since 2017, the number of users in social media has increased almost by one billion, reaching a total of 4.2 billion active users by 2021 . ICTs play a significant role in all aspects of modern life and have changed people's ways of communicating, finding needed information, citizens' interaction with others and government agencies on security . Has amplified watchfulness and surveillance practices. Voluntary citizens' participation in crime prevention leads to an increase in social support, the feeling of safety, and the active prevention of offences because it is digital monitoring . In theory, smartphone applications and social media channels make participatory surveillance practices accessible to all citizens. Attributed to the low threshold to communicate information to WhatsApp groups because it is easy to use the messaging applications already freely available and in use. These messages are not monitored by government institutions and are owned by commercial institutions. The conversations play out in an invisible and uncontrolled environment . Alongside everyday digital connections such as social media, citizens watch one another in person and spend time checking their environment and the behaviors of other citizens in that space. However, voluntary citizens are actively stimulated to take this greater responsibility within their communities to manage potential security risks. They internalize law enforcement strategies and use these in their community. Consequently, citizens become responsible for the safety and security of not only themselves but their communities and fellow citizens . These ICT applications and initiatives are increasingly becoming popular and signal a shift in relations between ordinary citizens and police practices. Through ICT applications and initiatives citizens in the neighborhood exchange warnings, concerns, and information about incidents, emergencies and allegedly suspicious situations. These exchanges lead to neighbors actively protecting and monitoring their area and sending messages about suspicious activities, increasing safety and social cohesion . The use of ICT applications and initiatives to manage security may vary from self-organized, citizen-led, do-it-yourself policing practices to police-initiated surveillance projects. When applying ICTs in community policing; There are no centralized universal guidelines-anyone can join and participate, it is he/she who feels it. Be aware and notice suspicious behaviors Alert the police Inform the ICT initiative group networks, and React safely. Malawi police service adopted community policing in 1994, but more of it has been conventional community policing. Characterized by low participation due to lack of resources. The result has been an increased state of decline influenced by a lack of citizen participation, poor consultations, coordination and collaboration between police and citizens . The use of social media has become pervasive in Malawi because of increased access to mobile devices in both rural and urban Malawi . There has been social media information sharing related to security, but how citizens participate in managing security through ICTs in community policing is important to know. Are people much more engaged in managing their security now than before ICTs became applicable in citizens' participatory activities in Malawi? The paper sought to analyze citizens' participatory activities in managing security in their neighborhood by applying ICTs in community policing and evaluating their effectiveness. This study is important because ICT-enhanced community policing may improve consultations coordination and corroboration enhancing security management in neighborhoods within local areas. Results will improve the citizens-police participation and interaction through digital and learn about the challenges of ICT-enhanced community policing in Malawi. To answer the research objective, analyze participatory policing practices that citizens do in their neighborhood security management and apply Information Communication Technology. The study came up with three specific questions. What participatory activities do citizens do to manage security in their neighborhoods, How are ICTs used to increase citizens participatory activities and How effective are ICT applications in citizens participatory activities? The rest of the paper is organized as follows: methodology which describes the study area, and methods of data collection and analysis. Then presents results on identified citizens participatory activities, how ICTs are applied in citizen participatory activities and ends by analyzing the effectiveness of ICTs in citizens participatory activities. --- METHODOLOGY --- Study Area The study was conducted in Muloza-Mulanje. The area is located at 36 K 792628.97 m E 8220322.32 m S. Muloza is on the east of Mulanje along the border between Malawi and Mozambique. The nearest province is Vila De Milange District to the East of Mozambique less than 10 km from the border. Muloza has unique social security challenges including the use of unchartered routes through the Muloza and Ruo rivers can be attributed to the fact that many border communities share common language and blood ties and claim reciprocal rights to access services. The existence of "international boundaries and territoriality" is often ignored. Formal migration and trade take place at the Muloza border post controlled by the Immigration Department and Malawi Revenue Authority . Mulanje mountain barriers between Muloza and the mainland of Mulanje and Phalombe. Police are challenged with a lack of resources. Hence criminals take the hard-to-reach areas as safe havens . --- Research Design The study adopted an exploratory sequential mixed method design. Mixed methods offset the weaknesses of the other. Data collected from multiple angles bring reliable conclusions. Qualitative aimed to collect data through oral interviews with focus group discussions, to explore and capture people's attitudes, beliefs, and motives. Quantitative aims to collect data through household surveys, to confirm and generalize qualitative results at a wider level . --- Data Collection Methods --- Focus group discussion and key informant interview The focus group discussions purposely identified four groups of participants within Muloza-Mulanje 10 police officers, 10 Community Police Forum members, 10 citizens and 10 business people. The study also conducted one key informant interview with police officers at the Regional Police Headquarters at Luchenza. The focus group discussions with participants from police, business persons, citizens and CPF members were more important because these are stakeholders who interact daily in community policing in Muloza and are actors in the phenomena. One key informant interview was conducted with police at Regional Headquarters because they are experts in guiding community policing. --- Household Survey The household individual survey used a structured questionnaire and randomly picked 432 participants in Muloza. The sampling technique was guided by Yamene's formula with a population size of 23,408 . The sample size was distributed as in Table 1 below. The structured questionnaire had questions on demographics, citizens participatory activities, how ICTs are applied in citizen participatory activities and the effectiveness of ICTs in citizens participatory activities. --- DATA ANALYSIS Qualitative interviews were transcribed, edited then uploaded in the NVivo software package. An initial coding scheme was developed with research questions providing a framework for labelling each category of data. Thematic analysis was conducted, utilizing nodes as the method of identifying emerging themes. Described this as a 'cross-sectional code and retrieve method', which is used to organize and highlight in a systematic manner emerging themes found in data. This method of analysis, however, is not without its critics with arguments made regarding a loss of context during the coding process. In an attempt to retain important contextual information where appropriate larger 'chunks' of data were included in the coding process which allows contextual information to be maintained . Data from structured questionnaires were analyzed by descriptive statistics using frequencies, percentages and charts. The data was analysed in Statistical Package for Social Sciences version 20. --- RESULTS AND DISCUSSION OF THE STUDY --- Demographic Profile of Respondents Based on the data collected from 432 respondents, there were 49.1% females and 50.9% males with an average age of 38.96 years. The marital status was that 5.3% were divorced, 7.2 % were separated, 10.2 % were widowed, 20.8 % were single, and 56.5% were married. It was found that 16.0% never attended and 2.5% attended adult school. The larger proportion of the respondents attained primary, secondary and tertiary education 32.2%, 38.9 % and 10.4 % respectively. The occupations of the respondents were 6.7 % school going, 7.6% casual labour, 16.2 % formal employment, 16.4 % skilled employment, 26.2 % farmer, 26.9 % petty traders/business with an average income of MK 103, 262.82. --- Citizens Participatory Activities In response to the question "What participatory activities do you engage to manage security in your neighborhoods?" Household survey results in Table 3 above reveal that 67% of respondents agreed to have participated in neighborhood watch patrol activities in managing their neighborhood security. Household survey results in Table 4 below show that 90.3% of respondents confirm they feel personally responsible for the safety and security of their neighborhood, and Table 5. below show that 78.5% agree that citizen participation in ensuring neighborhood safety and security is very important. Table 6. results indicate 71.2% of respondents are very satisfied with the level of citizen participation in neighborhood safety and security practices in the community. The interpretation is that citizens engage neighborhood patrols as participatory activity to manage security in their neighborhoods. Citens are more zealous than police on taking charge of security. Almost every person counts and this includes every gender. Citizens perceives, security on democratic roots . The implication is that police are not actively involved in neighborhood patrols. The evidence to show that citizens were engaged in neighborhood watch patrols is substantiated in recorded statements during focus group discussions as evidenced by following statements: "As CPF members, we always form neighborhood patrol groups of about 40 members. We do patrols to monitor the people who just move aimlessly at night with intentions to commit offences" . "The CPF members do patrols within the market area and if they encounter challenges, they make telephone calls to police for assistance on transport or manpower reinforcement" . "It is our responsibility to participate by doing patrols and communicating on social media with fellow citizens, CPF and police" . "Police stopped doing patrols in 2022, helping night neighborhood patrols. Police officers on police patrols were able to join the neighborhood patrols to beef up security" . "By the end of 2022, police officers did vehicle patrols once during night times. This means that if we may have police vehicle and police foot patrols together with our CPF neighborhood patrols there could be enhanced security" . Results are similar to Pridmore et al. that safety starts with citizens who take responsibility to joint efforts to manage their security in their neighborhoods. Mols Citizens are social actors, can aid police and make the social control process more effective because they stay in the community where crime occur and know suspects. Contrary to citizen's zealousness, Yero et al. claim community oriented-policing may bring back memories of Marxian revolutionary stages of the state where in the end the people will rise up to the challenges of their time and take charge of state affairs in the "government of the people" or rather the proletariat state. Source: Own survey --- ICT application used in citizen participatory activities In response to the question, "How are ICTs used to increase citizens participatory activities?" Quantified responses from focus group discussion fig 1 below, reveal that citizens mentioned Telephone calls for reporting and requesting transport on security information to police and CPF members at 25. CPF members mentioned Telephone for organizing patrols15 times. Telephone for mobilizing help and WhatsApp for reporting tied at the mention 10 times each. Police was found to use Telephone calls, SMS, WhatsApp, and Facebook because of nature of their job . To confirm and generalize qualitative results, household survey results in Table 7 below indicate that ICT applications used by citizens during participatory activities for neighborhood safety and security purposes were SMS application at 49.9%, seconded by WhatsApp application at 38.8% and least by phone call application at 24.2%. Table 8 below reveal 67.2 % of respondents used ICT applications to report safety and security incidents to law enforcement agencies. Table 9 below results indicate ICT applications allow citizens share information about suspicious activities on crime in the neighborhood at 79.2%, alert neighbors about potential security threats or emergencies at 67.5% and organizing community patrols at 26.4%. The interpretation is that, SMS, WhatsApp and telephone call applications have revolutionized security management where citizens were able to report and request transport on security information to police and CPF members, used for organizing patrols and mobilizing help during participatory activities in the neighborhoods. Implication is ICTs provide wide participation characterized by formal and non-formal communications, increased and enhanced citizen engagement, collaboration and partnership. More of neighborhood management is done by citizens. There is less participation by police in neighborhood patrols. Self-organized and do-it-yourself styles of virtual communication takes the lead. The evidence of enhanced consultations, collaboration and coordination through ICTs in managing security by citizens is evidenced by what respondents pointed out during interviews: "When I call a neighbor through telephone, neighbor also calls other many neighbors. In the end, we may have alerted each other about managing security within our neighborhood through Telephone calls where no person suspects us. If suspects have targeted my neighbor's house and I happen to see them, I would simply dial my telephone to alert others so that we can mobilize to intervene in the issue" . "At Makokola village within Gawani area. Criminals attacked the house of a community member. Neighbors managed to call through telephone to some of us alerting us of the development. I also called through telephone some four friends to come to our help. We managed to mobilize about 20 of us. The criminals who were outnumbered started running away" . "At Limbuli the citizens and CPF members mobilized through telephone calls and managed to overpower the criminals where they managed to arrest one suspect and kill one criminal" . "When I see any suspicious activity during the night, I always call through telephone to get assistance" . Results are similar to Mols and Pridmore that use of ICT applications in participatory activities vary from self-organized, citizen-led, do-it-yourself policing practices. No centralized universal guidelines to control joining and participation. --- Effectiveness of Icts in Participatory Activities In response to the question, "How effective are ICT applications in citizens participatory activities?" House hold survey results in Table 10 below indicate 84.5 % of respondents confirm use of digital platforms has positively affected citizen participation in neighbourhood safety and security responsibilities. Table 11. below reveal 71.2 % of respondents are very satisfied with the overall level of safety and security in their neighbourhood. This may be interpreted, ICTs have greatly enhanced citizens' participation in managing security within their neighbourhoods because ICT applications accorded real-time updates, real-time responses, and real-time capacity to access prompt assistance through virtual context. Implication, virtual platform accorded wide participation with formal and non-formal communication with authorities, increased social cohesion, collective efficacy and community guardianship. Citizens actively took ownership of their neighbourhoods, became more engaged, alert and reporting to police or neighbourhoods preventing crime and arresting of suspects. The evidence from interview statements is not limited to the following: "WhatsApp is more efficient because allows taking and sending of pictures and videos of evidence to police that is used to nab a notorious criminal" PFGD Reference 3). "Wide participation of citizens through WhatsApp and Facebook will allow citizens to share with anybody on sensitive issue which others might have been hiding. And once the issue has been widely deposited many audiences may have the chance to see and make a follow up, influencing reduction of corruption" . "When police have arrested suspect, we share through WhatsApp with other police formations for identification if the suspect could be wanted by them too. Social media defeats lack of resources such as transport and time" . "People send various security related issues on WhatsApp and law enforcers are able to learn that someone has violated someone's rights, immediately investigations may be instituted. For example, a bus driver was driving carelessly and recklessly and one passenger recorded a video then uploaded on WhatsApp group. Immediately, the members started commenting and some police officers too were on the group. They informed their traffic friends on duty along the road. They stopped and apprehended the driver" . "The advent of social media such as WhatsApp and Facebook have allowed other citizens who may be concerned to act by recording the activity and secretly sharing with other groups. In the long run there is no hiding of the issue" . "Sometimes, those who upload videos and picture on WhatsApp are forced to do so because they do not know where to complain. In short social media has allowed wide participation and does not follow formal structures" . "ICT applications have allowed police and citizens to build good criminal intelligence because of WhatsApp group initiative which has even gone beyond our borders in Mozambique" . "The child went missing at Kanjedza in Blantyre. Through Facebook and WhatsApp sharing the information between police of Muloza and Blantyre. The Muloza police also linked up with Mozambique counterparts and indeed the child was found in Mozambique. This shows how important social media is on prompt and wide participation" . "WhatsApp overrides differences between networks. We can share, and deposit security-related materials promptly and effectively" . "Last night a gang of criminals had invaded someone's house intending to steal from him. Through telephone calls, we mobilized and managed to arrest one suspect and also managed to kill one suspect" . Results are similar to Kumwenda et al, that digital platform has enhanced citizen positive engagement because of real-time updates, real-time responses, and real-time capacity to access hard-to-reach populations through virtual. Pridmore et al. surveillance and monitoring has become more accessible and desirable. --- Conclusion and Recommendations Results show citizens participate in neighborhood patrols and they feel personally responsible because it is very important for them and are satisfied with level of participation. Almost every person counts and this includes every gender in securing neighborhoods. Citizens use SMS seconded by WhatsApp and least Telephone calls, and have revolutionized communication for reporting and requesting transport on security information to police and CPF members, for organizing patrols and mobilizing help during participatory activities in the neighborhoods. ICTs have greatly enhanced citizens' participation in managing security within their neighbourhoods because they accorded real-time updates, real-time responses, and real-time capacity to access prompt assistance through virtual context. Virtual platform accorded wide formal and non-formal communication with authorities and increased social cohesion, collective efficacy and community guardianship, preventing crime and arresting suspects. --- Based on findings, police must take the leading role in citizen participatory activities and civic educating on the importance and advantage of public attention to messaging because can have different effects on crime, participating neighbors become more alert and more willing to contact the police or their neighbors, and that citizens take more measures to prevent crime. The study recommends more research on the challenges of the deployment of ICT applications for the detection and investigation of crimes in community policing. --- Declaration of Conflict of Interest The authors declare no conflicts of interest. Co-authors have reviewed and agreed to the manuscript, and there is no financial interest to report. The submission is original and not under review elsewhere.

The ubiquitous proliferation of mobile phones has made digital surveillance and monitoring both accessible and desirable. Citizens have actively engaged in managing their neighborhood security. The study aims to analyze participatory policing practices that citizens do in their neighbourhood security management and apply Information Communication Technology. To achieve the research objective, the study used mixed methods exploratory sequential design. Qualitative purposive sampling targeted four focus group discussions and one key informant guided by an interview guide. The household survey used randomly sampled 432 respondents using a structured questionnaire. Data collected was analyzed using descriptive statistics in SPSS version 20. Qualitative data analysis follows transcribing, coding, and grouping into sub-themes, themes answering to research objectives. Results show citizens engage neighborhood patrols and they feel responsible and are satisfied with level of participation. Almost every person counts and this includes every gender in securing neighborhoods. Citizens use SMS seconded by WhatsApp and least Telephone calls for participatory activities, have revolutionized communication for (i) reporting and requesting transport on security information to police and CPF members, (ii) for organizing patrols and (iii) mobilizing help during participatory activities in the neighborhoods. ICTs have greatly enhanced citizens' participation in managing security within their neighbourhoods because they accorded real-time updates, real-time responses, and real-time capacity to access prompt assistance through virtual context. Virtual platform accorded wide formal and non-formal communication with authorities and increased social cohesion, collective efficacy and community guardianship improving prevention of crime and the arresting of suspects.

Introduction Adolescent girls and young women aged 15-19 living in Sub-Saharan Africa are underserved by existing sexual and reproductive health service delivery [1]. Research consistently demonstrates adolescents' disproportionate risk of HIV acquisition and unintended pregnancy [2,3], and the related poor outcomes of maternal morbidity and mortality and complications of abortion under unsafe conditions [4,5]. Forty-five percent of pregnancies among African adolescents are estimated to be unintended , and nearly two-thirds are classified as having unmet need for contraception-meaning they are sexually active and report not wanting a child for at least 2 years, but are not using a contraceptive method [3]. In response to these indicators and the unique developmental and social circumstances facing adolescents, researchers, governments, and funders have advocated for a renewed emphasis on improving adolescent SRH [7]. With the goal of promoting adolescent SRH, the global public health discourse endorses targeting this population with interventions to increase contraceptive prevalence-or reduce unmet need-through voluntary and tailored family planning services [8]. Despite substantial investments in adolescent SRH and high contraceptive availability in many countries [8,9], however, contraceptive prevalence has been slow to rise among adolescents [10]. Multiple sociocultural and health systems barriers intersect to influence adolescents' contraceptive access and use [11]. Adolescent sexuality and contraceptive use are highly stigmatized in many communities [12][13][14], and discriminatory treatment from healthcare providers further deters SRH care-seeking [15,16]. Adolescents are susceptible to misconceptions about contraceptive risks, especially future fertility concerns, which are rooted in gender and social norms [17][18][19]. Furthermore, it is clear that unmet need for contraception-a frequently-used concept and indicator referring to fecund women whose stated desire is to delay or limit childbearing but are not using a contraceptive method [20]-does not equate with desire to use contraception [21]. The public health research and policy narratives have focused on the former, leading to a gap in knowledge about how adolescents in sub-Saharan Africa who are most likely to experience unintended pregnancy perceive their own pregnancy risk and contraceptive need, or make decisions about contraception within their lived realities and social networks. Thus, the reproductive priorities of adolescents classified as having an unmet need for contraception, and influences on their contraceptive use or nonuse, have not been adequately characterized. These needs and drivers of SRH must be framed by the complex and often non-binary nature of reproductive preferences. A growing body of literature has highlighted the limits of using the pregnancy planning paradigm, or dichotomous categories of planned/unplanned and intended/unintended, to understand contraceptive needs [22]. More reproductive preference data and measures specific to African contexts and adolescent populations are needed to guide interventions to support contraceptive decision-making. These research gaps are particularly germane to adolescents younger than 18; younger, unmarried adolescents are underrepresented in SRH research due to ethical constraints and parental consent requirements, though studies suggest they are even more vulnerable to poor health outcomes [23,24]. This study explores how Kenyan adolescent girls and young women aged 15-19 perceive their own contraceptive needs and make contraceptive decisions within their social contexts. We examine the social influences on contraceptive decision-making among adolescents at risk for pregnancy, aiming to provide nuanced insights into adolescents' contraceptive behaviors. Our research questions and approach were guided by constructs of Social Cognitive Theory [25] and Social Norms Theory [26]. Specifically, we focused on conceptualizing the dynamic interplay between adolescent behavior and their social environment, and adolescents' perceived self-efficacy in making contraceptive decisions [27]. We drew on insights of literature exploring how social norms affect reproductive health behaviors [26,28,29]. Through advancing understandings of adolescent contraceptive decision-making, this study aims to inform adolescent-centered efforts to better align reproductive preferences and outcomes for this underserved population. --- Materials and methods --- Study design and setting For this qualitative study, we selected in-depth interview and focus group discussion methods to provide data source triangulation, maximizing the benefits of IDIs for gathering data on complex and sensitive topics, and FGDs for examining questions within the context of social interaction [30]. We conducted 40 IDIs and 6 FGDs with adolescents aged 15-19 in the Nyanza region of Kenya between October 2017-March 2018. This region, which has the highest HIV burden in Kenya and high relative unmet need for contraception [31], was selected due to existing research infrastructure in the area. Study activities took place in the peri-urban Ahero community and the primarily rural area near Bondo town in Kisumu and Siaya counties, respectively. Luo is the primary ethnic group and language of the Nyanza region. All authors have had long-term research engagement in women's reproductive health research in the study region and have extensive collective experience working with adolescent populations in Kenya. --- Sampling and eligibility Prior to initiating community-based recruitment, a community advisory board was consulted to optimize communication about the study with prospective participants, parents of adolescents, and other community members. Our all-female data collection team consisted of two local qualitative interviewers with backgrounds in social science, assisted by two community mobilizers. The principal investigators trained the team, emphasizing ethical conduct of sensitive research among adolescents. The semi-structured guides were piloted, revised, and collaboratively translated to optimize question comprehension. The team worked alongside community health volunteers to recruit participants for IDIs or FGDs in various community venues, including markets, private homes, and youth gatherings. Using purposive and snowball sampling strategies, the study team recruited in rural and peri-urban locations in Kisumu and Siaya counties. Focus groups were disaggregated by age group to optimize group dynamics. Adolescents were eligible to participate if they were between the ages of 14-19; had a parent/guardian willing provide consent ; unmarried; spoke English, Luo, or Swahili; and were "at risk" for pregnancy. Adolescents who had ever been sexually active or used a family planning method, or were in a current romantic relationship with a male partner regardless of sexual activity, were considered "at risk." We excluded adolescents with a prior live birth in order to focus on adolescents with less representation in existing literature, but not those who had experienced spontaneous or induced abortion. Adolescents were approached in the community, and eligibility assessed privately by a study staff member. For adolescents 18 or older, written informed consent was signed. Written parental consent and adolescent assent were obtained for participants under 18. Participants received 400 Kenyan Shillings to ensure transportation cost coverage. This study was approved by the Kenyatta National Hospital Ethics and Research Committee, the Human Subjects Division at the University of Washington, and the County Health Management Teams in Kisumu and Siaya counties. --- Data collection The primary IDI/FGD domains represented in the semi-structured interview guide are listed in Fig 1 . Each participant completed a brief tablet-based socio-demographic questionnaire, which was administered verbally by study staff. IDIs were primarily conducted in private homes; when privacy could not be assured or when preferred by the participant, study offices located at two county health facilities were used. FGDs were held with permission in private areas of county health facilities. Staff were fluent in written and spoken English, Luo, and Kiswahili; IDIs and FGDs were conducted in the language preferred by participants. IDIs and FGDs were digitally audio-recorded and simultaneously transcribed and translated into English by the interviewers themselves; a different staff member reviewed each transcript for accuracy of transcription and translation. --- Data analysis We used an inductive, thematic approach to analyzing the qualitative data. Our iterative analytic strategy drew on contemporary grounded theory methods, described by Charmaz as "systematic, yet flexible guidelines" for qualitative analysis and theory-building [32]. The co-PIs agreed on a set of initial codes designed to reflect a priori domains of interest from the interview guides and exposure to the raw transcripts and field notes. The lead author , who has over 10 years of experience working in Kenya, and a Kenya-based expert qualitative data analyst coded 5 IDIs in parallel, after which they compared code application and meaning, adding new codes as needed to identify emerging themes. After a second round of parallel coding, investigators constructed a final codebook for both IDIs and FGDs, and re-coded or coded all remaining transcripts. Two-thirds of transcripts were double-coded, and investigators met regularly to discuss coding discrepancies and emerging themes. The analytic team identified the most significant codes, grouped similar and contrasting excerpts within and between codes, and wrote analytic summaries for each major theme. Investigators conducted all coding in Dedoose , an online qualitative analysis program facilitating data management and collaborative coding. The conceptual model was developed during the analytic process in order to organize and interpret themes emerging from the data. --- Results Study staff screened 200 adolescent girls and young women in community settings, of whom 115 were eligible and 86 were enrolled: 40 in IDIs and 46 in FGDs. The median age was 17, with 11 participants aged 15 or younger . The majority were currently students , romantically partnered , and reported ever being sexually active with a male partner . Most reported ever using a contraceptive method; 71% of these had used male condoms alone. No participants reported a prior pregnancy. Our findings synthesize the complex social factors at play as adolescents navigate pregnancy risk and decisions around contraceptive use. In the results and conceptual model below, we present the primary themes emerging from the data: adolescent pregnancy intentions and consequences, the apparent incongruence between pregnancy desires and contraceptive behaviors, and the major social influences on adolescent contraceptive decisionmaking, including various sources of stigma. --- Adolescent pregnancy: Intentions and consequences Understanding adolescents' pregnancy preferences is vital to gaining insight into their contraceptive decision-making. Respondents universally placed high value on future pregnancy and childbearing. The majority of girls resonated less with the concept of "planning" a pregnancy, and more with the idea that pregnancy is a "blessing" when it occurs at an ideal time, for example when education is complete and finances are stable. Participants viewed pregnancy while still in school or prior to financial independence extremely negatively. Many adolescents closely related adolescent pregnancy with lack of future educational attainment, which they explained led to living in poverty. Pregnancy among adolescents was seen as a burden to those close to them, and a challenge to meeting their basic needs for food and shelter. Speaking of a friend who had become pregnant, a young woman explained, "It was not a good life. . .it was a bad life because, you just see that grandmother has a problem, because she doesn't have a job and she also has to care for the baby and the mother, so it could force her to go to the rice fields. . .It was very sad for me [to see her get pregnant]. . . because she was very young, so I was thinking, now this girl, she is young and she is pregnant, When participants were asked to share stories of adolescents in their communities who had become pregnant, they described multiple situations in rich detail, requiring very little prompting from interviewers. They emphasized the bleak social realities pregnant adolescents face: shame of dishonoring parents, community and peer stigma, and social withdrawal. "She was 13 years old and she wanted to remove [have an abortion] and the boyfriend also denied the pregnancy, but I just advised her that removing it can kill you, so she just gave birth. . . She used to be stressed up because she felt like removing, and the friends are also talking badly, so she never used to walk, she just kept sleeping in the house. She even thought of committing suicide but I just advised her until she gave birth to the baby." Very few adolescents expressed ambivalence about pregnancy preferences. However, reactions to potential unintended pregnancy were extremely varied. For example, a 17 year-old participant expressed a strong preference to wait until she had finished school to become pregnant. However, when asked about how she would feel about becoming pregnant now, she imagined a relatively casual, unconcerned reaction, contrasting the temporary nature of pregnancy with the permanence of HIV infection: - --- "If I were to get pregnant now, I would just take it easy, befriend the nurses here [at the clinic]. . .I would be happy because pregnancy is not a disease that infects the body [like HIV], you only carry it for 9 months and you deliver and you go back to your old self." -IDI participant, Kisumu --- Incongruent desires? Desire to avoid pregnancy and contraceptive behavior All participants were previously or currently sexually active with a male partner or currently in a romantic relationship. While 20% of ever sexually-active adolescents declined to answer questions about frequency of sexual activity, 53% of those who responded reported sexual activity in the last month. Despite their sexual activity and their universally strong desire to avoid pregnancy while financially dependent on parents, hardly half of participants were using a contraceptive method; the majority of these were using condoms. Many participants stated that they would have an abortion rather than continuing a pregnancy at their current life stage. This was striking to the authors in the context of participants' broader narratives around abortion, which centered on numerous stories of peers and community members who experienced severe morbidity or death as a result of abortion under unsafe conditions. A notable, but not exceptional, example of this is a 17 year-old participant who recounted the story of her older sister, who had successfully aborted her pregnancy by drinking a local cola with tea leaves. The sister bled heavily, was eventually taken to the hospital, and survived. The participant then went on to talk about two classmates who had mixed laundry powder with water treatment concentrate and bleach to abort; both had died. She replied to the interviewer that "If I become pregnant now. . .in my mind I am thinking that I can abort it because I want to first study so that I can get a baby later." This respondent, a 17-year-old IDI participant in Kisumu, was using a contraceptive implant. However, the majority of those who said they would have an abortion rather than continue an unintended pregnancy in adolescence, often specifically acknowledging risk of death or serious injury due to unsafe abortion, reported using condoms inconsistently in the absence of other method use. Despite participants' pregnancy preferences and their eloquent articulation of the consequences of pregnancy, very few expressed unfulfilled desire for contraception. Adolescents were generally familiar with multiple methods, and were able to communicate where they could get a method; they named public and private health facilities and pharmacies, and some had received condoms from female community health workers. Rather, multiple competing priorities other than avoiding pregnancy emerged from the interviews and FGDs, including assuring future fertility, avoiding side effects and community stigma associated with contraceptive use, and preferences of sexual partners. The apparent incongruence between adolescents' pregnancy preferences and their contraceptive behaviors was not readily acknowledged by participants in the interviews and FGDs, even when specifically probed by interviewers. In section 3.3, we offer adolescent perspectives on the social influences that contribute to incongruent pregnancy intentions and contraceptive behavior. --- Social influences on contraceptive decision-making 3.3.1 Stigma: Sexuality and contraceptive use. Participants frequently used the term "spoiled" to refer to girls who are sexually experienced. While all but a few respondents felt that adolescent contraceptive use was generally more acceptable in their communities than adolescent pregnancy, both were considered stigmatizing and reflected poorly upon families. The minority felt that they would face fewer social consequences if they became pregnant, compared with openly using hormonal contraception. If a girl were discovered to be using a contraceptive method, many participants asserted, she could be labeled as "promiscuous" in the community. An 18 year-old participant, who used condoms consistently to prevent pregnancy but avoided what she called "family planning" explained: --- "They [people in the community] will come to many conclusions [if they see me going for family planning], maybe they will say that I have gone for family planning because I want to have sex." -IDI participant, Kisumu --- Community narratives of contraceptive risks. Adolescents spoke of pervasive community narratives of adolescent contraceptive use leading to future infertility or inability to give birth to a healthy baby in the future as a powerful influence on their decisions around contraception. Several participants brought up the term "family planning," explaining that women who already have a family are the only ones who should plan. An 18 year-old participant reported using condoms only for this reason: "People are saying that it [family planning] is not safe. . .let's say for example I decide to use family planning and yet I have never gotten pregnant, so people say that maybe I can end up being barren. . .That is the only thing I hear people [say] discouraging use. . .I think it is because there is no way I can just from nowhere start using family planning, I have never been pregnant even for one day, so I don't know whether I can get pregnant or not. So you should use after you have gotten pregnant." -IDI participant, Kisumu Some respondents reported not believing these widespread messages about family planning and infertility, citing examples of people they knew who had conceived after using a method. Others expressed uncertainty about what messaging to believe, but perceived family planning use prior to having a child as the lesser of two evils when compared with adolescent pregnancy: --- "If you are a good girl then you just take care of yourself [abstain from sex], you don't use family planning until you give birth one day. . .but nowadays because children start having sex early, it forces them to use family planning so that it can help them finish school. . .it is good if a girl. . .[uses family planning] until she finishes school then she can become pregnant. . .It is not a bad idea because for her she is securing her future." -IDI participant, Siaya, age 14 Concerns about contraceptive side effects figured prominently into adolescents' perceptions of their need for contraception. Changes in menstruation and body weight, as well as abdominal pain and reduced sexual urge were frequently mentioned, but were considered less troubling than the risk of infertility. Respondents also alluded to family planning stigma among adults and other adolescents in their communities, primarily associating use with sexual promiscuity and the seeking out of sexual pleasure among girls and young women. As one young woman put it, "I have heard that family planning if it doesn't suit you it can destroy your body. . .that it causes infertility. . .They say that family planning is not so good among young girls. Because some people once they are put family planning, so they just start becoming promiscuous." --- -IDI participant, Siaya, age 17 When participants were asked which specific people in their lives influenced their decisions to use or not use a family planning method, they most frequently named their mothers and boyfriends. Most described their mothers' advice or directive against using contraception other than condoms, usually due to concerns about effects on girls' health or future fertility: ". . .Sometimes maybe we are sitting in the house, our mother usually advises us to abstain from sex and not go for family planning citing that family planning spoils young girls. She normally advises us, especially me, she really advises me because I am the youngest in our house that I should not start using those things [FP methods] now. I should not even think of using them. . .If I start using now, in the future when I will want to have children it will be very hard for me to conceive. . ." --- -IDI participant, Siaya, age 15 Mothers had a high level of sway over their daughters' beliefs about contraception and their comfort accessing it. Condoms were not widely considered a family planning method among participants, however, and a young woman makes this distinction clear: --- ". . .For me, I can get advice from my mother. If I have to go for the family planning that is put on the arms [injection, implant] then I have to tell her because maybe this thing might bring complications, but for condom, I can just go without telling her-even if she will realize later. If it's about use of condoms then it is [up to] me to decide." -FGD participant, Kisumu, age 19 Female peers were main sources of influence on participants' sexual behavior, as well as information about contraceptive methods. Peers often propagated community narratives about contraceptive risks. A 14 year-old participant explained: "There is a woman I used to live with a long time ago, she is the one who told me that family planning is bad. . .that if you used family planning and you are a girl who has never had a child, then getting pregnant might be a bit difficult. . .When she told me like that I went and shared with my girlfriend and she told me that. . .she cannot use [family planning] because she has also heard that family planning is bad. . .." --- -IDI participant, Siaya Seeing peers get pregnant at school was particularly powerful for some adolescents, who often learned more about contraception from classmates than from parents: "Because maybe you are already spoiled [sexually experienced], and you know pregnancy never knocks at the door, it is like HIV, so that [school] is where we learned about family planning, because your parent can never advise you to go for family planning. . .Those whose parents were refusing that they go for family planning. . .could just sneak to the hospital." -IDI participant, Kisumu, age 17 --- Power in sexual relationships. While the median age of first sex among sexually active participants was 15 years, the majority of respondents reported that typical sexual debut in the community was between 10-14. Adolescents named peer and partner pressure as primary influences on sexual debut, and most discussed the need for financial support or material gifts from male partners, usually older adolescents or men, in order to have sex with them. A 17-year-old participant who reported first sex at age 13 described how she saw her peers acquiring food, clothing, and other material goods from sexual partners. Her own lack of these items motivated her to start having sex: ". . .maybe you see your friend has a nice dress and you don't have, so you know that boy will tell you that he is going to buy for you. So that is the reason she will say. . .let me go have sex with him so that I can get even inner wears [underwear]. -IDI participant, Kisumu She went on to talk about her first sexual partner, with whom she said she would not have had sex had he not bought her feminine pads among other things: "There is no need of someone to feel sweet on your body without helping you." Sexual partners both encouraged and discouraged contraceptive use. Some participants felt they were urged to access family planning by boyfriends to avoid the embarrassment of "impregnating" a young girl, or to stop using condoms. Others described pressure from boyfriends to stop using their method due to concerns around future fertility. Respondents described various dynamics around negotiating condom use, but most stated that boys put pressure on girls in general to have unprotected sex: --- "Most of them [girls] may want to use it [condoms] but now the boyfriend might refuse, and since she believes that she loves her boyfriend, she will just end up following what he says. . . [Boys] will say that you cannot eat [a] sweet with the cover [on]." -IDI participant, Siaya, age 17 Despite the diverse factors and experiences influencing their contraceptive choices, some adolescents expressed awareness of agency over their decisions to prevent pregnancy. A participant who had been sexually active since 14 recalled how a partner told her that condoms affected him negatively and refused to use condoms: "Long time ago I used to listen to him when he told me like that, but nowadays since I knew how this world is, if I have a man, I tell him we go for testing and he refuses, and I also tell him we use protection and he refuses. I just tell him my dear, walk with Jesus and you also let me walk with Jesus. . .. . . because that is someone who doesn't wish well for you, there is something he wants, he wants to spoil your future." --- -IDI participant, Kisumu, age 19 When asked who or what influences her decision to use contraception, another young woman who reported using condoms answered: "My future. . .Because if I don't give birth now I will finish school and I will have a better future and if I give birth now I will drop out of school and I wouldn't have a good future." -IDI participant, Siaya, age 18 --- Discussion Among the 86 adolescents who participated in IDIs and FGDs in this community-based study, we elicited a strong preference to delay pregnancy that was often uncoupled from a desire for contraception. While the majority of participants would be considered to have an "unmet need" for contraception based on the indicator's definition, their perception of contraceptive need and ultimately their contraceptive decisions were shaped by various factors and priorities within their social contexts. These factors are illustrated in the conceptual model . Informed by social norms and future life aspirations, participants viewed pregnancy in adolescence prior to relative financial independence as an unacceptable outcome. They then had to navigate interrelated "layers" of social messaging that ultimately influenced their contraceptive decision-making and ability to prevent pregnancy. Each "layer" had differing levels of influence on each adolescent. For example, for some, perceived risks of hormonal contraception causing infertility and stigma associated with contraceptive use outweighed pregnancy concerns and motivated a desire for condom use, while unequal power with sexual partners and need for financial support further influenced ability to prevent pregnancy using condoms. The curved arrows in the model represent the ability of adolescents who desire to delay pregnancy to traverse social norms and narratives, stigma, and unequal power dynamics that may intervene in their ability to prevent pregnancy. Use of a contraceptive method is a primary way that adolescents may exercise agency to prevent pregnancy, and several participants clearly articulated self-efficacy in demanding condom use or using contraception despite dominant social influence. Pulerwitz et al. propose a conceptual framework articulating the relationship between social norms and adolescent SRH that features social and gender norms as a central influence on health outcomes for adolescents [33]. The concept that power can be demonstrated in decisions to "adhere to social norms" [33] is relevant to our analysis. Contraceptive use was generally viewed as a transgression of social norms among our participants, but some adolescents made the decision to use a method in spite of social pressures. Furthermore, adolescents' agentic actions around contraceptive decision-making may not take the form of bold steps [28], but rather smaller and more socially-acceptable steps such as avoiding sex or inconsistent condom use. For example, since condoms are widely considered a mode of HIV prevention rather than a contraceptive method by adolescents, condom use carries less contraceptive stigma than hormonal contraceptive methods, which are known more widely as "family planning." While participants frequently reported elements of transactional sexual relationships, few adolescents reported other forms of sexual coercion, such as age-disparate relationships, or sexual assault. We discussed sexual coercion in the interviews, but our study was not focused on examining experiences of coercion and violence in detail. Studies suggest that many Kenyan adolescents have limited interpersonal power in sexual relationships [34], and approximately 20% of Kenyan girls and young women aged 15-24 have experienced sexual violence from an intimate partner in the prior year [35]. Additional research examining how a history of violence affects adolescent contraceptive decision-making is needed. Stigma emerged as a powerful source of adolescent influence in the present study, consistent with existing research [14], including a qualitative study of postpartum Kenyan adolescents from the same geographic region [17]. "Good girls" abstained from sex, didn't use contraception, and didn't get pregnant. The relevant stigmas can be conceptually tied together by participants' use of the word "spoiled," directly translated from Luo. Adolescents referred to themselves and peers who had begun sexual activity or were seen as promiscuous as "spoiled," futures were "spoiled" by adolescent pregnancy, and contraceptive use could "spoil" one's eggs and make one unable to have children in the future. Indeed, sexually-active adolescents risk experiencing stigma regardless of whether they actively prevent pregnancy through contraceptive use or become pregnant during adolescence. Understanding this context of multidimensional stigma may help to explain adolescents' apparently incongruent pregnancy preferences and contraceptive behaviors. Our study focused on contextual rather than on health systems factors influencing contraceptive decision-making, which limits our data's representation of stigma experienced by adolescents in the health care setting. However, prior qualitative and mystery-client studies describe provider bias towards sub-Saharan African adolescents, leading to restriction in method choice due to provider beliefs [36] as well as experiences of shame, embarrassment, and loss of confidentiality among adolescents seeking contraception [37]. Our data elicited little diversity in participants' current pregnancy preferences: they reported that an unintended pregnancy would be unacceptable. Adolescent pregnancy is common in Kenya: among women aged 20-24 in 2014, nearly 25% of women had given birth by age 18, and 46.8% by the end of the 19 th year. Among 15-19 year-olds, 18% were pregnant or had already give birth [38]. Given that context, we anticipated a wider spectrum of positive, ambivalent, or indifferent perspectives on pregnancy "now." Studies from the United States among young women advocate that unintended pregnancy is not a universally poor outcome; many unplanned pregnancies are welcomed and acceptable [39], and social context, including limited financial resources and social status, may make pregnancy planning less relevant for some [40]. The present study's narrow scope of pregnancy preferences likely reflects its young, 15-19 year-old population, and the exclusion of married and parous adolescents; it is also important to note that participants viewed future pregnancy extremely favorably. Interestingly, our findings are consistent with a recent quantitative study of ambivalence in pregnancy intentions in Kenya, where the youngest age group of women were the least likely to express ambivalence [41]. Social desirability bias may have influenced some adolescents' responses, especially in the FGDs, though many adolescents freely shared other behaviors and experiences that would be considered contrary to social norms. Additionally, as parental consent was required for participants under age 18, adolescents who did not feel comfortable involving their parents, perhaps due to parents' more conservative attitudes towards sexuality or contraception, may have avoided sampling. However, only 3 adolescents who were otherwise eligible for the study declined due to lack of parental availability, and no parent declined to provide permission. Quotes from IDI participants are primarily used in the results, as they tended to be more concise than the longer FGD conversations, but thematic elements were similar between data sources. Our study was designed to take advantage of the social nature of FGDs and anticipated that that the way peers conversed about the topics may uncover new insights. However, we did not find that the FGDs promoted interaction that elicited different perspectives than the IDIs. While our team thought carefully about word choice when translating interview questions from English into Luo and Kiswahili, there may have been challenges with managing the connotations and translation of various descriptors such as "unplanned" and "unintended." Despite translation quality control measures, it is possible that some meaning and nuance around pregnancy preferences were lost in translation. This study's conceptual framing distinguishes it from other studies assessing adolescent perspectives on contraceptive use. We explored norms and preferences around adolescent sexuality, pregnancy, contraception, and abortion to contextualize adolescent contraceptive behaviors and gain insight into their contraceptive decision-making. This approach allowed us to closely examine the incongruence between adolescent reproductive desires and contraceptive behavior through the lens of "spoiled" girls' narratives, beliefs, and life experiences. With this incongruence in mind, individual-and community-level interventions to prevent undesired pregnancy should focus on supporting adolescent decision-making agency rather than increasing contraceptive prevalence, with the goal of better aligning reproductive preferences and health behaviors to achieve those preferences. Such efforts might support adolescents in navigating various "layers" of social norms, such as providing adolescent-centered contraceptive counseling and decisional support, addressing misperceptions, and allowing adolescents to access contraception with peers or in locations not associated solely with contraception such as pharmacies [42,43] and other community-based venues. Clinical interventions must acknowledge and address pervasive social messaging linking contraceptive use with future infertility. Furthermore, efforts to engage adolescents and young adults, including boys and men, in shifting harmful social and gender norms at the level of communities [44,45] are also critical to enabling social environments for adolescent SRH care-seeking and healthy behaviors. The adolescent voices represented in our study support the perspective that contraceptive prevalence rate and unmet need for contraception are useful indicators to compare over time, but are not person-centered outcome measures [46] for programs and interventions aiming to improve adolescent SRH. Indeed, our findings and conceptual model identify domains that should be considered in such adolescent-specific measures of reproductive desires and preferences. Senderowicz calls attention to the mismatch between stated rights-based global family planning goals and the metrics currently used to evaluate family planning programs [21]. Contraceptive autonomy, defined as "the factors that need to be in place in order for a person to decide for themselves what they want in regards to contraceptive use, and then to realize that decision," is proposed as a novel family planning indicator [21]. Measures such as contraceptive autonomy, reproductive autonomy [47], and sexual and reproductive empowerment [48] should be tailored for adolescents and used to guide programs and funding priorities. Finally, contraceptive access and the ability to prevent pregnancy with contraception when desired are clearly critical to efforts to improve adolescent SRH outcomes. However, as was poignantly expressed in our participants' stories, a focus on prevention of undesired pregnancy alone is inadequate and unrealistic: along with contraceptive services, quality maternity care, safe abortion services, and post-abortion harm reduction interventions are urgently needed to reduce maternal morbidity and mortality and improve quality of life among adolescents living in Sub-Saharan Africa. --- Due to the small sample size and detailed personal narratives contained in interview and focus group transcripts, it is not possible to fully anonymize the data. While the ethics committees have not approved public release of the data, please contact the corresponding author and the Secretary,

Despite significant public health emphasis on unintended pregnancy prevention among adolescent girls and young women in Sub-Saharan Africa, there is a gap in understanding how adolescents' own reproductive priorities and the social influences on their decision-making align and compete. We examined the social context of contraceptive decision-making among Kenyan female adolescents.Using community-based sampling, we conducted 40 in-depth interviews and 6 focus group discussions among sexually-active or partnered adolescent girls and young women aged 15-19 in the Nyanza region of Kenya. We analyzed the data in Dedoose using an inductive, grounded theory approach, and developed a conceptual model from the data illustrating social influences on adolescent contraceptive decision-making.Participants viewed adolescent pregnancy as unacceptable, and described severe social, financial, and health consequences of unintended pregnancy, including abortion under unsafe conditions. Yet, their contraceptive behaviors often did not reflect their desire to delay pregnancy. Contraceptive decision-making was influenced by multiple social factors, centering on the intersecting stigmas of adolescent female sexuality, pregnancy, and contraceptive use, as well as unequal power in sexual relationships. To prioritize pregnancy prevention, adolescents must navigate conflicting social norms and power dynamics, and put their perceived future fertility at risk.

Introduction Occupations are an important outcome of previous life decisions and a determinant of life chances. Therefore, occupation is an important variable in social and economic research; it allows the analysis of labor market processes, social mobility, and status attainment, to mention just a few. Occupation is a complex construct that is difficult to measure and requires categorization. Most surveys aiming at collecting data on occupation include open-ended questions about the respondent's current or last job. Hence, responses are recorded verbatim. In order to be analyzed, these textual data on occupation must usually be coded, that is, unstandardized texts must be aggregated into pre-defined categorical systems, for example, the International Standard Classification of Occupations or the Standard Occupational Classification . Researchers often derive socioeconomic status scores, occupational prestige scores or social class positions , or occupational health hazard scales from these occupational codes. Because of the great importance of the occupation variable, it is essential to understand the coding process and to assess the quality of coding results. In the present article, we therefore address the following questions: What percentage of responses to openended questions about occupation are uncodeable? How reliable are occupational codings? With regard to the latter question we ask, in particular, to what extent coding results are affected by rules implemented by the coding agency . We explore how the application of different coding rules influences measures of socioeconomic status and occupational prestige. Finally, we analyze the relationship between the length of answers given by respondents and the reliability of coding across different agencies. In order to answer these questions, we use data from the German General Social Survey 2010 and the German field test of the Programme for the International Assessment of Adult Competencies carried out in 2010. Occupational information was coded into the 2008 version of ISCO by three agencies for ALLBUS and two agencies for PIAAC. This article is structured as follows: We begin by briefly reviewing relevant literature on the coding of occupations and discussing key challenges in this field. Based on this presentation, we develop hypotheses that will guide our empirical analysis. After a short description of the data and the coding procedures used in our study, we present our results. We conclude with a discussion of the main findings and implications of our study. --- Challenges of Occupation Coding and Previous Research Transferring textual information on occupations into numerical information is a multifaceted and demanding task. Errors can occur at several stages and impact the quality of coding. To date, both theoretical attempts to systematize and explain factors affecting the quality of occupational coding results and empirical research on this topic have been limited. Elias distinguished between 1) problems relating to the extent and quality of the data to be coded, 2) problems relating to the classification itself, and 3) problems relating to the formulation and application of coding rules and the coding process. In interviewer-administered surveys, interviewers must ask the questions and record the information provided by respondents. The first prerequisite to obtaining valid and reliable information on occupation is the use of adequate questions. The usual recommendation is to use at least two separate questions 2012, 55f.) to obtain both information about the job title and about the main tasks and duties performed in the job. Hoffmann et al. point out that the questions should be simple, containing familiar, widely understood terms. Furthermore, they underline the importance of specific design features. For example, including instructions and examples in the questions seems to have different effects depending on the education of respondents. They also show that the size of the text fields provided for the answers affects the answer length because it serves as a cue to the level of the detail expected . Hak and Bernts point out the fact that the interpretation of answers by the coders is a preliminary key step of the coding process. In particular, when respondents used vague or ambiguous terms or provided contradictory information, the assignment of a specific occupational code is mainly a matter of interpretation. One can try to improve the quality of the responses by appropriate interviewer training . Indeed, several authors have reported challenges encountered during the interview process that might result in problems when coding answers to the questions about occupation. For example, Schierholz et al. reported that respondents tend to provide incomplete or contradictory information, which influenced coding results. Similarly, Geis and Hoffmeyer-Zlotnik reported that typically 15% to 25% of answers cannot be completely coded to the most detailed level. Therefore, our first hypothesis is: H1: We expect the percentage of not completely codeable answers to be around 20%. The quality of coding is not only influenced by the question used or the interviewer training, but also depends on the precision, completeness, and clarity of the coding scheme. In this context, the descriptions of the categories play a crucial role in helping coders to find the correct codes. However, changes in occupational specialization reduce the applicability of coding schemes over time, and the typical lengthy updating phase of such classifications might be too long to capture ongoing changes in occupational structures. Coding schemes developed for cross-national comparison are even more challenging because they have to find the right balance between categories reflecting international comparative structures and national specificities. In this article, we cannot test the effect of the coding scheme itself on the quality of codings, as we use only one coding scheme in our empirical analysis. Basic criteria reflecting the quality of coding results are inter-coder reliability -that is, the extent to which the same code is assigned to a given text by different coders -and internal validity -that is, the extent to which the most appropriate code is selected. For our data, we do not have any codings that could serve as a "gold standard" to assess validity. Therefore, we will focus on reliability, keeping in mind that high reliability does not guarantee high validity, but certainly is a necessary requirement. Several measures of reliability are available, for example, simple percent agreement, Cohen's kappa, Scott's pi, and Krippendorff 's alpha . We will report Cohen's kappa -a coefficient measuring agreement adjusting for agreement occurring by chance. With an increasing number of categories, random agreement is negligible and Cohen's kappa is only marginally lower than percent agreement. One way to increase the reliability of occupational coding is to formulate and apply rules capturing the definitions and general guidelines of an occupational classification . Usually, the occupational classification manual includes some general rules. For example, in the publication International Standard Classification of Occupations: Structure, Group Definitions and Correspondence Tables, which is referred to in what follows as the "ISCO manual", ILO suggests that the following three rules should be applied when classifying jobs with a broad range of tasks and duties: If the tasks and duties require different skill levels, the job should be classified in accordance with the tasks and duties that require the highest level of skills. If the tasks and duties are connected with different stages of the production and distribution process, tasks and duties related to the production process should take priority. In cases with tasks and duties at the same skill level and at the same stage of production and distribution, the job should be classified according to the predominant, that is, most time-consuming, tasks performed. Detailed explanations of specific boundaries are included in the ISCO manual. Also, recommendations are provided regarding the use of job-related information other than job title and main tasks and duties actually performed. The developers of ISCO recommend that decisions should be based on the tasks actually performed, rather than on any other information. However, beyond general clarifications and rules, more specific operational rules are needed to determine the most appropriate occupational code for a given description of a particular job. When automatic dictionary-based coding is applied, the program that assigns codes to text can be considered part of these rules. In addition, to ensure consistency, previous coding decisions will usually be incorporated into agencies' coding rules. Finally, recurring incomplete or ambiguous answers require rules on how to deal with these cases. As an example, Ganzeboom and Treiman recommend providing coders with information on the numerical sizes of the specific occupational categories, and thus on the probability of category membership, in order to help them code ambiguous cases. Specifications on whether and how further job-related information are to be taken into account in case of doubt are also an important part of practical coding rules. This is especially relevant if important information about performed tasks is lacking in the input material. In the actual process of assigning codes to verbatim answers, the individual coders will -more or less systematically -use and "enrich" these rules in their ownmore or less idiosyncratic -way. This may be a source of bias, especially when the correlated coder variance is high because of high workload per coder . Belloni et al. studied coding errors in occupational data in the Netherlands using data from the Survey on Health, Ageing and Retirement in Europe . The authors recoded responses to open-ended questions about occupation for the Dutch sample of the SHARE data using software for semi-automatic coding: all cases above the certainty score threshold of 70 were coded automatically; all residual cases were coded manually by an expert coder. This coding was used as a benchmark and compared to the results of the coding according to the standard procedure implemented in SHARE . Inter-coder agreement at the one-digit ISCO level was 71% for current job and 72% for last job; at the three-digit ISCO level, it was 52% for current job and 56% for last job. Campanelli et al. reported results of two similar empirical studies, in which they compared results from different coders of the Office of National Statistics and applied manual, computer-assisted and computer-automated coding methods on data of the British Household Panel. Data were coded in accordance to the UK Standard Occupational Classification , a classification similar to ISCO. Reliability for manual codings was between 0.75 and 0.80. Validity, assessed by comparisons with expert coders, yielded agreement rates, ranging between .69 and .84. Only modest gains in reliability and validity could be obtained by using computer-assisted methods. Several studies have dealt with the question of how the quality of coding results is influenced by the rules applied during the coding process. For manual coding, Hak and Bernts argued that the effectiveness of training in terms of inter-coder reliability is improved, not only by "communicating coding instructions to coders ", but also by "socializing coders into practical rules" beyond the general coding instructions. Such practical coding rules are, at least partly, specific to the coding agency and, as such, reflect their experience and expertise. Thus, coding rules can have a negative effect, as they can lead to what Bushnell referred to in a different context as the "coding system bias." Depending on the coding agency and the rules in place, results of coding can differ systematically between agencies . The higher the differences between these agency-specific rules, the higher is the likelihood of house effects in occupational coding, that is, that the coding will differ systematically between agencies. Thus, although these house effects increase the reliability within one agency, they may reduce reliability between different agencies and jeopardize validity. Thus, our second hypothesis is: H2: The coding reliability between different coding agencies is lower than the coding reliability within agencies. In social research, occupational codes are often used to calculate occupational prestige and socioeconomic status scores . The agreement between prestige and SES scores derived from the codings of different agencies will be higher than the inter-agency agreement between the occupational codes. This is necessarily true because merging a large number of codes into a smaller number of categories results in higher agreements per se. Additionally, we argue that in the cases where the choice between different codes is hard to make, the occupations in question are more similar with respect to prestige and SES than a randomly chosen pair of occupations would be. In a study by Maaz et al. , occupational information was coded by professional coders and lay coders . The level of consistency between these codings was generally not as high as desirable. For example, for the ten onedigit codes of the International Standard Classification of Occupations of 1988 , the authors reported a Cohen's kappa of around 0.67. For the 390 more detailed fourdigit ISCO codes, k did not exceed 0.5. When occupational codes were converted into International Socio-Economic Index scores, correlations were generally higher . Therefore, our third hypothesis is: H3: Socioeconomic status and occupational prestige scores will show high levels of agreement. Finally, we are concerned with the relation of answer length and codeability. Previous research suggests that longer text strings do not necessarily result in more reliable codings. Conrad et al. found that longer descriptions of occupations were less reliably coded than shorter descriptions. For coders, texts that provide too much detail or are too complex can be difficult to interpret within the framework of an occupational classification and can make it difficult to decide on a unique occupational code. The Conrad et al. study corroborates Bergmann and Joye that ": : : the more detailed the information to be sorted into occupational groupings, the less reliably individual cases are assigned to categories." It is also in line with the findings of Cantor and Esposito . They found that coders who were asked to comment on interviewer recordings of occupations only rarely indicated that these should contain more specific information; some even criticized the fact that interviewers had provided too much information. Thus, our fourth hypothesis is: H4: Long answers to open-ended questions about occupation do not result in higher coding reliability. Before we present analyses on our hypotheses, we briefly discuss the data and approach we use in our empirical study. --- Data, Classification, and Coding Procedure --- Data To test our hypotheses, we use data from the German General Social Survey 2010 and the German field test of the Programme for the International Assessment of Adult Competencies conducted in 2010. Both ALLBUS and PIAAC were carried out as computer-assisted face-to-face interviews with randomly selected respondents from official population registers. We restricted the samples to respondents in gainful employment aged between 18 and 65 years. In our analysis, we considered the respondents' current or last occupations and their parents' occupations. We did not differentiate between these different types of occupations because coding reliability showed no substantial differences. Respondents' job titles and activities were measured with a two-part open-ended question in ALLBUS and a three-part open-ended question in the PIAAC field test . These questions were embedded in different background questionnaires that also A similar instruction was provided in ALLBUS in the general interviewer training. However, this instruction did not appear on the screen and was therefore not read out to the respondents. included other occupation-related questions. The ALLBUS respondents were first asked about their status in employment using a very detailed classification that distinguished white-collar workers according to their tasks, civil servants according to their career paths, blue-collar workers according to their qualifications, employers by number of employees, and farmers by the size of their utilized agricultural area . This was followed by the open-ended questions about occupation presented in Table 1. The first sub-question of ALLBUS asked for a description of work done in the job. Strictly speaking, only the second sub-question referred to the key information "job title", although in practice, many respondents reacted to the first sub-question by naming the title of their job. After these questions, several other occupation-related questions were asked. The original question wording in German can be found in the Supplemental data . The question about occupation in PIAAC was similar to, but more detailed than that in ALLBUS . There were three sub-questions about occupation, starting with a sub-question explicitly asking for the exact job title. In contrast to ALLBUS, all subquestions required an answer . Also in PIAAC, these questions were followed by several occupation-related questions . In both surveys, respondents who were currently not employed, but who had worked in the past, were asked about their last job. Furthermore, respondents were asked about their mothers' and fathers' occupations when the respondents were 15 or 16 years old. The wording was similar to that of the questions about the respondent's occupation, however only limited ancillary information was provided to the coders about the parents . To carry out the survey, the PIAAC interviewers were trained very thoroughly. Their training included instruction on how to retrieve and record information on occupation; they were instructed to ask the respondents in a way that the interviewers actually understood what kind of job the respondents were doing. Furthermore, they also had to practice how to ask the occupation questions. In ALLBUS there was no such training. However, the interviewer instructions for the study included explanations and examples to illustrate the level of detail necessary with respect to the open-ended questions on occupation. --- The ISCO-08 Classification Scheme ISCO-08 is the current version of the International Standard Classification of Occupations curated by the International Labour Organization, a specialized agency of the United Nations. It "provides a system for classifying and aggregating occupational information : : : [and] allows all jobs in the world to be classified into 436 unit groups" . For the purposes of ISCO-08, a job is defined as "a set of tasks and duties performed, or meant to be performed, by one person" . Jobs with very similar main tasks and duties are aggregated into "occupations". Different occupations form the most detailed level of ISCO, unit groups. Besides job, the second main concept underlying ISCO classification is skill, which is defined as "the ability to carry out the tasks and duties of a given job" . More specifically, in ISCO, occupations are categorized according to the typically required skill level and skill specialization. Skill level is differentiated into four ordered groups from low to high educational qualifications and accompanying levels of literacy and numeracy. As can be seen in Table 2, Major Group 9: Elementary Occupations is characterized by Skill Level 1; Major Group 3: Technicians and Associate Professionals by Skill Level 3; Major Group 1: Managers by Skill Levels 3 and 4; Major Group 2: Professionals by Skill Level 4; and all other major groups by Skill Level 2. Major Group 0: Armed Forces Occupations is an exception, in that it can contain a broad range of skill levels. Within the major groups, occupations are grouped by the type of skill specialization, that is, "the field of knowledge required; the tools and machinery used; the materials worked on or with; and the kinds of goods and services produced" . It is important to note that the concept of skill level and skill specialization refers to the requirements of jobs and occupations, rather than the skills or education of a specific job incumbent. Overall, ISCO follows a hierarchical structure where an increasing level of detail is expressed as one-to four-digit codes. As mentioned above, unit groups, which are denoted by four-digit codes, constitute the most detailed level of the classification. These unit groups are aggregated into 130 minor groups, expressed as three-digit codes, which in turn are grouped into 43 sub-major groups, denoted by two-digit codes. At the highest level, the classification comprises 10 major groups . The structure of the classification is illustrated in Table 3. For example, Unit Group 3112: Civil Engineering Technicians is part of Minor Group 311: Physical and Engineering Science Technicians, which in turn is part of Sub-major Group 31: Science and Engineering Associate Professionals, which belongs to Major Group 3: Technicians and Associate Professionals. --- The Coding Process For our study, a number of agencies were contracted to independently code all the occupational data. These agencies are experienced in occupational coding and offer this service commercially; all agencies offer their service roughly in the same price range. All agencies trained their coders according to their own procedures. Most of the coders were experienced in coding occupations, however, we do not have detailed knowledge about their background. The agencies were provided with respondents' answers to open-ended questions about occupation, which had been recorded verbatim by the interviewers. Furthermore, ancillary variables were provided for use by coders during the coding process, including age, gender, formal education, status in employment, public sector employment, supervisory tasks, and industry. Three agencies coded the responses to the ALLBUS open-ended questions about occupation; two agencies coded the PIAAC data. Whereas Agency A was the same agency in both cases, Agency B was not. Agency B refers, in fact, to two agencies: The first agency coded the PIAAC occupation data; its successor coded the ALLBUS data. As the first agency handed over all relevant material to the second agency, and the second agency also adopted the first agency's coding procedures, they are treated as one unit here. The agencies were not given any further instructions on how to code. Rather, they could apply their typical coding procedures. They could use either semi-automatic or manual coding, and they were allowed to use any coding tools they had available. Furthermore, the agencies themselves decided on the coding strategy to be adopted, that is, 1) the number of coders; 2) the additional material provided ; and 3) the organization of the coding process . For the coding process, Agency A used a software program that provided the text of the responses. There was no automatic coding involved in the process. However, the coding software offered coding suggestions via templates. These suggestions are based on the ISCO group labels . The main objective of employing this software was to reduce the complexity of the coding process for the coders by using a hierarchical coding strategy, providing the same information to all coders. The coders were not given the ISCO manual as a reference. The coding instructions, ancillary information, conventions, and known problems with suggested solutions were part of an interface in the software. Agency B used, as a first step, automatic coding based on an extensive dictionary. About 45 to 55% of the answers can usually be coded this way . Responses that were not automatically codeable were listed alphabetically, and alphabetically ordered blocks were randomly assigned to coders. In the case of the coding of the PIAAC data by Agency B, an expert coder reviewed the codes assigned, and if systematic errors were observed, coders were given additional training. Several rules were specified for the coders, for example: 1) If two different codes from different major groups are plausible, assign the occupation with a lesser degree of professionalization. 2) If two occupations were mentioned by a respondent, code the first-mentioned occupation. 3) If two different job titles were mentioned, code the more concrete title. The first rule seems to contradict the ILO rule cited earlier, whereby, in the case of several plausible alternatives, the occupation with the highest skill level should be coded. However, we are not sure how "professionalization" was operationalized. To our knowledge, this agency used auxiliary information on employment status in ambiguous cases. We do not know how the two other agencies handled this status variable during coding. The coding process at Agency C involved two steps: first, occupations were coded according to the five-digit German Classification of Occupations 2010 . This was initially done automatically, using a dictionary with around 100,000 entries. This coding process was supported by taking into account the ancillary information provided . The remaining responses were then manually coded. Several rules were applied, for example: 1) If two occupations were reported, code the first-mentioned occupation. 2) If the open-ended response is inadequate to determine the occupational code, use the ancillary information. 3) If two different codes are equally plausible, assign the code that occurs more frequently in practice. As a second step, the codes of the German Classification of Occupations 2010 were automatically mapped to ISCO-08 codes using correspondence tables provided by the German Employment Agency . To control the coding quality within the agencies, and to determine whether low coding reliabilities were due to house effects, we asked the two agencies coding the PIAAC data to code a subsample of answers a second time, assigning them to different coders. --- Results In the following sections, we describe our results in the order of our four hypotheses. First, we present results on uncodeable or not completely codeable answers, followed by the description of inter-agency and inter-coder reliability and coding differences. After that, we present consequences for socioeconomic status and prestige scores and finally, we report differences in answer length and reliability. --- Uncodeable or Not Completely Codeable Answers The number of uncodeable answers was low; to a large extent answers could be coded at the four-digit level . As can be seen in Table 4, the percentages of answers that could not be assigned to at least a major group by the three agencies that coded the ALLBUS data were 3.4%, 3.0%, and 2.2% respectively. For PIAAC, the respective percentages were 4.9% and 1.8%. These findings are in line with the findings of Hoffmeyer-Zlotnik et al. , who reported 3% of not codeable answers in a survey of the German Environment Agency in 1999. Looking closer at the answers that could not be coded, we found large differences between the agencies: For ALLBUS, 274 answers were classified as uncodeable by one of the agencies, 51 answers were classified as uncodeable by two agencies, but only 28 answers were classified as uncodeable by all three agencies. A clearer pattern could be observed for PIAAC, with Agency A showing a greater tendency to classify a given answer as uncodeable. Agency A could not code 134 of the cases that Agency B coded, whereas Agency B was unable to code only eight cases that Agency A coded. If we look at answers that were coded, but not to the most detailed fourdigit level, we find that these sum up to 16% and 19% for ALLBUS and 11% and 9% for PIAAC. These percentages were, thus, lower than we expected . Agencies A and B seem to follow the recommendation of the ILO to code vague responses to the most detailed level still supported by the information provided. In addition, we found surprisingly large differences with respect to which answers were classified as uncodeable. Here, the different rules of the agencies become obvious. To give an example, one agency classified the response "housewife and typist" as uncodeable , whereas the others assigned the code for typist . Agency C is a special case in this regard: because they first coded the answers into the Germany Classification of Occupations and then recoded these codes to ISCO-08 using the official cross-walk, they used the most probable ISCO unit group when answers were not completely codeable or when the KldB code was not unambiguously transferable. --- Inter-Agency and Inter-Coder Reliability and Coding Differences For the analysis of the reliability we included all codings of the three agencies. If an agency marked an answer as uncodeable or not codeable on the four-digit level, whereas the other one coded the same answer with a four-digit code, this could be handled as a mismatch in the sense of coding reliability. As can be seen in Table 5, coding agreement between the different agencies was far from perfect. We only report k, because checking the proportion of agreement showed that the differences between these measures are negligible. For ALLBUS, in a comparison of Agency A and Agency B, the inter-agency reliability of assigning an open-ended answer to the same major group was 0.72. With k ¼ 0.68, the inter-agency coding reliabilities for major groups between Agency A and Agency C and between Agency B and Agency C were even lower. The results are similar for PIAAC: The inter-agency reliability between Agency A and Agency B was around 0.76 for the major groups. As there are only ten major groups, these inter-agency reliabilities are not satisfactory. Because of the hierarchical structure of ISCO, once the first digit differs, all more detailed codes differ. Thus, reliabilities for sub-major, minor, and unit groups were even lower. In the end, the assigned four-digit ISCO codes differed in nearly half of the cases for ALLBUS , and inter-agency agreement was only slightly higher for PIAAC . This low level of reliability casts serious doubt on the usability of these data for further analysis. As Table 5 shows, the largest information loss occurred at the level of the major group. Taking the ALLBUS codings as an example, a glance at the distribution of the major groups reveals systematic differences resulting from the codings of the three agencies involved. In particular, frequencies of Major Groups 3, 5, 8, and 9 differ considerably. The difference is most extreme in Major Group 9, where the number of elementary occupations coded by Agency C was 1.5 times higher than that coded by Agency A. In what follows, we briefly explore some of the systematic discrepancies we observed in the assignment of the first digit by the different agencies . Because it is impossible to list all the differences we found, we concentrate on those differences observed most often and on those that have a large effect on the prestige and SES scores derived from the codes. We identified six such systematic differences: The first difference related to the handling of self-employed respondents . For ALLBUS and PIAAC, Agency B coded a large group of these respondents into Major Group 1: Managers, whereas the other agencies often assigned them to Major Group 5: Services and Sales Workers. To give some examples: if a respondent reported that he or she was a self-employed hairdresser, Agency B classified the occupation into Unit Group 1120: Managing Directors and Chief Executives, whereas the other agencies assigned it to Unit Group 5141: Hairdressers . A similar problem was caused by the answer "innkeeper". Agency B classified it into Unit Group 1412: Restaurant Managers; the other agencies assigned it to Sub-major Group 51: Personal Service Workers. ISCO defines "Operators of small cafe ´s, restaurants and bars to whom the management and supervision of staff is not a significant component of the work are classified in Unit Goup 5120 : : : " . The rules are clearly defined but the answer of the respondents are often not detailed enough to decide what the main task is. A second frequently occurring difference between the codings for ALLBUS is related to the classification of manual workers. Depending on the agency, these workers were coded either into Major Group 6: Skilled Agricultural, Forestry and Fishery Workers; Major Group 7: Craft and Related Trades Workers; Major Group 8: Plant and Machine Operators and Assemblers; or Major Group 9: Elementary Occupations. Agency C assigned bluecollar workers much more often to Major Group 9, whereas, whenever possible, the other two agencies assigned them to a major group other than Major Group 9 . Differences arose because many of the responses were not unambiguously codeable. Major Group 6, 7, or 8 would be appropriate if more complex tasks were performed, whereas Major Group 9 should be reserved for those occupations involving only simple and routine tasks. The third typical systematic coding difference which occurred both in the ALLBUS and the PIAAC codings related to the assignment of responses to Major Groups 3 and 4. Major Group 3 comprises technicians and associate professionals who "perform mostly technical and related tasks connected with research and the application of scientific or artistic concepts and operational methods, and government or business regulations" . Major Group 4 comprises clerical support workers who "record, organize, store, compute and retrieve information, and perform a number of clerical duties in connection with money-handling operations, travel arrangements, requests for information, and appointments" . Whereas occupations in Major Group 3 typically require completion of upper secondary education and possibly a higher education degree , most occupations in Major Group 4 require only completion of lower secondary education and possibly vocational training . Frequent and typical answers that caused the coding problem in question were "secretary", "clerk" , and "accountant" . To give an example, secretaries may, on the one hand, be what are referred to in ISCO-08 as "administrative and specialized secretaries", who "provide organizational, communication and documentation support services, utilizing specialized knowledge of business activity of the organization in which they are employed" . In this case, they should be classified into Major Group 3. On the other hand, secretaries may be "secretaries ", who "transcribe correspondence and other documents, check and format documents prepared by other staff, deal with incoming and outgoing mail" and so on, in which case they should be classified into Major Group 4 . Without more detailed information about the respondents' tasks and duties, the occupations cannot be coded with certainty. However, the coding of these answers is of great relevance because of their high frequency and the low inter-agency agreement in relation to these two major groups . The fourth systematic coding difference occurred -again for both surveys -when assigning responses to Major Group 7: Craft and Related Trades Workers, and Major Group 8: Plant and Machine Operators and Assemblers. The main difference between these two major groups is the operation or use of machines at work. Although in the modern industrial world, most crafts require the use of machines, it is not clear from the respondents' answers whether or not the work was done predominantly with or without machines. Thus, the occupation "metal worker" may be classified into Sub-major Group 72: Metal, Machinery and Related Trades Workers or into Minor Group 812: Metal Processing Plant Operators, depending on the stage of development of the industry in the national context. The fifth systematic coding difference occurred when coding occupations such as educators working in early childhood education and care or youth welfare services. Agency A assigned these occupations mostly to Major Group 5: Services and Sales Workers, Unit Group 5311: Child Care Workers; Agency B assigned them to Major Group 2: Professionals, Unit Group 2342: Early Childhood Educators; and Agency C assigned them to Major Group 3: Technicians and Associate Professionals, Unit Group 3412: Social Work Associate Professionals. At least part of the confusion probably stemmed from the difference between the educational requirements for these occupations in Germany and elsewhere. Whereas in Germany these occupations usually require qualifications corresponding to Skill Level 3, most other countries appear to require Skill Level 4. The ISCO manual is quite clear on this point, stating that "occupations that require the performance of similar tasks should be classified in the same group" and that primary and pre-primary teachers "should all be classified in Major Group 2" . However, this innovation introduced for ISCO-08 does not seem to have been acknowledged by all coding agencies. The sixth coding problem was caused by a special national situation in Germany, where -as in some other countries -one type of vocational education and training is "dual vocational training", which comprises theoretical and practical elements. Most careers in crafts start with such a training program. After successful completion of this program, further training is possible in order to obtain the qualification "Meister" . This type of career is very common in Germany. When asked what their occupation was, many respondents answered that they were, for example, a master carpenter or a master electrician. Although no additional information was available on the tasks performed by these master craftspersons, Agency C coded such answers into Unit Group 3122: Manufacturing Supervisors, whereas the other two assigned a code from Major Group 7: Craft and Related Trades Workers that described the craft the respondent was working in. However, the required skill level in Major Group 7 is lower than that in Major Group 3. The reason for the different handling of this type of occupation may result from the coding in KldB and transferring the codes to ISCO: in KldB 2010 all job titles containing the word "Meister" are coded as supervisory tasks, which results in the 3122-coding. Once again, the difference in coding shows that agencies create specific rules to handle national specificities that are not reflected in ISCO. In this case, Agency C applied a standardized rule using the public and official German classification as their standard. As there is no concerted effort to standardize these rules, agency-specific approaches result in different codings and lower inter-agency reliability. The reliabilities reported in Table 5 reflect not only differences between the agencies' coding rules, but also differences between coders. To disentangle house effects from coder effects, we asked the two agencies that coded the PIAAC data to code a subsample of the occupational information twice, assigning the answers to different coders. We interpret the difference between the inter-coder and inter-agency reliabilities as a house effect that reflects differences in the coding procedures and rules. Inter-coder reliability of the four-digit codings of our PIAAC subsample was 0.84 between the two coders deployed by Agency B, and between 0.68 and 0.74 at Agency A, where coding was carried out by five coders. As these reliabilities were considerably higher than those between the agencies , this result supports our assumption that the coding procedures and rules applied by a coding agency affect ISCO coding to a considerable extent. This is in line with hypothesis H2. --- Consequences for Socioeconomic Status and Prestige Scores As mentioned previously, occupational prestige or SES scores are often derived from ISCO codes. Therefore, we analyzed the extent to which the reliability of these scores was affected by the reliability of the ISCO codes. Our analysis focused on the International Socio-Economic Index and the Standard International Occupational Prestige Scale , both of which were developed for cross-national research . Using the syntax proposed by Ganzeboom and Treiman , we calculated these scores separately for each of the coding results of the agencies. In less than 5% of the coded occupations, it was not possible to assign an SES or occupational prestige score. Although the inter-agency reliability of the four-digit ISCO codes was only around 0.5, the consistency between the derived socioeconomic status and prestige scores was much higher than expected. For ISEI, the correlation between each pair of agencies was 0.90, which is higher than the correlations reported in the study by Maaz et al. . For ALLBUS, the correlation of SIOPS between Agency A and Agency B was 0.84, between Agency B and Agency C it was 0.82, and between Agency A and Agency C it was 0.85. For PIAAC, the correlation between Agency A and Agency B was 0.84. The high correlations of both SES and prestige scores confirm hypothesis H3. --- Differences in Answer Length and Reliability In a final step, we analyzed the length of the answers given to the question about occupation in the two surveys. Because of the differences on how the questions on occupations were phrased in PIAAC and ALLBUS , we assume that PIAAC answers were more detailed, that is, longer. We measured answer length by counting the total number of words after combining all sub-questions. One drawback of this method is that we were not able to identify word repetitions. Thus, if respondents repeated words, these were counted each time. Our analysis showed that the questions in PIAAC were answered, in fact, in more detail, leading to answers that were about twice as long as the answers in ALLBUS. The average number of words recorded by interviewers was 2.70 for ALLBUS, with a minimum of one and a maximum length of 23 words. For PIAAC, the average number of words recorded was 5.52 , with a minimum of one and a maximum of 48 words. Furthermore, when systematically checking very common answers, such as "secretary" and "clerk", it became obvious that in PIAAC, as compared to ALLBUS, most respondents had given more details than just a general job title. For example, PIAAC respondents more frequently added information such as "personal assistant to the CEO" to the general job title "secretary", or "office clerk in accounting" to the job title "clerk". However, the crucial question is whether longer answers result in better -that is, more reliable -codings. To investigate this question, we calculated reliability coefficients based on length of answer . We found that the inter-agency reliability did indeed vary with the length of the answer. However, contrary to what one might intuitively expect, reliabilities decreased as answers became longer, thus confirming hypothesis H4. This is in line with the finding by Conrad et al. . A closer inspection of long answers did not yield any specific clues as to the reasons why they appear to be more difficult to code. We suspect that fewer words might result in incomplete occupational descriptions but that, for frequently used occupational titles, coders have internal coding rules for handling these answers. If the answers are longer, coders have more scope for decision-making, and simple rules can no longer be applied. Less consistent coding of longer answers may also be due to the fact that coders are reluctant to read longer texts and therefore base their coding on only parts of the answer. Although this result is true for both ALLBUS and PIAAC we found, in general, that occupations in PIAAC were more reliably coded than in ALLBUS. For example, the reliability between Agency A and Agency B for one-word ALLBUS answers was 0.54, whereas it was 0.63 for one-word PIAAC answers . Interviewer training and a more detailed question seem to have led to longer, more precise answers. Finally, if longer answers are more difficult to code, it could be assumed that they are also more often uncodeable or not completely codeable. However, our findings do not support this assumption. Uncodeable or incompletely codeable answers were, on average, shorter than codeable answers: the means for completely codeable answers were 2.76, 2.71, and 2.71, while the means for not completely codeable answers were 2.26, 2.39, and 2.30 for Agencies A, B, and C respectively. --- Conclusion To better understand how the process of occupational coding affects outcome quality, we studied occupational data from two surveys, each coded by several agencies. We found that the share of uncodeable answers was comparatively small, ranging from 1.8% to 4.9%, depending on the coding agency and data source. However, a closer look revealed that what was deemed "uncodeable" varied between coding agencies. In a next step, we examined the coding results more closely and discovered characteristic, systematic differences in the way agencies interpreted ISCO and defined relationships between answers to be coded and ISCO categories. Notwithstanding the comparatively low agreement between occupational codings from the different agencies, the correlations for SES and occupational prestige scales were satisfactory . Finally, we observe a negative correlation between length of answers and coding reliability both in ALLBUS and PIAAC. At the same time, coding reliability between the two agencies that coded data from both surveys was higher for PIAAC than for ALLBUS. This difference remained when we controlled for length of answer. In line with Elias we discuss our findings in relation to the following aspects: the data to be coded, the coding rules and coding process, and the classification. With regard to the data to be coded the seemingly contradictory results concerning answer length are the most interesting. Several factors may have contributed to these findings. First, the intensive interviewer training in PIAAC may not only have increased answer length -with a general negative effect on reliability -but might also have led to "better" answers because interviewers were thoroughly trained on what kind of information was useful for occupational coding. Especially with respect to some frequently reported occupations that are difficult to code without additional information the interviewer training may have had, on balance, a positive effect on reliability. The higher percentage of four-digit codes for PIAAC data also points to the effect of interviewer training. The second factor that may have contributed to the higher inter-agency reliability of PIAAC codings are differences in the way occupational information was collected. In particular, the first sub-question in PIAAC explicitly asking for the exact job title, may have elicited answers that were more suitable for ISCO coding. In addition, the detailed question about employment status that preceded the open-ended question about occupation in ALLBUS may have played a crucial role. This may have affected the answers given to the open-ended questions because respondents had just given some of this information in the preceding question on their employment status. Besides effects caused by the resulting differences in the core input material for occupational coding, it seems reasonable to assume an interaction effect of data provided and coding rules. The use of this ancillary information for occupation coding seems to have varied in extent or manner between the agencies involved. This could be an important factor that explains the lower inter-agency reliabilities observed for ALLBUS compared to PIAAC, where the question on employment status only differentiated between employed and self-employed. Taking all these factors together, our results concerning the ALLBUS-PIAAC differences suggest that some of the inter-agency coding disagreements result from various differences in the survey instrument and the interviewer training, and the ensuing differences in the occupational data to be coded -partly in interaction with coding rules applied concerning this data. Therefore, identifying and understanding those characteristics of question wording and interviewer training that lead to the most suitable input material for occupational coding would be a promising field for further research. Turning to the coding rules and coding process, we argue that the systematic deviations that we observed between codings from different agencies indicate that the rules and procedures laid down in the ISCO manual do not suffice to completely cover the coding process. This results in low inter-agency reliability and low validity of the results. From a quality perspective, reliability is important, but validity of codes is essential. Hak and Bernts pointed out that the validity of the coding process depends on the quality of coding instructions, whereas inter-coder reliability depends on the implementation of these instructions. This implies that insufficient reliability and validity of occupational codes have common sources. To elicit both reliable and valid information, interviewers should be trained to filter as little information as possible and to ask for an occupational title. Moreover, they should be familiarized with basic coding procedures . To improve coding quality, agencies engaged in occupational coding create rules of their own that reflect their interpretation of ISCO, and these house rules lead to the large differences in coding decisions that we found. The shortcomings of the ISCO manual and the secondary, agency-specific coding rules are a serious threat to the validity of occupational codes. We assume that this threat is even larger in the cross-national context where these agency-specific rules most likely reflect particularities of national labor markets. To overcome this situation, we suggest that coding agencies state the rules they apply as clearly as possible and make them public. A publicly accessible body of rules would allow more systematic discussion and development by those engaged in occupational coding. For the time being, survey practitioners commissioning coding services should ask agencies about the coding rules they apply and, where necessary, negotiate such rules. One may think that the problems resulting from differences in rules may be overcome by applying completely automatic coding routines. Indeed, such methods have been proposed, for example, based on statistical learning . However, up to now these approaches are not able to code considerably more than half of the occupations, thus limiting their usability. With respect to the classification, some problems related to coding ISCO originate from the national context in which it is applied, for example, "Meister" in Germany. It can be expected that similar problems arise in other national contexts, as ISCO can be seen as a compromise between different national views and peculiarities . We have also seen that agencies make fundamentally different decisions when coding occupations. We were particularly surprised to see that there are considerable differences in the use of major groups, pointing to "boundary problems". A further improvement of the ISCO manual, by including even more precise definitions and more detailed explanations, could certainly be helpful in dealing with problematic distinctions. Given the multitude of jobs actually performed in the real world, rules and procedures laid down in a manual will never suffice to completely cover the coding process. Approaches supporting the coding of ISCO directly in the field may help to avoid some of the problems. The idea behind in-field coding is that respondents have better and more detailed knowledge of the type of work they carry out. Based on additional variables, such as employment status, public sector employment, or supervisory tasks, they could be asked for more details depending on their answer to the first question about occupation. Initial results of such approaches are promising. When discussing developments in the way occupational data is assessed and coded, it is also important to look into the effect that new approaches have on the survey process. In-field coding would introduce additional burden on respondents and interviewers, and increase survey costs. Nevertheless, improving results of occupational coding should be a central aim of efforts to improve the overall quality of surveys.

As occupational data play a crucial part in many social and economic analyses, information on the reliability of these data and, in particular on the role of coding agencies, is important. Based on our review of previous research, we develop four hypotheses, which we test using occupation-coded data from the German General Social Survey and the field test data from the German Programme for the International Assessment of Adult Competencies. Because the same data were coded by several agencies, their coding results could be directly compared. As the surveys used different instruments, and interviewer training differed, the effects of these factors could also be evaluated. Our main findings are: the percentage of uncodeable responses is low (1.8-4.9%) but what is classified as "uncodeable" varies between coding agencies. Inter-agency coding reliability is relatively low k ca. 0.5 at four-digit level, and codings sometimes differ systematically between agencies. The reliability of derived status scores is satisfactory (0.82-0.90). The previously reported negative relationship between answer length and coding reliability could be replicated and effects of interviewer training demonstrated. Finally, we discuss the importance of establishing common coding rules and present recommendations to overcome some of the problems in occupation coding.

Plain English summary Lived Experience refers to what people know from their experience. It goes beyond formal education or professional experience to provide unique insights that can improve public health research and service provision. This paper discusses the role of LE in a UK city-region government's gambling harms reduction intervention. Internal and external stakeholders, including people with and without LE, reflected on the value of LE and challenges to its effective involvement in interviews and focus groups. The evaluation team summarised the data collected into four themes: personal journeys to LE involvement ; the value added by LE to interventions for addressing gambling-related harms ; emotional impacts on people with LE ; and collective LE and diverse lived experiences . The paper concludes that LE should be understood as many diverse lived experiences. Resources to support people to manage the emotional tensions raised by LE activities are needed, alongside routes to LE involvement that facilitate self-development and improve representation in positions of authority and decision-making. --- Background --- Lived experience in public health Lived Experience , or "knowing from experience", refers to people's direct or indirect experiential knowledge of an issue or service [1]. LE involvement in mental health services and research is longstanding [2,3] and recognised as important in public health [4][5][6][7], where it echoes established approaches to public involvement and asset-based community development [8]. The emphasis on knowledge in LE distinguishes it from Patient and Public Involvement and Engagement , although PPIE can be an engagement route for involving 'experts by Lived Experience' , or 'people with Lived Experience' . The developing research base on LE involvement in public health intervention design and delivery highlights its importance for ensuring relevance and practical impact [9,10], and for reducing the risk that an intervention may unintentionally exacerbate harms [1,9]. Gambling-related harms are complex and span a continuum [11]. They include harms to relationships, resources, and health [12], such as intimate partner violence [13], crime [14], costs to the economy [15], and suicide [16], with wide-ranging consequences for individuals, families, communities and society [17]. Research is increasingly revealing the extent of commercially-driven normalisation of gambling and its harmful products, as well as the tactics of the gambling industry, which echo other harmful industries, in circumventing regulation and shaping associated discourses [18]. LE involvement in GRH research and practice is growing, with further research required to more fully understand its contribution [12,19]. This paper explores the value of LE as part of a regional public health intervention for addressing GRH that incorporated multiple opportunities for input by PLE. It provides insights into different facets of LE involvement in the gambling harms reduction sector and examines the contributions of LE, as well as the tensions that can arise in practice. --- Lived experience in the gambling sector Gambling has been discussed as a public health issue for more than twenty years [12,20,21], yet there is no national public health framework for GRH reduction in the UK, and a comprehensive, socio-ecological public health perspective on gambling remains nascent [22][23][24]. The development of a robust public health approach is further complicated by the complexity of the UK policy environment, in which the influence of the gambling industry, rather than statutory public services, is wellentrenched in guiding health promotion, prevention and treatment activities [24]. In the absence of UK government leadership, GRH continues to be a 'niche' issue, with local initiatives sometimes implemented without the resources to ensure sustainability [25]. Engaging PLE in facilitating recognition and awareness of GRH is an emerging area of study [12,19,26,27] and in the meantime, GRH treatment and education are funded and, in some cases, directly provided by the gambling industry, rather than statutory public health bodies [18,28]. The development of LE involvement in the gambling sector is distinct because of this policy context, which is also important for understanding the barriers that PLE face when they get involved in the sector. LE has much to offer public health policy analysis and development [29], but LE-led advocacy work on gambling reform contends with outdated UK legislation around a public health response to GRH. LE-led grassroots campaigns in the UK and further afield [24,33] have sought to challenge the influence of the gambling industry over education and harms messaging framed as "responsible" or "safer" gambling [30] that push the responsibility for GRH onto individuals [31,32]. The delayed UK Government review of the Gambling Act 2005 -to render gambling regulation fit-for-purpose in today's digital world [33] -has so far been limited to a white paper [34] and various consultations, but this does recognise a need for public health-framed messaging and an end to industry-funded campaigns. Regulatory developments and product innovation and diversification in the industry [21] mean it is timely to advance applications of LE within a public health approach specific to gambling [11,35], and in the context of public health discourse around commercial determinants of health [36]. Research suggests that strategies to address GRH are more effective when they incorporating LE-based perspectives [12]. In particular, PLE can contribute unique knowledge based on their experiences of how the gambling industry, its products, practices, and discourse, harms people [37]. LE involvement has already sought to influence language and inform service development, as in the significant pushback from PLE in response to the dominant framings of the industry based on the way in which such narratives shift the focus away from the problematic nature of addictive gambling products and onto individuals [37]. However, the lack of evidence-based frameworks for appropriately engaging PLE in GRH reduction activities, which can be attributed to the sensitivity and stigma surrounding gambling, has only recently begun to be addressed [38]. As online gambling continues to be a target area for growth and precision marketing, it is imperative that public health researchers and practitioners engage with communities to raise awareness of GRH [1] and develop their own and others' critical health literacy [39] specific to gambling. --- Communities addressing gambling harms: a public health intervention for gambling-related harms reduction The Communities Addressing Gambling Harms intervention was delivered across a city-region in the UK through 12 diverse community projects and a LE advisory panel that advised the community projects. It aimed to facilitate regional understanding of and action on GRH, and presented several opportunities for LE involvement: from advising the city-region government and the community projects on GRH reduction interventions, through to more active involvement in the projects . A process evaluation of the CAGH intervention was approved by university ethics no. [anonymised] and the main evaluation is described in detail elsewhere [40]. The current paper broadens its focus beyond the CAGH intervention to investigate LE involvement in the GRH reduction sector more generally, in recognition that LE in CAGH is influenced by experiences external to it. This was necessary to fully understand LE in CAGH and reflected the complex reality of LE involvement on-theground, with PLE in CAGH often engaging in gambling harms reduction activities outside of the intervention. A PPIE panel of public representatives contributed to the evaluation at every stage, including providing insight into local developments; advising on the data collection tools; assisting with recruitment; and theorising emergent data. The PPIE panel was recruited via a public call for expressions of interest. Two PPIE panel members are co-authors of this paper. --- Methods --- Sample Three groups of people who were involved in the intervention were sampled to inform the evaluation: Sampling was based on individuals' potential 'information power' , or diversity of experience [41,42]. The 'without LE' status of project staff and senior stakeholders is qualified by 'declared' above, in recognition that while these groups contained no people who self-reported LE specific to GRH, distinctions between PLE and people without LE are not clear-cut. Further details of the sample are reported below: 1. People with LE : People with LE of their own or someone else's gambling, i.e., as an Affected Other . Ensuring the terms used to describe the expertise and experiences of PLE reflect their preferences is important [10]. PLE does not define people by their LE, but recognises LE as an ongoing part of people's expertise and daily life [26]. Here, LE relevant to GRH is defined as 'Having personally experienced the suffering and destructive consequences of gambling related to own or someone else's gambling' [26]. This definition explicitly includes AOs -people in an individual's social network who are affected by the individual's LE or LE testimony [11,35,43]. Compared to PS and SS, PLE held complex and varied roles in the intervention and contributed to the evaluation in multiple ways. The involvement of PLE is highlighted and summarised in Table 1. PLE --- Data collection Two methods of data collection were used: in-depth, semi-structured interviews and focus groups . Both were selected because of their potential to elicit rich storytelling from participants, and in recognition of the role of storytelling within LE activities as an accessible engagement strategy that can facilitate the sharing of testimonies in ways that resonate with listeners by building empathy and challenging preconceived narratives [10,26]. Interviews have previously been used to understand GRH [44,45]. The combination of the interviews and FGs was complementary and enabled in-depth investigation of LE involvement in and beyond CAGH, while also capturing the views of people who did not have LE but did have experience of working with PLE in gambling harms reduction efforts. PLE, PS and SS were given a participant information sheet and confirmed their consent to being audio-recorded and quoted. All recordings were transcribed verbatim. The semi-structured interviews were conducted online in two sets: a first set of interviews at the midpoint of the CAGH intervention, and a second set of followup interviews at the end-point. Interviews ranged from 40 min to 1 h 27 min, and involved PLE, PS and SS. Three PLE who were interviewed also sat on the PPIE panel. Two PLE and two PS were interviewed twice: at the outset of the evaluation, and later as a follow-up. Not everyone was interviewed . In total, 13 interviews were conducted with PLE; 16 with PS; and four with SS. Following the first set of interviews, three FGs were held online to more fully explore the different types of LE activity that had been identified and to discuss how best to facilitate effective LE involvement. FGs had an average duration of 70 min. FG participants were all PLE and included people external to the CAGH intervention who were recruited via purposive sampling, based on their expertise in a LE activity that formed the topic of each FG. The FG topics were: --- FG1: LE in campaigning and awareness-raising. FG2: LE in education and consultancy. FG3: LE in peer support interventions. Participants with roles that spanned the above topics were welcome to join more than one FG, resulting in some overlap. FG participants were invited to accept a voucher to thank them for their time and contributions. --- Data analysis The Framework Method [46] was used to organise data into a framework that informed the onward development of themes [47]. The Framework Method proceeded through an initial categorisation phase to an analytical framework that was then refined. Theoretical perspectives and concepts from the LE literature also informed the analysis [46]. Abductive reasoning, which allows themes to be developed from initial codes for ongoing exploration alongside prior theory, underpinned engagement with the literature [48,49]. Data were analysed thematically using a coding apparatus collaboratively developed via team coding sessions using NVivo 12, Microsoft OneNote , and in-person discussion and diagramming. A first round of analysis was carried out after the first set of interviews and the FGs. A second round of analysis was carried out after the follow-up interviews [50]. Both rounds of analysis fed into the identification of themes in the data. --- Results Four themes were identified: Theme 1: Personal journeys to Lived Experience involvement. Theme 2: The value added by Lived Experience to interventions for addressing gambling-related harms. Theme 3: Emotional impacts on people with Lived Experience. Theme 4: Collective Lived Experience and diverse lived experiences. The themes interlink with each other: for example, Theme 1 shapes the contributions made by LE identified under Theme 2. Theme 3 affects how people navigate the individual journeys of Theme 1, and considers the range of emotions that LE involvement brings to the fore for them and others. Theme 4 encompasses the diversity of people's journeys and the activities that PLE are engaged in, as well as the importance of diversity in who is represented in LE involvement and the LE community in the gambling harms reduction sector. --- Personal journeys to lived experience involvement: 'lived experience still seems a little bit of a title' This theme focuses on the role of personal journeys in how PLE become involved in GRH reduction efforts; the tensions arising from the professionalisation of LE involvement; and the new understandings around addressing GRH that result from LE involvement across informal and formal domains. PLE become involved in the gambling harms reduction sector through unique, individual journeys that are ongoing and frequently non-linear. Journeys to LE involvement intertwine with people's ongoing journeys following their experiences of GRH as they develop working relationships in the gambling harms reduction sector, or tap into the spontaneously-emerging networks that underpin the LE community. Without a structured pathway to follow, PLE innovate and craft their own trajectories through this challenging "space": The Lived Experience space is a walking contradiction because it is one of the most inclusive environments to be in, but it is also one of the most difficult to navigate. PLE forge their own techniques and strategies for harnessing their LE to address GRH in meaningful and impactful ways, often in response to gaps in services. For example, one participant described how their LE motivated them to train as a counsellor specialising in GRH: --- the counselling that I experienced was quite outdated and was with people that really didn't have any idea or understanding of what gambling addiction is […] So, I kind of had to go on my own journey […] And that led me to thinking, 'Right, [what] would actually be [helpful for] other people' […] And then [I] set myself up as a private practitioner, initially -or primarily -to work with people who were experiencing gambling harm. Informal domains both provide a springboard to LE involvement, and benefit from the agility and freedom of LE to work around bureaucratic "red tape": --- I'm not saying that people are unprofessional in the Lived Experience space but the things that they can do and add […] the way they're able to form, means that they can do things that we [formal services] can't do. For some PLE, their personal journeys were intertwined with self-directed efforts to craft roles that reflected past professional experiences and skills; other PLE sought and acquired new professional skills that they could use in their efforts to address GRH. The diversity of past professional experience was viewed positively by the LEAP, and meant that different areas could be covered by different people: [We have] experts in certain areas, and there's areas where I don't particularly get involved. Like my background was accountancy before, I was an accountant for 12 years, and then there's someone who works in the NHS, there's a former copper […] people have their areas of expertise. One PLE spoke about their surprise in being able to contribute their design skills in this context. Moreover, while they no longer enjoyed football , their in-depth knowledge of the sport meant they could respond to a request for information. They reported that it was 'great' to draw on past knowledge in this way: --- I created the logo for [VCFSE organisation], so it was good to just use some completely different skills that I didn't think would come in handy […] I'm [also] on a call next week, someone with [redacted] University's looking at gambling advertising and football: she just wants to do some sort of research on that. And that was one of my main things, was football and things, so it's great to have a bit of input on that. As PLE in the gambling harms sector become more established, opportunities for formalising their activities into salaried roles are presented. Professionalisation, in both the statutory and VCFSE sector, has positive and negative implications for PLE. Formal posts with training and qualifications attached, such as counselling and peer support roles in NHS gambling services, can be beneficial to PLE in providing recognition in the form of pay, organisational structures that safeguard self-care and personal boundaries, or prevent the burden of work being overwhelmingly devolved on a single individual. The below account describes a personal journey into the LE space that blurred the boundaries between personal and professional social media use, with largely positive results: --- At the beginning I was keen to keep my personal Twitter and my work's Twitter separate, but […] they started to blend […] That helped my recovery, we all know Twitter and social media can be a toxic space as well, there's good and bad in Twitter, but my experiences were good in that I've built up my network […] it's got me in employment. But the 'incremental professionalisation' whereby LE becomes 'a little bit of a title' , can impose unwanted constraints. PLE needed to know how to fit one's own LE journey into uncompromisingly linear ways of working: --- you're on your own, trying to hold all that up and be a representative for peer support […] Lived Experience is not enough in roles like [peer support] because there's everything else that comes with it, of understanding a data system and learning how to use it and going on six months' worth of training that you have to [have] to be at a certain level […] so that can be a barrier. Some PLE resisted professionalisation and the titles and training that this entailed, but still sought to be involved. Furthermore, some were concerned that they were being incorporated into statutory health services in ways that was 'de-professionalising' , suggesting a form of exploitation: --- There's an economic rationale for including and involving people with Lived Experience because […] it is de-professionalising […] perhaps for economic reasons […] [S]ome of the workload and responsibilities have been pushed onto volunteers and lower roles […] I don't want to use the word 'exploitation': I will use it though. I'll use the word 'exploitation' softly. There was a sort of creeping 'taking advantage of ': exploitation. --- The value added by lived experience to interventions for addressing gambling-related harms: 'I see gambling adverts for what they are' The value of LE is linked to its variety: it is 'not just one thing' . LE contributions to public health span two complementary categories of activity that PLE undertake: supporting people, and driving social change. This theme provides examples of the value LE adds to both sets of activity in informal and formal domains: The value of LE in educational and training interventions lies in the understanding of GRH that LE contributes, and the means of engagement through which awareness is raised. By sharing their stories, PLE can humanise GRH statistics, or make the harms experienced by individuals relatable for people who have not had those unique experiences. Storytelling can be tailored to suit the purpose of the encounter and the audience: a careful balance between seeking maximum impact, and avoiding additional harms . One LEAP member described the value of their input in an educational intervention that showed their LE story on film while they were physically present in the room: 1. And that film is loosely based on my story and I'm in the film […] And it is a nice moment when the kids in the room watch the film […] and, all of a sudden, my face pops up and I start saying, 'Oh, yes. I had a really good upbringing' and then they all look and realise, 'Oh, it's you. That Storytelling combines effectively with developing GRH "literacy", or critical health literacy: a skillset that equips people to critique if and how messaging serves industry interests, and take action to address the wider determinants of health associated with gambling. Strong counter-industry narratives featured in LE accounts. This, a public health professional observed, was unique to the sector and aligned with a public health approach: the gambling lived experience community actually, in my opinion, are advocating for quite upstream measures around gambling, which isn't always what people are looking for, say, from the drug and alcohol perspective. My experience is that people are more advocating for treatment and better access to treatment. Which, in gambling, there's a need there as well but what people are actually asking for is much more upstream. Which --- is quite interesting and slightly different from other areas. And, obviously, those upstream interventions align very closely with what you'd want from a public health perspective. The following data extracts show how such counterindustry framings can enhance critical health literacy, here enabling people to withstand the pull of gambling advertisements: --- I didn't even really know that these products were designed to guarantee profit, I just thought I could win […] that sort of [counter-industry] messaging if I'd have seen that, would have had more impact than just hearing my own story […] If I'd have heard my Lived Experience story at 16, 17, I'd have probably said, 'Oh well, unlucky but I know what I'm doing still' . Development of critical skills in relation to gambling was built-in to some educational interventions, for example one in which a slot machine is dismantled to demonstrate its inner workings. Involvement in CAGH also contributed to developing a critical mindset in the case of a community project that had been promoting a charitable lottery in its email signature, but subsequently recognised this as being 'contradictory with the aims of the gambling [harms reduction] project' . A shared LE understanding of the importance of upstream interventions was, however, not echoed in a shared perspective on strategy -notably on whether PLE should work with the gambling industry. This issue remained contentious and emotive for the LE community, with different perspectives apparent: [gambling] is a business, it is there to make a profit, but it's how they make that profit and who they make that profit from, that's where the argument sits for me. And so that ' --- s why I think dialogue and conversation with the industry is important because those that are in the Lived Experience community that complain about people talking to the industry, working with the industry, well, I've no time for that because what's the point, if you're not going to have a conversation with the industry you're always going to be poles apart and there's never going to be any progress, you have to have conversation. I wanted this project as a whole to really disrupt, and one of the disrupting things would be that we won't work with any organisation directly funded by the gambling industry and that's not happened with this. Emotional impacts on people with lived experience: 'it was like I'd given blood and never had anything to replenish me' LE involvement in gambling harms reduction efforts exerts ongoing emotional impacts on PLE. Conflicting emotions are raised, which can range from frustration with policy inertia and having their experiences contested following media appearances, to more positive impacts: a sense of purpose and fulfilment, hope and passion for change, empathy and emotional intelligence, and new understandings of how to address GRH. This theme describes the emotional impacts on individual PLE who contribute to this sector, and the personal and professional benefits that can accrue. PLE expend considerable personal time and effort in LE involvement activities around GRH, which can make such involvement exhausting. Exhaustion is exacerbated when organisations engage with PLE superficially, or when responsibilities are devolved on individuals without adequate support in place: --- that's why I left my [peer support] role […] it was like I'd given blood for nine or ten months and never had anything to replenish me. To safeguard PLE's personal resources of self-care from depletion, supervision -as practised in counselling -and peer-to-peer support is important in providing space for individual reflection and replenishment. In the absence of this, the 'goodwill' on which LE involvement is run can assume an ever-present availability on the part of PLE. This erodes the professional boundaries put in place to protect PLE's energy and time, because 'you're not going to say 'Sorry, can you ring me back during office hours': that isn't how it works' : --- there is a risk sometimes that you forget that you yourself are in recovery because you spend so much time focused on other people and trying to help them -and quite rightly, because you do that because you want to do [it] -but you can sometimes get this feeling of invincibility. Not a conscious one, but subconscious. You can almost forget that, hang on a minute, this still applies to me. you've been that person with no one to pick up the phone to […] That's what the Lived Experience gives you, because it is those things where you think, I know they need me on the phone right now […] There are people we know who are living and breathing today because they've had those 3am calls, because people like us have not respected our boundaries. The authenticity of LE, of value in disseminating LE insights, is also what makes PLE potentially vulnerable: 'it brings that emotion, your voice will crackle and stuff like that, that is what it is: it's authentic' . While LE involvement can provide a sense of meaning and purpose to PLE, it can also be emotionally and physically draining if adequate support is not in place. Adequate support and safeguarding are, therefore, vital to protect PLE and sustain their involvement: If people are recounting their story and bringing up, in some cases, quite painful memories, then they really should then have somewhere that they can go, where they can check themselves out and make sure they're safe […] I almost feel a little bit worried for people if they are recounting their story on a daily basis and bringing up things that maybe are still unprocessed or uncomfortable and they haven't got anywhere to take that. Some PLE in the sample were active campaigners who advocated for the adoption of a national public health framework. They reported that such campaigning could be extremely challenging, with their stories being frequently contested by people hostile to arguments for industry regulation -adding to the emotional pressures on PLE. Some believed that this was unique to the gambling harms reduction sector because, they argued, LE perspectives are not scrutinised or challenged in the same way in other sectors, where there is less campaigning among respective LE communities for upstream national policy reform. One PLE stepped back from campaigning because of this pressure, but sought to support those who continue to campaign and urged them to engage in self-care: [ --- I]t's okay to take a step back […] I admire the [names of prominent gambling harms campaigners] who are relentlessly out there. You can be attacked if you're out there, you can be criticised, you can be judged […] If someone does say […] [to me], 'At the moment this is really tough: I'm feeling this is impacting me personally' , the first advice I give to that person is 'Think about you and only you' […] It's not being selfish […] [to be] mindful of your self-care. [PLE11, interview] Differences in perspective on whether to engage directly with the gambling industry created internal tensions within the LE community, sometimes resulting in intense disagreement on social media. At the root of this issue was a debate about the funding of services for addressing GRH, with some PLE contributing to gambling industry-funded treatment services and educational campaigns: --- [T]he biggest one is the funding element of it all […] The reason that it can create some conflict is because a lot of [people with] Lived Experience are out there working […] [for] organisations [that] are funded through the industry. I have no views on it at all in the sense of [how I view] those individuals and what they do, but that can create conflict within Lived Experience […] [P]eople can take offence and then it creates these divides. The only people that are loving that is the industry itself. [PLE11, interview] --- Collective lived experience and diverse lived experiences: 'a collective voice of lived experience is better than a singular one' As a term, 'LE' is frequently discussed as if it constitutes a single, authoritative experience. This is at odds with the plurality of perspectives that together create the LE community: diverse lived experiences which, in their variety, provide the whole with a greater potential influence and legitimacy than could be achieved by individuals alone. This theme considers the role of LE in representing, and doing justice to, the diversity of people's experiences . The lack of ethnic diversity in LE within the gambling harms reduction sector is significant, and visibility of some populations over others can undercut the relatability that is central to the power of LE: 'all the Lived Experience panels that I've been involved in are just 90% white men […] it does restrict you in getting to those other communities and hearing other voices' . There is recognition in the data of the risks involved in over-reliance on individuals, and on single accounts of LE, at the expense of scalable interventions: --- the results that we get are better when it's someone with Lived Experience [delivering] […] but it is quite a big worrying question that I think about a lot, is have we created a programme that focuses so much on me but what do we do when I'm not doing it? In this sector specifically, it is important to recognise that, beyond diversity in the communities affected by GRH, diversity also has broader relevance. Diversity in this context therefore includes the diverse forms of gambling that are available, 'in a betting shop or a casino or online slots' ; the diverse harms associated with gambling; and the diverse strategies mobilised to address those harms: --- There is another point of diversity that I think is more important than actual diversity [of protected characteristics]: is the diversity of experiences […] gambling is changing. And I don't think we're reflecting new people's experiences […] like the rise of day trading, the rise of crypto, the different products, the different practices of the industry, the way the industry's regulated. I think we need to say up-to-date with what's going on, and that involves bringing in the people with Lived Experience who are more current. So diversity's not just about gender and religion and race -whatever -it's about experiences too. There is a role for public health in cultivating connections with diverse communities to facilitate their LE involvement. Diversity in GRH means more than a collection of representative "voices" -it also supports a community or collective, within which there is room for diverse experiences of GRH and diverse viewpoints, including critique of LE involvement itself. As one participant expressed it, 'a collective voice of Lived Experience is better than a singular one' . This distinguishes LE in the collective sense , from LE understood as encompassing a range of perspectives. Tensions can arise between the two, particularly around the divisive question of whether to engage with industry: --- you need to have those conversations with [the gambling] industry to say, 'this is the reality' […] There needs to be more cohesion and more conversations and understanding that each other exists, and each other has a value and a reason to exist and not just an outright 'No, we're not going to engage' . Some PLE may be open to 'a healthy two-way discussion' on such tensions, but the structural power imbalance in decision-making in this space undermines diversity of opinion: --- you can give a perspective that nobody else can give [but] I think sometimes it's just been used as a bit of window dressing -oh, we've got a Lived Experience group, but they're not given any authority to make decisions […] I was part of a four-person Lived Experience panel that was scoring bids for tender […] and our four votes only counted as one vote, so they watered down what we did. Of note here is that 'window dressing' , used here to refer to tokenistically involving PLE, is distinct from another image used in the data: the 'tick box exercise' whereby LE involvement is viewed as something to wheel in and "get done", before swiftly moving on: 'just been brought in to tick a box or to be consulted on, and then nothing goes anywhere' . --- Discussion --- Contributions of lived experience to a public health approach for reducing gambling-related harms The key insight from this evaluation is that involving LE can contribute substantially to a PH approach to gambling harms reduction, and in varied ways. The themes outlined above demonstrate how LE contributions contend with tensions in the application of LE involvement. LE is really many lived experiences, as different domains and activities related to LE involvement give rise to different contributions and challenges. The themes confirm and add to the existing literature on LE in public health approaches to GRH reduction. They demonstrate that a range of participation options for PLE in the GRH sector is important [51], and they develop understandings of the positive and timely contributions that LE involvement can make to guide public health interventions. To aid the identification and harnessing of these contributions and their implications for public health, a typology of LE involvement specific to the aims associated with gambling harms reduction based on the data has been developed and is presented below . PLE undertake many activities that support public health. Some of these are delivered through formal job roles that offer opportunities for PLE to utilise their unique insight and motivation to support people or affect social change. Underpinning formal involvement, there are also self-organising networks in which people support each other and engage in social campaigns to deliver change. Figure 1 maps these activities. Broadly speaking, the activities have two core purposes or 'functions': supporting people and social change. The form that these activities take on, however, changes depending on whether they are enacted in informal or formal domains. Thus, the figure presents a four-fold typology of LE activities relevant to public health. The four types are fluid and can overlap with each other. Examples of each of the types in practice are provided. The typology visualises the interrelated activities that PLE undertake across informal and formal domains in support of a public health approach to reducing GRH. These activities contribute to supporting people and driving social change, and are steered by the agency of PLE. Each forms part of the contributions that PLE make to GRH reduction efforts. The various activity types have an impact, both on the person "doing" LE and those who "receive" it. Immersion in spontaneously emerging friendship and support networks can help people through the sharing of experiences, understanding, empathy and hope. Similarly, for PLE who are involved extensively in the field, formal job roles provide opportunity for recognition, pay, meaning and purpose. Each activity type can present challenges, whether it is located in informal or formal domains. Figure 2 follows sequentially on from Fig. 1 in mapping the impacts of LE involvement, and the emotions such involvement raises in PLE. It also considers the different skills developed and deployed across informal and formal domains. As in Fig. 1, the importance of diversity in the LE community is central. Figure 2 presents the positive and negative impacts of LE involvement. LE adds something distinct to the activity types in Fig. 1: it makes formal services approachable, Fig. 1 A typology of lived experience activities in the gambling harms reduction sector and it adds authenticity to social campaigns. But there are corresponding possible negatives for the person. These include exhaustion from filling in gaps in statutory support services, as it is not easy to decline requests of help from people in need. Social campaigns take time and are frustrating when policymakers and senior leaders do not respond to evidence-based demands , or treat LE voices as 'window dressing' , rather than genuinely listening to their views. As shown in Sect. 3.3 and 3.4 above, there are also differences of opinion among the LE community on strategic questions: some PLE urge engagement with the gambling industry and work for, or receive funding from, gambling operators, while others emphasise the importance of independence from industry and aspire to work for or with organisations that are funded from independent sources. This can lead to tensions within the LE community. Involvement in LE requires support to assist PLE to manage the associated tensions as well as continuing personal and professional development opportunities that accompany the translation of local experiences for use in wider contexts. Safeguarding is necessary to sustain the life-blood of LE involvement and to harness and repurpose the emotions associated with it for positive ends while continuing to protect PLE who contribute their experiences. The development of formal pathways to LE involvement is challenged by the encroaching professionalisation of LE work and the narrowing of the expertise of PLE to a single subject. The "vitality" of LE that is threaded throughout the data -the energy and responsiveness in how LE is spoken about, and how its value is perceived -is sapped by institutional workflows that require PLE to somehow fit the messiness of LE into linear career trajectories. At the other end of the scale, informality can lead to eroded boundaries and burnout for emotionally-exhausted PLE. Both figures prompt consideration of what public health can do to enhance the positives and lessen the negatives, while retaining the value of LE's informality. The perspectives of PLE can contribute important knowledge to inform a social harms perspective on gambling and insights into how interventions to address GRH can be designed and delivered effectively [17]. A LE collective that welcomes and accommodates people's diverse lived experiences and expertise beyond their LE is needed. Such a collective might best be facilitated by a public health approach to reducing GRH that supports PLE and public health professionals to develop their and others' critical health literacy relevant to gambling. The entanglement of LE involvement in gambling harms reduction efforts with the commercial determinants of those harms complicates LE in this sector by implicating it in continued health-harming industry operations [28]. LE encompasses, and should do justice to, diverse experiential knowledges. LE spans unique, personal, and ongoing journeys that can be shared to advance gambling harms reduction, chiefly through two complementary Fig. 2 A mapping of the positive and negative impacts of LE involvement in the gambling harms reduction sector areas of work: supporting people and driving social change. For LE involvement to be effective, full consideration must be given to the emotional impacts of LE roles and the multiplicity of experiences being lived through by those who take on such roles. LE contributes a valuable alternative way of looking at things that public health needs to learn from and use if it is to accommodate non-linear LE journeys and facilitate the translation of LE for public health purposes. A collective LE stance is needed that can overcome differences of opinion on strategy within the LE community, while also recognising the value of this plurality and the reflections it prompts on the part of PLE and the organisations that they advise. --- Implications for lived experience involvement in public health This study has implications for what LE involvement in public health should look like to ensure that it is sustainable, fully-supported and applied to its full potential. It is vital that public health professionals value the existing expertise of PLE and LE-led organisations specialising in GRH to help facilitate work in this area. Done right, the involvement of PLE is meaningful and can make a difference. In the UK, the proposed adoption of a statutory levy to fund work in the gambling harms reduction sector provides a recent example of why listening to PLE is important: some public health actors critique the levy because of its continued links with industry [50], but more work is needed to understand whether this reflects the differing viewpoints of PLE. The evaluation team consulted the PPIE panel, which included PLE of GRH, about a set of recommendations from this evaluation. Many of the recommendations align with previous research [26], but the following are additions specific to this research: --- Limitations Our sample of PLE was not diverse. This reflected the reality of LE involvement on the ground, as the PLE who participated in the gambling harms reduction intervention were predominantly White British. As we explore in theme 3.4, this local weakness of representation was recognised, and the PLE sought to connect with and recruit from diverse ethnic groups from the region. The public health professionals involved in CAGH had a role in this, as they arranged opportunities for the PLE to speak at community and faith events, but ultimately it proved challenging to recruit. Ensuring diversity in LE involvement should, therefore, be viewed as an ongoing objective and future research should prioritise efforts to recruit a diverse sample to explore the barriers to LE involvement. PLE who were motivated to participate in this study also did not represent the spectrum of harms engendered by gambling, and tended to be individuals with previous experience of LE involvement , rather than those less-experienced or new to this sector [26]. Stigma significantly affected recruitment, particularly of community project participants, given the LE community around GRH in the UK is of a size that individuals are potentially identifiable [32]. The decision not to hire PLE to conduct the interviews or coordinate the FGs is a limitation: a peer-interviewer approach, with trained PLE interviewing PLE, could have helped reassure participants as well as providing a way to give back to the LE community [52]. The PPIE panel were however consulted throughout the project, and a free training session in evaluation methods was offered to interested participants post-intervention. Research based on LE can face difficulties in claiming to be generalisable, because LE is so unique [29]. However, foregrounding LE, as this evaluation aimed to do, can offer lessons for other public health systems by drawing comparisons with industry-led versions of events [29]. --- Conclusion This study has highlighted the importance of integrating LE into public health approaches to address GRH. Public health professionals should ensure that LE is included in intervention design, delivery and evaluation processes, and that support is in place to facilitate LE activities related to reducing GRH. Harnessing the contributions of LE in all their diversity is essential to inform the direction and responsiveness of the gambling harms reduction sector in the UK as it develops at pace. A LE-informed public health approach to gambling harms reduction requires local access to involvement for PLE via diverse and equitable routes that are free from stigma and that can contribute to decision-making. LE involvement in this sector needs to enable PLE and people without LE to develop the affective and critical skills necessary to navigate the tensions inherent in co-existing with industry-funded LE involvement programmes. --- Data availability No datasets were generated or analysed during the current study. --- Abbreviations --- AO ---

Background Lived Experience (LE) involvement has been shown to improve interventions across diverse sectors. Yet LE contributions to public health approaches to address gambling-related harms remain underexplored, despite notable detrimental health and social outcomes linked to gambling. This paper analyses the potential of LE involvement in public health strategy to address gambling-related harms. It focuses on the example of a UK cityregion gambling harms reduction intervention that presented multiple opportunities for LE input. Methods Three focus groups and 33 semi-structured interviews were conducted to hear from people with and without LE who were involved in the gambling harms reduction intervention, or who had previous experience of LE-informed efforts for addressing gambling-related harms. People without LE provided reflections on the value and contributions of others' LE to their work. Data analysis combined the Framework Method with themes developed inductively (from people's accounts) and deductively (from the literature, including grey literature).Four themes were identified: (1) personal journeys to LE involvement; (2) the value added by LE to interventions for addressing gambling-related harms; (3) emotional impacts on people with LE; and (4) collective LE and diverse lived experiences. Two figures outlining LE involvement specific to gambling harms reduction in the UK, where public health efforts aimed at addressing gambling-related harms coexist with industry-funded programmes, are proposed. Conclusions Integrating a range of LE perspectives in a public health approach to gambling harms reduction requires local access to involvement for people with LE via diverse routes that are free from stigma and present people with LE with options in how they can engage and be heard in decision-making, and how they operate in relation to industry influence. Involving LE in gambling harms reduction requires enabling people to develop the affective and critical skills necessary to navigate complex emotional journeys and a challenging commercial and policy environment.

Introduction Children who grow up in families where the parents have alcohol problems are at increased risk of several negative consequences, including poor school performance, poor mental health, and early onset alcohol use [1][2][3][4][5][6]. Parentification may also occur, where children assume adult roles even though they are not developmentally or emotionally ready [7]. The consequences are often long-term [8], and they augment the likelihood of other disorders, for instance, mental disorders such as major depression [9]. Furthermore, studies have demonstrated that when parents have alcohol problems, their offspring are at increased risk of alcohol-related hospitalization and mortality, including suicide [10,11]. Currently, international studies have estimated that the prevalence of children with parents who have alcohol problems is 4-29% [12][13][14][15][16][17]. The primary reason for this broad range is that parental alcohol problems are defined and assessed differently in different studies. For instance, some studies examined hazardous drinking among parents and others examined parental alcohol use disorder. Furthermore, some studies were based on self-reports from either the children or parents, and others were based on surveys, psychiatric interviews, or registry data. Drinking patterns vary across countries, and there may also be differences in how alcohol problems are defined. --- of 10 In Nordic countries, only a handful of scientifically determined estimates are available, and the estimated prevalence varies. A web-based survey distributed to Swedish youth, 16-19 years old, concluded that 20.1% of the sample had at least one parent with an alcohol problem [13]. In that study, perceived alcohol problems were assessed with the short version of the Children of Alcoholics Screening Test [18]. Another survey, which was distributed to a nationally representative sample of Swedish adults, 17-84 years old, assessed alcohol problems with the Mini International Psychiatric Interview, derived from the Diagnostic and Statistical Manual of Mental Disorders, fourth edition. They concluded that 3.7% of children had at least one parent with a current alcohol use disorder [12]. Another study, based on Danish registry data, concluded that 4.5% of children had parents that had been hospitalized due to an alcohol-related illness [17]. A recent Danish study, based on 75,853 high-school and vocational school students, reported that 7.3% of the surveyed students perceived that they had at least one parent with alcohol problems [19]. A Norwegian study, based on reports from parents of teenagers, found that 15.6% of fathers and 4.7% of mothers were defined as individuals that misused alcohol [20]. However, these figures may not be generalizable to parents with younger children [21]. The scarcity of data on the prevalence of children who have parents with alcohol problems in Norway calls for further studies. Early adversity may have a negative impact on many aspects of life, including socioeconomic indicators, such as education, employment, and income [22]. However, to the best of our knowledge, no studies have explicitly investigated whether there exists a social gradient connected to parental alcohol problems in non-clinical populations. Moreover, although it is important to understand how widespread parental alcohol problems are, it would be valuable to have estimates based on the perceptions of the children or adult children. Therefore, this study aimed to estimate the prevalence of parental alcohol problems during childhood in a general population of Norwegian adults, and to investigate associations between parental alcohol problems during childhood and lower socioeconomic status in adulthood. --- Materials and Methods This cross-sectional study included a random sample of 75,191 individuals, aged 18 years or older, that resided in the region of Agder . The sample was drawn from the Norwegian Population Registry, and e-mails or telephone numbers were obtained from the contact registry of the Agency for Public Management and eGovernment . Individuals who had declined to participate in surveys, individuals registered as deceased, those with unverified contact information, and those with an address outside the region were removed. Thus, in 2019, 61,611 inhabitants were invited to participate in the Norwegian Counties Public Health Survey. The respondents participated by completing a questionnaire online. The questionnaire included questions related to health, well-being, childhood, living conditions, local environments, accidents, and injuries. Participants gave online consent to participate when they answered the survey questions, and provided their age and sex to confirm their identity. Of the 61,611 invited individuals, 28,047 completed the questionnaire; the response rate was 45.5%. --- Ethics Informed consent was obtained from all subjects involved in the study. All personal identification variables were removed before the researchers obtained the dataset. Data were handled in compliance with applicable personal data protection regulations. The Norwegian Institute of Public Health is responsible for the health survey. The survey was approved by the Norwegian Data Inspectorate, and it adhered to the regulations of the Personal Health Data Filing System Act. In addition, a Data Protection Impact Assessment was performed by the Norwegian Institute of Public Health. Ethical approval for the current study was obtained from The National Committees for Research Ethics in Norway , and from the Faculty Ethics Committee at the University of Agder. --- Measures The questions, response categories, and definitions used in the survey are shown in Table 1. Vocational training/middle school /upper secondary/junior college 3. University/college <4 years 4. University/college ≥4 years 1 = low education 2 = intermediate education 3 and 4 = higher education --- Financial capabilities For one-person households, consider your total income. If you live with others, consider the total income of everyone in the household. How easy or difficult is it for you to make ends meet day to day with this income? 1. Very difficult 2. Difficult 3. Relatively difficult 4. Relatively easy 5. Easy 6. Very easy 7. Do not know 1-3 = low economic capability vs. --- 4-7 = middle/high economic capability --- Employment status What is your current status concerning employment etc.? 1. Full-time 2. Part-time 3. Self-employed, 4. On sick leave 5. Unemployed 6. Receiving disability pension/work assessment allowance 7. Receiving social assistance benefits 8. In retirement/early retirement 9. Pupil/student 10. Undertaking national/alternative civilian service 11. Homemaker 1 = ≥ 32 h/week vs. not2 = < 32 h/week vs. not3 = Self-employed vs. not4 = On sick-leave vs. not6 and 7 = Receiving welfare benefits vs. not The six-item CAST-6 instrument was used to estimate perceived parental alcohol problems [18]. Respondents could answer yes = 1 or no = 0 to each question, and the total score ranged from 0 to 6. The CAST-6 demonstrated high internal consistency and concurrent validity , compared to the original 30-item CAST for adults [18,23,24]. Moreover, it showed good to excellent test-retest reliability for both adults and adolescents [23][24][25]. In the present study, the scale showed excellent reliability . Two alternative cut-off scores are commonly used with the CAST-6. One cut-off score is more inclusive and the other is more conservative [18,24,26,27]. The more conservative cut-off score was used in the current study. Data on socioeconomic factors were collected with questions related to education, economic capability, employment status, and whether respondents received welfare benefits . Participants' age and sex were provided through the national population registry. In addition, participants were asked about their marital status. --- Statistical Analysis Data were analysed with SPSS version 25 . Descriptive statistics for the overall sample were estimated for key demographic and socioeconomic variables. Pearson's χ 2 analyses were performed to evaluate associations between the overall distribution of parental alcohol problems and the demographic and socioeconomic variables. Multivariable logistic regression was performed to investigate the association between parental alcohol problems and measures of low socioeconomic status, adjusted for age and sex. Results are expressed as odds ratios with 95% confidence intervals . A p-value < 0.05 was considered statistically significant. --- Results Descriptive characteristics of the sample are provided in Table 2. * Multiple responses could be selected; education level and employment status are defined in Table 1. Table 3 shows that, overall, 15.6% of the respondents had experienced problematic alcohol use among their parents during childhood. This experience was significantly more prevalent among females than among males . The proportion of individuals who reported experiences of problematic parental alcohol use varied among different age groups. The lowest prevalence was observed for respondents aged 67 years or older. Moreover, this experience was less common among respondents that were married or had a registered partner, compared to those with another relationship status. We also observed a consistent social gradient in associations between parental alcohol problems and various socioeconomic variables. Parental alcohol problems were more prevalent among those with a lower education level, compared to those with intermediate or high education levels; among those with low economic capability, compared to those with middle/high economic capability; among those on sick leave, compared to those not on sick leave; and among those who received welfare benefits, compared to those who did not receive welfare benefits. 1. Results from the multivariable logistic regression are displayed in Figure 1. Findings revealed consistent associations between parental alcohol problems and all measures of low socioeconomic status. The strongest association was found between parental alcohol problems and the need for welfare benefits . Other forms of marginalization within the work force, such as being on sick leave or being unemployed, were also associated with parental alcohol problems . The experience of parental alcohol problems was also significantly associated with no college/university education . forms of marginalization within the work force, such as being on sick leave or being unemployed, were also associated with parental alcohol problems . The experience of parental alcohol problems was also significantly associated with no college/university education . --- Discussion We found that, among an adult Norwegian sample randomly drawn from the general population, 15.6% had experienced problematic parental alcohol use during childhood. To the best of our knowledge, no previous studies have estimated the prevalence of parental alcohol problems in the Nordic context based on a broad age range of adult offspring from the general population. A previous Norwegian study analysed self-reported problems with alcohol use among parents of teenagers. They found that 15.6% of the fathers had alcohol problems , but the proportion of mothers in this category was significantly lower . In our study, we did not group individuals based on parental sex; thus, our finding that 15.6% of parents had problematic drinking behaviours included fathers, mothers, or both. Other international estimates of the prevalence of parental alcohol problems have varied greatly [12][13][14][15][16][17]. This variation might partly be explained by differences in the samples and measures used in different studies. A Swedish study included adolescents aged 16-19 years, and also used the CAST-6. They found that the prevalence of respondents that reported perceived parental alcohol problems was 20.1% [13], which was somewhat higher than our estimate. This difference might be explained by the difference in respondents' age between studies. Our results indicated that the oldest age group was least likely to report parental alcohol problems. This result could be explained by several factors. First, the questions were retrospective in nature, and recall bias could be a prominent issue [28]. Second, it has been shown that adverse childhood experiences, such as parental alcohol problems, were associated with impaired health [29] and elevated mortality [10]; therefore, the oldest respondents who experienced problematic parental alcohol problems could have been underrepresented. Third, alcohol consumption among Norwegian adults increased after the second world war [30]; thus, it is plausible that the prevalence of parental alcohol problems was, in fact, relatively low during the era that the oldest participants grew up. --- Discussion We found that, among an adult Norwegian sample randomly drawn from the general population, 15.6% had experienced problematic parental alcohol use during childhood. To the best of our knowledge, no previous studies have estimated the prevalence of parental alcohol problems in the Nordic context based on a broad age range of adult offspring from the general population. A previous Norwegian study analysed self-reported problems with alcohol use among parents of teenagers. They found that 15.6% of the fathers had alcohol problems , but the proportion of mothers in this category was significantly lower . In our study, we did not group individuals based on parental sex; thus, our finding that 15.6% of parents had problematic drinking behaviours included fathers, mothers, or both. Other international estimates of the prevalence of parental alcohol problems have varied greatly [12][13][14][15][16][17]. This variation might partly be explained by differences in the samples and measures used in different studies. A Swedish study included adolescents aged 16-19 years, and also used the CAST-6. They found that the prevalence of respondents that reported perceived parental alcohol problems was 20.1% [13], which was somewhat higher than our estimate. This difference might be explained by the difference in respondents' age between studies. Our results indicated that the oldest age group was least likely to report parental alcohol problems. This result could be explained by several factors. First, the questions were retrospective in nature, and recall bias could be a prominent issue [28]. Second, it has been shown that adverse childhood experiences, such as parental alcohol problems, were associated with impaired health [29] and elevated mortality [10]; therefore, the oldest respondents who experienced problematic parental alcohol problems could have been underrepresented. Third, alcohol consumption among Norwegian adults increased after the second world war [30]; thus, it is plausible that the prevalence of parental alcohol problems was, in fact, relatively low during the era that the oldest participants grew up. We also found that parental alcohol problems were reported slightly more frequently by females than by males. Although this result was puzzling, other studies have shown similar findings [24,25]. Havey and Dodd [25] have suggested that females, compared to males, may be more sensitised toward substance use and related issues within the family, and that they also may be more prone to express concern about a family situation in a self-reported questionnaire. Overall, our findings showed that perceived parental alcohol problems were most prevalent among socioeconomically disadvantaged groups . The largest proportion of respondents that experienced parental alcohol problems comprised those who received a disability pension, work assessment allowance, or social assistance benefits . In this group of respondents, 25% experienced parental alcohol problems during childhood. This finding remained significant after adjusting for age and sex in the multivariable analyses. Although alcohol consumption in Norway was found to be highest among adults with a high education level [31], we found that the respondents' childhood experiences of problematic parental alcohol use were inversely associated with the respondents' education level. Other studies have also found socioeconomic inequalities in the distribution of individuals that experienced alcohol-related harm [32]. Although we lack studies that have specifically addressed socioeconomic differences in the distribution of individuals with parental alcohol problems, other studies have shown that adverse life experiences are socially patterned in childhood [33]. Therefore, the social gradient that we observed among our adult respondents could be related to the socioeconomic disadvantage present in childhood. However, adverse childhood experiences can also reduce educational attainment; indeed, Houtepen et al. [34] found that this relationship remained significant after controlling for family socioeconomic variables. Possible explanations of these relationships are likely complex. Exposure to chronic stress may induce changes in the developing brain and impact a range of important functions that interfere with learning and other skills needed to succeed in education or the workplace [35]. Childhood adversities such as parental alcohol problems could also increase health risk behaviours, physical and mental health problems, and developmental disruptions [36] which may also contribute to economic marginalisation. --- Study Strengths and Limitations This study has expanded existing knowledge by contributing estimates of perceived parental alcohol problems, based on reports from a large adult sample of 28,047 individuals drawn randomly from the general population. Our outcome was based on the CAST-6, which is a validated instrument [24]. Item four of the CAST-6 presumes the presence of two parents, which could influence the score for respondents who grew up in single-parent families. Sensitivity analyses excluding this item did not alter the findings significantly. Our findings shed light on the socioeconomic patterns associated with the prevalence of parental alcohol problems, which were rarely studied in previous research. This study was limited by its retrospective design. Moreover, responses could be prone to recall bias and the risk of measurement error [28]. The validity of retrospective assessments of childhood experiences has been debated; however, a comparison between retrospective reports and prospective results did not reveal a bias in the retrospective assessment of difficult childhood experiences [37]. Additionally, cautiousness regarding the generalizability of the findings is necessary due to possible non-response bias. Finally, this study was based on cross-sectional data; therefore, the results should be interpreted with caution when considering causality. --- Implication for Practice The CAST-6 was not designed to identify diagnostic criteria; instead, it identifies individual perceptions of problematic parental alcohol use. Previous studies that investigated adverse outcomes related to parental non-dependent alcohol use had mainly focused on offspring substance use [38]. However, several studies have identified other negative outcomes related to parental non-dependent drinking patterns [39][40][41]. These dysfunctional patterns often continue into the next generation. To break the patterns, early support interventions should be available. However, support might be available to varying degrees. For instance, in Sweden, the vast majority of municipal social services provide support to children growing up with parental substance use problems, most often in the form of individual counselling or support groups but, at the same time, support only reaches a small proportion of the targeted children [42]. Several organizations identify children in need and offer support, including the adult substance-use treatment services, psychiatric care, and social services. However, studies have shown that, in most cases, those organizations did not determine whether the clients had children [43]. The situation in Norway appears to be similar: only about one fifth of the professionals working in substance use treatment facilities offered support to their clients' children, and about half of the professionals never assessed whether the clients had children [44]. One obvious arena to identify children in need of support is the school setting. Since these children often are neglected, schools could work with policy documents and action plans and inform and train their staff about this vulnerable group. Previous research has shown that policy documents increase the likelihood of school staff to receive training in this issue, which in turn increases the likelihood of identifying these vulnerable children in the school setting [45]. Digital interventions represent a promising approach for increasing the availability of support. However, currently, only a small number of digital interventions are currently being tested that target this group of individuals [46][47][48][49]. For instance, in Sweden an online chat group program has been developed [47], based on a Dutch program [48]. The program consists of eight weekly sessions, each 60-90 min long, focusing on themes such as 'your role in the family', 'social networks', and 'substance use, tolerance, and heredity'. Each session is moderated by a trained counsellor. The program is currently being evaluated but has the potential to reach a large number of adolescents and young adults. --- Conclusions This study showed that one in six adults reported problematic parental alcohol use and, among disadvantaged sub-groups, this prevalence increased to one in four. It is imperative to make both universal and selective prevention interventions available at an earlier age if we expect to break family patterns of problematic alcohol consumption. In addition, we need better methods for early detection, for instance by identifying burdened children when parents are in contact with general or more specialized health care [43,50]. Furthermore, we should ensure proper support and follow-up for these children and their families. --- Data Availability Statement: Restrictions apply to the availability of these data. Data was obtained from Norwegian Institute of Public Health and are available at https://helsedata.no/en with the permission of NIPH. --- Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. ---

The aim of the study presented here was to estimate the prevalence of parental alcohol problems during childhood in a general population of Norwegian adults, and to investigate associations between parental alcohol problems during childhood and lower socioeconomic status in adulthood. This cross-sectional study recruited 28,047 adults (≥18 years) to an online health survey (Norwegian Counties Public Health Surveys). We evaluated demographic and socioeconomic measures and responses to a shortened version of the Children of Alcoholics Screening Test (CAST-6) scale to assess whether respondents perceived parental alcohol consumption during childhood as problematic. Respondents reported parental alcohol problems at a rate of 15.6%, but the experience was more prevalent among adults with a low education (20.0%), compared to those with intermediate (16.4%) or high educations (13.8%, χ 2 (2) = 87.486, p < 0.001), and it was more common among respondents with low economic capabilities (21.1%) compared to those with middle/high capabilities (14.2%, χ 2 (1) = 162.089, p < 0.001). Parental alcohol problems were most prevalent among respondents that received welfare benefits (24.5%). Multivariable logistic regression analyses revealed associations between parental alcohol problems and low socioeconomic status in adulthood; odds ratios (95% confidence intervals) ranged from 1.33 (1.25-1.42) to 1.89 (1.72-2.06). From a public health perspective, children who grow up with parental alcohol problems should be reached through both universal and selective interventions.

D esign thinking methodology is a framework to foster creative and interdisciplinary approaches to innovations. The aim is to develop effective solutions to current problems using creative thinking and a human-centered approach . Successful implementation of design thinking in health care can improve clinical practice, quality of care, and patient outcomes . This methodology was used to address a knowledge-to-action gap in health care for lesbian, gay, bisexual, transgender, queer, intersex and Two-Spirit people. This article presents the authors' experience at a design jam hosted by the Canadian Institutes of Health Research to design a tool for knowledge transfer and exchange in nursing education. This article aims to describe design thinking, critique our experience, and highlight implications for creative development of nursing tools. --- PROBLEM DESCRIPTION Historically, research about health care for gender minorities has focused on LGBT people, with a dearth of literature about barriers to care for queer, intersex, and Two-Spirit people. LGBT people experience distinct health disparities and are one of the largest underserved populations in health care . Barriers contributing to health disparities include discrimination, lack of provider experience and knowledge, and a deficiency of services . Lack of knowledgeable health-care providers has been identified as a key barrier to culturally sensitive health care for the LGBT population . Content about caring for LGBTQI2S people is lacking in baccalaureate nursing curricula, with a median time of 2.12 hours devoted to the topic in U.S. nursing programs . Similarly, programs in Canada provided limited, inconsistent, non-standardized content in LGBTQI2S health and wellness . Rondahl explored nursing students' understanding of the LGBT population and found that their knowledge was inadequate, with only 10% of students achieving a passing mark on a LGBT knowledge questionnaire. Faculty who feel unprepared to teach LGBTQI2S content have been identified as a major barrier for including it in nursing curricula . Recommendations have been made regarding key content for curricula to provide a strong foundation for cultural Pdf_Folio:118 humility with LGBTQI2S patients , yet there is no evidence that these recommendations have been implemented in nursing curricula. Providing nurses with a better understanding of LGBTQI2S health issues and inequalities would reduce barriers faced by this population when accessing health care . The CIHR Institute of Gender and Health, in conjunction with Hacking Health, a volunteer-run organization that hosts design thinking events, created a 2-day KTE development initiative based on the theme of Innovative Thinking to Support LGBTQI2S Health and Wellness . Trainee awards funded the authors and other graduate students and researchers from across Canada to attend this event. We were teamed with local health-care providers, LGBTQI2S community members, and marketing experts to develop campaigns, services, or products to disseminate research evidence to improve health outcomes . --- DESIGN JAM PROCESS The format of this event was a design jam, a modality within design thinking methodology very similar to a hackathon. Design thinking involves five steps: 1. Empathize: Consider the population experiencing the health challenge. 2. Define: Focus on the specific problem. Within the design thinking process, design jams are a modality created by the computer science and technology industries . In a design jam, a focused problem is framed by the organizers and sponsors, and a solution product is developed by interdisciplinary teams . Hackathons, similar technology-focused events, include a competitive element . The design jam we attended included prizes that supported project development, but the competition was not the main focus. Our design jam team consisted of two graduate nursing students , a graduate student in health policy , a facilitator from a major LGBTQI2S advocacy group, and a graphic artist. --- Empathizing and Defining The design jam started with an orientation to design thinking and LGBTQI2S health. Defining the problem more clearly was informed by the orientation, our expertise in LGBTQI2S health, and communication with the LGBTQI2S community members present. Our group decided that creating a product for nurses to increase knowledge and skill in providing positive and safe care to LGBTQI2S patients would address the knowledge-to-action gap for LGBTQI2S care. --- Ideating and Prototyping Brainstorming activities facilitated the description of the population who would use our product, what they would need and want, and possible means by which we could impart the necessary knowledge. Our group envisioned the diversity of nurses who might engage with a KTE product to learn more about providing safe care to the LGBTQI2S population. We designed an online educational tool kit with learning objectives, curated resources, virtual simulation games, self-assessments, and reflection opportunities. We pitched this product to the judges and stakeholders at the end of the design jam. --- Testing We tested our prototype on other health-care participants at the design jam. We engaged with members of the LGBTQI2S community represented in the product plan to ensure that the format and content were accurate. --- Recommendations have been made regarding key content for curricula to provide a strong foundation for cultural humility with LGBTQI2S patients, yet there is no evidence that these recommendations have been implemented in nursing curricula. --- Pdf_Folio:119 The graphic artist admitted that she was drawing on popular television drama representations of nurses in her descriptions of young, stylish, energetic feminine nurses who were working in obstetrics or pediatrics, single, and dating heterosexually. The tool kit consists of four case-based scenarios and four rapid-fire scenarios . After each scenario the learner is presented with strategies for debriefing and self-reflective questions. --- Ongoing Project Development In order to allow for continual professional development, the tool kit also included curated resources that can be used in practice . --- Critique of the Design Jam Process Our team reflected on the facilitators and barriers contained in the design jam process. Our experience did confer a number of facilitators and benefits common to design jams and hackathons: informal learning and practice in collaboration, communication, delegation, fast-paced and time-sensitive project management, pitching, and teamwork, as well as networking opportunities . All of the teams at the design jam produced workable outputs in the form of a knowledge translation product prototype and a plan for development . The event was well run, including a welcoming environment, play and growth opportunities for the participants that embraced a beginner's view of design thinking and creative products, non-intimidating project development, and use of prize incentives . Author reflection on participation: It was a good networking opportunity to meet other people working diligently to identify and address issues surrounding LGBTQI2S health and wellness and healthcare provision issues . . . I found it a great way to get empowered in the esoteric area of LGBTQI2S health promotion, and hugely beneficial to learn and know that good things are happening across Canada [and] to develop a way to improve LGBTQI2S educational content in nursing. Part of the design process involves envisioning the typical end user of the product. Our experience was that the qualities and characteristics we were asked to describe risked stereotyping health providers. Relying on stereotypes would also risk creating products that did not fit the reality of the target audience. Author reflection on participation: The exercise was to fully imagine the end user. Describe them in rich detail: their age, what they looked like, their job, friends, car, hobbies, family. We had decided that our end user was a nurse. It was notable to me how the depictions reflected many nursing stereotypes. The graphic artist admitted that she was drawing on popular television drama representations of nurses in her descriptions of young, stylish, energetic feminine nurses who were working in obstetrics or pediatrics, single, and dating heterosexually. The research and testing phases of the process relied on a small number of community members local to the event, in a major Canadian city with a well-developed LGBTQI2S community and a strong health and social services net. The experiences of community members there would not necessarily translate across Canada. Considerations for LGBTQI2S people in smaller cities and rural and remote areas should be considered. Similarly, there were few middle-aged, elderly, or Two-Spirit community members. Pdf_Folio:120 Design thinking innovations involving community collaborations may be more acceptable and effective compared to more expert-driven methods. A series of scaffolded design thinking events could focus attention on various processes in knowledge translation product development : focused networking events such as unconferences for fulsome community engagement, design jam engagement for solution ideation, and finally a hackathon for product prototyping, then testing by community members. As this process occurs over a longer time period, reflection and adjustments can be made to correct oversights. Follow-up support on product development, project management, and dissemination is a step seen as crucial by design thinking experts . Author reflection on participation: It was a whirlwind experience. Fast paced with not a lot of time to process or reflect on the decisions being made. The event was held over two very long days. It was a challenge to learn about the design thinking process, develop a protocol, and present a sales pitch to "sell" our proposed product in less than 48 hours. Although we got it all done, it would have been better if it was not condensed into the short timeframe. --- IMPLICATIONS AND USE OF DESIGN THINKING IN NURSING Health care requires innovations that are appropriate, evidence based, and that represent creative and imaginative solutions to problems . Design thinking allows nurses to creatively develop solutions to improve practice as well as creative KTE products. Design thinking innovations involving community collaborations may be more acceptable and effective compared to more expert-driven methods . Device designing, improving interventions and patients' experiences, teaching, and developing community health programs have all been implemented in health care using design thinking . --- CONCLUSION Design thinking methodology has potential to develop innovative and creative products for nursing practice and education. This article explored our experience at a design thinking event to address the knowledge-to-action gap in health care for LGBTQI2S people. Through the design jam process, we were able to create an online education tool kit that aims to provide nurses with the knowledge and skills to provide safe, quality care for LGBTQI2S people. --- Author reflection on participation: The research portion was informed by a group of local community members. As we needed to talk to anyone, an announcement with the particulars of the community members sought was made: "Any trans men or women who transitioned as youth please go to table 5 for research." Many of the announcements were repeated as it became obvious that there were no community or researcher participants that fit the description being requested. Pdf_Folio:121

Design thinking methodology is a collaborative strategy with the potential to create innovations. Design thinking is being used increasingly in health care. Design jams are interdisciplinary events that bring together experts and community members to collaborate on creative solutions to health-care problems. This article describes the design thinking process and includes reflection on the authors' participation in a design jam event aimed to address the knowledge-to-action gap that exists in health care for (LGBTQI2S) people.

Introduction Along with the advances in communication and information technology and the popularity of smartphones and other mobile terminals, massive quantities of social media data are produced and propagated [1][2][3]. Social media sites such as Twitter, Facebook and Weibo provide venues for the public to share opinions, exchange insights and vent emotions without intermediaries or delay. On these sites, individuals, organizations and communities can share information in real time in the virtual world via their posts and sharing of content [4,5]. Therefore, researchers increasingly regard social media data as a new type of data to study human behavior at an unprecedented scale [6][7][8]. A mass gathering, which is a typical human behavior, is defined as an event attended by a large crowd of spectators and participants [9,10]. An emergency event at a mass gathering is a sudden, urgent, and usually unexpected event at a crowd gathering that can very easily become a crowd disaster. An example of this is the Shanghai New Year's Eve stampede on 31 December 2014, when 36 people were killed and some 47 others were injured [4]. Similar to most other crowd disasters, one of the major reasons for this tragedy was that management officials incorrectly estimated the number and density of participants, and the impact of psychological factors, such as crowd mood, was not properly considered [4,11,12]. According to Berlonghi [13], the safety of a mass gathering event largely depends on crowd management, which is a challenging task because both human flow control and psychological monitoring must be simultaneously considered. Human stampedes, as a major cause of mortality in mass gatherings, are incompletely investigated complex phenomena, and prevention strategies remain insufficient [14,15]. A large quantity of social media data makes it possible to detect emergency events or the status of an assembled crowd. Leveraging social media big data for emergency response and disaster management has attracted considerable attention [16][17][18]. Social media data associated with temporal and geographical information can reveal human behavior patterns [19][20][21], and human emotion changes can be reflected by social media content [22][23][24][25]. Thus, many researchers have applied social media methods to study natural disasters such as earthquakes [26,27], forest fires [28,29] and floods [30,31]. Zheng et al. [32] aimed to detect urban emergency events such as fires, fights, and crashes based on social media data but did not discuss events such as crowd gatherings, which were treated as a process rather than an emergency. In addition, an increasing number of studies have explored ways to monitor crowds by predicting the daily routine trajectories of individuals. However, such an approach cannot detect crowd anomalies that will lead to crowd disasters in an emergency because crowd anomalies are generally caused by infrequent collective activities [11]. Zhou et al. [11] used big data from the Baidu map to establish early warnings of crowd disasters; however, they based the study mainly on users' query data and did not consider users' emotions. Ngo et al. proposed a crowd-monitoring framework [33] that combines an emotional analysis of Twitter data and widely-adopted crowd models. Martínez-Castaño et al. proposed a platform incorporating a huge amount of social media data for real-time analysis of the incidence of depression in the population. Nevertheless, these frameworks might be applicable only for English or Spanish posts, and they do not analyze the influence of spatiotemporal neighborhoods [34]. In general, most of the literature has focused on the effects of human activities on relatively extreme types of natural disasters or on daily routine behavior. Additionally, research on Chinese social media remains rather limited compared with studies of English-based services like Twitter and Facebook. We employed social media big data to identify patterns, trends and associations relating to the 2014 Shanghai New Year's Eve Stampede. Social media data-based approaches have advantages in accurately and multidimensionally tracking human mobility, reflecting crowd sentiment and identifying human behavior in an emergency. Techniques including spatial autocorrelation analysis, topic modeling and sentiment analysis in the Chinese context are applied in this research. To the best of our knowledge, the proposed work is the first to study crowd detection in mass gatherings based on social media data in Chinese, with spatiotemporal information. This work demonstrates the potential for further applications of data-driven emergency response and management. This paper is structured as follows: Section 2 introduces the study area, experimental data and methods used in the research. Section 3 outlines the results of spatial autocorrelation, topic modeling and sentiment analysis. Section 4 summarizes contributions and suggests future work. Finally, Section 5 draws a conclusion. --- Methodology --- Study Case Description Shanghai, located in the Yangtze River Delta in the middle portion of the East China coast, had a population of more than 24 million in 2014. The administrative boundaries of the city are shown in Figure 1. The Shanghai New Year's Eve stampede occurred near Chen Yi Plaza on the Bund on 31 December 2014. Approximately 310,000 people had gathered for the New Year celebration. The stampede was a disaster that killed 36 people and injured 49 others [4]. The Bund is a waterfront area in central Shanghai . Centered on Zhongshan Road , this area faces the modern skyscrapers of Lujiazui in the Pudong District. The Bund usually refers to the buildings and docks along this stretch of road, together with some adjacent areas. Chen Yi Plaza, lying in the middle of the Bund, links to the Huangpu River platform by a large and scenic staircase, which attracted a huge visitor flow, causing the highest population density during the New Year celebration [35]. In our research, we established a 10-km buffer zone for the Bund as representative of the Bund area . At approximately 20:00 on 31 December, more people were arriving at the Bund than were leaving. There was a continuous flow of visitors onto the platform, and they gradually huddled. According to the estimations from the government, there were approximately 120,000 people on the Bund at approximately 20:00, 160,000 from 21:00 to 22:00, 240,000 from 22:00 to 23:00 and 310,000 from 23:00 until the incident occurred. Finally, at 23:35, under pressure from the crowd, people at the bottom of the stairs were knocked to the ground, which triggered the eventual deadly stampede. As shown in Figure 2, the check-in density varied significantly over time, and the density from 18:00 to 24:00 on 31 December, when the stampede occurred, was apparently higher than at other times . The Bund is a waterfront area in central Shanghai . Centered on Zhongshan Road , this area faces the modern skyscrapers of Lujiazui in the Pudong District. The Bund usually refers to the buildings and docks along this stretch of road, together with some adjacent areas. Chen Yi Plaza, lying in the middle of the Bund, links to the Huangpu River platform by a large and scenic staircase, which attracted a huge visitor flow, causing the highest population density during the New Year celebration [35]. In our research, we established a 10-km buffer zone for the Bund as representative of the Bund area . The Bund is a waterfront area in central Shanghai . Centered on Zhongshan Road , this area faces the modern skyscrapers of Lujiazui in the Pudong District. The Bund usually refers to the buildings and docks along this stretch of road, together with some adjacent areas. Chen Yi Plaza, lying in the middle of the Bund, links to the Huangpu River platform by a large and scenic staircase, which attracted a huge visitor flow, causing the highest population density during the New Year celebration [35]. In our research, we established a 10-km buffer zone for the Bund as representative of the Bund area . At approximately 20:00 on 31 December, more people were arriving at the Bund than were leaving. There was a continuous flow of visitors onto the platform, and they gradually huddled. According to the estimations from the government, there were approximately 120,000 people on the Bund at approximately 20:00, 160,000 from 21:00 to 22:00, 240,000 from 22:00 to 23:00 and 310,000 from 23:00 until the incident occurred. Finally, at 23:35, under pressure from the crowd, people at the bottom of the stairs were knocked to the ground, which triggered the eventual deadly stampede. As shown in Figure 2, the check-in density varied significantly over time, and the density from 18:00 to 24:00 on 31 December, when the stampede occurred, was apparently higher than at other times . At approximately 20:00 on 31 December, more people were arriving at the Bund than were leaving. There was a continuous flow of visitors onto the platform, and they gradually huddled. According to the estimations from the government, there were approximately 120,000 people on the Bund at approximately 20:00, 160,000 from 21:00 to 22:00, 240,000 from 22:00 to 23:00 and 310,000 from 23:00 until the incident occurred. Finally, at 23:35, under pressure from the crowd, people at the bottom of the stairs were knocked to the ground, which triggered the eventual deadly stampede. As shown in Figure 2, the check-in density varied significantly over time, and the density from 18:00 to 24:00 on 31 December, when the stampede occurred, was apparently higher than at other times . --- Sina Weibo Microblog Check-In Data Sina Weibo is a popular social networking website in China and has many similarities to the Twitter microblog. In addition, Weibo has been widely used in disasters and emergencies in China due to its sufficient data availability [36][37][38]. Individual users can post an emergency in near-real time, and the administrators utilize this information to broadcast warnings, notices and urgent news in a timely manner. We obtained the Weibo data from Weiboscope at the University of Hong Kong [39]. Check-in Weibos, accounting for approximately 1% of posts, are Weibo posts with geotags. Check-ins on social media involve users confirming their location and safety when an incident occurs at a specific location [40][41][42]. In addition, check-ins can involve users automatically sharing their location when they transmit messages on websites [43]. Sina Weibo check-in data include both types of location information. Thus, the data can provide geographic coordinate records of user activities and the context of user posts over a specific period. During the Shanghai New Year's Eve stampede, thousands of Weibos were posted. In this work, we collected all Weibos with a location in Shanghai, China, between 24 December 2014, and 7 January 2015 . After filtering duplicate posts, 46,193 Weibos were used in the study. --- Proposed Methods The analysis framework used in this paper is shown in Figure 3. The Weibo check-in data in this research were acquired from the data source by selecting the points within Shanghai between 24 December 2014, and 7 January 2015. We focus on the check-in data in Shanghai Bund for further investigation. The check-in data from 31 December 2014, at 20:00 to 1 January 2015, at 12:00 are analyzed hour-by-hour. The location information from Weibo check-in data over time are aggregated to analyze the temporal and spatial evolution of the stampede, mainly based on spatial autocorrelation analysis with Moran's I. The text content of Weibo posts is analyzed by the topic modeling method, from which we derive ten stampede-related topics and identify the percentages of and temporal changes in them. Sentiment analysis is also applied to the text content of Weibo posts to categorize them into five groups, ultimately determining the percentages of and temporal changes in negative posts. In our work, this framework is applied to assess the practicality of social media data for crowd detection in mass gatherings. --- Sina Weibo Microblog Check-In Data Sina Weibo is a popular social networking website in China and has many similarities to the Twitter microblog. In addition, Weibo has been widely used in disasters and emergencies in China due to its sufficient data availability [36][37][38]. Individual users can post an emergency in near-real time, and the administrators utilize this information to broadcast warnings, notices and urgent news in a timely manner. We obtained the Weibo data from Weiboscope at the University of Hong Kong [39]. Check-in Weibos, accounting for approximately 1% of posts, are Weibo posts with geotags. Check-ins on social media involve users confirming their location and safety when an incident occurs at a specific location [40][41][42]. In addition, check-ins can involve users automatically sharing their location when they transmit messages on websites [43]. Sina Weibo check-in data include both types of location information. Thus, the data can provide geographic coordinate records of user activities and the context of user posts over a specific period. During the Shanghai New Year's Eve stampede, thousands of Weibos were posted. In this work, we collected all Weibos with a location in Shanghai, China, between 24 December 2014, and 7 January 2015 . After filtering duplicate posts, 46,193 Weibos were used in the study. --- Proposed Methods The analysis framework used in this paper is shown in Figure 3. The Weibo check-in data in this research were acquired from the data source by selecting the points within Shanghai between 24 December 2014, and 7 January 2015. We focus on the check-in data in Shanghai Bund for further investigation. The check-in data from 31 December 2014, at 20:00 to 1 January 2015, at 12:00 are analyzed hour-by-hour. The location information from Weibo check-in data over time are aggregated to analyze the temporal and spatial evolution of the stampede, mainly based on spatial autocorrelation analysis with Moran's I. The text content of Weibo posts is analyzed by the topic modeling method, from which we derive ten stampede-related topics and identify the percentages of and temporal changes in them. Sentiment analysis is also applied to the text content of Weibo posts to categorize them into five groups, ultimately determining the percentages of and temporal changes in negative posts. In our work, this framework is applied to assess the practicality of social media data for crowd detection in mass gatherings. --- Spatial Autocorrelation Spatial autocorrelation generally describes the correlation among variables considering the relative position dependence of spatial units [44]. If adjacent units tend to have similar values, a positive spatial autocorrelation appears, and if adjacent units tend to have different values, a negative --- Spatial Autocorrelation Spatial autocorrelation generally describes the correlation among variables considering the relative position dependence of spatial units [44]. If adjacent units tend to have similar values, a positive spatial autocorrelation appears, and if adjacent units tend to have different values, a negative spatial autocorrelation will appear [45]. The basic measure of spatial autocorrelation is the spatial autocorrelation coefficient. Moran's I is the most common index, defined as the ratio between the local and global consistencies, as shown in Formula : I = n n i=1 n j=1 w ij x j -x n i=1 n j=1 w ij n i=1 2 where n denotes the number of spatial units, x i and x j denote the values of the ith and jth units, x denotes the mean of x i , and w ij denotes the spatial weight matrix between the ith and jth units. In our work, to quantify the distribution of check-ins, we establish regular grids to divide the study area and determine the quantity of check-ins in each grid. The grid size is 2 × 2 . We regard each grid as a basic spatial unit to calculate Moran's index I. --- Topic Modeling and Sentiment Analysis in Chinese Topic modeling aims at discovering hidden semantic information or knowledge in a large amount of text data. Latent Dirichlet allocation , as one of the most common topic models, is a generative probabilistic model and is specifically intended for usage on text corpora [46]. The model assumes that the document is a bag of words and analyzes the main discourse in the text corpus and divides the concurrent words into topics. Latent semantic analysis is another important method to identify topics in a database of text. In this research, we select LDA instead of LSA because these two methods exhibit highly overlapping topics and LDA outperformed LSA in large test cases [47,48]. In addition, LDA has proved its adaptability for Chinese documents [49,50], while related work with LSA is limited. In LDA models, each document can be regarded as a collection of latent topics, and each of the topics can be further illustrated by a set of keywords. Specifically, we use D to denote a text corpus, which is an assembly of documents, and N D to denote the number of documents in D; V is a set of words in documents, and N V is the number of words; N T is the number of topics. The generation process of the LDA model is as Algorithm 1 [51,52]. Algorithm 1 The Generation Process of the LDA Model for all topics k ∈ [1, N T ] choose a word distribution → ϕ k ∼ dirichlet for all the documents d ∈ [1, N D ] choose a topic distribution → θ d ∼ dirichlet for word n ∈ [1, N d ] choose the topic of the n th word: Z d,n ∼ Multinomial choose the n th word: W d,n ∼ Multinomial In this procedure, → ϕ k is the word distribution for topic k; We select the open-source toolkits Jieba and Gensim for Chinese text segmentation and faster implementation of LDA, respectively. These toolkits are based on Hoffman [53], allowing both LDA model estimation from a training corpus and inference of topic distribution. As a result, the probability distribution of each Weibo post associated with the top 10 topics is computed and rescaled. In this work, we choose the topic that has the maximum probability as the Weibo topic. Sentiment analysis is a natural language processing task involving the computational treatment of opinions, sentiments and subjectivity in text [54][55][56]. In this research, we use the BosonNLP toolkit with an engine that uses the Weibo corpus for annotation and training. The BosonNLP toolkit provides a sentiment analysis application programming interface especially for Weibo data. --- Results and Discussion --- Results of Spatial Autocorrelation We calculated the number of check-ins in Shanghai and near the Bund each day from one week before to one week after the stampede , as shown in Figure 4. The red line in Figure 5 illustrates that the number of check-ins in Shanghai from 31 December 2014, to 1 January 2015 was almost two times the peak value at other times. Additionally, the blue line, which represents the number of check-ins around the Bund, shows that there was a peak in Weibos during the disaster period. Furthermore, both the red line and the blue line show that the number of check-ins on New Year's Eve in 2014 apparently exceeded the number of check-ins on Christmas Day. --- Results and Discussion --- Results of Spatial Autocorrelation We calculated the number of check-ins in Shanghai and near the Bund each day from one week before to one week after the stampede , as shown in Figure 4. The red line in Figure 5 illustrates that the number of check-ins in Shanghai from 31 December 2014, to 1 January 2015 was almost two times the peak value at other times. Additionally, the blue line, which represents the number of check-ins around the Bund, shows that there was a peak in Weibos during the disaster period. Furthermore, both the red line and the blue line show that the number of check-ins on New Year's Eve in 2014 apparently exceeded the number of check-ins on Christmas Day. Based on the method in Section 2.3, we calculate Moran's I for the number of check-ins and plot the corresponding distribution over time. As shown in Figure 5, the red line is Moran's I in Shanghai, with an average value of 0.601. The distribution of check-ins displays strong spatial autocorrelation with the municipal scale. In addition, we can observe several high-spatial aggregation areas in Shanghai. In particular, as indicated by the red line, the peak value of Moran's I appeared on 1 January 2015. Combined with Figure 5, which shows a peak in the number of check-ins, we can conclude that the population density on New Year's Eve was higher than at other times. The blue line in Figure 6 shows Moran's I around the Bund, from which we can conclude that there was considerable spatial aggregation in the Bund area , but this finding is not as obvious as that at the municipal scale. However, Moran's I on New Year's Eve was apparently higher than the index values on the days before and after the stampede, as well as the average value of the blue line. Based on the method in Section 2.3, we calculate Moran's I for the number of check-ins and plot the corresponding distribution over time. As shown in Figure 5, the red line is Moran's I in Shanghai, with an average value of 0.601. The distribution of check-ins displays strong spatial autocorrelation with the municipal scale. In addition, we can observe several high-spatial aggregation areas in Shanghai. In particular, as indicated by the red line, the peak value of Moran's I appeared on 1 January 2015. Combined with Figure 5, which shows a peak in the number of check-ins, we can conclude that the population density on New Year's Eve was higher than at other times. The blue line in Figure 6 shows Moran's I around the Bund, from which we can conclude that there was considerable spatial aggregation in the Bund area , but this finding is not as obvious as that at the municipal scale. However, Moran's I on New Year's Eve was apparently higher than the index values on the days before and after the stampede, as well as the average value of the blue line. In general, Moran's I is higher for festivals and on weekends, suggesting that crowd spatial aggregation is more likely to occur on non-workdays. Combined with the crowd anomaly shown in Figure 4, we can intuitively detect the crowd flow and spatial aggregation distribution by analyzing the number and Moran's I of check-ins. In general, Moran's I is higher for festivals and on weekends, suggesting that crowd spatial aggregation is more likely to occur on non-workdays. Combined with the crowd anomaly shown in Figure 4, we can intuitively detect the crowd flow and spatial aggregation distribution by analyzing the number and Moran's I of check-ins. --- Results of Topic Modeling We extracted the posts made in the Bund area from 31 December 2014, at 18:00:00 to 1 January 2015, at 12:00:00 for topic modeling. In the preprocessing stage, we removed words related to the topic "happy New Year," which were dominant at this special event and could potentially influence the topic modeling results. After preprocessing, 1280 posts were deemed valid and used to extract topics. To avoid a high degree of similarity between topics, while maintaining the clarity of each topic, we set the number of topics to 10. We defined a topic to be related to the stampede if at least one of the most frequent keywords was related to a crowd, stampede, or mass gathering. Seven of the 10 topics are related to the studied stampede, accounting for 72.4% of all posts. The topics are as follows. Topic 1: There was a stampede in Shanghai Bund, I was fortunate that I missed it yesterday. Topic 2: Hello, Shanghai Bund. There are many people walking here. Topic 3: We are having fun among many people. Topic 4: Good morning everybody. Wish you good luck. I feel blessed that I stayed with family yesterday. Topic 5: I still feel happy despite the cold wave. Topic 6: Happy days are coming with a very nice breakfast. Topic 7: I will start working hard from the early morning. Topic 8: Sharing songs for the dead. May the dead rest in peace. Topic 9: Many people died tonight. We should stay calm in mass gatherings. Topic 10: Hoping the dead from the stampede rest in peace. The extracted 10 topics can be divided into four classes, and the percentage of each class is shown in Figure 6. The first class includes Topic 2, Topic 3 and Topic 9, which mainly express complaints about the huge crowds around the Bund. The key words for these topics are "huge crowd", "Bund" and "death." The posts associated with these topics account for 30.1% of all posts. The second class contains Topic 1 and Topic 4, and a common theme is people expressing relief about not going to the Bund. "Stampede", "gloat" and "being alive" are words appearing frequently in these topics, and 24.3% of the posts are associated with such topics. The third class involves blessings and includes Topic 7 and Topic 10. The most important words for these topics are "rest in peace" and "pray for the deceased and injured." Posts associated with these topics account for only 18.1% of all posts. The rest of the topics are not related to the stampede, including Topic 5, Topic 6, and Topic 8. Figure 7 shows the usage frequency of each class in each hour from 31 December 2014 at 18:00:00 to 1 January 2015 at 12:00:00. The line chart in Figure 7 shows the number of check-ins in each hour. As shown in Figure 7, omitting three hours with very few check-ins, the proportion of the first class from 31 December at 20:00 to 1 January at 3:00 was 31.77%, higher than those at other times. The proportion of the first class from 1 January 2014, at 1:00 to 1 January 2015, at 2:00 was the highest, reaching 44.73%. Thus, people complained about crowding beginning at 20:00, which basically matches the process described in the report [4]. For the second class, the proportion of posts was higher after 1 January at 7:00 than on 31 December, suggesting that most people talked about the absence of celebration around the Bund the day after the event. For the check-in data outside the Bund in Shanghai, only one topic is related to the stampede. This low proportion also reflects the significant spatial autocorrelation characteristics of the stampede based on check-in data. 20:00, which basically matches the process described in the report [4]. For the second class, the proportion of posts was higher after 1 January at 7:00 than on 31 December, suggesting that most people talked about the absence of celebration around the Bund the day after the event. For the checkin data outside the Bund in Shanghai, only one topic is related to the stampede. This low proportion also reflects the significant spatial autocorrelation characteristics of the stampede based on check-in data. --- Results of Sentiment Analysis In our research, we calculate a score between 0 and 1 for each post from 31 December 2014, at 18:00:00 to 1 January 2015, at 12:00:00 around the Bund. A high score indicates a positive post, and vice versa. For convenience in determining changes in sentiment, we divide the sentiment score into five groups with an interval of 0.2. Group 1 includes scores between 0 and 0.2, which is the group with the most negative sentiment; Group 5 includes scores between 0.8 and 1, which is the group with the most positive sentiment. The histogram in Figure 8 shows the usage frequency of each group in each hour, and the line chart shows the number of check-ins in each hour. Excluding the three hours with very few check-ins, we can conclude that the proportion of negative sentiment increased in two periods: from 31 December at 20:00 to 1 January at 2:00 and on 1 January from 10:00 to 12:00 . The first period was when the stampede occurred. In addition, Group 3 accounted for a minor proportion of posts during the event, suggesting that people shared an apparent sentiment inclination. For comparison, the proportion of negative sentiment from 31 December at 18:00 to 31 December at 20:00 is 27.83%. --- Results of Sentiment Analysis In our research, we calculate a score between 0 and 1 for each post from 31 December 2014, at 18:00:00 to 1 January 2015, at 12:00:00 around the Bund. A high score indicates a positive post, and vice versa. For convenience in determining changes in sentiment, we divide the sentiment score into five groups with an interval of 0.2. Group 1 includes scores between 0 and 0.2, which is the group with the most negative sentiment; Group 5 includes scores between 0.8 and 1, which is the group with the most positive sentiment. The histogram in Figure 8 shows the usage frequency of each group in each hour, and the line chart shows the number of check-ins in each hour. Excluding the three hours with very few check-ins, we can conclude that the proportion of negative sentiment increased in two periods: from 31 December at 20:00 to 1 January at 2:00 and on 1 January from 10:00 to 12:00 . The first period was when the stampede occurred. In addition, Group 3 accounted for a minor proportion of posts during the event, suggesting that people shared an apparent sentiment inclination. For comparison, the proportion of negative sentiment from 31 December at 18:00 to 31 December at 20:00 is 27.83%. --- Discussion In this study, we propose an innovative framework from two dimensions of Weibo check-in data, geographic attributes and blog content, to explore the application of geotagged social media data in crowd detection. Social media data have outstanding advantages in tracking crowd change. This study demonstrates the potential of using geotagged social media data for analyzing population spatiotemporal activities, especially in emergencies. First, social media data are not limited by sparse sensor coverage. For example, in this study, only Weibo check-in data are used. Everyone in the stream is aware of the entire event timeline --- Discussion In this study, we propose an innovative framework from two dimensions of Weibo check-in data, geographic attributes and blog content, to explore the application of geotagged social media data in crowd detection. Social media data have outstanding advantages in tracking crowd change. This study demonstrates the potential of using geotagged social media data for analyzing population spatiotemporal activities, especially in emergencies. First, social media data are not limited by sparse sensor coverage. For example, in this study, only Weibo check-in data are used. Everyone in the stream is aware of the entire event timeline through the timestamp, geographic location, and semantic information in the social media data. Thus, processes such as event feedback do not require additional sensors or costs. Second, social media data are real-time data. The Weibo check-in data used in this paper can reveal the aggregation situation of the crowd in near-real time. Social media data can be obtained in real time through the API provided by the social media operator. At the hourly and daily scales, social media data have proved effective for detecting anomalies in real time [57]. Finally, social media data can be used in multidimensional analyses. For example, Weibo check-in data provide multiple features, such as space, time, and semantic features, to comprehensively analyze crowd change. Notably, the analysis of Weibo posts reflects the influence of user psychology and activities before and after the stampede. Performing such tasks is difficult based on video crowd detection [58] and mobile phone crowd detection [59]. Social media data can provide a new perspective for crowd gathering detection. Based on these characteristics, social media data, especially geotagged social media data, can provide different dimensions for crowd detection and help the relevant government entities make good decisions. However, the application of social media data in crowd detection still retains some challenges and requires further research. The main points are as follows. Improved data-filtering methods and more powerful NLP models are needed to improve the accuracy of the results. This research considers only Weibo posts and check-ins and does not include Weibo comments. In the future, we can add Weibo comment data and relationship chain information from comments to the analysis. For example, the impact range of the stampede event can be evaluated in combination with the comments. In cases of disaster recovery, this approach can enable more in-depth analysis of the event process. To efficiently extract relevant user information and classify microblog users, the current framework does not take user preferences into account. In future work, the features of different users can be extracted by using models, assigning different weights to Weibos or input features directly in an NLP model as parameters, and eliminating bias associated with user preferences in social media. --- Conclusions In this paper, we analyzed the crowd aggregation level, topic changes and sentiment changes of a crowd based on Weibo check-in data from the 2014 Shanghai New Year's Eve stampede. For this purpose, we introduced Moran's I in spatial autocorrelation analysis to quantitatively reflect the aggregation level and performed an analysis of posts in Chinese to reveal the psychological changes of users in the crowd. In addition, we established a 10-km buffer zone for the Bund and consequently differentiated the study area from the greater Shanghai area and the Bund area for analysis purposes. In addition, we divided the study area into equivalent grids to calculate Moran's index I. We also considered changes in a given time period-from a week before to a week after the event-to examine people's reactions in time and space. The findings of this study are as follows. First, combined with the number of check-ins, Moran's I can reveal the crowd aggregation level. The case study shows that a high index with an unusual number of check-ins suggests that a mass gathering has likely occurred. In this case, we can quantitatively detect the crowd flow and spatial aggregation distribution by calculating Moran's I of check-ins. Second, the results of topic modeling show that the proportion of the crowd that complained about the huge crowds in the Bund area increased beginning 31 December at 20:00 and remained at a high level until 1 January at 3:00. Moreover, topics about the absence of celebration increased in proportion at 7:00 on 1 January. Finally, sentiment analysis based on the posts indicated that people with negative sentiments were most active from 31 December at 20:00 to 1 January at 2:00 . In addition, the results demonstrate that, during the event, people generally showed an apparent sentimental inclination. We believe that the huge flows of people and the arrival of the new year triggered crowding, which led to both complaints and celebration. The results show that it is feasible to detect crowd aggregation by social media data. On the one hand, the geographical attributes of social media can directly reflect changes in the flow of people over time; on the other hand, through analysis of Weibo posts, the effects of people's psychological states before and after an incident can be identified. We believe that the proposed framework will be further applied in data-driven emergency response and management. Timely collected data via social media with geotags from cell phones was used to derive a real-time population distribution map and its spatiotemporally changing mode. Such massive accurate and multidimensional positioning data can be used to supervise the flow direction of crowds, control traffic, evacuate crowds and rescue wounded individuals. Additionally, using natural language-processing methods to identify people's emotions in emergencies using social media data will aid in providing mental health care to help people overcome difficulties and recover from the disasters. Although general patterns can be reflected by social media data, there are still challenges in assessing the psychological status of a population with high precision. This is an open question in data-driven emergency response and management. ---

The Shanghai New Year's Eve stampede on 31 December 2014, caused 36 deaths and 47 other injuries, generating attention from around the world. This research aims to explore crowd aggregation from the perspective of Sina Weibo check-in data and evaluate the potential of crowd detection based on social media data. We develop a framework using Weibo check-in data in three dimensions: the aggregation level of check-in data, the topic changes in posts and the sentiment fluctuations of citizens. The results show that the numbers of check-ins in all of Shanghai on New Years' Eve is twice that of other days and that Moran's I reaches a peak on this date, implying a spatial autocorrelation mode. Additionally, the results of topic modeling indicate that 72.4% of the posts were related to the stampede, reflecting public attitudes and views on this incident from multiple angles. Moreover, sentiment analysis based on Weibo posts illustrates that the proportion of negative posts increased both when the stampede occurred (40.95%) and a few hours afterwards (44.33%). This study demonstrates the potential of using geotagged social media data to analyze population spatiotemporal activities, especially in emergencies.

Introduction In low and middle-income countries, the prevalence of alcohol uses at least once in the past 30 days was 8.4% among early teenagers aged 12-13 years, [1]. The Youth Risk Behaviour Survey 2013 showed that 16% and 8.7% of male and female teenagers respectively have started alcohol consumption before 13 years of age. In a longitudinal study to delay alcohol use and sexual initiation among learners, alcohol consumption increased from 34.6 to 39.9% between the two points [2]. The study further revealed that the proportion of learners reporting alcohol consumption was 45.6% and 26.4% among males and females respectively. Studies elsewhere have shown that testing and sipping of alcohol began between the age of 10 to 14 or even younger [3,4]. There is evidence that the strongest association between underage drinking and alcohol availability was found among children aged 12-14 years [5,6]. Alcohol remains one of the main three leading health factors for global disease burden [7], with underage alcohol consumption presenting grave health problems. Three million death was attributed to alcohol use annually in 2016 [8]. Studies have associated alcohol drinking with mental health disorders, suicidal behaviour [9,10], Chronic liver disease [10,11] and violence [12][13][14]. A national longitudinal study of adolescents from grade 7 to grade 12, showed that early initiation of alcohol use translated to a higher level of alcohol use in adulthood [15]. Other studies have reported alterations in the maturation and functioning of the adolescent brain because of early exposure to alcohol use [16,17]. Advertising alcohol has been recognised as an instigator of alcohol misuse and health risk to drinkers. Several studies have associated exposure to alcohol advertisements with alcohol consumption among adolescents [18][19][20][21][22]. In South Africa, a qualitative study conducted among 40 high school learners showed that alcohol consumption was motivated by alcohol advertisements in Eastern Cape [23]. In another study of alcohol availability and marketing among rural and urban youth in South Africa, [24] reported that consumption of alcohol and binge drinking were induced by alcohol advertising and marketing. A systematic review of alcohol adverts and prospects for alcohol consumption in developing countries showed that alcohol advertising predicted initiating alcohol drinking among children and youths [25]. Similar evidence of alcohol adverts leading to alcohol consumption among children and adolescents has been documented in the US [26], Thailand [27,28] and Australia [22]. Self-reported advertisement exposure on four types of media was associated with increased drinking over time among 1877 youths in multilevel analytical studies in which other covariates were controlled [29]. However, Gunter et al. [30] did not find any association between exposure to any type of alcohol advertising and general alcohol consumption. In a randomized control trial study, exposure to alcohol advertisements in grade six did not predict drinking behaviour amongst grade six drinkers or grade eight non-drinkers [31]. In addition, Jones demonstrated that other factors play a more decisive role in alcohol initiation among young people than alcohol advertisements in Australia [22]. Furthermore, in many parts of the world, the locally-made alcohol that people consume was never advertised [32]. These suggest that the association between alcohol advertisement and consumption may be mediated through other factors. There is evidence that a spectrum of other factors in the environment of young people promotes alcohol consumption. Risk factors studies have documented parental drinking behaviour [28,[33][34][35] the community where the adolescents were living [28] gender [28] and direct peer group influences [28,30,36,37] were associated with adolescents' alcohol use. Other studies have demonstrated an association between alcohol consumption and ethnicity [19,32]. Taking together the association between alcohol advertisement and alcohol consumption appeal to a multifaceted investigation among children and early teenagers. Exposure to media alcohol advertisements has been documented to influence alcohol consumption among young people in South Africa. Qualitative empirical studies [23,24] exist on the interaction between alcohol advertising and drinking among young people in a few provinces of South Africa. However, qualitative studies were limited in accounting for the association between alcohol adverts and alcohol use among children and early teenagers in South Africa. Morojele et al. [38] examine the role of demographic factors, exposure to alcohol marketing and linking of alcohol advertisements in predicting the use of alcohol in the past 6 months among 408 males and 461 females aged 16 and 17 years old in Tshwane Municipality. They found a positive association between the mode of alcohol exposure with alcohol use in the past 6 months. However, the study population did not account for early teenagers aged 12 to 14 years. Furthermore, the geographical coverage and 6 months assessment of the study was very limited and a short period to account for a nationally representative population of alcohol exposure and underage drinking. Hence, these studies cannot stimulate policy on alcohol adverts and underage drinking in South Africa. Thus, there exists a paucity of information between alcohol adverts and alcohol consumption among children and early teenagers. South Africa has regulations that forbid the selling of alcohol to people below the age of 18. Furthermore, the Bill to ban alcohol advertisements to improve the health of children and teenagers below the age of 18 has not been passed into law because of various competing interests. Some are concerned that banning alcohol adverts would affect economic activities and subsequently lead to job losses [39]. On 26th June 2022, twenty-one teenagers comprising 12 girls and 9 boys aged 13-17 were found dead at the Enyobeni Tavern in Scenery Park in East London on account of alcohol-related. Sacrificing the health and lives of early teenagers on the altar of economic viability and job sustenance has been a difficult stand for both researchers and policymakers concerning alcohol advertisement. There is evidence that alcohol advertisement in media affects alcohol consumption. However, this evidence is sketchy for children and early teenagers. Furthermore, it is not clear whether all forms of media advertisement for alcohol mediate alcohol consumption among these early teenagers. In addition, other background factors have been shown to be more predisposing to underage alcohol consumption. The study is a secondary data analysis extracted from the South African National HIV Prevalence, HIV Incidence, Behaviour and Communication Survey 2017. The survey is conducted every three years among individuals of all ages. This enables exploration of shifts over time on demographic and other variables, as well as allowing for investigation of new research areas such as exposure to underage alcohol exposure and consumption. Hence the dataset has robust information on the study population and dynamics in alcohol use. The study aimed to examine the association between exposure to alcohol advertising media and underage alcohol consumption while controlling for the background characteristics in South Africa. --- Methods The South African National HIV Prevalence, HIV Incidence, Behaviour and Communication Survey 2017 was a cross-sectional, population-based household survey conducted using a multi-stage stratified random cluster sampling approach [40]. SABSSM 2017 was the fifth wave in a series of national cross-sectional surveys that collected information from all population segments using households and hostels as visiting points from December 2016 to February 2018. Educational institutions, old-age homes, hospitals, and uniformed-service barracks were excluded from the sample. The 2017 survey used small area layers as the primary sampling unit instead of the enumeration areas to bypass the confidentiality issues associated with EAs having a few VPs. A SAL consists of one or more EAs and is better defined geographically than the EA [41]. Four questionnaires used in SABSSM 2017 comprised a Household questionnaire , a questionnaire for parents and guardians of children aged 0-11 years, a questionnaire for children aged 12-14 years and a questionnaire for people aged 15 years and older. However, the present study used a dataset from the questionnaire for children aged 12-14 years. The participants in the present study were 3833 early teenagers aged 12 to 14 years old. Nearly half of the participants were males. Information collected on race was categorised into Africa, coloured, Indian, and white. Over twothirds were African. Information on the nine South African provinces was recategorized into rural and urban provinces which showed that over a quarter were residing in urban designated provinces. Participants were asked about where they have seen alcohol adverts: Whether on television, poster/billboard, social media and Events. These were coded as . Posters and billboards were combined, and the components of social media were also merged. They were further asked whether they have ever consumed alcohol . In addition, recent alcohol consumption was accessed by asking about alcohol consumption in the past 12 months . --- Statistical analysis The background characteristics, alcohol exposure and alcohol consumption were examined with the age of the respondents. The relationship between alcohol advertising and ever alcohol consumption and recent alcohol consumption was examined by logistic regressions. Unadjusted model 1 for each type of alcohol advertising media and alcohol consumption was examined. In model 2 all the media of alcohol advertisements and alcohol were examined. In model 3, exposure to all alcohol advertising media and ever alcohol consumption and recent alcohol consumption were examined while controlling for age, gender, race, and province. The analyses at the multivariate level were carried out on the weighted sample to account for the population of children and early teenagers in South Africa. The results were presented as unadjusted and adjusted odds ratios with p values < 0.05 indicating statistical significance. All the analyses were carried out using IBM SPSS version 27. --- Results Table 1 presents the media of exposure to alcohol advertisement and alcohol consumption among the early teenager sample. Seeing alcohol advertisements on television was reported by nearly one-third . Over three quarters and nearly all participants indicated that they have seen alcohol advertisements on posters/billboards and events. About 9 in 10 have seen alcohol advertainments on social media. Over three quarters indicated that they have consumed alcoholic drinks while in the last 12 months about half have consumed alcohol. About 7 in 10 have engaged in sexual intercourse. --- Association between media of alcohol adverts and ever-consumed alcohol Table 2 shows seeing alcohol advertisements in television and posters/billboard was associated with alcohol consumption which intensity from [OR = 1.64 to AOR = 3.28 ] for television and from [OR = 1.38 to AOR = 1.52 ] for posters/billboard. However, reduced odds of consumption of alcohol 35% and 30% were reported among those who indicated seeing alcohol advertisements during events and on social media, which increased to 42% and 37% respectively after adjusting for background characteristics. Those aged 12 and 13 years old were more likely to report alcohol consumption compared to those aged 14 years. Furthermore, males showed 35% reduced odds of reporting alcohol consumption compared to their female counterparts. Compared to the African, Coloured, Indian, and White were less likely to report alcohol consumption. Teenagers residing in rural provinces were more likely to report alcohol consumption than their urban counterparts. --- Alcohol advertainments and alcohol consumption in the last 12 months Table 3 shows alcohol consumption in the last 12 months and the association with exposure to alcohol advertisements. Exposure to alcohol advertisements on television and posters/billboards was consistently associated with alcohol consumption in the last 12 months. Conversely, exposure to alcohol adverts during events and on social media showed reduced odds of reporting alcohol consumption. In the last 12 months, teenagers who were below the age of 14 years and males were less likely to report alcohol consumption. Compared to the African teenagers, the coloured and Whites were more likely to report alcohol consumption in the last 12 months. --- Exposure to alcohol adverts and alcohol consumption according to age Table 4 shows that after adjusting for the demographic characteristics, there was an association between alcohol consumption and exposure to alcohol advertising media. For all ages, alcohol adverts on television were associated with alcohol consumption [AOR = 3.46: ] for 12 years; [AOR = 1.50: ] for 13 years and [AOR = 7.61: ] for the 14 years old. A positive association was reported among those who reported exposure to alcohol adverts through posters/billboards. Alcohol advertising during events was associated with a reduced likelihood of alcohol consumption by age [AOR = 0.18: for 12 years; [AOR = 0.71: ] for 13 years and [AOR = 0.69: ] for the 14 years old. Reduced odds 79%; 35% and 48% for consuming alcohol were reported by 12, 13 and 14 years old respectively for those who were exposed to alcohol adverts on social media. Males were less likely than females to report alcohol consumption across the ages. By racial divides, all were less likely to report alcohol consumption compared to the Africans among those aged 12 and 14 years. For those aged 13 years, apart from the Whites, other races were less likely than Africans to report alcohol use. Across all ages teenagers in the rural provinces were more likely than their urban counterparts to report alcohol consumption. --- Discussion Early teenagers consume alcohol despite regulations that forbid them from consuming alcohol in South Africa. These early teenagers have reported that they have been exposed to alcohol advertisements on television, on posters/billboards, on social media and during events. The present study has shown the patterns of influence different media of exposure to alcohol advertisements has on underage drinking in South Africa. The nature of association differed by the media of alcohol advertisements. The result of the study showed that alcohol advertisements on television and posters/billboards were associated with ever-consumed alcohol and recent alcohol consumption. This finding is consistent with reports linking TV and billboard alcohol advertisements to adolescents' alcohol consumption in Zambia [42], Chile [43] and developing countries [25]. However, disagreed with studies in Australia [22] which did not find an association between alcohol adverts on posters/billboards and alcohol consumption in the last 12 months among teenagers aged 12-15 years. Furthermore, contradicts the findings that exposure to television advertisements was associated with reduced odds of alcohol consumption in the last 12 months in Australia [22]. Our results further revealed that exposure to alcohol advertisements on social media and during an event showed lower tendencies of ever-consumed alcohol and recent alcohol consumption. The possible explanation for this observation could be attributed to the small proportion of early teenagers that have been exposed to alcohol advertisements during events and on social media. Previous studies reported only 2.4% had never been exposed to alcohol-related content, on media [44] compared to 92.2% observed in the present study. Secondly, given the ages of these children, they may lack despotic opportunities to attend events and limited access to social media which impacts negatively their exposure to alcohol advertisements and subsequent prevention of underage drinking. Our results concerning social media contrast with the assumption that anything that enhances exposure to alcohol marketing may also lead to alcohol use [19]. Other studies have reported that social media exposure to adolescent alcohol use was mediated by normative beliefs or perceived norms [20,45]. The patterns of the effects of alcohol advertisement exposures during events or through social media did not differ by age despite adjusting for other factors. These suggest that other factors play stronger roles in underage alcohol consumption than exposure to alcohol advertisements during events and social media. These findings are pertinent for an integrated strategy for the prevention of underage alcohol consumption. The result of our findings is in line with a previous report in Thailand [28] that reported reduced odds of alcohol consumption among males. This is surprising given that males are always associated with alcohol drinking. This suggests that in the early teenage, there is an inverse association between alcohol consumption and the male gender. Our finding is further in consonance with the report that current drinking in the past year was associated with teenagers aged 14 years [28] compared to those aged 12 and 13 years. Overall rural Provinces influenced early teenage alcohol consumption, a finding that concurs with a recent report in Spain [46] and Germany [47]. It is anticipated that the teenagers from the rural provinces would show --- Conclusion Our study on media exposure to alcohol adverts among children and early teenagers is novel various media examined were based on the national representative population of South Africa. To the best of our knowledge, no similar study has been conducted in South Africa. The findings of the study showed patterns of influence on alcohol consumption with the medium of exposure. The present study revealed that advertising alcohol through television and posters/billboards needs to be revisited for additional restrictions. On the other hand, exposure through social media and during events appeals for continuous monitoring as they currently do not pose threat to underage drinking. Further actions to deter children and early adolescents from alcohol consumption should --- Limitations The current study provided information on the impact of different types of alcohol advertisements and underage drinking. However, interpretation should be done in light of some limitations. Given that the study population were children who are entirely under parental guidance, we did not have variables to measure parental factors in the study. We do not know to what extent parental guidance has played a role in the whole gamut of alcohol use. However, a systematic review of studies documented several parental factors such as parental modelling, limiting the availability of alcohol to the child, disapproval of adolescent drinking, general discipline, parental monitoring, parent-child relationship quality, parental support, and general communication [45,46] were deterrents to underage drinking. A recent study has also reported that parental support and monitoring influence adolescent alcohol use [47]. Variables to examine peer influences were also missing from the survey data. Furthermore, knowledge about alcohol use was not assessed, hence we were unable to account for the knowledge gap in the study. However, future research will take note of parental factors, peer influences and knowledge.The study was a cross-sectional design which limited our ability to establish causal inference. The findings of the study remain extant in the current environment because children and early teenagers are constantly exposed to alcohol advertisements in South Africa. • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year --- • At BMC, research is always in progress. --- Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: --- --- Author contributions GO: Conceptualization; data curation; formal analysis; investigation; methodology. CO: Conceptualization; data curation; formal analysis; investigation; Reviewed the manuscript. GC formal analysis and review the manuscript. All authors read and approved the final maunscript. --- --- Ethics approval The survey protocol was approved by the Human Sciences Research Council Ethics Committee , the CDC Division of Global HIV and TB and the CDC's Center for Global Health . The survey adhered to international ethical standards and complied with the South African Children's Act of 2007. Parental/guardian approval was obtained by completing the informed consent form for learners who were below 18 years. Children's assent to participate in the study was obtained. All methods for human participants were carried out following relevant guidelines and regulations. --- --- Competing interests The authors declare no competing interests. ---

Background There is a paucity of information on alcohol adverts on alcohol consumption among early teenagers. The study aimed to examine the association between exposure to alcohol adverts and alcohol consumption. Methods A sample of 3833 early teenagers aged 12-14 years were extracted from the South African National HIV Prevalence, HIV Incidence, Behaviour and Communication Survey (SABSSM) 2017. Participants answered questions related to their exposure to alcohol adverts across television, Poster/billboard, events, and social media. Alcohol consumption was assessed on ever consumed alcohol and in the previous 12 months. Information on demographic characteristics was collected. Descriptive and inferential statistics were used to process the data. Result Exposure to alcohol adverts increases from 5.2% for Events to 77.0% on television. All alcohol media of adverts were associated with alcohol consumption by age. The results mirror studies that demonstrated that exposure to alcohol advertisements was associated with alcohol consumption.The association between alcohol consumption, exposure to alcohol adverts, and demographic characteristics need the urgent attention of policymakers and interventionists. The focus of action should be on protecting these early teenagers from exposure to media of adverts and risk background factors.

Brief Overview of LGBTQ Population and Early Healthcare Policy History The National Academy of Sciences' Institute of Medicine discusses two historical ways of thinking about LGBTQ population's healthcare needs. These two main contexts are as follows: that the LGBTQ population's "lifestyle" was seen as an illness itself and that the HIV/AIDS pandemic shaped how the LGBTQ population was viewed and treated for their health care needs . I will first discuss the emergence of the term homosexuality in medicine and how the medical community initially treated homosexuality and gender identity as a mental illness. I will then explore a brief history of the LGBTQ population's health care policies and practices in the USA. --- The Medical Community's Early Stance on the LGBTQ Community Societal norms during the late nineteenth century deemed homosexuals and gender non-conforming individuals as "inverts," meaning having sexual inversions, placing these groups of people under the same category of "perverts" and social deviants . Even though human same-sex attractions have been seen across human history, the modern term of homosexual did not appear until the end of the nineteenth century . In fact, the term homosexuality was "invented" as a concept by Viennese-born human rights activist Karl Maria Kertbeny in 1868 as a progressive term to take away stigmatizing prior terms . Theological doctrine and laws at the time were the main influencers of policies and social norms surrounding sexuality and gender, and anything outside the purview of procreational marriage was considered a sin . Homosexual acts, such as sodomy, were considered an illegal crime , and even today, religion is still the strongest social determinant to explain the rejection of homosexuality . Theology, laws, and social norms influenced the domains of medicine and psychology during this period, affecting how the LGBTQ community was treated in terms of their health care needs. Even though this initial doctrine spread into the medical community at the time, medicine was seen as progressive. Homosexuals and those with non-gender conforming identities were not seen as "sinful," but rather as victims of a disease that needed curing . This view became the dominant thought process throughout the twentieth century in medicine and psychiatry . In opposition to this "standard of care," the modern term of "sexual orientation" is credited to Sigmund Freud, who is the first person among the medical community that uses the word homosexuality in his work . Freud asserted that someone who is homosexual will move through psychotherapy, eventually settling on a heterosexual identity, and this shift would "cure" the depression and anxiety that resulted from the way society treats non-heterosexuals . Freud stated that someone who was simply homosexual with no pathology did not need to be cured . This statement suggests that Freud may not have seen a non-heterosexual sexual orientation as a disease to be rectified, but rather the resulting neurosis and pathology that might occur from having this identity in our society . This directly runs parallel with current work on minority stress theory , suggesting that LGBTQ people experience minority stress due to stigma and discrimination in our society. Minority stress can cause negative mental health consequences and can affect the overall health of LGBTQ people . Freud may have been ahead of his time in acknowledging that homosexuality was not a disease to be cured but was missing the crucial element of society needing to change its attitude towards sexuality that deviated from the "norm" of heterosexuality. Freud's positionality may just reflect how in the field of psychology, discussions around how structure affects an individual's health remains a missing element. As noted above, the medical community in general disagreed with Freud , stating that homosexuality itself was a disease to be cured, not the pathology that resulted from having a homosexual identity . Lieberman , in reviewing the book Recollecting Freud , discusses that Freud's opponent in the psychological field, Isidor Sadger, believed that homosexuality itself was a curable disease using psychoanalysis, in line with dominant medical field views on homosexuality at the time . According to Dundes and Sadger , despite Freud's objection to defining homosexuality as a curable disease, Bayer homosexuality was considered a mental illness until its removal from the DSM in 1973 . Drescher goes on to discuss how gender identity disorder was being debated at the time of his article, the main discussion being "What is normal sexuality or gender?" and should they be considered a disorder just because they are a deviation from the norm? Eventually siding with Freud, the medical community now embraces acceptance of diversity and difference in humans, rather than labeling people with deviations from the norm as a disease to be cured . This progress of thought in the medical field helped lead to gender identity disorder also being removed from the DSM in 2013 . This lengthy history of LGBTQ people being seen as a disease to be cured by psychiatry set the stage for how the HIV/AIDS epidemic was handled and continues to affect LGBTQ older adults today during the current pandemic. HIV/AIDS before it was known what it was. In fact, one of the early terms researchers started using for HIV/AIDS, published in The New York Times in 1982, was gay-related immune deficiency , which deepened the association with gay men and the disease . This history of the disease being associated with gay men in the USA from the start points to the fact that blaming of the LGBTQ community for HIV/AIDS was inevitable. Martos et al. explains further that when the CDC discovered this disease, the Reagan administration ignored the illness soon to be known as HIV/AIDS altogether, causing deaths of thousands, remaining silent on the issue for all of the early 1980s . Reagan had stated publicly that being gay was "an abomination," not even mentioning HIV/AIDS until after his reelection , demonstrating with his ignoring of the disease a political move on his administration's part to win the election. The first time Reagan publicly mentioned HIV/AIDS was not until Reagan's Hollywood friend Rock Hudson dies of the disease in 1985. Reagan finally had the surgeon general speak about HIV/AIDS to the public in 1986 . This silence from the president and his administration caused the lack of clear policy direction within multiple branches of the USA, including uncertainty within the Food and Drug Administration , the Surgeon General , and the Centers for Diseases Control . The US government, specifically the, even went as far as leaving out vital information on safe sex practices within HIV/AIDS education materials meant for the public, as they felt it promoted being gay . This was due to the flip-flopping policy recommendations about sex education by the CDC. At first, in 1984, the CDC was quietly funding community organizations that supported HIV/AIDS work, but the Reagan administration put a stop to this in 1985, citing that they did not believe that the government organization should be telling gay men how to participate in "sodomy" . The CDC was silenced by this retraction of funding and politics . This general censorship of the problem promoted by the US Government and the Reagan administration eventually led to the poor containment of the disease, discrimination of people with HIV/ AIDS overall , who lost their jobs, health insurance, and were subject to violence . Goh argues that by the early 1980s, the media also has influence, portraying HIV/AIDS as being a result of the LGBTQ community's choice of "lifestyle," which led to the blaming, victimization, and stigmatization of the LGBTQ community. Singer et al. discusses in their book, Erotic Welfare: Sexual Theory and Politics in the Age of Epidemic, that HIV/AIDS affected the USA's culture by prompting the need to blame others for disease in their panic, and that this blaming can be more harmful, even exceeding the damage done by the actual contagion. This blaming of disease ultimately leads only to the oppression of the group that is blamed, instead of leading to the unity of the public to contain the disease itself . Shah expands on this discussion in her book, Pandemic, by stating that the public was resistant to accept the contagious nature of HIV/AIDS and instead engaged in scapegoating the LGBTQ community, declaring the disease a gay cancer instead. This scapegoating of the LGBTQ population and those most at risk of catching the disease only impeded early containment efforts , backing up claims by Singer et al. . Overall, blaming, combined with discrimination against LGBTQ people with HIV/AIDS, led to the collective trauma in the LGBTQ community , the hindering of treatment, and cure of HIV/AIDS. This collective trauma brought on by a history of blaming LGBTQ people, from looking at LGBTQ people as a disease to be cured , and causing a disease , has an intertwining affect with outward and inward stigma of the LGBTQ community . When people are outwardly stigmatized in their communities, they end up turning that world view onto themselves, making that individual feel less than, weak, and exposed . This shame leads to feelings of being traumatized, while also contributing to physical and mental health problems . This outward and inward stigma combined, along with trauma that this population faces, shows a need for further intervention. I will first discuss how LGBTQ people intervened for their own people, the social work profession's response, and finally the eventual general medical community's response during the HIV/AIDS epidemic. --- Community Responses to the HIV/AIDS Epidemic --- LGBTQ Population's Response to Disease Model Treatment In 1969, as noted by Carter in their book entitled: Stonewall: The Riots that Sparked the Gay Revolution, members of the LGBTQ community banded together in what is now known as the Stonewall riots, where LGBTQ people rioted against the police at the Stonewall bar in New York City. These riots are the first time LGBTQ people protested the insufferable situation with police in New York City . The Stonewall riots are noted as the beginnings of the origins of LGBTQ health services . These services increased community mobilization and started the rejection of homophobia . The Stonewall riots also sparked the first appeals for same-sex marriage . From this literature, it can be concluded that the Stonewall riots lit a fire under the LGBTQ community to come together to fight for policy change. Even though the Stonewall riots began the fight for same-sex marriage, it was not until the HIV/AIDS epidemic that this became a priority of the LGBTQ rights movement. The HIV/AIDS epidemic created a greater sense of community among LGBTQ people that had not previously existed, which forced the LGBTQ community to band together to fight HIV/AIDS for their own survival . HIV/AIDS was rapidly becoming responsible for the death of many gay and bisexual men in the USA , which led to the LGBTQ community launching their own social service agencies. Organizations such as the Gay Men's Health Crisis in 1982 and ACT UP were started by the artist and playwright Larry Kramer, who screamed until he was heard about how HIV/AIDS needed to be addressed . These organizations helped to address issues of sex education, food, transportation, physical assistance, and emotional support for those dealing with HIV/AIDS . LGBTQ people fought for same-sex marriage then not only as a basic human right, but for reasons during the HIV/AIDS crisis that were related to medical care. Examples of these medical care needs include the right to visit their romantic partner in the hospital, gain access to health insurance, or to be able to divide up family assets . Since same-sex marriage is uniquely tied to these medical care rights, this was a motivator for LGBTQ activists to fight for this cause . Social workers ended up being one of the first professions to assist the LGBTQ community in advocating for those affected by HIV/AIDS. --- Social Work Response to Disease Model Treatment: Advocacy for Change Rice and Willinger's book about the social work profession's response to the epidemic of HIV/AIDS, called A History of AIDS Social Work in Hospitals: A Daring Response to an Epidemic, discusses the early days of the disease. Social workers emerged as leaders from the first days . Mayer et al. notes by the 1980s, because of the HIV/AIDS crisis, there were clinics, mental health groups, and providers that started to focus their attention on sexual minority health. Social workers established group interventions at these health facilities so that those diagnosed with HIV/AIDS could speak to each other about their commonalities, including feeling socially isolated and experiences of stigma . Social workers also created interdisciplinary care management teams that would care for individuals being diagnosed and living with this disease . This social work advocacy for those suffering from HIV/AIDS was helpful in establishing assistance for people living with the disease and helped to advocate for change at the local level. Social work advocacy also established the infrastructure for administering grants from the National Institutes of Health and the Centers for Disease Control and Prevention for HIV/AIDS research. Mayer et al. notes that national LGBT organizations, such as the Gay and Lesbian Medical Association, the National Gay and Lesbian Task Force, and the Human Rights Campaign, recognized the importance of advocacy for further research focused on sexual and gender minority health and service delivery. San Francisco social workers expanded on this research, establishing one of the first networks for social workers working during the HIV/AIDS crisis in 1985 . This network, entitled the Social Work AIDS Network , provided social workers with peer support, education, and advocacy platforms concerning sexual and gender minority populations . SWAN advocated for policy measures and helped those social workers who worked with HIV/AIDS populations to make sure professional standards were being met . By the 1990s, there were over 100 organizations that formed the National Coalition for LGBT Health, whose primary goal was to advocate for sexual minority health and reduce health disparities . Social work networks and conferences on HIV/AIDS helped make change happen on a larger scale, bringing more recognition for the need to assist LGBTQ populations. --- Eventual Medical Community Response: Current Model of Health Care Equity Previous advocacy from the LGBTQ community and the social work profession's advocacy helped influence the medical profession's move towards a health care equity model instead of looking at LGBTQ people as a disease to be cured. The LGBTQ population was eventually added to the federal government's research agenda in the year 2000 by the creating their decennial health disparities research report, Healthy People 2010: Understanding and Improving Health. This report showed that health care advocacy for LGBTQ people and recognizing their health disparities had become a federal health issue , marking the move from a disease model to a health equity framework. The health equity model relies on concepts from the minority stress model. The minority stress model discusses that sexual minorities experience greater health disparities due to discrimination against themselves as a minority population . National Academy of Sciences adds that the "person in environment" framework backs the minority stress model, noting that intersectionality is important since the year that an LGBTQ person was born can often largely affect their health outcomes. Fredriksen-Goldsen et al. notes that both frameworks inform how health care practitioners and policy makers should currently address health care needs of the LGBTQ population and their sub populations, such as LGBTQ older adults, who have pertinent needs now in the COVID-19 pandemic. Existing Health Disparities for LGBTQ Older Adults in the USA SAGE et al. discuss that in 2017, the estimated population of LGBTQ people over age 50 was 2.7 million people in the USA, or about 2.7% of the population, and that the population for those 65-plus was 1.1 million people . Fredriksen-Goldsen notes that this population is estimated to more than double, with estimates of approximately seven million LGBTQ adults over 50 by 2030 . The growth of this population makes it even more imperative to do further research regarding this population's health. Adams et al. argues that sexuality should be considered a health determinant and that practitioners need to consider the needs of sexual minorities while LGBTQ people live in a heterosexist world. This argument points out the need for the health care industry to adequately address pervasive stigma and discrimination related to LGBTQ individuals' marginalized identities . LGBTQ people in general experience unique health care disparities due to this stigma and discrimination . Baptiste-Roberts et al. , Colpitts and Gahagan , and Mayer et al. establish that existing research in health disparities for sexual minorities has shown that LGBTQ populations, compared to heterosexual/ cisgender counterparts, have a higher prevalence of many health ailments. These ailments can be attributed to nonequitable treatment and social inequities in society and not biomedical explanations of health . Compared to their heterosexual/cisgender peers, LGBTQ older adults have higher rates of substance use, drinking, obesity, arthritis, diabetes, high blood pressure, chronic obstructive pulmonary disease , colon cancer, lung cancer, higher rates of mental health disorders, suicide attempts, greater disability rates, and overall lower rates of functional health when compared to heterosexual/cisgender peers . These populations also have disparities to tangible health care access and experience economic insecurity, employment discrimination, housing discrimination , thin support networks, and fear discrimination when accessing aging services; all of which can be tied to health care needs . Transgender people and people with HIV positive status are at the most risk . These disparities in health were present before the COVID-19 pandemic and have been exacerbated during this crisis. --- Amplification of LGBTQ Older Adult Health Disparities During the COVID-19 Pandemic The COVID-19 pandemic has been particularly difficult for older adult populations. In the USA, older adults are more likely to become seriously ill or pass away due to the disease . Older adults are also more likely to suffer long term physical effects of the virus due to the increased risk of having comorbidities, with 95% of deaths from COVID-19 being adults over 50 years of age . Other serious conditions from having to socially distance are social isolation, loneliness, and the loss of caregiver support . There has also been "evidence of openly ageist discourses" which "complicates the experiences of living through Covid-19 for older people" and "contribute to feelings of worthlessness" . Older LGBTQ people have seen similar discourses as their heterosexual peers, but with further complications due to the stigma and discrimination they face that is described above. Certain health disparities have been exacerbated within different subgroups among the LGBTQ population , such as older adults, and have been amplified during the COVID-19 pandemic . Drabble and Eliason and Jen et al. have found that during COVID-19, LGBTQ older adults have had elevated rates of chronic conditions , higher prevalence of anxiety/depression, greater substance abuse, higher economic insecurities, limited community resources, and limited access to health care services. Other health determinants such as isolation and loneliness have also been intensified . Salerno et al. add that since LGBTQ older adults in general are four times less likely to have children, and twice as likely to live alone, and are more likely to be estranged from family compared to heterosexual/ cisgender peers, concerns about solitude and mental health are even more troubling due to differences in social network size. This may make needing to socially distance a complex issue that may result in longterm mental health issues . LGBTQ older adults also have a difficult time accessing the health services they need due to discrimination in medical settings, as well as lacking proper health insurance, which has been magnified by COVID-19 . Chatterjee et al. adds that along with this fear of discrimination, getting the COVID-19 virus has been a threat, leading to less screening of chronic diseases. Drabble and Eliason take into consideration also that due to lacking health statistics of LGBTQ people, especially during the COVID-19 pandemic, we will have limited information on long-term affects and outcomes of LGBTQ older adults who have suffered from the COVID-19 virus. More research needs to be done in this area for LGBTQ older adults with concern to long-term mental and physical affects from the COVID-19 pandemic. Specific populations of LGBTQ older adults, such as those with HIV/AIDS, suffer even more from these concerns, with higher rates of posttraumatic stress disorder and isolation, lower physical health, and premature aging . Despite this, some older cohorts of LGBTQ people who have HIV/AIDs have shown to be more resilient during the COVID-19 pandemic due to their experiences of living through the HIV/AIDS epidemic . Quinn et al. interviewed gay men who lived through the HIV/AIDS epidemic, asking them how they relate that epidemic to the current pandemic. Adults in their study made connections from having to wear condoms during the HIV/AIDS epidemic to having to wear masks now, needing to do so to save themselves and others. This research helps us understand that LGBTQ older adults have previously lived through a disease crisis, and the historical context noted above informs current research about the COVID-19 pandemic for this population. These connections back up the claim by Kline that the COVID-19 pandemic represents "a critical time to emphasize root causes of health inequality" for LGBTQ people, making it pertinent to address these above concerns through policy, practice, and research implications. Cahill et al. calls for a need for non-discriminatory health care for LGBTQ populations on all levels of policy, practice, and research implications and is pertinent in the COVID-19 pandemic. By discriminating against LGBTQ people in public health, we are causing further health disparities and compounding the issues noted above . Advocacy for LGBTQ people and their subpopulations needs to be done to fully address public health. Directives suggest that policy makers should pay special attention to vulnerable sub populations, such as LGBTQ older adults, as well as those who have chronic conditions , and look to these populations as partners in addressing health care issues within their communities. --- Addressing Needs of LGBTQ Older Adults Now During the COVID-19 Era: Equitable Health Policy, Practice, and Research Implications --- Policy Implications Discrimination should be prevented as it is seen to be one of the causes of the health disparities noted above . One of the ways to lessen discrimination and address health disparities in public health is for a person's identification as part of the LGBTQ community be recorded, thus ensuring the counting of this population through the generation of health statistics . COVID-19 has shown us how important legal protections are at the federal policy level and how this is connected to public health policies . If LGBTQ people are not counted in public health care records, how can policy makers have the information to serve this vulnerable population if this information is not known? Kline discusses also how LGBTQ people in general have learned from the HIV/AIDS crisis and can bring this knowledge to the current COVID-19 pandemic. Blame in a pandemic only leads to poor health care, and Quinn et al. suggest calling for collective activism among LGBTQ people and their allies. Activism can provide help, not only for LGBTQ older adults, but for all minorities facing health disparities. This collective activism among LGBTQ communities and allies' shines light on how this population has been through a pandemic before and advocated together to create real change . Kline echoes these sentiments and speaks about how LGBTQ people were blamed during the HIV/AIDS crisis for the disease, just as Americans are blaming people of Asian descent for COVID-19, and how the broader community must put policies in place to fight against systematic discrimination as it leads to unfairly blaming whole communities for disease transmission. Advocacy needs to be done in this area since blaming does not lead to controlling the disease as shown above, but only harming the community it is targeting. LGBTQ people should be seen as the expert advocates, as this is now the second pandemic they are living through. Policy makers and service providers can be allies by reminding LGBTQ populations of their strength and community togetherness, giving this population power to make real change in their communities . The broader community will have to actively join the LGBTQ community's lead in creating this change. Jen et al. discuss that all health professionals currently and after the pandemic should practice care that is "person-centered, strength-based, and trauma-informed" . Drabble and Eliason make suggestions for hospitals to become person centered: ensure that health providers are trained to provide respectful and competent care and increase access to LGBTQ health care community supports. de Los Reyes and Collict highlights the need to train practitioners to work with sexual minority patients as it increases positive attitudes towards LGBTQ populations and attitudes toward diversity in general. Rosa et al. adds that training can help professionals give care that is always inclusive of LGBTQ people's needs, as it may be unknown who identifies as being part of this population, and to practice active listening skills to meet patients' needs. Bristowe et al. states that to increase access to LGBTQ community care, providers should promote LGBTQ visibility, such as partnering with LGBTQ organizations. These suggestions combined would aim to provide care that is inclusive of LGBTQ people's needs and would be person centered. --- Practice Implications For a strength-based approach, Jen et al. note that LGBTQ older adults' narratives should be guided as a resilient population. Older adults with HIV/AIDS are a particularly resilient group . Despite the resilency of this group, we are reminded that they still need the extra support during the COVID-19 pandemic . This support can be provided safely during the current pandemic specifically, with practice delivery and remote program options . Gorenko et al. also found online programming for older adults helpful with mental and psychological distress during COVID-19. Healthcare providers must care and advocate for LGBTQ older adults in a way that limits transmission of illness, but also makes sure we are meeting the needs of a population that may be socially isolated during this time and need to maintain a sense of connection, as well as community . In this manner, we can support LGBTQ older adults to be able to play off their strengths, while providing support. For a trauma informed practice, practitioners must understand barriers to health care access, as getting access to begin with can be filled with stigma and discrimination that can be traumatic for this population. Mayer et al. and Bristowe et al. discuss the barrier of tangible healthcare access further with three main causes: reluctance of LGBTQ populations to disclose their sexual orientation or gender identity to their providers due to discrimination, need for culturally competent medical care and services, and health insurance access. All these barriers to tangible health care access make it difficult for older adults to receive the care they need, let alone deal with the health disparities they face. Chatterjee et al. discuss that fear of discrimination and stigma in health care settings, as well as COVID-19 as a threat, have led to people not seeking care. If people do not have access to health insurance due to discriminatory policies, they will not seek care. If a practitioner and the services they provide are not welcoming, a person who is LGBTQ will not visit their health care provider, leading to delayed diagnosis and more serious illness. If people do not feel comfortable disclosing their sexuality, it is difficult for practitioners to meet their specific needs. And finally, if a practitioner and the services they provide are not welcoming, a person who is LGBTQ will not visit their health care provider. These barriers need to be addressed further to meet the specialized healthcare needs of LGBTQ people. --- Research Implications Since we have limited research documenting this population, the main public health research directive for the LGBTQ older adult population should be to use a community approach. Community approaches aim to reduce discrimination/stigma while making new policies and directives, as LGBTQ people are experts on their own communities. Marchia describes that the making of LGBTQ history and activism comes from a grassroots activist lens, in line with policy implications indicated above. Marchia goes on to state that researchers should model the grassroots approach by being activists themselves, focusing on research that aims to increase access to equal treatment, resources, and combat stigma, since we know these contribute to better health outcomes. LGBTQ people had to band together to fight for their basic rights, such as getting access to treatment for HIV/AIDS, visiting their sick partners in the hospital, and obtaining health insurance for their partners . LGBTQ people have been their own researchers and activists, so for this population, a ground up approach for research would be the most ethical method in obtaining information. In line with an activist, ground up approach, Wallerstein and Duran discuss how to address health disparities using a collaborative approach called a communitybased participatory action research approach. This approach has the community become a co-collaborator with the researcher, as they are the experts on their own community's needs . By conducting research in this manner, the researcher is gaining the proper knowledge needed to address health disparities directly from the community it aims to research. To conduct research specifically with LGBTQ older adults, the researcher should partner with an LGBTQ aging advocacy community organization, identify LGBTQ older adults who receive the services at the agency, find out LGBTQ older adults needs during and post the pandemic by interviewing both groups in focus groups, and work in tandem with them to meet their needs. To find out their needs, the researcher should run focus groups with the partner organizations and LGBTQ older adults receiving the services, in line with a CBPR approach. This information should then be used to create a survey to ask LGBTQ older adults in the broader community about their needs and struggles during the pandemic. Questions that should be posed to agencies and LGBTQ older adults that they serve are the following: By asking questions in this manner, researchers can gain the knowledge needed to come up with research questions that can be asked to even begin to find the answers to assist this population. If a researcher does not know what this population is experiencing related to the COVID-19 pandemic, how can the actual research questions be developed? After obtaining important knowledge from the LGBTQ older adult community agencies about their needs, the researcher should work in tandem with this population to obtain the answers to problems related to their health care needs. --- Conclusions In summary, this paper discussed a brief healthcare history of the LGBTQ population in the USA and how in this history, blaming LGBTQ people for disease played a major role. Blaming fueled the discrimination and shame felt then and has affected the LGBTQ older adult population's health outcomes to this day. The response of the LGBTQ population, the social work profession's involvement, and eventual late response from the medical community during the HIV/ AIDS crisis resulted in a general movement from a disease model of health care for LGBTQ persons, to that of a health equity model approach. This past knowledge helps to inform our current pandemic crisis of COVID-19 by assisting health care workers, including social workers, in how to specifically care for the healthcare needs for LGBTQ populations and their sub populations now. LGBTQ older adults have many health disparities due to poor past treatment by the medical community, before and during COVID-19, that have attributed to health inequalities discussed above. Key suggestions for policy makers such as counting LGBTQ people in health care statistics, learning from the HIV/AIDS crisis, and using LGBTQ people as experts in developing their own policies are suggested. Practice with LGBTQ older adults should be person-centered, strengths based, and utilize trauma care approaches. Policy makers and practitioners should combine these suggestions when working with LGBTQ older adults. Researchers going forward should ask this community directly for help with their research on LGBTQ older adults as these older adults are the experts on outlining their needs. Further research efforts regarding LGBTQ older adult populations must be made to address these current and past health concerns. By combining policy, practice, and research approaches, the USA can make an impact to improve the overall health of LGBTQ older adults. Change must be made and implemented now by LGBTQ persons and their allies to make sure the health of LGBTQ older adults is a priority. There is no more pertinent time than during the current COVID-19 pandemic. --- Availability of Data or Materials Availability of research is available upon request. Code Availability No code was used for this manuscript. --- --- Conflict of Interest The authors declare no competing interests. --- Author Contribution Breana Bietsch researched these ideas and findings independently and wrote the manuscript. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

U.S. policy has, once again, overlooked the health care needs of older adults who are lesbian, gay, bisexual, transgender, and queer or questioning (LGBTQ). This population is estimated to more than double in the USA (Fredriksen-Goldsen, in Generations 38(4), 86-92, 2015), with estimates of approximately 3 million LGBTQ adults over 50 currently and 7 million by 2030 (Services and Advocacy for GBLT Elders in SAGE, New York, 2018). The healthcare model for addressing the needs of LGBTQ persons has historically been that of a disease model of care, particularly during the HIV/AIDS crisis, and has recently moved to that of a health equity model in the past 20 years. The LGBTQ community, social work profession, and general medical community worked to create the health care equity model we have today and this paper will discuss how this evolved. The health care equity model addresses the health needs for older adult LGBTQ populations. Older LGBTQ adults are more likely to experience elevated rates of chronic conditions (such as HIV, cancer, diabetes), higher prevalence of anxiety/depression, greater substance abuse, higher economic insecurities, limited community resources, and limited access to health care services compared to heterosexual/cisgender counterparts. This paper discusses how health disparities among this minority population and heterosexual/cisgender individuals have been exacerbated during the COVID-19 pandemic. Further, this paper will address policy, research, and practice implications to understand how to assist this vulnerable subpopulation of LGBTQ persons.This paper explores the health care history literature, past and present, of the lesbian, gay, bisexual, transgender, and queer or questioning (LGBTQ) populations in the USA, with a focus on LGBTQ older adult populations. The acronym LGBTQ will be used as an over-arching umbrella term aimed to encompass the broad identities of this unique and vulnerable minority population. I will explore how this history in the USA has affected and helped to create LGBTQ older adults' health disparities today. The justification for examining this topic currently is because the COVID-19 pandemic continues to impact older adult populations (Brooke & Jackson, 2020), as well as exacerbate health disparities among minorities, such as the LGBTQ population (Kline, 2020), which calls for special attention to older LGBTQ adults. Drabble and Eliason (2021) discuss this intersection and note that LGBTQ older adults have been particularly vulnerable, with increased health risks during the pandemic. Marchia (2018) notes that there are gaps in the literature for this vulnerable sub population of older adults that need to be addressed regarding health disparities. This paper will make suggestions to address these gaps through policy, research, and practice implications during COVID-19.

Introduction As a critical part of Canada's continuing health care system, long-term care services serve a predominantly older population . These services are facility based, meaning that residents live in communal dwellings with 24 hour nursing care, which is provided by staff primarily, but is also complemented by family members and formal and informal volunteers. Although LTC is effectively a "home" for residents, facilities are highly regulated and often hospital-like. Efforts to improve LTC residents' living conditions have focused on quality of care standards derived from monitoring residents' medical, social, physical, and emotional status . As LTC is not included in the Canada Health Act, these standards of care are set provincially and enforced by provincial regulating agencies, resulting in provincial variations and frequent regulatory process tensions. Although LTC institutions exert considerable influence over how LTC relationships are organized, provincial regulatory and government-endorsed policy documents constitute the basic frameworks within which institutional policies are developed. Quality of care is related to the technical processes involved in health care delivery as monitored and regulated by government. A medical perspective is used to conceptualize and address LTC quality of care problems ; however, in the last 20 years, researchers, residents, and advocacy groups have called for provincial policy fails to address the roles of volunteers. What language does exist has conflicting role interpretations and tends to limit these roles , rather than exhorting LTC institutions to facilitate practical, creative, and unique avenues and mechanisms for volunteer engagement . This happens primarily through: omitting volunteers from most regulatory policy, likening volunteers to supplementary staff rather than to caregivers with unique roles, and overemphasizing residents' "safety, security and order," rather than volunteers' relational activities with residents. Research shows that the roles of and desire for LTC volunteers is growing and that we will likely soon see expanded language describing volunteers in Canadian LTC policy. Therefore, this is a critical time for policy analysis to inform these policy changes. Rather than simply highlighting promising policy, however, we suggest that changing regulatory policy for LTC volunteers will not necessarily enhance residents' QoL. Regulatory procedures in and of themselves may be inadequate or ill-suited to address the cultural, social, and structural changes needed for volunteers to enhance QoL in LTC settings. Following a review of literature on volunteers working with older people in LTC facilities, we examine related four jurisdictions' regulatory policies, analyzing the QoL domains supported in each policy text and noting emerging crossjurisdictional trends. We then consider how interpretations of these texts may enhance or thwart volunteers' capacity to improve LTC residents' QoL, and conclude by highlighting promising new policy directions for LTC volunteers and offering suggestions for future policy work and research on policy development and implementation processes, which are beyond the scope of this article. --- Literature The growing body of literature on volunteer contributions to the quality of care and QoL of older people focuses on volunteer roles, activities, and motivations, and identifies the continuing care sector's challenges in recruiting, training, coordinating, and retaining volunteers. In 2013, Morris, Wilmot, Hill, Ockenden, and Payne published a literature scan of volunteers' contributions to end-oflife services, noting that volunteers who work with people with dementia and/or those at the end of their lives often draw on previous experiences of loss and are motivated by deeply personal desires to support others going through similar experiences. Some researchers have argued that older volunteers' unique skills and life circumstances are an important under-tapped resource for volunteer recruitment and programming in LTC; palliative and hospice care should be specifically targeted to older retired volunteers with requisite life experience. However, there is growing interest in the potential of younger volunteers to enhance LTC intergenerational relations . Expanding volunteer roles, services, and programming is generally understood as a cost-saving strategy that "adds value" to existing services and offers a good investment . Therefore, well-developed volunteer programs are understood to "make good business sense" in LTC institutions and often serve as a stopgap response to funding cuts and staff shortages in non-profit facilities . Nevertheless, some research frames volunteers as playing crucial roles beyond resource investments and supplementary staff labour. Morris et al. , for example, suggests that volunteers should be "used in innovative ways" through programming that enriches services and creates new ways of thinking about QoL and quality of care for older people. Innovative programming and exploratory approaches are currently concentrated in dementia care or palliative and hospice care research, where volunteer programming seems most developed . Candy, France, Low, and Sampson argue that much more needs to be investigated regarding the "the mechanisms or aspects of the volunteers' role that may lead to beneficial effects" . --- Volunteers' Roles Volunteers play diverse roles in supporting many LTC activities. In Canada, research reveals that formal LTC volunteer work can include personal or "friendly" visiting, mealtime assistance, administrative duties, fundraising, special programming , or organizing special events . Research tends to concentrate on the volunteer roles of LTC ombudsmen in the United States, and more generally, in spiritual care, mealtime assistance, and other community-based programming, often tailored for specific faithbased or ethno-cultural communities or people with specific psychosocial needs . The roles of volunteers working with older people also vary depending on the location of services, type of care provision, and funding arrangements of the service provider. In non-profit LTC settings, several studies have argued that volunteers help "fill the gap" in staff shortages , particularly providing support for residents during mealtimes . In palliative care settings, Watts warns that improperly supported volunteer roles risk positioning volunteers as "handmaidens to the professional care team" , thus engendering resentment from professional staff. In LTC, Tingvold and Skinner found that lack of clarity about volunteer roles and limited opportunities for staff to learn about volunteer activities led to a variety of coordination challenges and staff conflict. Similarly, Keith and Nelson, Netting, Borders, and Huiber both underscore the significance of administrative support for LTC volunteer ombudsmen to minimize their conflict with staff, enhance the efficacy of their work, and motivate their continued volunteering despite inevitable challenges. Hunter et al. found that, with proper supports and training, staff are enthusiastic about volunteers' complementary LTC roles. Similarly, Hurst, Coyne, Kellet, and Needham argue that volunteers with LTC residents with dementia not only benefit residents, but also staff and the larger organization; however, clarification about volunteer roles is key for enhancing these contributions. In an American context, Falkowski noted that non-profit and for-profit LTC institutions had distinctly different volunteer roles. Non-profit institutions tend to have much larger volunteer pools and more frequent volunteer visits. These volunteers were "flexible," serving primarily to support/supplement staff and sometimes allowing staff to spend more time with residents, although activities were not organized around residents' interests per se. Forprofit settings, on the other hand, tended to recruit volunteers for specific programs geared towards residents' "socialization." These volunteer activities were heavily organized and often restricted volunteers' scope of activities, which placed additional strains on LTC staff. In the United Kingdom, Johnson and Cameron argue that LTC volunteers are usually conceptualized as "a spare pair of hands" to fill the gaps in care, or as the "cherry on the cake" to augment existing services. However, in care settings that are strapped for resources and funding, the line between volunteers and paid care workers was blurred . Similar studies that contrast volunteers in non-profit and for-profit LTC institutions have suggested similar trends. Several studies attempt to clarify volunteers' roles vis-à-vis staff and family members. For the most part, this research emphasizes that volunteers tend to excel in relational care rather than taskbased medical care provided by staff . In the United Kingdom, Hussein and Manthrope showed that, largely, volunteers did not supplement staff but rather supported LTC residents with counseling, support, advocacy, and advice, which are increasingly out of the purview of paid staff. Technology-based programs, administered exclusively by volunteers, seem also to enhance residents' autonomy in specific ways . Similar trends are found in in Norway, where volunteer activities complemented staff duties by focusing on "cultural, social and other activities aimed at promoting mental stimulation and well-being" . Andfossen argues that managed volunteers in Norway serve a highly organized, "non-personal" role unique from that of family and unmanaged volunteers, which should be better reflected in LTC policy and programming in order to maximize its potential to improve LTC. Some researchers have argued that volunteers' altruistic motivations distinguish volunteers from both staff and family and uniquely situate them to enhance QoL for families, staff, and residents. Mellow notes that direct service hospital volunteers have more control of their time, so they can engage in affective dimensions of care that is crucial to improving quality of care and dignity in highly regulated care environments. Guirguis-Younger and Grafanaki argue that palliative care volunteers play an integral role because they are motivated to share their "emotional resilience" with families and people who are dying. Volunteer choice and flexibility appear to be integral to volunteers' abilities to make specific and valuable contributions in palliative care. Weeks, MacQuarrie, and Bryanton make similar observations in hospice palliative care settings , arguing that volunteers provide a "unique care link" between paid and unpaid caregivers, indirectly benefitting HPC users. Although Weeks et al. note that volunteer roles are highly context specific, in general, volunteers play an "in-between-role", allowing for family confidences and contributing to sustained relationships even after death, while also acting as a buffer between family members and staff. This role differed significantly from that of paid staff, even in non-profit settings. Many of these roles are difficult to "train" for or standardize, as they are frequently nuanced and often informed by volunteers' years of lived experience caring for older people . Further, Ferrari and Funk and Roger both note that despite trends towards LTC volunteers' routinization and regulation , volunteers seem to value autonomy and control in their activities and may be deterred by heavily formalized volunteer environments. Nevertheless, research has shown that volunteers benefit from organizational support and clear roles and expectations. --- Challenges of Managing Volunteers Numerous studies emphasize the importance of appropriate LTC volunteer training, coordination and support , even as such strategies for volunteers are becoming more complex2 . In the United States, Thompson and Wilson suggest that management strategies should be tailored to older volunteers, who seem to do particularly well supporting older LTC residents. Their recommended strategies include having wellresourced, centralized volunteer agencies, incentives, and mechanisms allowing volunteers to play active and creative LTC roles. In the United Kingdom, research suggests that volunteer programs should recruit from relevant voluntary sector organizations , so that recruitment efforts target volunteers with prior experience and organizational support. In Norway, Tingvold and Skinner noted that staff should be actively engaged in volunteer programming and training to increase understanding of volunteers' value and exert some control over coordination with staff activities. In Canada, efforts have long been underway to develop "best practices" for volunteers and quality indicators for accrediting volunteer agencies . Strategies to enhance volunteer experience and motivational recruitment and retention tools for volunteers working with older people in LTC have also grown increasingly sophisticated over the last 30 years . Despite these initiatives, Tingvold and Skinner note that there are often limited time and few operational mechanisms in place to support effective LTC volunteer integration and coordination, which is reflected in high volunteer turnover and conflict between staff and volunteers. Notably, these challenges appear to manifest in ineffective volunteer policies, or in implemented policies that fail to account for LTC realities. In many cases, policy change is not the best response to challenges experienced by LTC volunteers. In fact, many LTC volunteers do not want their roles formalized in policy and procedures , as these can undermine their experiences with residents and create discriminatory recruitment practices . Other research warns that formalization might be negatively correlated with what volunteers do best: provide nuanced relational care . Mellow , for example, describes how direct service hospital volunteers must work creatively around regulatory constraints, which focus primarily on instrumental tasks, in order to provide affective, intimate care and maintain dignity in their work. Finally, Banerjee and Armstrong and Kane suggest that moves towards further regulating LTC are more reflective of an increasingly risk-averse care culture, rather than an effective enhancement for QoL or quality of care. They argue that further regulation may strain staff resources and obscure systemic problems in LTC. With these cautions in mind, the diverse volunteer roles and activities we discussed here have the potential to enhance all 11 of Kane's QoL domains. Yet, how volunteers are recognized and supported varies considerably across Canada's expanding patchwork of LTC policy. In the following sections, we detail our analysis of how formal volunteer roles-those that require management, regulation, screening and training-are defined in four provincial regulatory contexts in ways that might inhibit or enhance LTC residents' QoL. --- Methods This present study is part of a larger policy analysis associated with a Pan-Canadian multi-method research project, Seniors -Adding Life to Years . SALTY aims to enhance QoL for LTC residents in Canada during their later years using an integrated knowledge translation approach. The SALTY research team involved non-researcher stakeholders, including policy makers and health professionals, and LTC end users such as volunteers, family members, residents, and people with dementia. These stakeholders assisted in research design and analysis to ensure that our research addressed priority areas for those most impacted by policy changes . It was also these stakeholders who identified a volunteer perspective or "lens" on LTC policy as a priority research area. The policy analysis team collected data by scanning public repositories in four provincial jurisdictions-Alberta, British Columbia, Nova Scotia, and Ontario-to identify policy documents related to residential, long-term, and end-of-life care. No policy documents were collected after July 2017, with the exception of the Minister of Justice's Framework on Palliative Care in Canada Act and the Government of Alberta's Resident and Family Council Act, which was approved and operational in Alberta. These were included as a result of feedback from the project's policy stakeholders. The initial search resulted in 350 policy documents by numerous non-governmental organization authors that encompassed legislative, best practice, and strategic papers with varying goals. In order to refine our policy data pool, inclusion/exclusion criteria were applied in a series of stages to ensure that the policy document related to the overarching research question: How does existing policy enable or inhibit the QoL of residents in LTC facilities? First, the documents needed to refer to current LTC residents 65 years of age and older , second, the document was to be prescriptive in nature , and, finally, the document had to relate to facility care or be inclusive of facility care. The projects policy stakeholders also made recommendations to remove a couple of the initially included documents because their texts resembled hiring criteria more than policy direction. After applying the inclusion/exclusion criteria and consulting with project stakeholders, a total of 139 policy documents were selected, constating our base policy library. After finalizing our policy library, we described and categorized the documents using deductive content analysis . All 139 documents were categorized into six regulatory power levels to determine LTC facilities' degree of compliance: level 1 policies are compulsory LTC-specific provincial or federal regulations, whereas level 6 policies are voluntary, non-LTCspecific best practice documents. After consultation with project stakeholders, the decision was made to anchor our policy analysis in the highest regulatory levels . Each document was scanned for keywords related to various LTC policy lenses, including the "volunteer lens"; the term "volunteer" was used to search policy documents for inclusion in the "volunteer lens" analysis. Initially, only regulatory policies were searched and included, yielding nine documents. In one of these documents, the volunteer-relevant text excerpts were entirely duplicated; therefore, the older of the two documents was eliminated as being redundant, leaving eight texts. This scant data set prompted the researchers to expand their search to include levels 3 and 4 , which yielded an additional 4 documents . Figure 1 shows how the overall policy library was refined and organized to facilitate a volunteer focus. Inclusion/exclusion criteria, coding, data categorization, and interpretation were discussed and refined regularly at research team meetings. /Next, text excerpts from the 12 documents were inserted into an Excel spreadsheet for inductive interpretive analysis. These excerpts were coded according to Kane's 11 QoL domains, to determine which domains are best supported in existing provincial policy. We followed Kane's QoL domain descriptions in including a sense of: "Safety, Security and Order" meaning that residents can trust that their living environment is benevolent and organized by ordinary ground rules; "Physical Comfort," meaning that residents are free from physical pain and discomfort caused by symptoms or the environment; "Enjoyment," through programming and physical settings; "Meaningful Activity," according to personal preferences; "Reciprocal Relationships" with anyone living, visiting, or working in the LTC facility; "Functional Competence," so that the resident is as independent as possible, depending on impairments; "The Resident's Dignity" or unique humanity must be respected; "Resident Privacy," which means having control over when one is alone and what information is shared about oneself; "Individuality," through residents expressing identity and having a desired continuity with the past; "Autonomy/ Choice" that enables residents to have some direction in their respective lives; and "Spiritual Well-Being," which includes, but is not limited to, religiousness. To guide our coding, we used a modified objective hermeneutics method to interpret policies according to QoL domains. We sought to uncover not necessarily the intent or the potential outcomes of the policy text, but rather how it might be interpreted based on the text alone. For example, if the policy excerpt did not explicitly refer to privacy issues , it was not coded as relevant to the "Privacy" QoL domain. At least two researchers analyzed each text excerpt independently and then compared excerpt coding to ensure consensus on the direct link between the excerpt and particular QoL domain. Text excerpt coding was discussed and finalized through consensus during larger research team meetings. Once coding was complete, emerging themes and key findings were discussed at subsequent team meetings and shared with project stakeholders. Policy data and applicable domains are summarized in Table 1. As mentioned earlier, there are numerous regulatory tensions caused by the decentralized, patchwork nature of Canada's LTC policy, which exist at multiple levels and are sometimes poorly aligned. Therefore, linking policies with QoL domains was often complicated . Ultimately, however, linking QoL domains to these policy text excerpts enabled us to develop a clearer idea of which domains have more leverage in various jurisdictions and which ones might require more work at the policy level to ensure clarity and consistency as to how LTC volunteer activities are coordinated. --- Findings --- Alberta Four pieces of policy were identified in Alberta: two at the regulatory level, and two that are government endorsed. The regulatory policies are categorized as level 2 . They include language mandating that LTC facility operators ensure that volunteers are trained, that volunteers have limited involvement in residents' personal affairs, and that operators maintains a physical environment comfortable for everyone who occupies the facility, including volunteers. The level 2 document Continuing Care Health Services Standards mentions volunteers sparingly as one kind of caregiver requiring proper training before using the facility's equipment, technology, and supplies. This language applies only to the "Safety, Security and Order" domain. The Accommodation Standards and Licensing Information Guide is much more expansive. Six excerpts from this document state that LTC operators must ensure a clean, comfortable environment, including for volunteers; operators must ensure that volunteers receive training in security, communication, and emergency call systems; volunteers must complete a criminal record check; volunteers must be familiar with policies designed to maintain residents' privacy and personal information; LTC facilities must have policies documenting staff and volunteer involvement in residents' personal affairs; and written policies must be provided to residents, their representatives, staff, and volunteers. Altogether, these texts suggest that volunteers are expected in the LTC facility, their physical comfort is prioritized, and they might receive similar training to staff regarding equipment use and technology, privacy, and personal information. The policies so infrequently mention volunteers, however, that volunteers' role in improving residents' QoL remain unclear. The QoL domains coded here are still predominantly "Safety, Security and Order", and touch on "Physical Comfort" and "Privacy" domains. In each instance, however, volunteers are treated as counterparts of paid staff and therefore are subject to much of the same training, policy oversight, and policy documentation. This might be interpreted as LTC volunteer "support", yet it is minimal and does not guarantee an increased or clarified role for volunteers. Despite more frequent mention of volunteers in the Guide, the policies have a narrow focus, addressing only three of the 11 QoL domains. We read this language as restricting volunteers' role rather than encouraging volunteer involvement and support through the creation of opportunities. Non-regulatory, government-endorsed documents suggest more relational and distinctive roles for volunteers. A New Vision For Long Term Care Mirosh Report frames volunteers not as staff counterparts, but rather as outside sector representatives from community organizations and corporate agencies that might contribute funding for older people's service delivery. The document recommends that "the volunteer sector be considered an integral part of the long term care system", and that "every effort should be made" to ensure that 1999), outlines a potentially specific role for volunteers-facilitating good communication with residents by serving as language interpreters-if staff or other caregivers cannot communicate with the resident. Although this certainly enhances the "Relationship" QoL domain, it also has the potential to enhance QoL for residents within the "Functional Competence" and "Individuality" domains. Here volunteers are asked not only to play a resident-centered role, but also to support staff and possibly enhance relationality across various caregivers. Despite these policy excerpts being more relationship oriented than those in Alberta, we note that they enhance residents' QoL in a very small number of domains, with relatively limited scope in each. There appears to be no indication that volunteers might play a broader or more central role in augmenting residents' overall QoL. --- Nova Scotia Long Term Care Program Requirements is the only relevant regulatory policy in Nova Scotia that includes language pertaining to LTC volunteers. In its first statement, volunteers are grouped with employees and are subject to policies and procedures limiting their involvement in residents' personal affairs. This language suggests that, in Alberta and Nova Scotia, there is no overlap between family or close friends and volunteers. In other words, volunteers play a formal rolesimilar to that of staff-distant from residents' personal lives. The second text from this policy document reads that "volunteers are supported and supervised and do not replace paid staff" . Here it is clarified that volunteers might be subject to some staff policies and regulations; however, they are distinct from staff and should not engage in the same work. Nevertheless, volunteer roles and duties are not defined, and it is not clear from this latter excerpt how volunteers might be "supported" or "supervised" or how they differ from staff. There is no mention of volunteers' relationship with, or role vis-à-vis, residents or their families. These policy excerpts differ from British Columbia and Ontario policy in which volunteers play explicit roles in resident councils. The policy language suggests that volunteers are understood more to be workers and distinct or discrete from friends and family; as such, the volunteer role is limited in a number of ways. We coded this policy as potentially enhancing QoL in the "Relationships", "Privacy", and "Safety, Security and Order" domains, although it is unclear, based on our hermeneutics interpretation, how this might happen. One additional text was included at policy level 3: Viral Illness Outbreak Control in LTC . This document lists volunteers as one of several groups that should be included in data collection and training procedures aimed at preventing illness Overall, Nova Scotia regulatory policy and government-sponsored documents provide little indication of how volunteers might have distinctive roles in LTC that differ from, yet complement, those of staff and family caregivers. --- Ontario We found four LTC-specific policies in Ontario that made explicit mention of volunteers. The level 1 documents-Long-Term Care Act Regulations , Long-Term Care Homes Act , and Long Term Care Homes Funding Policy Eligible Expenditures provide relatively rich supports and resources for LTC volunteers and cover several QoL domains. The LTCA Regulations outline the necessity for volunteers to be screened and given some orientation or training before they work with residents. This language is most like that from the other provinces and addresses similar QoL domains: "Safety, Security and Order", "Physical Comfort", "Enjoyment", "Relationships", "Functional Competence", and "Privacy." The LTCHA , however, is by far the most expansive, including several statements emphasizing the importance of supporting communication and healthy relationships across diverse care networks as being necessary to maintaining and improving residents' QoL across 8 domains: "Safety, Security and Order", "Physical Comfort", "Relationships", "Functional Competence", "Individuality", "Meaningful Activity", "Enjoyment", and "Spiritual Well-Being." This document addresses many themes that are present in other previously noted provincial documents, but makes use of different language. For example, Ontario's LTCHA mandates that licensees explicitly invite volunteers to help develop and revise mission statements for the LTC homes and collaborate with resident and family councils. This language articulates significantly more active and creative roles for volunteers than exist in the other jurisdictions in our study, and acknowledges volunteers' contributions as being not only valuable, but perhaps also unique from those of other caregivers. The LTCHA also mandates "an organized volunteer program … that encourages and supports the participation of volunteers in the lives and activities of residents" . This policy text is the most explicit about recognizing the importance of volunteer supports, and highlights the value of volunteers in Ontario. Another interesting level 1 policy feature in Ontario is the listing of "volunteer co-ordinators" as an eligible expenditure under Long Term Care Homes Funding Policy Eligible Expenditures , provided that they "improve the quality of life of residents" . This demonstrates Ontario's strategy to invest material resources in supporting volunteers because of their important role in improving residents' QoL. Without a clear definition of QoL and how volunteers might contribute to it, however, the volunteer coordinator role is discretionary. Nevertheless, this resource allocation is supported in the level 3, government-sponsored Commitment to Care: A Plan for LTC in Ontario , which states that every LTC facility must employ a volunteer coordinator and recruit volunteers through liaising with other community centres, programs, and schools to facilitate intergenerational activities, events, and programming. Facilities should also develop best practices manuals for these volunteer engagement activities. In contrast to other provinces' disciplinary measures, particularly reflected in Alberta's and Nova Scotia's policy documents, Ontario seems to focus on finding resources and structures to support volunteers and mandating a collaborative role for them in resident care. Ontario moves beyond language indicating general volunteer screening and training laws. Rather than peripheral afterthoughts in policy development, volunteers in Ontario are an integral part of LTC vision, even helping to formulate LTC institutions' mission statements. --- Discussion Several themes emerge from the volunteer lens applied to provincial policy documents outlined. Table 2 compares these themes and their applicable QoL domains across all four jurisdictions. The most common themes relate to screening practices, limitations to volunteers' involvement in residents' personal affairs, and supporting relationships among volunteers, staff, and family. These themes are not surprising given increasing licensing requirements, demands for further volunteer regulations to manage facility liabilities and risk, as well as federal regulations to protect "vulnerable" residents through screening and privacy protection legislation. Perhaps unsurprisingly, "Safety, Security and Order" was the most commonly applied QoL domain, showing up in almost every theme identified here, except for the themes I which volunteer roles are particularly well developed and conceptualized as being unique from staff and family roles. "Relationships" is another widely addressed domain in the policy documents that we collected, particularly in Ontario and British Columbia. This is consistent with our literature review, which underscores the critical role that volunteers play in relational care and as go-betweens in complex and diverse care arrangements. This is also in line with the increasing influence of new LTC policy governance , wherein volunteers, as non-state actors, are increasingly seen as integral participants improving LTC quality. There were, however, some revealing absences. Volunteer roles in palliative care policy were given scant attention in the policy documents that we reviewed. Based on our literature review, we know that volunteer roles are particularly developed and crucial in palliative care approaches, yet these are not reflected in provincial regulations or government-endorsed documents. In 2017, Canada's federal government established palliative care policy as a key priority . In our general policy scan, we found 28 government policy documents relevant to palliative care; only one-Palliative and end of life care: Alberta provincial framework -mentioned volunteers. If we consider palliative approaches to care as critical for improving LTC residents' QoL , this is an important area in which volunteer roles could be, but are not currently, well supported by policy. Moreover, QoL domains "Spiritual Well-Being", "Individuality", and "Dignity", were coded very rarely and the "Autonomy/Choice" domain was not coded in any policy documents. From our literature review and stakeholder feedback, we know that volunteers can and do play important roles in enhancing these QoL domains, even if this is not well reflected or supported in policy. It is possible that if volunteers were better represented in palliative care policies, these QoL domains would be better supported. What appears in provincial policy documents is not necessarily representative of all that volunteers do to enhance LTC residents' QoL. Nor do these policy documents represent all the levers shaping volunteers' involvement in LTC. Indeed, institutional policies and procedures and national, non-regulatory frameworks, such Accreditation Canada's Residential Homes for Seniors Standards , certainly exert considerable influence.3 However, they do emphasize QoL domains with the most regulatory leverage in LTC and show which volunteer roles are reinforced and supported through regulation. Overall, we noted that very few documents outline unique volunteer roles. Table 3 sketches how each of the policy documents we included represents volunteer roles by jurisdiction. Many text excerpts were vague, listing volunteers as one of many parties to whom policies apply. There are some interesting provincial differences conceptualizing volunteer roles. In most policy documents, volunteers are only mentioned vis-à-vis staff-directed policies. This is particularly the case in Nova Scotia, where no policy document outlined a unique volunteer role distinct from those of staff. Alberta policy documents4 sometimes conceptualize volunteers as NGOs, corporate sponsors, or community agencies involved in LTC; these specific roles were not captured in other provinces' policy documents. In British Columbia, volunteer roles are largely unclear, except that volunteers might be used as resources for facilitating residents' communication and must be included in collaborative relationship-building processes with staff, family, and residents. In Ontario, all four policy documents recognized volunteers' unique role in LTC and volunteer programming, and through mandatory resource allocation. Moreover, although British Columbia's policy notes that volunteers might play a role in resident and family councils, in Ontario, family and resident councils are invited to collaborate with volunteers. It is unclear whether classifying volunteers in these ways effectively supports their potential to enhance residents' QoL. Although best practice handbooks and documents might exist , there is very little regulatory policy that mandates programming that supports volunteers' unique relational roles in LTC as they are outlined in our literature review. If volunteers are to be regulated, policy language that recognizes and supports these roles through programing might leverage volunteers' opportunities to better address all of Kane's 11 QoL domains. Not only is there scant attention to the specific roles of volunteers, but also very few LTC policy texts mention volunteers at all. Notably only 12 policy documents, out of our pool 139 coded policies, mentioned volunteers. This scarcity might be explained in several ways: volunteer roles are largely casual and/or not well defined or understood by policy makers5 ; volunteers are not considered to be key stakeholders when LTC regulatory policy is being developed ; there have been few documented problems with LTC facilities' volunteers; therefore, there has been little incentive to regulate their behavior further; and/or, volunteers are largely managed through more general policies, such as Accreditation Canada standards, which were not part of our formal analysis. Certainly, these regulatory tensions and lack of role clarity may also restrict rather than enhance volunteers' capacity to improve residents' QoL. Further research is necessary to determine the leading factors leading to scant attention to volunteers in LTC regulatory policy. Enhanced clarification around volunteers' roles and how these various roles may be supported through creative collaboration with other LTC users should be reflected in policy documents. Some of the more promising policy texts, from British Columbia and Ontario in particular, address these issues. Volunteers are explicitly mentioned in relation to British Columbia and Ontario resident and family councils, giving them a voice that seems largely absent in Alberta and Nova Scotia. Rather than framing volunteers as potential risks to residents or caregivers who must be carefully managed, the language in many British Columbia and Ontario policy texts positions volunteers as making rich and unique contributions that improve residents' QoL overall, and emphasizes volunteers' complex relationality within diverse social networks. Ontario, in particular, is a leader in this regard, putting mechanisms in place to encourage collaboration with volunteers and enhance "volunteer voice". Collaboration opportunities give volunteers meaningful and creative, rather than restrictive, roles to play. Allocated funding in Ontario for volunteer coordination also gives volunteers direction and perhaps minimizes the likelihood that care/nursing staff will have to take on the additional labour of volunteer training and direction. Ontario's comparatively rich policy framework for volunteers owes something to the comprehensive nature of their LTCHA, which as Lai argues, reflects a strong adherence to a new governance approach in improving LTC quality. A similar policy framework seems to be emerging in Alberta as well, as is evidenced by the expanded role of volunteers in Palliative and end of life care: Alberta provincial framework , which emphasizes better integration into existing services and systems rather than restricting or inhibiting volunteers' capacity to enhance QoL. These expanded and unique volunteer roles in provincial policy texts show how volunteers might be better supported from the top down . Nevertheless, additional regulation or changes to current regulations may not address these problems. As demonstrated in Tingvold and Skinner 's research on Norwegian volunteer and staff work, there is often insufficient time and attention spent on thoughtfully coordinating volunteer activities mandated or encouraged in policy frameworks into daily LTC processes. Moreover, Funk and Roger suggest that many volunteers do not want to be regulated and that LTC regulations themselves present a barrier for volunteers to improve residents' QoL. Volunteers' restrictions and specific obligations may conflict with volunteer motives and incentives for altruistic relationship building and a sense of meaning or purpose . This leads to questions about the extent of volunteer regulation. Heavily regulated volunteer roles may also contradict goals to make LTC less "institutional" through mutually beneficial relations that encourage more spontaneity and creativity in LTC. These goals may be supported through policies that allow LTC residents to take some risks in their relationships with volunteers. Funk and Roger helpfully suggest a scale of volunteer guidelines that might modify resident and volunteer relationships based on mutually determined terms rather than predetermined or overly prescriptive terms. Finally, even promising LTC policies cannot enhance QoL when austere budgets and opaque decision making dominates. Structural issues around LTC funding and ownership models may shape volunteers' roles in ways that cannot be addressed through the provincial LTC regulations that we analyzed here. For example, Watts argues that austere health care funding tends to follow a business model, incorporating overly managed volunteer roles that run counter to many QoL improvement measures discussed here. Lowndes et al. argue that the increasing reliance on volunteers does not necessarily reflect a recognition of their unique contributions, but rather reflects the need for supplementary labour as LTC staffing levels diminish. Ensuring adequate staff compensation and resources to provide more relational care to residents may ease tensions between staff and volunteers and enhance collaboration . Given this structural context, better supporting volunteers through LTC care policy is just one of many strategies required to make concrete improvements to LTC residents' QoL in Canada. --- Limitations and Future Research Our focus on regulatory and government-endorsed documents that explicitly use the term "volunteer" does not present a comprehensive analysis of who volunteers in LTC facilities, what institutional policies support or inhibit volunteers, or what gaps exist between written policy and practice. Moreover, as one of our project stakeholders stated, provincial policy may be silent in areas that are reflected in national non-regulatory frameworks. Although we have noted differences across jurisdictions, a detailed analysis of the specificities of each regulatory context is beyond the scope of this analysis but warrants further investigation. Questions that require further research include: How much LTC volunteer care is formalized through screening and training as opposed to "unmanaged" or casual roles? Are certain QoL domains better supported through unmanaged volunteer roles? How much of the Canadian volunteer force is motivated by faith-based, ethnocultural, or program-specific LTC activities, and to what degree can these activities enhance the overall QoL of all LTC residents? Is there a role for regulatory policy in enhancing volunteer diversity to better attend to the changing LTC resident population? How does current policy reinforce gender divisions in informal care provision and how might this change at the policy level? Finally, more Canadian research is needed on policy's role in changing relationships among LTC volunteers, staff, and management, and on how these changes could impact volunteers' capacity to enhance LTC residents' QoL. --- Conclusion Demand for LTC volunteers is increasing in Canada and existing literature suggests that the contributions volunteers make to resident QoL are important and unique. Despite limitations, our analysis offers some timely insights into the role of Canada's regulatory policy in both enabling and inhibiting volunteers to enhance LTC residents' QoL. We analyze several gaps in policy, particularly palliative policy, where volunteers currently lack regulatory support for their contributions to residents' QoL, notably in the "Autonomy/Choice" domain. Although we discuss the limitations of relying too heavily on LTC regulatory change for enhancing residents' QoL, we also highlight some promising policy language uncovered in our analysis. This language clarifies and expands unique LTC volunteer roles, prioritizes concrete resources to support these roles, and offers guidance on how to facilitate volunteers' active collaboration with LTC staff, families, and residents to better enhance QoL in multiple domains. Developing such regulatory policy in each jurisdiction provides consistent clarity and support for volunteers that may result in tangible improvements to resident quality of life.

La recherche a mis en évidence le rôle important des bénévoles en soins de longue durée (SLD), particulièrement pour l'amélioration de la qualité de vie des résidents plus âgés, par leur apport unique en soins relationnels. Nous avons utilisé une méthode d'herméneutique objective modifiée et inspirée des domaines de la qualité de vie identifiés par Kane pour analyser la représentation et les appuis associés aux rôles uniques des bénévoles dans les politiques provinciales en Alberta, en Colombie-Britannique, en Ontario et en Nouvelle-Écosse. Nous avons constaté que les politiques définissent étroitement les rôles des bénévoles, ce qui peut limiter la qualité de vie des résidents. En effet, cela se traduit par (1) des politiques réglementaires qui, en grande partie, omettent de mentionner les bénévoles (2) les assimilent à du personnel supplémentaire plutôt qu'à des soignants ayant un rôle spécifique, et (3) mettent trop d'emphase sur la sécurité des résidents et l'ordre. Nous proposons dans cette étude un aperçu d'orientations politiques provinciales prometteuses pour les bénévoles en SLD. Cependant, nous soutenons aussi que l'augmentation de la réglementation visant les bénévoles pourrait constituer une réponse inadéquate ou mal adaptée, considérant les changements culturels, sociaux et structurels qui seraient nécessaires pour que les bénévoles puissent contribuer à l'amélioration de la qualité de vie des résidents en SLD.

I. INTRODUCTION Human behavior is profoundly affected by the influenceability of individuals and the social networks that link them together. Well before the proliferation of online social networking, offline or interpersonal social networks have been acknowledged as a major factor in determining how societies move towards consensus in the adoption of ideologies, traditions and attitudes [1,2]. As a result, the dynamics of social influence has been heavily studied in sociological, physics and computer science literature [3][4][5][6][7]. In the sociological context, work on diffusion of innovations has emphasized how individuals adopt new states in behavior, opinion or consumption through the influence of their neighbors. Commonly used models for this process include the threshold model [8] and the Bass model [9]. A key feature in both these models is that once an individual adopts the new state, his state remains unchanged at all subsequent times. While appropriate for modeling the diffusion of innovation where investment in a new idea comes at a cost, these models are less suited to studying the dynamics of competing opinions where switching one's state has little overhead. Here we address the latter case. From among the vast repertoire of models in statistical physics and mathematical sociology, we focus on one which is a 2-opinion variant [10] of the Naming Game [11][12][13][14][15] and that we refer to as the binary agreement model. The evolution of the system in this model takes place through the usual NG dynamics, wherein at each simulation time step, a randomly chosen speaker voices a random opinion from his list to a randomly chosen neighbor, designated the listener. If the listener has the spoken opinion in his list, both speaker and listener retain only that opinion, else the listener adds the spoken opinion to his list . The order of selecting speakers and listeners is known to influence the dynamics, and we stick to choos-ing the speaker first, followed by the listener. It serves to point out that an important difference between the binary agreement model and the predominantly used opinion dynamics models [4,6,[16][17][18] is that an agent is allowed to possess both opinions simultaneously in the former, and this significantly alters the time required to attain consensus starting from uniform initial conditions. Numerical studies in [10] have shown that for the binary agreement model on a complete graph, starting from an initial condition where each agent randomly adopts one of the two opinions with equal probability, the system reaches consensus in time T c ∼ ln N . Here N is the number of nodes in the network, and unit time consists of N speaker-listener interactions. The binary agreement model is well suited to understanding how opinions, perceptions or behaviors of individuals are altered through social interactions specifically in situations where the cost associated with changing one's opinion is low [19], or where changes in state are not deliberate or calculated, but unconscious [20]. Furthermore, by its very definition, the binary agreement model may be applicable to situations where agents while trying to influence others, simultaneously also have a desire to reach global consensus [21]. Another merit of the binary agreement dynamics in modeling social opinion change seems worth mentioning. Two state epidemic-like models of social "contagion" suffer from the drawback that the rules governing the conversion of a node from a given state to the other are not symmetric for the two states. In contrast, in the binary agreement model, both singular opinion states are treated symmetrically in their susceptibility to change. Here we study the evolution of opinions in the binary agreement model starting from an initial state where all agents adopt a given opinion B, except for a finite fraction p of the total number of agents who are committed agents and have state A. Committed agents, introduced previously in [23], are defined as nodes that can influence other nodes to alter their state through the usual prescribed rules, but which themselves are immune to influence. In the presence of committed agents adopting state A, the only absorbing fixed point of the system is the consensus state where all influenceable nodes adopt opinion A -the opinion of the committed agents. The question that we specifically ask is: how does the consensus time vary with the size of the committed fraction? More generally, our work addresses the conditions under which an inflexible set of minority opinion holders can win over the rest of the population. The effect of having un-influencable agents has been considered to some extent in prior studies. Biswas et al. [24] considered for two state opinion dynamics models in one dimension, the case where some individuals are "rigid" in both segments of the population, and studied the time evolution of the magnetization and the fraction of domain walls in the system. Mobilia et. al. [25] considered the case of the voter model with some fraction of spins representing "zealots" who never change their state, and studied the magnetization distribution of the system on the complete graph, and in one and two dimensions. Our study differs from these not only due to the particular model of opinion dynamics, but also in its explicit consideration of different network topologies and of finite size networks, specifically deriving how consensus times scale with network size for the particular case of the complete graph. Furthermore, the above mentioned studies do not explicitly consider the initial state that we care about -one where the entire minority set is un-influencable. A notable exception to the latter is the study by Galam and Jacobs [26] in which the authors considered the case of "inflexibles" in a two state model of opinion dynamics with opinion updates obeying a majority rule. While this study provides several useful insights and is certainly the seminal quantitive attempt at understanding the effect of committed minorities, its analysis is restricted to the mean-field case, and has no explicit consideration of consensus times for finite systems. --- II. COMPLETE GRAPHS A. Infinite network size limit We start along similar lines as [26] by considering the case where the social network connecting agents is a complete graph with the size of the network N → ∞. We designate the densities of uncommitted nodes in states A, B as n A , n B respectively. Consequently, the density of nodes in the mixed state AB is n AB = 1 -p -n A -n B , where p is the fraction of the total number of nodes, that are committed. Neglecting correlations between nodes, and fluctuations, one can write the following rate equations for the evolution of densities: dn A dt = -n A n B + n 2 AB + n AB n A + 3 2 pn AB dn B dt = -n A n B + n 2 AB + n AB n B -pn B Before interaction After interaction The terms in these equations are obtained by considering all interactions which increase the density of agents in a particular state and computing the probability of that interaction occurring. Table 1 lists all possible interactions. As an example, the probability of the interaction listed in row eight is equal to the probability that a node in state AB is chosen as speaker and a node in state B is chosen as listener times the probability that the speaker voices opinion A . The fixed-point and stability analyses of these mean-field equations show that for any value of p, the consensus state in the committed opinion is a stable fixed point of the mean-field dynamics. However, below p = p c = 5 2 -3 2 A A → A A -A A A → B A -AB A A → AB A -A B B → A B -AB B B → B B -B B B → AB B -B AB A → A A -A AB A → B AB -AB AB A → AB A -A AB B → A AB -AB AB B → B B -B AB B → AB B -B 3 5 + √ 24 -1 2 - 3 2 3 5 - √ 24 -1 2 , two additional fixed points appear: one of these is an unstable fixed point , whereas the second is stable and represents an active steady state where n A , n B and n AB are all non-zero . Fig. 1 shows the steady state density of nodes in state B obtained by numerically integrating the mean-field equations at different values of the committed fraction p and with initial condition n A = 0, n B = 1 -p. As p is increased, the stable density of B nodes n B abruptly jumps from ≈ 0.6504 to zero at the critical committed fraction p c . A similar abrupt jump also occurs for the stable density of A nodes from a value very close to zero below p c , to a value of 1, indicating consensus in the A state . In the study of phase transitions, an "order parameter" is a suitable quantity changing from zero to a non zero-value at the critical point. Following this convention, we use n B -the density of uncommitted nodes in state B -as the order parameter appropriate for our case, characterizing the transition from an active steady state to the absorbing consensus state. In practice, for a complete graph of any finite size, consensus is always reached. However, we can still probe how the system evolves, conditioned on the system not having reached consensus. Fig. 1 shows the results of simulating the binary agreement model on a complete graph for different system sizes . For p < p c , in each realization of agreement dynamics, neglecting the initial transient, the density of nodes in state B, n B , fluctuates around a non-zero steady state value, until a large fluctuation causes the system to escape from this active steady state to the consensus state. Fig. 1 shows these steady state values of n B conditioned on survival, for several values of p. As expected, agreement of simulation results with the mean-field curve improves with increasing system size, since Eq. represents the true evolution of the system in the asymptotic network size limit. Accordingly the critical value of the committed fraction obtained from the mean-field equations is designated as p c , although, for brevity, we refer to it simply as p c throughout this paper. The existence of the transition as p is varied and when the initial condition for densities is can be further understood by observing the motion of the fixed points in phase space. Fig. 1 shows how the stable fixed point and the unstable fixed point move in phase space as p is varied from 0 to p c . The active steady state moves downward and right while the saddle point moves upwards and left. At the critical value p c the two meet and the only remaining stable fixed point is the consensus fixed point. A similar observation was made in the model studied in [26]. The fact that the value of n B converges to ≅ 0.65 and does not smoothly approach zero as the stable fixed point and the saddle point approach each other, explains the origin of the first-order nature of the phase transition. --- B. Finite network size: Scaling results for consensus times Even though consensus is always reached for finite N , limits on computation time prohibit the investigation of the consensus time, T c , for values of p below or very close to p c . We therefore adopt a semi-analytical approach prescribed in [27] that allows us to estimate the consensus times for different N for an appreciable range of p including values below p c . We start with the master equation which describes the evolution of the probability that the network of size N has n uncommited nodes in state A . We denote by c, the number of committed nodes, and by l, the number of uncommitted nodes in state AB. dp nm dt 1 N = 1 N 2 -p nm + 2lm + mc) + p n-1,m 3 2 + p n+1,m 2 + p n-2,m 2 + p n,m-1 3 2 + p n,m+1 2 + p n,m-2 2 The factor of 1/N in the LHS comes from the fact that a transition between states takes place in an interval of time 1/N . The transition rates in each term are the product of two densities, which is responsible for the overall factor of 1/N 2 in the RHS. The probabilities are defined over all allowed states of the system and the allowed transitions from any point {n, m} in the interior of this state-space are {n, m} → {n, m ± 1}, {n, m} → {n ± 1, m}, {n, m} → {n, m + 2}, {n, m} → {n + 2, m}. We know from the mean field equations that in the asymptotic limit, and below a critical fraction of committed agents, there exists a stable fixed point. For finite stochastic systems, escape from this fixed point is always possible, and therefore it is termed metastable. For a finite system, the probability of having escaped to the metastable fixed point as a function of time is P e = 1 -P s where P s is the survival probability. The surviving fraction is constrained to be in the allowed region of n, m quadrant excluding the true fixed point {N -c, 0}. If the number of committed agents is far lower than p c N we expect that this surviving fraction will occupy configurations around the metastable fixed point, and the occupation probabilities p n,m will be peaked around the metastable fixed point. In systems which exhibit such long lived metastable states in addition to an absorbing fixed point, applying a quasistationary approximation has been found to be useful in computing quantities of interest [27][28][29]. This approximation assumes that after a short transient, the occupation probability, conditioned on survival, of allowed states excluding the consensus state, is stationary. Following this approximation, the distribution of occupation probabilities conditioned on survival can be written as, pnm = p nm /P s [27] and using this form in the mas-FIG. 1: The steady state density nB of nodes in state B as a function of committed fraction p for complete graphs of different sizes, conditioned on survival of the system. Simulation results are from 100 realizations of the binary agreement dynamics. Movement of the stable fixed point and the saddle point in phase space as a function of committed fraction p . The location of these points in phase space was obtained through a stability analysis of the mean field equations Eq. . The inset shows the density of nodes in state B at the stable and unstable fixed points as p is varied; the expressions from which these numerical values are obtained are given in the Appendix. ter equation ), we get: dP s dt pnm = - P s N pnm + 2lm + mc) -pn-1,m 3 2 -pn+1,m 2 -pn-2,m 2 -pn,m-1 3 2 -pn,m+1 2 -pn,m-2 2 Considering transitions from states {N -c -1, 0} and {N -c-2, 0} to the absorbing state {N -c, 0}, we obtain the decay rate of the survival probability dP s /dt: dP s dt = -P s pN-c-1,0 3 2N + pN-c-2,0 2 N (4 ) Substituting Eq. into Eq. , we finally obtain a condition that the occupation probabilities conditioned on survival must satisfy [30] : pnm = Qnm W nm -Q0 where, Qnm = Q nm /P s is obtained through explicit consideration of the terms in the master equation: Qnm = pn-1,m 3 2 + pn+1,m 2 + pn-2,m 2 2 + pn,m-1 3 2 + pn,m-2 2 2 + pn,m+1 2 and Q0 = pN-c-1,0 32 + 2 pN-c-2,0 is the term arising from the decay of the survival probability ). W nm is the coefficient of p nm within the brackets on the right hand side of Eq. ) and is equal to the transition rate out of state {n, m} times N 2 . Eq. indicates that the survival probability decays exponentially with a rate λ = Q0 /N . Since the mean lifetime of an exponentially decaying process is the inverse of the decay rate, it follows that the mean consensus time is T c ≅ 1 λ = 1/ pN-c-1,0 3 2N + pN-c-2,0 2 N (7 ) Thus, knowledge of pnm s would allow us to calculate T c through Eq. . In order to obtain pnm , we adopt the iterative procedure proposed in [30]. Following this procedure, we start with an arbitrary initial distribution p0 nm , and obtain a new distribution using: pi+1 nm = α pi nm + Qi nm W i nm -Qi 0 , where 0 ≤ α ≤ 1 is an arbitrary FIG. 2: Trajectories ) in the phase-plane show the nature of flows from different regions of the phase-plane into existing fixed points for a) p = 0.05 and b) p = 0.1 parameter, and Qi nm , W i nm and Qi 0 are all obtained using the probability distribution at the current iteration, pi nm . With sufficient number of iterations, this procedure is expected to converge to a distribution that satisfies Eq. and which is thus, the desired QS distribution. In our case, we obtained acceptable convergence with a choice of α = 0.5 and 30000 iterations. Following the above method, we obtain the QS distribution, and consequently the mean consensus times T c for different values of committed fraction p and system size N . Fig. 3 shows how the consensus time grows as p is decreased beyond the asymptotic critical point p c for finite N . For p < p c , the growth of T c is exponential in N , consistent with what is known regarding escape times from metastable states. For p > p c , the QS approximation does not reliably provide information on mean consensus times, since consensus times themselves are small and comparable to transient times required to establish a QS state. However, simulation results show that above p c the scaling of mean consensus time with N is logarithmic ). A snapshot of the QS distribution near p c for a system of size N = 100 shows clearly the bimodal nature of the distribution, with the two modes centered around the stable fixed point, and the consensus fixed point. The precise dependence of consensus times on p can also be obtained for p < p c by considering the rate of exponential growth of T c with N . In other words, assuming T c ∼ expN ), we can obtain α as a function of p. T c ∼ exp ν N ) This exponential growth is presumably modulated by factors of log N which become dominant only when p = p c . Above p c , the dependance of T c on p as seen from simulations is negligible . --- III. SPARSE NETWORKS Next, we present simulation results for the case when the underlying network topology is chosen from an ensemble of Erdős-Rényi random graphs with given size N and given average degree k . The qualitative features of the evolution of the system in this case are the same as that of the complete graph, although the critical fraction p c displays some dependence on k . For small k and fixed N , the drop in consensus times occurs slightly earlier in p for ER graphs than for a complete graph of the same size, as shown in Fig. 5. However for p > p c , a complete graph has shorter consensus times than an ER graph of the same size. Above p c , the difference between consensus times for a graph with average degree k and the complete graph, ∆T c , decays approximately as power law with increasing k ). The deviation from a perfect power law is likely due to other weaker k dependent terms, presumably logarithmic in k . We also performed simulations of the binary agreement model on Barabasi-Albert networks ), and found similar qualitative behavior as observed for ER networks including the difference from mean-field behavior. We leave a detailed analysis of the dependence of the critical fraction p c and the consensus times T c on the average degree k of sparse networks, for future work. --- IV. SUMMARY In closing, we have demonstrated here the existence of a tipping point at which the initial majority opinion of a network switches quickly to that of a consistent and inflexible minority. There are several historical precedents for such events, for example, the suffragette movement in the early 20th century and the rise of the American civil-rights movement that started shortly after the size of the African-American population crossed the 10% mark. Such processes have received some attention in sociological literature under the term minority influence [26,31]. Our motivation here has been to study this process in more detail through semi-analytical methods and simulations for finite-sized and sparse networks, within the realm of a particular social influence model -the binary agreement model. There are several open questions and extensions of this work that are worth studying, in our opinion: for example, given a network with non-trivial community structure, what is the optimal scheme for selecting committed agents that would minimize consensus times, and reduce p c ? Secondly, extensions of the model to include utilitydriven opinion switching by agents may be useful in designing optimal incentive schemes for opinion spreading. Here, we analyze the mean-field equations for the existence of fixed points. To simplify notation we use the notation x = n A and y = n B . Thus for a fixed point of the evolution given by Eq. : -xy + 2 + x + 1.5p = 0 -xy + 2 + y -yp = 0 which can be reduced further to: x = [ 2 -9p 2 /16]/ y = 2 Substituting the expression of x into the expression for y, and denoting z 2 = y we get zz + p/2 + 1) = 0 For any value of p, z = z 0 = 0 is a solution to the above equation. In other words, for any value of p, the meanfield equations admit a stable fixed point, n A = x 1 = 1 -p, n B = y 1 = 0 which represents the network having reached a consensus state where all nodes have adopted the opinion of the committed agents. The remaining fixed points are roots of f = z 3 -z + p/2 + 1 = 0. (12 ) In order to find the criterion which has to be satisfied for valid roots )) of the above equation to exist, we analyze the extrema of the function f which are given by: Thus, the extrema occur at : f ′ = 3z 2 -2 + p/2 = 0 z 1,2 = ± 2/3 -p/6 It can be seen from Eq. that f is increasing, decreasing and increasing again in the intervals ,, respectively. Consequently, f achieves a maximum at -1 < z 1 = -2/3 -p/6 < 0 and a minimum at 0 < z 2 = 2/3 -p/6 < 1. Furthermore, since f = -p/2 -3 < 0 and f = 2 > 0, one root of f = 0 occurs in the interval -2 < z < -1. Since f is positive at z 1 , decreasing from z 1 to z 2 where it achieves a minimum, and increasing thereafter, it follows that a necessary and sufficient condition for more roots of f = 0 to exist, is that f be less than zero: f = z 3 2 -z 2 + p/2 + 1 < 0 Denoting z 2 = q and p = 4 -6q 2 ) yields the following inequality for q as a condition for more roots of f = 0 to exist: f = q 3 + 1.5q 2 -1.5 > 0 . Analyzing the derivative of f enables us to glean that the inequality Eq. is satisfied for q > q 0 where q 0 is the solution of the cubic equation f = 0 and is given by: q 0 = 3 5 + √ 24 + 3 5 - √ 24 -1 /2 Thus, the original fixed point equation Eq. has at least one valid root besides z = 0, so long as p is less than or equal to: p c = 5 2 - 3 2 3 5 + √ 24 -1 2 - 3 2 3 5 - √ 24 -1 2(16 ) which using standard computer algebra software is evaluated to be p c = 0.09789. Using, z 2 = y = q 0 and Eq. , we obtain the state of the system at p c to be {n A , n B } = {0.0957, 0.6504}. It also follows from the expression for f , that f > 0 and therefore if f is negative, Eq. has two roots on the positive line when p < p c . Thus there are two fixed points of Eq. when p < p c . The exact expressions for these fixed points , obscure their dependence on p. We therefore adopt an approximation which exhibits a much clearer dependence of the fixed point values on p, and numerically yield values close to those obtained from the exact expressions. Substituting z = t 2 -p 2 in Eq. reduces it to: t 3 -t + r = 0 (17 ) where r = 3 for 0 ≤ p < p c , our range of interest. Function g = t 3 -t is monotonically decreasing for t < -1 and g = 0, while g < -1 < -r. Hence, there is a real root t 1 ∈ to Eq. which is a monotonically decreasing function of r, but which clearly does not yield a valid fixed point. This root can be expressed as t 1 = -2 √ 3 + ǫ, where ǫ is monotonically decreasing from less than 0.0106 to 0 over the range of our interest for r. Substituting this expression back into Eq. 17 and neglecting powers of ǫ higher than unity, we get an approximation of ǫ in terms of r, and consequently an approximation for t 1 : t 1 ≈ - 16 9 √ 3 - r 3 (18 ) with relative error of less than 0.01%. Now, we can factorize the LHS of Eq. and write it as . Equating this factorized expression with t 3 -t + r, gives us b and c in terms of r. Thus, two more roots of Eq. are obtained in terms of r by solving the quadratic equation t 2 + bt + c = 0 which yields: t 2,3 = 8 9 √ 3 + r 6 ± 17 81 - 8r 9 √ 3 - r 2 12 (19 ) Finally, we can obtain the values of z associated with the above roots, and therefore the values of x and y written in terms of these roots are derived as: y 2,3 = t 2 2,3 4 -p 2 x 2,3 = 2 -9p 2 8p + 16 The stability of these fixed points can be checked via linear stability analysis. Following the standard procedure, the stability matrix is given by: From the expression for the eigenvalues we numerically determine that the real part of both the eigenvalues is negative for over the range 0 ≤ p < p c indicating that that is a stable fixed point. This is however, not the case for , making it unstable. Similarly, the consensus fixed point is found to be stable for 0 ≤ p ≤ 1. Finally, we note that as p → 0, the stable fixed point converges to n A = 0, n B = 1, while the unstable fixed point converges to n A = n B ≅ 0.38.

We show how the prevailing majority opinion in a population can be rapidly reversed by a small fraction p of randomly distributed committed agents who consistently proselytize the opposing opinion and are immune to influence. Specifically, we show that when the committed fraction grows beyond a critical value pc ≈ 10%, there is a dramatic decrease in the time, Tc, taken for the entire population to adopt the committed opinion. In particular, for complete graphs we show that when p < pc, Tc ∼ exp(α(p)N ), while for p > pc, Tc ∼ ln N . We conclude with simulation results for Erdős-Rényi random graphs and scale-free networks which show qualitatively similar behavior.

Background From 2000 to 2019, suicide deaths nearly halved; however, after 2006, the decline of the suicide rate was broken and stagnated for several years [1]. After 2010, suicide mortality fell steadily and substantially again: the declining trend remained stable, and the suicide rate decreased by more than one-third which was remarkable even from an international perspective. Following these favourable trends, the COVID-19 outbreak appeared at the end of 2019 in Wuhan, Central China. On 11 March 2020, the World Health Organisation declared the outbreak a pandemic. The first case and first death in Hungary were officially registered in early-and mid-March 2020, respectively. In addition to the damaging effects of COVID-19 on the central nervous system [2], there are some social effects of the pandemic that increase the stress levels of individuals [3]: isolation, entrapment, disruption of normal routine, interpersonal conflicts, fear of infection, unemployment. The increased workload in the whole health care system impairs access to mental health services for chronic psychiatric patients, which can also lead to worsening of their condition. In Hungary, the COVID-19 pandemic increased the number of excess deaths. Similarly, during the COVID-19 epidemic , there were almost 11% more suicides reported in Hungary than in the same period the year before [4]. Additionally, the number of male suicide deaths increased to an even greater extent during the same period in Hungary. The effect of the COVID-19 pandemic on suicide mortality by gender was examined in detail in a recent study [5]. However, regional inequalities have always characterised the distribution of suicide mortality in Hungary. We might hypothesise that the excess mortality could be also observed in suicide mortality compared to the 10 years before the pandemic. Our aim was to investigate this hypothesis in Hungary by gender, age, educational attainment, and region, as well. --- Methods --- Study population and suicide data Data on the population were obtained from the published nationwide population register operated by the Hungarian Central Statistical Office . The annual mid-year population estimates were used since there were no monthly population data available. As trend stability in the period is crucial in interrupted-time series, the 10 years between 2010 and 2019 were considered in this analysis since the annual trend in suicide rates for the whole population remained stable during this period, unlike in previous years. The data on suicide deaths were also available on the online HCSO database [4]. These data were classified according to the International Classification of Diseases, 10 th revision ; codes concerning "intentional self-harm" were X60-X84 and Y87. Suicide rates were expressed per 100,000 population per year using the annual mid-year population estimates for the relevant year. The suicide rates were directly standardised by age [6] using the Revised European Standard Population published in 2013 [7] to facilitate a comparison of rates over time by removing the effect of age composition. Population and suicide deaths were initially broken down by age group as follows: 0-19 , 20-34 , 35-49 , 50-64 and over 65 years . The distribution of the RESP was as follows: 0-19 years 21.5%, 20-34 years 18.5%, 35-49 years 21%, 50-64 years 19.5%, and over 65 years 19.5%. Nonetheless, due to the lower number of cases in the under-20 age group, the first two groups were analysed together. The second level of the NUTS 2013 classification [8] served as the foundation for the territorial units. The seven regions of Hungary were as follows : Central Hungary , Central Transdanubia , Western Transdanubia , Southern Transdanubia , Northern Hungary , Northern Great Plain and Southern Great Plain . Educational attainment was categorised according to the ISCED-97 system [9,10]: less than eight years of primary school , eight years of primary school , vocational qualification without a secondary school-leaving certificate , secondary school-leaving certificate and higher education degree . Population data were only available for the group aged 15-74 years; in addition, the first two groups were aggregated as "at most primary school" due to their low number of cases and the "unstable" population of the first group . --- Statistical analyses The observed SR during the pre-vaccination period of the COVID-19 pandemic was compared with the expected SR based on the pre-COVID-19 period . Since the Hungarian SRs decreased in the last years, an interrupted time series analysis was conducted to control for annual trends and seasonal variations [11]. In the absence of "intervention" , this trend would remain "unchanged". This is a common method to apply in such situations. In the current phase of the analysis, we only investigated the effect of the "intervention", no other structural indicators were considered. Accordingly, the following segmented regression impact model was used: where Y t is the outcome at time t , T is the time since the beginning of the study , X t is a dummy variable indicating either the pre-intervention period or the postintervention period . The baseline level at T = 0 is represented by β 0 , the change in outcome associated with a time unit Y t = β 0 + β 1 T + β 2 X t increase is represented by β 1 , and the level change after the intervention is represented by β 2 . Due to over-dispersion , we conducted a negative binomial regression analysis to identify suicide trends in the pre-COVID-19 period, with the log-transformed mid-year population as offset variable . Seasonal effects were adjusted by harmonic terms using pairs of sine and cosine functions : where m represents the month . cos 2kπm 12 , sin 2kπ m 12 , k = 1, 2 To test whether SRs during the COVID-19 pandemic were out of line with the declining trend of previous years, expected and observed suicides of the COVID-19 period were compared by calculating incidence rate ratios and 95% confidence intervals . Analyses were carried out overall and then separately by sociodemographic factors , as stratified by levels of a single factor. Consequently, the models mentioned above included the corresponding stratumspecific population as an offset variable. To ensure the robustness of our results, we performed sensitivity analyses for a shorter period, starting in January 2015 . In favour of completeness, suicide rates during the prevaccination period were also compared directly to the same period of the previous year . p-values less than 0.05 were considered statistically significant. All analyses were performed using R . --- Results Table S1 displays summary statistics for suicide cases and the corresponding population at risk during various periods. The number of population and suicides were expressed as monthly averages to make periods comparable. Percentages were expressed as proportions of the total number . During the pre-COVID-19 period between January 2010 and February 2020, the average number of suicide deaths in Hungary was 164 per month. There were 23 fewer suicides per month between January 2015 and February 2020 ; this indicates a decrease in the monthly suicide rate of approximately 13%. In addition, rates of suicide decreased at each level of each factor. During the pre-vaccination period of the pandemic , 1438 suicides were registered in Hungary; this was about 11% more than in the same period of the previous year . Additionally, suicide rates increased in almost all subgroups: the regions of Western and Southern Transdanubia were the only exceptions. The interrupted time-series analyses were performed for the whole population and separately by gender, age group, educational attainment, and region, respectively . The COVID-19 period showed a significant rise in suicide deaths compared to the projected numbers : overall, among men and people aged 35 to 49 , respectively. The estimated IRRs were 1.167 for the general population, 1.185 for males, and 1.328 for the group aged 35-49 years, respectively. The results of sensitivity analysis were nearly identical to those previously indicated . Regarding educational attainment, significant increases in suicide mortality were revealed among those who graduated from vocational school and those who completed at most primary school. The corresponding IRRs were 1.261 and 1.178 , respectively. However, in the sensitivity analysis, no significant increase for the latter group was detected . There were also differences by region . The increase was significant in the case of. Central Hungary , Central Transdanubia and the Southern Great Plain ; the corresponding IRRs were 1.273 , 1.222 and 1.193 , respectively. In the sensitivity analyses, similar significant increases were observed in the regions of Central Hungary and Central Transdanubia but non-significant in the Southern Great Plain . Nonetheless, it might also be worth noting that a decrease was only observed in the case of Southern Transdanubia . Referring to the yearly comparisons, suicide rates increased significantly overall , among males , in the age group over 65 years , and in the region Central Hungary , respectively . --- Discussion --- Main findings The number of suicides increased significantly by 16.7% compared to the hypothetical trend that would have occurred without the appearance of COVID-19; significant rises were also observed in the male and aged 35-49 years subpopulations; furthermore, within the age group 15-74 years, among vocational school graduates . There were also some differences by region: the suicide rate increased significantly only in two regions, in Central Hungary and Central Transdanubia . --- Comparison with other studies Since suicide mortality can increase during infectious disease outbreaks [12], it was hypothesised that the COVID-19 pandemic might also elevate suicide mortality. However, a preliminary international study conducted in April-July 2020 that investigated 21 higher-income countries concluded that there was "showed no evidence of a significant increase in the risk of suicide since the pandemic began in any country or area" [13]. Despite this, it is still unclear how the COVID-19 pandemic has affected suicide deaths or at least can vary by country. Recently published papers have reported static [14], decreasing [15], and even ascending [16] suicide rates in the context of the pandemic. A recent Hungarian study [5] revealed a significant increase during the first year of the pandemic. One of the possible reasons for this is that social acceptance of suicide is still high, so Hungarian people tend to turn to suicide as a 'solution' in a crisis like the pandemic. In addition, 'normal' health care had to be cut back due to the pandemic , and psychiatry in Hungary suffered particularly: in many hospitals, beds and specialists were removed from here at the earliest and returned here at the latest. Gender differences might be explained by social factors . Moreover, although men are generally more exposed and susceptible to social and psychological stress, their compliance with the therapy is poorer than female patients' . Conversely, the increase in suicide mortality during the COVID-19 pandemic was more marked for females in Japan [17]. However, it is worth mentioning that from 2010 to 2019, the extent of the decline in suicide rates for men was nearly one and a half times higher than for women in Hungary . A significant increase in suicide rates was only found in the age group of 35-49 years. This might be attributed to the multiple burdens of being an adult during the pandemic: the members of this age group can be active workers , parents , and children of their ageing parents at the same time . In contrast to this, a study in Taiwan found that suicide rates decreased in the middle age group [18]. Again, it can be noted that from 2010 to 2019, the suicide rate declined to the largest extent in this age group. Our previous report revealed that among the levels of educational attainment, the greatest percentage drop in suicide mortality had been seen in the group of people graduating from vocational school and the largest male-to-female ratio had been also observed in this group between 1998 and 2017 in Hungary [19]. Moreover, the strict measures of the first wave temporarily limited the physical presence in the workplace which primarily affected manual workers and craftsmen. Furthermore, in the sectors of accommodation and catering, fishery, and agriculture, the proportion of people who are not able to work from home is over 80 per cent [20]. All the above could have contributed to the finding of a significantly increased suicide rate for vocational school graduates during the pre-vaccination period in Hungary. Previous studies have reported significantly higher suicide mortality rates in the Northern and Southern Great Plain compared to Central Hungary, which includes the capital city of Hungary . Pre-intervention period: white background; post-intervention period: coloured background; observed rates: points; fitted rates/trends : continuous lines; counterfactual scenario : dashed line. Notes: Only those groups were displayed here for which the effect of the pandemic on suicide rates was significant both in the main and sensitivity analyses. Regions were displayed on the map . However, a significant increase in suicide mortality was observed in Central Hungary during the COVID-19 pandemic. This might be caused by the decline in the 'bustling metropolitan life' due to the restrictions. In the case of Central Transdanubia, the increase may have similar reasons , supplemented by the decline in the automotive-and processing industry . Consequently, the regions of Central Hungary and Central Transdanubia were hardest hit by the first wave of COVID-19. At the peak of the first wave, Budapest and the county of Pest were among the most infected counties of Hungary, together with the counties of Fejér and Komárom-Esztergom [23]. Moreover, the registered unemployment rate has significantly increased in Budapest and its agglomeration, along with the settlements around Lake Balaton [24,25]. Interestingly, these two regions recorded the lowest mortality rates due to COVID-19 during the study period in Hungary . Additionally, according to the available data, divorce rates were highest in these two regions between 2012 and 2019 . In summary, our study has described the relationship between the impact of the COVID-19 pandemic and suicide mortality in Hungary during the pre-vaccination period. --- Strengths and limitations As far as we are aware, this is the first study analysing the suicide trends in Hungary concerning the COVID-19 pandemic by age group, educational attainment, and region. The longest study period was used in the interrupted time-series analyses. Although the vital statistics performance index of Hungary is one of the best in the world [27] and the percentage of garbage-coded deaths is also quite low in Hungary [28], there can be cause-of-death biases. Moreover, only deaths from "intentional self-harm" were considered in our analyses ; however, there are other death causes which can qualify for deaths by suicide [29]: "poisoning of undetermined intent" , "other events of undetermined intent" and "late effects of other events of undetermined intent" . Consequently, the undercounting of suicide deaths cannot be ruled out. However, we are confident that our results do reflect real trends. Some cells in the online HCSO database contained protected data. We considered these 'missing' fields to be 1 . This can cause some discrepancies in marginal numbers, but these differences are negligible. Overdispersion did not influence our results, as the NB regression method was employed in the analyses. We applied a divided age-group structure and thus calculated age-standardised mortality rates to make suicide rates comparable over time. Although the age variable could have been simply included in our models to control for age composition, we decided to standardise rates instead . Notwithstanding the above, age-standardisation did not offer any real advantages in our analyses as populations at risk were relatively stable over time in most cases. However, it should be mentioned that age-standardisation was not possible in the case of educational attainment where the concerning populations were particularly less stable in some cases . The HCSO's public mortality data have a lead time of two years; that is, there have been no available data for suicide deaths in Hungary during 2021 and 2022 yet. Consequently, we have only been able to investigate suicide deaths till December 2020. Accordingly, as the pandemic progresses, the patterns described here may shift. In addition, it is essential to keep in mind that vaccinations began in Hungary on December 26, 2020; as a result, we can say that the "pre-vaccination period" was analysed. The possibility of ecological fallacy is another limitation of our analysis: individual-level associations may not always be reflected in ecological-level associations. --- Conclusion Our findings demonstrate how the COVID-19 pandemic affected suicide mortality in two of Hungary's most affluent regions before vaccination. Although the underlying causes are unclear, economic variables may help to partially explain this . To our knowledge, this has been the first study to examine so comprehensively the effect of the COVID-19 pandemic on suicide rates in Hungary. In general, there was a significant rise in the number of suicides in Hungary before the vaccination period . The pattern of suicide related to the COVID-19 pandemic has been described in our study by gender, age group, educational attainment, and region; these patterns were different from one another. Since the identification of groups at higher risk during the pandemic may be crucial to suicide prevention , these findings may prove useful for preventive strategies. However, further research is required to determine their causes. • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year --- • At BMC, research is always in progress. --- Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: --- --- Abbreviations --- --- --- Funding Open access funding provided by University of Szeged. No funding was received for conducting this study. --- --- --- Competing interests The authors declare that they have no conflict of interest. ---

Background: From 2010 to 2019, suicide mortality fell steadily and substantially in Hungary: the declining trend remained stable, and the suicide rate decreased by more than one-third which was remarkable even from an international perspective. However, despite the declining trend, regional inequalities have always characterised the distribution of suicide mortality in Hungary. Following these favourable trends, COVID-19 appeared in Hungary on the 4 th of March 2020 which might lead to an increase in suicides. We aimed to investigate this hypothesis in Hungary by gender, age, educational attainment, and region, as well. Methods: To test whether the pandemic changed the declining trend of Hungarian suicide rates, the observed number of suicides during March-December 2020 (pre-vaccination period) was compared with the expected numbers (without the appearance of COVID-19). An interrupted time-series analysis was conducted by negative binomial regression using monthly data from January 2010 to February 2020 (pre-pandemic period). Results: Suicide mortality increased significantly compared to the trend during the pre-pandemic period: overall (by 16.7%), among males (18.5%), in the age group 35-49 years (32.8%), and among vocational school graduates (26.1%). Additionally, significant growths in suicide rates were detected in the two regions (Central Hungary and Central Transdanubia) with the lowest COVID mortality rates (by 27.3% and 22.2%, respectively).Our study revealed reversed trend in suicide mortality during the pre-vaccination period compared to the pre-pandemic period in Hungary. There were significant differences in the pattern of suicide rates by gender, age group, educational attainment, and region during the pre-vaccination period in Hungary, which might be attributed to the socio-economic effects of the COVID-19 pandemic. These findings could prove useful in preventive strategies as the identification of groups at higher risk may be important for suicide prevention; however, further investigations are needed to explore the reasons.

Introduction Prosocial development depends on the cultural system in which one grows up [1], and support for this argument comes from a series of cross-cultural studies. For example, residents in India are on average less cooperative than US residents in one-shot interactions in the laboratory setting [2][3]. Previous research has also shown that Indians are more spiteful than Americans when making decisions about real monetary allocations between themselves and another anonymous participant [4]. This empirical evidence suggests that culture plays an important role in prosocial behavior, yet few studies have examined the influences of cultural values on prosocial behavior/development within one specific culture. Thus we tried to explore the effects of cultural values on prosocial behavior based on a sample of Chinese adolescents. As traditional Chinese culture emphasizes hierarchy, and people are socialized to comply with the obligations and rules attached to their roles in the hierarchical system [5], values like power distance, authority, humility, and wealth are highly important in Chinese culture. In addition, establishing and maintaining interpersonal harmony via prosocial interactions is highly endorsed by Chinese culture , and this cultural norm is particularly in the case when people tend to establish positive contacts with powerful people for meeting individual needs . Therefore, we aimed to examine the effect of power distance value on adolescents' prosocial behavior toward powerful people. On the other hand, with the enormous economic progress and increasing globalization it has made over the past forty years, Chinese society has been changing dramatically. The economic and social changes correspond to the introduction of individualistic values from the western cultures [9]. These values have been gradually accepted by many Chinese people, especially in the younger generation [1,10]. In this regard, the subject of the present study would shed some light on understanding adolescents' social functioning and adjustment in the changing society. Moreover, little research has been conducted to examine how adolescents' cultural values might be shaped/influenced by social behaviors, such as prosocial behavior. The culturally guided social interaction processes including prosocial behavior toward powerful people likely serve as an important socialization agent of children's cultural value development [11]. Thus we tried to extend the existing literature by examining the longitudinal and bidirectional relations between adolescents' power distance value and prosocial behavior toward powerful people in the present study. --- The role of power distance value on prosocial behavior among Chinese adolescents Power distance has been defined as the extent to which the less powerful members of organizations and institutions accept and expect that power is distributed unequally [12]. Individuals with higher power distance value are more likely to view inequality as a result of different social positions as natural and even desirable than those with lower power distance value [13]. In line with this definition, individuals with higher power distance value usually have more positive attitudes toward the powerful and also higher willingness to interact with the powerful than people with lower power distance value . Thus, it may be logical to assume that adolescents with higher power distance value would be more prosocial toward powerful people. In the present study, powerful people refer in particular to parents, teachers, and others who have capacity to influence or control adolescents by providing valued resources [15]. Psychological processes and behaviors are attuned to the particular cultural meaning system with which the individual develops [16]. Growing up in a high power-distance culture, Chinese people are more culturally guided to and then are more inclined to build and maintain positive relationships with powerful people for desired resources or outcomes . Because responding prosocially to powerful people is an efficient way to build or maintain positive relationships with them , we supposed that, within the Chinese culture adolescents who hold higher power distance value might be more likely to help powerful people than those holding lower power distance value. --- The role of prosocial behavior on Chinese adolescents' power distance value Humans interact with and seize meanings from their cultures, and then internalize specific cultural values by these processes [16]. From a very young age, resulting from different experiences, children within one culture might behave differently, think in different ways, and gradually acquire some cultural values to different extents. In line with this reasoning, frequent prosocial interactions with powerful people might help adolescents obtain more desired resources , and have more positive relationships with the powerful . In this way, adolescents might be more likely to view power differentials as acceptable and socially normative, and consequently hold higher power distance value. Furthermore, self-perception theory [19] provides a theoretical guidance for us to understand how prosocial behavior toward powerful people might impact adolescents' power distance value. The theory supposes that people observe their own actions to make inferences about inner states, such as attitudes and values, and it is particularly likely to occur when the behavior is perceived as voluntary. For instance, empirical studies have shown that adolescents' prosocial practices in the family context could predict increases in familism values [20], and helping strangers is positively correlated to adolescents' subsequent moral values [21]. On the other hand, intervention research has demonstrated that encouraging pursuing goals that stand in opposition to materialism could effectively inhibit materialistic values among adolescents [22]. All these findings might point toward an assumption that adolescents' behavior could influence their value development. According to the self-perception theory and the instances, we assumed that adolescents who are more willing to help powerful people would take power distance more seriously in order to rationalize their behavior. With more and more prosocial behaviors toward the powerful, the adolescents might view the inequality among persons in different positions as a natural aspect of the social order, and hold positive beliefs about the power distances between people. Taken together, we expected to see a positive relationship between adolescents' prosocial behavior toward powerful people and subsequent power distance value. --- The current study In the present study we were interested in exploring the longitudinal and bidirectional relationships of adolescents' power distance value and prosocial behavior toward powerful people based on a Chinese sample. This attempt is meaningful in terms of understanding the nature and significance of the interplay between cultural values and adolescents' social behaviors. We hypothesized that power distance value would be positively and longitudinally associated with adolescents' prosocial behavior toward powerful people, and in return, earlier prosocial behavior toward powerful people would elicit higher power distance value among Chinese adolescents. --- Methods --- Ethics statement This study was approved by Institutional Review Board of School of Sociology and Psychology at Central University of Finance and Economics. Written informed consent was obtained from all the parents of the participants. --- --- Measures Power distance value. Adolescents' power distance value was assessed using the 4-item power distance subscale of the cultural value orientation scale [24]. The example items included "I believe older people in the family should make the important decisions and younger ones should not question their decisions" and "I believe students should obey their teachers, not question them". Participants rated the items on a five-point response format from 1 to 5 . The reliabilities at Time 1 and Time 2 were .69 and .75. Latent variables for power distance value at two time points were created. Measurement weak factorial invariance across time was tested [25], and invariance was reached = 6.78, p > .05). Factor loadings ranged from .38 to .80. The only item with a factor loading below .40 was "I believe superiors should be able to make important decisions without having to ask for the opinions of others ". This relatively low factor loading might be due to that adolescents were not so familiar with the "manager-subordinate" dyads, and then failed to give consistent answers. Prosocial behavior toward powerful people. This construct was measured by a modified version of the kindness and generosity subscale of the Values in Action Inventory of Strengths [26]. The present study adapted six of the items to assess prosocial behavior toward powerful people . Respondents answered on a 5-point Likert-type scale from 1 to 5 in terms of how much they disagreed or agreed with statements about themselves. Latent variables for prosocial behavior toward powerful people at two time points were created. Measurement weak factorial invariance across time was reached = 3.84, p > .05). The reliabilities at Time 1 and Time 2 were .91 and .92 with factor loadings ranging from .59 to .94. Demographic information. Adolescents' gender, age, whether being the only-child in the family or not, parents' educational levels, and monthly family income were investigated. As for parents' educational levels, the participants answered on a 5-point scale . Following the approach of a previous study [27], the present research measured monthly family income via a 5-point scale . --- Attrition analyses Attrition analyses were conducted via a series of t-tests and chi-square tests. As described previously [23], boys or those with less educated fathers and mothers were more likely to drop out of the investigation at Time 2. Regarding the study variables, adolescents with less prosocial behavior toward powerful people were more likely to drop out . As the participants were in the 6 th grade at Time 1 and were in the 7 th grade at Time 2, some non-resident students had to transfer back to the middle schools in the place of their own residence, which resulted in the attrition. These non-resident students who were absent at Time 2 usually had less educated parents and lower social status, which might influence their socialization, and then had less prosocial behavior toward powerful people. Whereas no significant differences on power distance value, and other demographics were found. In sum, though the final sample was more prosocial toward powerful people, more from highly-educated families, and became more female, it did not experience significant attrition in terms of power distance value, participant's age, and monthly family income. --- Results --- Descriptive statistics and correlations Descriptive statistics and correlations of the study variables are shown in Table 1. The results indicated that adolescents' power distance value and prosocial behavior toward powerful people at Time 1 and Time 2 were positively correlated with one another. We then ran repeated measures ANOVAs to test the mean differences over time and the interactions of time point and gender on the study variables. The main effects of both time point and gender on power distance value were significant = 79.86, p < .001, partial eta 2 = .16, 1-β = 1.00; gender: F = 7.53, p = .006, partial eta 2 = .17, 1-β = .78). Adolescents hold lower power distance value at Time 2 than at Time 1 , and boys had higher power distance value than girls . No interaction effect between time point and gender was found for power distance value = 0.04, p = .84). For prosocial behavior toward powerful people, only the main effect of time point was significant = 15.58, p < .001, partial eta 2 = .04, 1-β = .98; gender: F = 0.80, p = .37; interaction effect: F = 0.16, p = .69). The adolescents were less likely to help powerful people at Time 2 than at Time 1 . --- Cross-lagged models Cross-lagged models were implemented using Mplus software with the two waves of adolescents' power distance value and prosocial behavior toward powerful people. Given the gender differences found above, multiple-group comparisons across adolescent gender were conducted. The Δχ 2 test suggested that constraining the cross-lagged paths to be identical across gender groups did not elicit a significant decrease in model fit = 1.81, p = .40), thus a singlegroup model was run as the final model. Because of the sample attrition and the differences on the demographics between the retained adolescents and those who were absent at Time 2, gender, parents' educational levels, as well as monthly family income and whether being the only- child in the family or not were treated as the control variables. The final model had good model fit = 384.49, p < .01, CFI = .97, TLI = .97, RMSEA = .04, SRMR = .05). As shown in Fig 1, adolescents' power distance value at Time 1 was not significantly associated with prosocial behavior toward powerful people at Time 2 . Whereas prosocial behavior toward powerful people at Time 1 was positively associated with power distance value at Time 2 . Both the stability paths of power distance value and prosocial behavior toward powerful people from Time 1 to Time 2 were significant. None of the associations between the control variables at Time 1 and the study variables at Time 2 was significant . --- Discussion The purpose of the current study was to explore how specific cultural value and adolescent social behavior would influence each other over time. By examining the longitudinal and bidirectional relationships between power distance value and prosocial behavior toward powerful people among Chinese adolescents, we found that earlier prosocial behavior toward powerful people was positively correlated to subsequent power distance value, but not vice versa. The findings point toward an understanding of the important role of adolescent social behavior on his/her cultural value development. --- Relation between earlier power distance value and subsequent prosocial behavior The current research did not find the expected positive effect of power distance value on adolescents' prosocial behavior toward powerful people. Perhaps it was due to the relatively less salient role of power distance value on predicting prosocial behavior. According to Ajzen's Theory of Planned Behavior [28], the best predictor of a behavior is intention, which is determined by three factors: the person's attitudes/values toward the behavior, perceptions of important others' approval of the behavior , and perceptions of the ease or difficulty of performing the behavior . Although values motivate behavioral choices and actions [29], power distance value might be not one of the dominated factors in predicting prosocial behavior, as prosocial behavior seems to be highly associated with adolescents' dispositional traits and other-oriented values like emphasizing promoting others' welfare . Moreover, with the rapid economic and social changes in China, Chinese traditional values have been merging with the western individualistic values and ideologies such as independence, which is especially the case for children and adolescents . This changing nature of Chinese cultural values might also partly explain the nonsignificant correlation between earlier power distance value and subsequent prosocial behavior toward powerful people. On the other hand, as adolescents' values are somehow malleable and are still developing, power distance value during early adolescence might have an emerging effect on prosocial behavior, but the relatively short-term one-year design of the present study did not allow us to see significant changes induced by power distance value. Further research is needed to explore more waves of data to examine these patterns into middle and late adolescence. --- The effect of prosocial behavior toward powerful people on power distance value Consistent with our assumption, prosocial behavior toward powerful people at Time 1 was positively correlated to power distance value at Time 2. Based on the self-perception theory and given the formation nature of values during adolescence, adolescents are inclined to internalize and develop their values by observing and interpreting their behaviors. Because Chinese adolescents are socialized to treat superiors seriously and rightly [32], conducting prosocial behavior toward powerful people can serve as an important external cue for them to reason their attitudes and values about power differentials. By helping powerful people, adolescents might gain potential profits from the powerful ones. For instance, those students always doing some favors for teachers would be more likely to be liked and trusted by their teachers [33][34], which is very common in Chinese schools. Gradually, those adolescents having frequent prosocial interactions with the powerful would believe that treating powerful people prosocially would bring benefits and superiorities in many aspects [35]. Finally, adolescents who preferred to help powerful people would get used to the differences between people with different power levels, and be more willing to obey the cultural norms emphasizing power differentials. In this way, prosocial behavior toward powerful people facilitated the socialization of adolescents' power distance value. --- Gender difference regarding prosocial behavior toward powerful people No significant gender difference on prosocial behavior toward powerful people was found in our research, though most of the former studies have shown that females are significantly more prosocial than males. For example, several meta-analyses indicated that females are more altruistic in Dictator Game experiments [36][37], are more cooperative in social dilemmas [38], and are more honest in Sender-Receiver games [39] than males. However, there is also empirical evidence showing that adolescent boys and girls have the same levels of prosocial behavior toward family, whereas girls are more prosocial toward friends and strangers than boys [23]. One explanation of the mixed findings is that the role of gender on prosociality might vary across the target toward whom prosocial behavior is directed. In this regard, future research is desirable to test how gender differences in prosociality would vary as a function of target. --- Limitations and contributions Limitations of this study are addressed. First, attrition analyses showed that the retained adolescents were more prosocial toward powerful people than those who were absent at Time 2, which restricted to some extent the generalization of the findings. Second, as the current study was based on a sample of Chinese adolescents who were growing up in a relatively high power distance culture, such studies need to be carried out in other cultures with different levels of power distance to generalize the findings. Even so, the present research has important contributions by figuring out the longitudinal relations between power distance value and prosocial behavior toward powerful people among adolescents. It provides clear evidence for the significant role of prosocial behavior toward powerful people on the development of adolescent power distance value. More broadly, the findings would enlighten future research in terms of understanding the nature and significance of the interplay between cultural values and individual's social behaviors. --- All relevant data are within the manuscript and its Supporting Information files. --- Supporting information --- S1 File. The raw data of this study. --- --- Funding acquisition: Xinyuan Fu. Investigation: Xinyuan Fu. Methodology: Xinyuan Fu, Yichen Lv, Zhixu Yang, Xiaoxia Yu. Writing -original draft: Xinyuan Fu, Yichen Lv, Zhixu Yang, Xiaoxia Yu, Rongrong Wang. Writing -review & editing: Xinyuan Fu, Zhixu Yang, Rongrong Wang.

We were interested in how specific cultural value and adolescent social behavior would influence each other over time. Thus the present study explored the longitudinal and bidirectional relations between adolescents' power distance value and prosocial behavior toward powerful people over a year. A sample of 434 Chinese adolescents participated in the investigation (initial mean age = 11.27; 54.15% females). The results based on cross-lagged models showed that, earlier prosocial behavior toward powerful people was positively correlated to subsequent power distance value, but not vice versa. The findings point toward an understanding of the important role of adolescent social behavior on his/her cultural value development, and also shed light on future research in terms of the interplay between cultural values and individual's social behaviors in other cultures.

Introduction In their groundbreaking contribution to understanding international migration at the end of the millennium, Massey et al. in 1998 stated their credo "development creates migration". The authors aimed to point out that a global surplus in socioeconomic terms contributes to intensifying migratory movements. Since then, the number of international migrants worldwide has increased from 173 million in the year 2000 to a pre-pandemic estimated figure of 281 million in 2020, representing 3.6 percent of the world's total population . With a view to the transition contexts of Eastern and Southeastern Europe, we can easily rearrange the aforementioned quotation and determine that transition creates migration. The dissolution of the Eastern Bloc contributed to a massive post-socialist mobilisation that continues in some countries to this day. The main drivers of this trend were the migrants' search for employment and income in the face of massive socio-economic disparities, particularly between the global North and South. With this in mind, we will ask two questions about the contribution. Firstly, and concerning transition contexts, to what extent does migration create development? As a result, the so-called migration-development nexus, an approach that underlines the relationship between migration and development due to transnationalism, emerges . The assessments of the nexus so far range from migration as a "new development mantra" and a warning against another "round of enthusiasm" to the effects of migration in regions of origin. The positivistic and more critical assessments depend on changing phases of global development dynamics, including economic crises. What is interesting for our consideration in discussions like these is that migration is not understood solely as a problem but also as a resource, i.e. as part of a possible solution to a problem people face. Secondly, and this is the central question for our considerations, the extent to which migration thus represents a means of social resilience must be clarified, as this would provide a new perspective for future analysis. The study will show that changes in individual living conditions -concerning material existence, family situation or personal well-being -stimulate migration. Even if there is little aspiration towards mobility, people may often migrate . In this respect, aspirations, decisions, decision-making and the characteristics of movements are central factors in the resilient strategies of individuals who use migration to counter permanent risks and uncertainty or other threats. The individual context of migration and migrants' strategies are therefore the main focus of this study in empirical terms in order to develop a new perspective on the relationship between migration and resilience. With this aim in mind, we will first explain the conceptual and empirical approach in the paper, and then use findings from the examples considered to support the thesis of migration as a means of social resilience. Taking these findings into account, the next part will provide further detail on the theoretical foundation. --- 2 The migration-resilience nexus as a new perspective Adger defines social resilience as "the ability of groups or communities to cope with external stresses and disturbances resulting from social, political and environmental change". A more differentiated definition distinguishes three dimensions, comprising coping capacities , adaptive capacities and transformative capacities, i.e. the power "to craft sets of institutions that foster individual welfare and sustainable robustness towards future crises" . Both studies, like many others, interpret social resilience using an actor-related approach and address the resilience of migrant groups after migration. In line with this, many scholars analyse how migrants adapt to the new situation following arrival with needs such as social integration, education and access to the labour and housing market. This shows that the topic of "resilience of migrants" is well established in relation to "a wide range of issues including social, economic and psychological stressors associated with settlement and integration process" . Last but not least, to broaden the range of very different perspectives on the view, the field of political science sees resilience as a "neoliberal form of governmentality that emphasises individual adaptability" . This "way of mobilising social agents" leads to a more dynamic, agent-based reading of what resilience could be. Revisiting the dimensions by Keck/Sakdapolrak , actors in transition societies have various options to cope with the crisis beyond migration. In contrast, the "adaptive capabilities" mentioned above are absent in the case of Albania and "transformative capacities" are at best questionable in some countries of East and Southeast Europe, including both Albania and Latvia, at least as regards sustainable robustness in times of crises such as the pandemic or the war in Ukraine. As part of theoretical reflections such as these, and based on our previous research and our own current empirical studies , we propose adding a new approach by switching perspectives to understand migration as a kind of 'a means to an end'. This new paradigm, the migration-resilience nexus, interprets migration as a 'tool' for social resilience, i.e. the strategy for solving an individual or collective problem. Thus, it is an element of migrants' agency, keeping in mind at the same time that "To be resilient is not to have eliminated vulnerability" . We see the paradigm in line with de Haas's aspirations-capabilities framework, an approach that also emphasises "the instrumental dimension of migration [that] reflects the role of migration as a way to achieve other […] goals such as increased income, education, living standards or, in the case of refugees, personal safety" . In this perspective, migration is the reaction of individuals to personal threats and/or socio-economic instability. The latter is true for millions of inhabitants of East and Southeast Europe in subsequent times of societal and/or turbulent economic transition. In many cases, this meant a drastic destabilisation of the living conditions in most parts of the former socialist world. Often enough, for those affected, these processes were accompanied by a deterioration in quality of life, de-skilling and devaluation, marginalisation and a sense of having been left behind, and not least the loss of identity. All these aspects are related to the blind spot of migration studies, i.e. "Psychosocial Wellbeing" , which also plays a prominent role in migration and decision-making. This was also true in the two countries we selected for our case studies, with all their similarities and differences described in the next section. --- 3 The conceptual and empirical approach In this section, our aim is to underline the migration-resilience nexus as a new perspective, using empirical evidence from research conducted in two spatial contexts from Eastern Europe's periphery. Our study refers to the Baltic region and the Western Balkans or, more precisely, to Latvia and Albania. This enables us to view the migration-development nexus through the lens of both directions, i.e. emigration and return. From the conceptual point of view, this is a necessity for social resilience related to mobility. Furthermore, we use the comparison and counterposition of experiences of migrants with different origins to better understand the compound consequences of movements in the short and long term. This means that we have included two countries with diverse economic, political and historical framing, an explanation of which is required. --- The Albanian migration society: from Kurbet to Asylum Generally, every geographic phenomenon in Albania is connected to migration: The country is living with and from migration. For good reason, King labelled the country a "laboratory for migration studies". Albania is a "migration society" . Migration culture is deeply rooted in Albanian history, which can be seen if we go back to the 15th century and think about the "Kurbet" when Arbëresh migrated to Italy due to the expanding Ottoman Empire . Kurbet is still present in the Albanian language and means living in exile or, respectively, diaspora , which is the norm rather than an exception. Even the chosen path of consecutive self-isolation and Stalinist socialism after the Second World War, when all forms of migratory movement were tightly regulated by the state, could not alter this in any way. Consequently, emigration was a repeated motive immediately after the end of communism and the lifting of mobility restrictions, when several thousands of migrants boarded the vessel Vlora in 1991 during the so-called "Exodus", and, the post-1990 mass migration of Albanians to Italy and Greece started . Migration is still a tool for social resilience in Albania. One of the striking results of the census in 2011, two decades after opening the borders, was that almost half of the population had had a personal international migration experience . Figure 1 shows that the exodus has continued unabated since then. With fluctuating migration volumes, the annual balance was negative and amounted to another 246,000 people, equivalent to 8.7 percent of the country's total population of around 2.8 million. According to official data from 2019, 1.64 million Albanians live abroad . Most emigrants stay in Greece and Italy . Other destinations are the United States, Canada and Germany. The main driver for emigration is the search for employment and income . World Bank data report annual remittances of an estimated USD 1.5 billion on this basis, equivalent to 10 percent of the country's gross domestic product and substantially impacting the national economy. Other non-economic motivations are family reunification and education . Albania has remained a candidate for membership of the European Union since 2014, while Latvia has been a member since 2004, offering its citizens the privilege of full access to the EU labour market -one of the striking differences between both countries in terms of labour migration. --- The case of Latvia Latvia, in the 21st century, is a country of emigration. The 20th century was characterised by a changing "migration landscape", especially after the Second World War. It was described as forced migration when many refugees went to the West or were deported to Siberia but massive immigration also took place. Latvia was annexed to the Soviet Union during the second half of the 20th century and experienced four decades of constant immigration, mainly from the Russian Federation . As a result, the dissolution of the Soviet Union provoked a significant wave of return migration in the early 1990s, followed by a substantial increase in migration movements and demographic change over the next 30 years . Net migration has constantly been negative since then. Enlargement of the EU in 2004 significantly increased emigration from Latvia, particularly during the 2009-12 economic downturn . Return migration by Latvian nationals accounts for a sizeable portion of immigration flows, for example, 68.1 percent and 53.5 percent of all immigrants in 2013 and 2020 respectively . Significant modifications to migratory patterns were brought about in 2004 by Latvia's accession to membership of the EU. This created numerous options for migration, but the unrestricted movement of labour was particularly advantageous. At first, only the UK, Ireland and Sweden -three of the 15 so-called old EU Member States -offered similar options to the new Member States . In 2011, Germany and Austria finally lifted limitations on Latvia and the other EU-8 states. The 2008 global economic crisis had a significant impact on migration. As a result of the economic downturn, Latvia's GDP fell by 18 percent, and unemployment rose sharply in all regions of the country . This, in turn, contributed to an increase in emigration. In the post-crisis period, migration volumes also stabilised as the economic situation improved . The UK and Ireland were the primary destinations in the early years of EU enlargement. The UK began to take the lead during the global financial crisis, with fewer people travelling to Ireland on account of the economic hardship that it was experiencing. As of May 2011, when work and residency permits for nationals of the EU-8 were no longer necessary, many Latvians relocated to Germany. Although migrant flows to Germany rose significantly post-crisis, the UK remained a key destination . Furthermore, they increased in Norway while decreasing again in Ireland. This is a product of Ireland's strong social networks, which are seen as a source of support despite the country's dismal economic position. Individuals seeking a secure base are drawn to Norway because of its independent financial stability and potential prosperity . The study's further examination of the distribution of respondents found that it also accurately depicted the geographic location of migratory movements. --- The comparative approach Regardless of the diverse framing in Albania and Latvia, "international migration in the periods of transition and crisis" intensified at the start of the 1990s. Since then, both countries have displayed a pattern of continuous outmigration. In fact, and this is the connecting bracket in our comparative analysis, both countries belong to those parts of the former socialist world which were massively affected by long-term outmigration in times of transition. This makes them interesting for geographic migration studies, as related effects such as a widespread diaspora, population decline, brain drain and "peripheralisation" seem comparable. The authors used this argument in an earlier migration-related comparative study on transnationalism and transregionalism . For good reason, Lulle draws a similar Balkan-Baltic comparison with a focus on underlying peripheralisation processes at a later stage. Thus, we consequently analyse how far migration patterns fit with the perspective of migration as a tool for social resilience by following hypotheses. First, after socialism, there were fewer limits on immigration, which increased mobility and occasionally perpetuated earlier migratory trends. Second, emigration is a person's reaction to the existential crises they encountered throughout the shift and an effort to improve their circumstances. Third, migration may be used as a social resilience tactic under some conditions. Migration as social resilience thus covers individual needs in the short and medium term. It may contribute to protecting the shortcomings of the labour market in the receiving context; vice versa, the nexus can lead to long-term development impulses through transnational lifestyles and possible return. Research on the first component of the working hypothesis is best done using qualitative studies on Albania's transition, and on the second component using quantitative empirical data from Latvia. At the same time, findings on the third component are derived from both parts of the study. This needs detailed explanations of research methods, which are given in the following section. --- Research methods The study uses a mixed methods approach. Findings on Albanian migration are gained through qualitative fieldwork in several research projects which started in 2006 . These surveys are based on semi-structured in-depth interviews. During the study on migration and entrepreneurship in 2006, we gathered information on a total of 34 companies and entrepreneurs , while the study on return migration in 2012 contained 40 biographical interviews with returnees and addressed, additionally, relevant institutions in southern Albania . The third study addressed asylum seekers from Albania and Kosovo who were temporarily housed in a refugee reception centre in Bamberg in 2016. There, we conducted 20 qualitative interviews in the migrants' native language. Additionally, we had the chance to carry out another interview with a family that had returned from Bamberg to Elbasan in May 2016. In contrast, findings on the Latvian case are gathered using a quantitative approach to investigate mainly the emigration motivations in the sense of migration as a tool for social resilience. Evidence is based on a one-of-a-kind dataset, collected and analysed from an online internet poll of 2,567 Latvian individuals living in five key destination countries, namely the United Kingdom , Ireland, Germany, Sweden and Norway. We utilised a customised data collection strategy to reach an appropriate target group. The survey took place in 2016 and was carried out on a Latvian social media site which serves as a well-known communication platform among Latvian migrants. When the first survey was conducted in the UK to investigate Latvian residents living abroad, it was done so in collaboration with draugiem.lv . The top 5 migration destinations were the subject of the second survey conducted in 2012. New question blocks have been added to each subsequent survey while potential bias has been recognised and the sample selection has been improved. Participants for the survey were chosen based on the profile of the typical user in the destination country . The data gained reflect the concentration of Latvian immigrants in other countries, as their geographical location or IP address recognised registered social media site users at the time of the poll, and thus represents a significant source of information . The increasing availability of information and communications technology and the current state of human mobility are inextricably linked . The widespread use of ICT expands the possibilities for more forms of international migration and adds a new dimension to migration research . Sample size and selection vary depending on the general population: the possibility of bias in internet survey mode, which excludes non-internet users a priori, must be considered. In contrast to statistical data obtained at the macro level, primary and unique data from the survey have an advantage. They provide information about the participants' motives for departure and reveal specific concerns not addressed by conventional statistics. The completed study sample totalled 2,567 people, including 1,028 Latvians living in the UK, 500 in Ireland, 500 in Germany, 239 in Sweden and 300 in Norway. Young people have typically accounted for a sizeable share of intra-EU migration. This is also the case with Latvia. Young adults form the group with the highest mobility . The respondents' ages range from 16 to 69 years, with the 16-34 years age group having the highest proportion of emigrants. The questionnaire was completed by 976 men and 1581 women . Two interconnected factors may explain the higher proportion of female respondents. Women are not only more active users of digital social media platforms , but they are also more likely to be involved in the scientific investigation . --- 4 The case studies 4.1 Voices from Latvia: Reaction to the economic crisis shifts from absorptive to transformative response Migrants who have moved overseas must decide whether to stay or return to their native country. This query is relevant to both the life trajectory and future migration plan. Social resilience is undeniable in demanding times and takes, according to Dagdeviren and Donoghue , three primary forms. These are agentic responses to the economic crisis, each with a different temporal frame: absorptive, or coping capacities, in which people make short-term adjustments to deal with the shock of a sudden worsening of their circumstances, frequently at the expense of their wellbeing; adaptive capacities, where people make decisions to safeguard and stabilise their well-being over the long or medium term; and transformative capacities, when individuals alter their lives in ways that improve them and make them less susceptible over time. In their study, Czaika et al. 2021 emphasise that "migration decisions are made at significant turning points in people's lives, determine and are determined by longterm life trajectories, and bring lasting consequences for the decision-maker and people affected by the decisions" . In this sense, our data on Latvian migration is exceptional for studying return intentions because they offer insight into migratory behaviour and are distinctive regarding migrant characteristics. The study asked questions relating to the respondents' demographics, their socioeconomic standing, reasons for emigrating, advantages and particular livelihood aspects. Respondents were asked to select three main reasons for emigrating to other European countries . The authors contend that emigration is a person's attempt to improve their circumstances and respond to the existential crisis experienced throughout the shift to address the hypothesised question. The motivation for leaving was made explicit by indicating three key factors behind individuals' decision to emigrate. Latvian ex-pats most frequently cited low pay, a shortage of jobs at home, and an inability to make repayments as the main reasons they left. As seen in this figure, the inability to make daily payments and the lack of employability were mentioned as the main reasons by both migrant groups who intended to stay in or return to Latvia. Poor wages were regarded as the primary reason for both migrant groups, with family motives also playing a crucial role. Uncertainty about the future is one of the factors that both migrant groups value differently . Differently valued opportunities for new experiences are also mentioned more frequently by migrants who intend to return to Latvia. Unemployment has been mentioned more frequently by potential returnees as a reason for leaving. Uncertainty about the future reflects the impact of the crisis on the migrant's psychosocial well-being and is one of the reasons that discourages them from returning to their homeland. Similarly, Ķešāne , through an analysis of post-Soviet emigration from Latvia towards the West, establishes the relevance of the inequality-emotion link in migration on the migration-emotion nexus . It marks both an absorptive, adaptive, and transformative response. Respondents were asked to list the top three advantages of emigration. Figure 4 demonstrates the numerous advantages of investing in human capital. Both potential returnees and stayers mentioned that their ability to speak a new language had improved, and they had acquired new skills and knowledge. For potential returnees, earning money for a specific purpose is a decisive benefit . This money could be used, for instance, to repay a loan. Opportunities for professional development and new life experiences are also important . Although earning money for a particular cause is an absorptive solution, it may also be a guarantee for a transformative one. A number of different family situations are also mentioned, with family matters becoming more essential attributes. Besides having greatest relevance for young people, the aspiration to travel and seek out new experiences demonstrates the importance of transnational links between countries. These elements correspond to international migration traits and show that lifestyle migration has become more common in addition to economic and family migration . Polish immigrants had a similar experience, and non-economic causes are becoming more significant . In the context of return, we should also pay attention to psychosocial wellbeing, which underlines the importance of contacts, social and emotional harmony, and the individual experience . Binary regression analysis was used to determine the degree to which the traits of potential return migrants differed from those of stayers . The study's model is based on the reasons for and advantages of migration for likely stayers and returnees. The model's findings indicate that men are more likely than women to return to Latvia. Compared with other age groups, younger migrants may return more frequently. A partner in Latvia increases a migrant's likelihood of returning. --- Fig. 4: Benefits to be gained from a stay abroad The same applies to rural migrants, particular when compared with migrants living in urban areas. A comparison of destination countries clearly demonstrates that Germany and Sweden have lower return rates than the United Kingdom. Less likely to return are those who left due to unemployment, lack of employment prospects and low-income levels. On the other hand, there are no differences between those who cited a lack of prospects and those for whom migration is a repeated experience. Comparing this with prior studies , it is unique. Those who used to be jobless in Latvia were also able to return because of significant increases in employment rates since the end of the economic crisis. Nevertheless, low salaries deter people from returning to Latvia. However, relatives and friends abroad may diminish the likelihood of returning to Latvia. The possibility of leaving Latvia can also increase with acquired language skills, new professional skills and education abroad. Migrants who left during the crisis are more likely to return than those who left during the post-crisis period. According to the findings, post-crisis migrants appear to feel less obliged to stay abroad and the time spent since the emigration may have had an influence; the longer a person stays abroad, the less likely they are to return. Overall, migration motivations that are influenced by the economic context are driven by necessity; those who move to another country for a specific short-term goal or to save money for a particular purpose are more likely to return than those who benefit from new knowledge and skills abroad. Psychological stress and future uncertainty affect crisis migrants more than other groups, thus illustrating the relationship between migration and social adaptability. A lack of employment prospects nevertheless means that there is less of a desire to return to Latvia. The model results show that migration can be used as a tool for social resilience or as a response to a shock that demands swift action, thereby endorsing the earlier hypothesis. When someone loses his or her job or source of income, they cannot make daily or credit payments on time. The person decides to travel abroad as a result of this. The possibility of competitive pay and, consequently, the capacity to earn for a specific purpose, attract migrants. It supports the hypothesis that emigration is both a personal response to the crisis and a point of departure while looking for a short-term solution. Additionally, migration is seen as a strategy to protect and stabilise their wellbeing in the medium or longer term. People look for better working conditions, career options, and prospects abroad because of low income, a lack of job opportunities, and worries about the future. The advancement of a migrant can be viewed as an investment in new knowledge and language learning that makes them better off in the longer term. --- Voices from Albania: Effects of three decades of a migrant society Generally speaking, Albanian emigrants are well integrated into the host countries, both economically and socially. However, the majority of emigrants continue to have a strong relation with their country of origin, whether through visiting relatives and friends during the summer holidays and attending family celebrations or 'merely' virtual contacts through social media. Many Albanian emigrants are actively embedded in transnational networks. That kind of mass emigration is often connected with complex back-and-forth movements and bears a high potential for an eventual return. In fact, Albania is also a good example of migratory countermovements as a lot of migrants have returned in the past 30 years. The respective spatial and temporal context is decisive, not only as regards to form, intensity and direction, but also, and in particular, in terms of socio-economic effects. The returnees of the 2000s were able to take advantage of the windows of opportunities in Albania, whereas the crisis migrants from Greece and Italy between 2010 and 2013 were confronted with problems on the labour market and challenges of reintegration, especially among the returning second generation. The same applies to those returning from the failed bid to seek asylum in Germany in 2015/16, when many Albanians left in order to follow the large number of refugees on their migration to Europe. As far as resilience is concerned, return was often just the starting point for a new migration project. Below, we will draw on three examples to explain the shifting aspects of migration as a tool for social resilience in the case of Albania. The summary of the first biographical example will give an impression not only of the highly volatile, liquid and elusive character of migration in the 1990s and 2000s but also of the economic effects in the sense of the migration-resilience-nexus of return in that period: The interviewee from Vlorë is the co-owner of a medium-sized furniture company. His migration biography is a sequence of back-and-forth movements. After having earned a university diploma in engineering in the late 1980s, he worked in the state-owned oil industry for a short period but became unemployed following the fall of the socialist system. Like many others, he began running an informal kiosk selling small goods. Gradually, he started importing food and, later on, wooden chipboards. In 1997, he opened a bigger shop, which failed due to the riots in Albania. Two years later, in 1999, he began to trade internationally once more, specialising in furniture parts. In between all of these different activities, the interviewee reported several temporary yet limited periods of time spent abroad where he was mainly employed in construction in countries such as Italy, Spain and Dubai. On returning to Albania permanently, he invested the money that he had saved to set up his own company in 2003. The fact that the company still existed almost ten years later, when we repeated our visit, may indicate the success of that migration-based start-up. This previous study, based on entrepreneurs' biographies, clearly showed that private financial transfers from abroad are crucial in the foundation of small and medium-sized enterprises in the service and secondary sectors . Only one out of 34 entrepreneurs in that sample was not in a state of emigration prior to starting a business in Albania. Preparation for the foundation of a company usually comes in the form of a stay of up to ten or more years abroad, mainly in Italy or Greece. The first decade of the 21st century was a rather dynamic period for Albania's national economy, not least due to the mushrooming migration-based private entrepreneurship , one of the characteristic elements of the nexus between migration and development. The determinism of entrepreneurship and migration is a rule in Albania. Migration, as described above, is a tool of social resilience in a dual respect: leaving home and seeking employment of any kind abroad is the reaction to critical living conditions following a change in system, while the decision to stay abroad -often in precarious working conditions with low wages and social deprivation -is made with a view to becoming self-employed upon returning at a later date. It is a truism that a migration project's success or failure depends largely on the socio-spatial and temporal context. In our first example, the migrant was able to decide voluntarily the moment of return. Back home, he was met by a rapidly evolving environment where he could contribute to the economic dynamism, together with many other returnees. In contrast, almost a decade later, a specific proportion of migration and return took place in less favourable circumstances, as our example below shows: The interviewee had just returned to a village near Himarë in southern Albania. As was commonplace in the early 1990s, he had left for Greece alone, but was followed by his wife and child after he had found a job. A daughter was born and the family moved into an apartment in Athens and integrated into society. The son earned a university degree and our interviewee moved into a better professional position. In 2011, with the emergence of the financial crisis in Greece, our interviewee accepted the compensation offered and returned alone to Albania. The son was already working and remained in Greece. The wife continued working in the service sector but had to accept several wage cuts due to the crisis. The striking argument as to why to stay for as long as possible, without spending too much of the savings before returning, was the daughter. She was born in Greece, was still attending school and only spoke a little Albanian. The parents wanted her to graduate from school so as to give her the chance of a brighter future abroad. The returnee himself, now in his mid-50s, runs a small-scale farm on his property in Albania, like his ancestors did and like he himself did before emigrating. "With agriculture you always have something to eat and might have your livelihood, even if the crisis worsens," he remarked, almost fatalistically, at the end of the interview. Both migratory movements were required for economic reasons, but were also necessary for resilience. The first one was an expression of a profound existential crisis. In contrast, the second migration, in the form of a "split return" associated with an emotional burden, is tantamount to returning for retirement and for the lower costs of living at the place of origin. This equates to a coping strategy to avoid the family spending too much of their savings. A third strategy of migration as an example of social resilience in post-socialist Albania is related to the European "refugee crisis" around 2015, when migrants from Western Balkan countries made up around 25 percent of the total number of refugees in Germany , but failed to meet the relevant criteria as set out in the Geneva Convention. 12.2 percent of the arrivals came from Albania. Many of the nearly 54,000 individuals came indirectly from Greece and, to a lesser extent, Italy, with a kind of stop-over in Albania before arriving in Germany. Like our second interviewee, they were affected by the economic crisis in their previous destination countries, but in contrast, they had no resources to make a living in Albania then. A case study from Elbasan may serve to illustrate the problem briefly: The couple is in their late 40s and have three sons. One of the sons lives permanently in Italy while the other two left to seek asylum in Germany but were forced to return in April 2015. Nevertheless, the parents moved to Germany in May, where they spent a couple of months in Bamberg. The main driver at the beginning was to be part of this mass movement. As asylum seekers, they received the request to leave in October, and they left the country on their own but made plans for re-entry simultaneously. "We only came to Germany to work. We do not want social assistance. We came for economic reasons. We are dissatisfied here [in Albania]. We liked it so much [in Germany]. But next time, we want to pay more attention to the rules." In retrospect, the 2015 migration of Albanians "into asylum" was potentially one of the biggest misunderstandings in the age of migration. With only a few exceptions, it was a desperate and often spontaneous but ultimately failed attempt to ascribe value to migration in terms of social resilience, following a common coping strategy established in the decades and centuries beforehand. Most Albanian asylum seekers found themselves back home a couple of months later, facing the same problems as before and thinking about another migration project, but this time with even more limited resources . These cases may confirm the thesis of returnees as potential future emigrants, as return in this particular context represents another stage in sequential migration biographies rather than a genuine prospect. At present, especially among younger people and those who have gained experience abroad , the belief that one can 'do one's thing' in Albania seems to have been lost. Statistical data from Figure 1 show a sharp increase in emigration following a smooth interruption in 2021 due to the pandemic, and may underline the argument that the common pattern of migration is being perpetuated. Between 2021 and 2022, the number of emigrants almost doubled. The decrease in inward migration at the same time indicates that Albania does not offer an attractive environment for remigrants, especially if they have to establish a livelihood and earn a living there. This could be different in a few years' time, when the returnees include retired people who may be able to live well on small pensions from abroad, given that the cost of living in the country is still relatively low. Return migration could then once again become part of migration as a tool for social resilience. --- Findings and discussion In the last 30 years, a broad range of demographic groups participated in emigration from Latvia. Emigration and working abroad have always been popular strategies employed by Latvians, especially during economic crises and periods of instability . The example of Latvia shows that migration as an immediate solution is considered to be a tool for social resilience. The same is true for Albania, but with a slightly different background regarding two contextual factors. The first is the state of isolation during socialism, with the result that, despite a strong affinity towards migration, there was no experience of dealing with international migration during the 1990s. This led to spontaneous, chaotic and sometimes tragic migratory movements. It is no exaggeration to say that this was the re-birth of Albanian emigration as a means of social resilience, now in the postcommunist transition crisis . We see a similarity of motives, symbols and narrations over the centuries, with the exception being a period of almost five decades of socialism during which migration was highly controlled. The second factor is the still unresolved status of EU membership and the subsequent limited freedom of movement and access to labour markets. Both factors made Albanian migrants one of the most vulnerable groups on the European migratory landscape. Through our short empirical sketches, we demonstrated the ability of migration systems to cope with internally and externally caused shocks, such as the economic downturn or the new freedom to travel , to re-organise during these processes in a novel and sometimes creative manner and, if necessary, to define a new social system which offers more stability. From an individual perspective, migration is a robust resource and one of the means with which people cope with the problems they face. Our empirical findings show that this is true, especially under conditions of rapid change, when migration may serve as a part, albeit not the only part, of the solution in the short term. Then, migration is an appropriate answer to uncertainty and threats. However, suppose we return to the general question of the degree to which migration creates development. In that case, the answer must be differentiated, especially if we think beyond macroeconomic factors such as wage gaps or remittances. From an economic point of view, migration definitely supports development in both countries of origin and destination. This positive assessment is put into perspective when we also consider the manifold social costs of migration, i.e. the intangible effects. The consequences of transnational lifestyles are complex and can be associated with high psychological stress and social costs resulting from family separation or lack of integration , even following an eventual return. One of the tailored social resilience solutions, especially in the short term, is one-way or, if necessary, repeated back-and-forth migration with shorter periods of time spent outside the country of origin. Both of the systems analysed are characterised by a widespread multi-local social network in which migration, individual migration experiences and migration culture play a leading role. In our study, migrants are interpreted as agents of societal transformation, adapting social practices from different economic, social and cultural contexts. Factors like social practices and social capital are becoming engrained in Albania and Latvia. All in all, this approach enables us to introduce the concept of social resilience into an agent-based interpretation of the consequences of migratory movements in both origin and host societies. Such resilience as a performed social practice is based on various factors: experience, knowledge and other individual skills, as well as the personal capacities of the network participants. The hypothesis of migration as a tool for social resilience can be assumed to be a rule under conditions of uncertainty at the macro level and individual threats at the micro level, as the cases from Albania and Latvia clearly showed. This research is partly funded by the Latvian Council of Science project DemoMigPro No. VPP-LETONIKA-2021/4-0002 and the University of Latvia. --- Published by

Following the fall of socialism in East and Southeast Europe, widespread destabilisation of living conditions was accompanied by immense skill and cost mismatches. Both of these factors continue to contribute to substantial levels of brain drain, brain waste and de-skilling. We propose and discuss the migrationresilience nexus as a new paradigm that emphasises the instrumental dimension of movements and migrants' agency in terms of the aspiration-capabilities framework. In this paper, we look at migration-specific contexts in two countries suffering from long-term emigration for different reasons. Migratory movements, including emigration and circular and return migration, are interpreted as "tools for social resilience". In many cases, migrants do not necessarily have the aspiration to migrate. Nevertheless, they can do so when conditions in their individual situation, such as material income, individual well-being or family status, change. Thus, in contrast to the few studies that have looked at migration and resilience so far, we focus on aspirations, decisions and movements as fundamental elements of a resilience strategy adopted by individuals to cope with permanent existential risk, constant harassment, socio-psychological stress or other threats. Our analysis pursues a comparative empirical approach. To cover the broad scope of this phenomenon, we chose Latvia and Albania as the study's examples. Data on Albania is gathered using qualitative methods, while a quantitative approach is adopted in Latvia.

Introduction Europe has been a world leader in the development and utilization of IVF from the first successful birth in the UK in 1978. By 2008, the five largest European countries were responsible for 273 000 cycles of IVF, compared with 148 000 cycles performed in the USA that same year . Nonetheless, utilization of IVF varies widely across the European Union , ranging from 2687 cycles per million inhabitants in Belgium to 150 cpm in Moldova in 2008, the most recent year for which data are available . These differences in utilization rates do not appear to be attributable to epidemiological variations among European countries. Indeed, the need for infertility treatment appears to be similar across countries and regions of the world. The prevalence of impaired fertility is difficult to estimate because self-reported estimates are subject to biases, times to conception are difficult to measure and clinical examination and tests are not always conclusive. The current worldwide prevalence of infertility is estimated at 9% of females in the reproductive age group who are in a marital or consensual union and currently not conceiving within one year while not using contraception . Infertility may be a result of male or female factors, a combination of factors in either or both partners or an unknown cause. While IVF is not the only available treatment for infertility, it is often recommended in cases of tubal obstruction and severe male factor infertility, as well as persistent infertility after failure of other treatments, such as fertility drugs, surgery for endometriosis or tubal blockage and IUI. In order to estimate the need for IVF treatment in a population, assumptions must be made about how many infertile people will seek treatment, how many will be candidates for IVF treatment, how many of those are willing to undergo IVF and how many cycles will be necessary for conception. Based on a conservative estimate of the number of IVF cycles needed to meet expected demand, assuming an uptake of 50% with one cycle per couple, health systems would need to provide 1500 IVF cpm each year . Yet only a third of the EU15 countries met or exceeded that level according to the most recent available data: Belgium , Denmark , Finland , Greece and Sweden . As a health-care good among others in the benefits basket, IVF occupies a unique place subject to a range of characterizations-from discretionary good to fundamental human right-that affect how it is regulated. Beyond the clinical and economic considerations that generally affect access and coverage decisions are complex social, historical and political dimensions that are sometimes cloaked in scientific and ethical rationales. For example, IVF in France is regulated pursuant to its bioethics law, which insists that a couple's infertility be pathological in nature and medically diagnosed. The law incorporates social eligibility criteria as it describes the 'man and woman making up the couple', thereby excluding single women and homosexual couples regardless of pathology . In a welfare state, it may be easier for policy-makers to limit access to treatment via laws that are external to the general health system rules rather than to do so within the system of public health coverage. The clinical context also distinguishes IVF from other treatments. Unlike most maladies, infertility or subfertility is most often diagnosed in the context of a couple, not an individual patient. The medical intervention may involve a patient who herself has no identified pathology, as in the case of infertility linked exclusively to male factors. Age may simultaneously be a clinical factor underlying a patient's subfertility and also a criterion that excludes treatment. Such restrictions may impede upon medical discretion in determining the appropriate treatment for an individual patient. When resort to donor gametes or embryos is necessary, the interests and protection of additional parties, including donors and the children who may be born as a result, are at issue, Regulation and utilization of IVF in Europe leading to regulations regarding whether donations may be made anonymously or not, which may have social, legal and potentially medical implications. It is against this complex backdrop that countries have enacted laws regulating access to, and public coverage of, IVF. In the early days of IVF, European governments had to play regulatory 'catch up', as uptake of this innovation outpaced the legislative process. More recently, evolving social forces, budgetary constraints exacerbated by the economic crisis, increased cross-border treatment seeking and supranational regulation have challenged previously settled paradigms. Regulatory responses to IVF in Europe have evolved in differing ways, a phenomenon described by Pennings as 'legal mosaicism' . The aim of this article is 2-fold: first, to describe the regulatory landscape in Europe by identifying the various forms of access regulation and public financing and examining the patterns that emerge; and secondly, to explore whether a link exists between regulatory patterns and access to IVF, as measured by the level of utilization. --- Methods and Materials The EU15 countries were selected for comparative analysis because these countries have well-established health-care systems with extensive experience in providing IVF treatments and monitoring via national, European and international registers. The comparison was further limited to those countries that affirmatively regulate and publicly finance IVF. Thirteen countries were identified as meeting the selection criteria for this analysis: Austria, Belgium, Denmark, France, Finland, Germany, Greece, Italy, the Netherlands, Portugal, Spain, Sweden and the UK. Ireland and Luxembourg were excluded because neither country has IVF-specific regulations. In addition, IVF is not covered by Ireland's national health system and Luxembourg does not report IVF data via the European IVF-Monitoring Consortium. 'Regulation' as used in this article encompasses both legislative and executive enactments, including authority to impose restrictions devolved to the regional, local or treatment levels, which were in force in 2009. The descriptive analysis of the European regulatory landscape focused on restrictions on access to IVF treatment, in terms of general eligibility as well as public financing and the scope of available services. The selection of specific regulations for comparison was based on those that were common across countries and that potentially could affect utilization of IVF services. The latter was informed in part by recent research regarding the reasons patients seek crossborder reproductive care, which found that it is most often driven by legal restrictions in the home country . The eligibility criteria we reviewed included age limits, social criteria and medical diagnosis requirements. Because fertility declines with age, soft or strict age limits may be imposed in an attempt to distinguish women with pathological infertility from those whose have reached the natural limits of fecundity. Restrictions regarding the civil status and/or the sexual orientation of treatment seekers may be seen as a means of enforcing traditional family norms by limiting access to heterosexual couples, although broader social trends in Europe suggest a greater openness to alternative family arrangements. Some countries require a medical indication as a pre-condition for IVF treatment, based either on infertility/subfertility or to prevent the transmission of a serious disease. Such requirements may also operate as a proxy to exclude the so-called 'socially infertile' individuals who lack a formal medical justification for treatment. Our review also examined the regulation of public coverage based upon eligibility criteria and level of coverage as a measure of financial access to IVF. In terms of eligibility, both medical diagnosis requirements and age limits may be more narrowly defined for publiclyfinanced IVF, with the ostensible justification of limiting expenditure of public health-care resources to those with demonstrated medical need and statistically better chances for success. In addition, we looked at the percentage of treatment costs officially covered by public health insurance or other public financing for eligible individuals, as well as the maximum number of treatment cycles covered, which can affect drop-out rates and the likelihood of success. Finally, we surveyed the range of reproductive technologies and services, from the availability of PGD to the legality of surrogacy arrangements. We decided to limit this part of the analysis to the regulation of donor gametes and donor embryos because donor restrictions can limit access and constitute a significant motivating factor for those seeking cross-border IVF . Beyond the issue of the availability of donor gametes and embryos is the question of donor anonymity, which can affect donor recruitment as well as recipients' willingness to use them. In order to explore the patterns of regulation across countries and to determine whether there was an impact on utilization of IVF, we undertook empirical legal research, which employs quantitative methods to investigate the effects of laws, regulations and policies on a particular outcome. We used multiple correspondence analysis , a statistical tool that allows a joint observation of a number of categorical variables-in this case, the different forms of IVF regulation in a number of countries-in order to identify systematic associations among them. MCA is a geometric approach that projects the data of an n-dimensional space onto a succession of two-dimensional planes. The row and column points are projected in the space created by two axes, which represent part of the total inertia . The two-dimensional display is appropriate for purposes of interpretation if the cumulative percentage of variance explained by the first two axes of inertia is sufficiently large . The multidimensional graphical map allows a global view of the trends within the data. The strength of association between the modalities of variables is defined by their spatial closeness and their positions within the quadrants: the closer the modalities are to each other, the stronger the association among countries sharing these features, and the greater the distance of a modality from the mean centroid , the greater its significance in the interpretation of results. In addition, it is possible to project supplementary variables onto the map to provide, in effect, a 'visual regression'. Utilization, the outcome investigated here as a supplementary variable, is one of the most common proxies for measuring access to health-care services. In order to perform the MCA, the regulations were first categorized. Given the small number of countries in our analysis , it was necessary to limit the number of categorical variables to as few as possible. Moreover, variables that were highly correlated or that represented services with low utilization were eliminated. The statistical analysis was carried out in two steps. First, each of the variables for the eligibility regulations and one of the treatment variables were categorized as follows: age limit ; only medical indications ; only couples eligible ; only heterosexuals eligible ; donor anonymity . Each of the selected categories, as well as the overall regulatory profiles for each country, were combined into an indicator matrix and then mapped. Four levels of utilization , ranging from less than half to in excess of the anticipated demand of 1500 cpm, were projected onto the map as supplementary variables. Secondly, the IVF public financing variables were categorized, using the same categories for age limit and medical/social indications as for the eligibility analysis. Binary modalities were defined for the level of coverage and number of cycles . These categories, along with the overall country profiles, were then mapped, again supplemented with the four levels of utilization. The MCA was carried out using SAS software version 9.1 for Windows . Data regarding access and coverage regulations came from a study by the European Society of Human Reproduction and Embryology of regulatory frameworks for medically assisted reproduction as of 2009 among the EU27 and were supplemented with additional legal research. IVF utilization data are from the European IVF-monitoring reports for 2007 and 2008 and from the International Committee for Monitoring Assisted Reproductive Technology world report for 2003 . --- Results The 13 EU countries studied reflect the diversity of regulatory approaches to eligibility and public financing of IVF, as well as to donor issues. Moreover, certain criteria underscore the fact that IVF is often subject to greater restrictions than other health-care goods. --- Eligibility for IVF Patterns emerged with respect to certain eligibility criteria, in particular the personal characteristics of those entitled to treatment, reflecting a clear division between countries with more restrictive versus more liberal regulations . In seven of the 13 countries, a medical indication must form the basis for a demand for IVF, regardless of whether the treatment is publicly covered. Austria, Germany and Italy are particularly restrictive, limiting treatment to diagnosed infertility, while France, the Netherlands, Portugal and Sweden also permit the use of IVF to avoid the transmission of serious diseases, such as human immunodeficiency virus or genetic disorders. Of countries that insist on a medical indication, all but two countries also require treatment seekers to be heterosexual couples. This latter restriction based on sexual orientation and relationship status is unique to fertility treatments among healthcare goods. Six countries have taken a more liberal approach to access to IVF. These countries do not require a medical diagnosis of infertility for access to IVF treatment, nor do they exclude single women and lesbians. The Netherlands and Sweden fall between the two extremes, with both requiring a medical indication while allowing access to treatment for lesbians who are otherwise unable to conceive. Sweden excludes single people from access to treatment, while providing treatment to lesbian couples. Regulatory patterns were less clear with respect to upper age limits. Of the six countries with no set age limit, three were among the countries with more liberal access , while the other three had more restrictive policies . The remaining seven countries imposed either strict or soft limits for women to access IVF independent of coverage. Sweden's soft limit also applies to men, requiring that they be sufficiently young to parent an infant through childhood. In countries with soft or no age limits, discretion to determine access based on age generally rests with the clinic or doctor, who may rely upon other clinical indicia, such as the patient's ovarian reserve. --- Public financing of IVF Regulations governing access to publicly financed IVF coverage are generally more restrictive than the general eligibility criteria as a means of limiting the covered population-and thus the budgetary outlays-based upon health system priorities . Indeed, all but two countries condition coverage upon the existence of a medical indication. In contrast to the general eligibility regulations, four of the six countries with no age limit for access to IVF established age limits only for publicly financed treatment, while two with age limits reduced it for covered treatment. Overall, more than half of the countries set strict age limits for coverage, including two that impose age limits for both women and men . In the other countries with strict upper age limits for women, the range is from ,40 years as in the UK to ,50 years in Greece, with an average of ,42. In countries with no fixed age limit, age criteria may vary by region, the size of the waiting list or be set by the center. In terms of financial access to IVF, policies regarding the level of coverage and the number of cycles covered vary significantly. Four countries offer full coverage to eligible couples or individuals: Belgium, France, Netherlands and Sweden. 'Full coverage' in this context is defined as 100% coverage of at least one cycle of IVF on a national basis. In Sweden, full coverage is only available in public clinics. Countries offering 'partial coverage' include those that impose additional fees and/or cost-sharing beyond what is generally applicable to other maladies, as well as those countries with regional differences in access to or level of coverage of IVF. For example, in the UK access to IVF is determined by primary care trusts, which has resulted in a 'postcode lottery' for services, with differing conditions for access to publicly financed treatment and no IVF offered in some areas. Variations in coverage can be seen in other countries as well, depending upon waiting lists , the particular social security organization providing coverage or region . In countries providing partial public coverage, cost sharing for drugs is seen in Denmark, Finland and Portugal. Austria has a special fund outside of the health-care system that covers 70% of IVF costs, while Germany requires 50% cost sharing for IVF. Another aspect of IVF coverage is the number of treatment cycles covered, which can in turn affect the cumulative birth rate following treatment as well as the drop-out rate for financial reasons. European countries have taken differing approaches to this issue. Some do not specify a maximum number of cycles, leaving the coverage decision to regional or local authorities , social security funds or individual doctors . However, most countries have established coverage limits of three or four cycles. In France, for example, a treatment cycle constitutes a fresh stimulated cycle, and any subsequent transfers using frozen embryos do not count against the four covered cycles. Moreover, if treatment results in a live birth, the 'counter' is reset to 0, so that another four cycles may be covered. Belgium has the most generous plan, covering up to six cycles provided the treatments are in compliance with its singleembryo transfer policy to reduce multiple births. --- Regulation of donor IVF All of the countries studied except for Austria and Italy allow donor gametes to be used in IVF, although Germany limits donor IVF to semen only . Austria's restriction on the use of donor gametes in IVF was challenged under the Convention for the Protection of Human Rights and Fundamental Freedoms on the grounds that it was discriminatory for Austria to allow donor semen for IUI while completely prohibiting the use of donor semen or oocytes in IVF. The European Court of Human Rights found that there was no violation of the Human Rights Convention based in part on its contention Related to the issue of donor gametes is the question of the donation of frozen embryos remaining after IVF treatments have ended. Couples may opt to use donor embryos, owing to infertility factors affecting both parties, as a last resort because of unexplained failure of IVF using their own gametes, or to avoid long waits for donor oocytes. Of the 11 countries that permit the use of donor gametes, three forbid embryo donation . The reasoning in Denmark and Sweden is similar, based on the idea that the child should have a genetic link with at least one of his parents. Although France prohibits 'double donation' of gametes, it nonetheless permits embryo donation subject to judicial oversight. Germany's prohibition of donor oocytes effectively precludes the use of donor embryos. An additional factor that may affect the decision-making of both donors and recipients is whether the donation is anonymous or not. The 11 countries that allow gamete donation are split on this matter, with 6 insisting on anonymous donations , and the other 5 requiring or permitting non-anonymous donations . Interestingly, this dichotomy is not consistent with the countries' overall liberal or restrictive regulatory policies regarding IVF. --- Regulatory patterns and impact on utilization The first mapping of the MCA combined the IVF eligibility criteria with the donor anonymity regulations. The regulation of access to donor gametes was not included in this analysis because this information was captured by the regulation of donor anonymity. The donor embryo variable was excluded because it represents a very small portion of overall IVF utilization ). The first two axes of inertia combined explained nearly two-thirds of the weighted variance, providing a sound basis for analysis, particularly the first axis, which accounted for 49% of the variance . The map of the first axis revealed that the restrictive and liberal modalities of the eligibility criteria were in direct opposition as we saw in the descriptive analysis, and two distinct groups of patterns of regulations and countries emerged. The countries with more restrictive policies limiting IVF eligiblity to couples with medical indications were grouped on the right side of the map. Within this grouping degrees of restriction are apparent, ranging from Sweden with its policy of extending eligibility to lesbian couples on the left side of the group to Italy at the far right side of the map because of its outright prohibition of donation. The countries with more liberal policies all extended IVF eligibility to single women and lesbians and were grouped on the left side of the map. The Netherlands was a bit closer to the center of the map as a result of requiring a medical indication for treatment. We can note the association of strict age limits with otherwise liberal regulatory regimes and soft age limits with the more restrictive policies. The second axis of inertia explained a smaller proportion of the variance and was essentially tied to the donor anonymity and age limit modalities. Age limits and the availability of anonymous donations were grouped in opposition to no age limits and prohibition of anonymous donations. The countries, however, were not clearly grouped but rather dispersed from Finland, Germany and the UK associated with no age limit and non-anonymous donation to France, Belgium, Denmark and Greece at the other extreme. Examining the relationship between eligibility regulations and utilization, we found a clear negative relation between the degree of restriction of IVF eligibility and the level of IVF utilization, with utilization diminishing as the criteria became more restrictive and vice versa . However, no link was evident between age limits and donor anonymity and utilization. Regarding access to coverage and level of public financing of IVF, the second MCA mapping also explained two-thirds of the variance , but the picture was less clear in defining groups of countries with similar patterns of regulation. On axis 1 , the countries were spread across a continuum from more restrictive access to coverage that appear to be associated with more generous policies regarding the extent of coverage to more flexible regulations in terms of age limit and medical criteria for access, which were closer to less generous policies offering only partial coverage and covering a smaller number of cycles. On axis 2 , the regulations were much closer to each other, except for the soft age limit and the countries applying this criterion . Overall, no relationship was found between the generosity of coverage and the degree of utilization. Morever, no link could be found between the eligibility and coverage mappings, underscoring the fact that there is no clear relationship between the policy approach of a country with regard to the restrictivenes of its IVF eligibility criteria and its generosity in terms of public financing. Figure 2 MCA of regulation of access to IVF: variance explained by axis 1. The strength of association between the modalities of variables is defined by their spatial closeness and their positions within the quadrants: the closer the modalities are to each other, the stronger the association among countries sharing these features, and the greater the distance of a modality from the intersection of the axes, the greater its significance in the interpretation of results. Note: the size of the circles reflects the relative weight of the regulatory modalities in defining axis 1. MCA of regulation of access to IVF: variance explained by axis 2. The strength of association between the modalities of variables is defined by their spatial closeness and their positions within the quadrants: the closer the modalities are to each other, the stronger the association among countries sharing these features, and the greater the distance of a modality from the intersection of the axes, the greater its significance in the interpretation of results. Note: the size of the circles reflects the relative weight of the regulatory modalities in defining axis 2. Figure 3 MCA of regulation of coverage of IVF: variance explained by axis 1. The strength of association between the modalities of variables is defined by their spatial closeness and their positions within the quadrants: the closer the modalities are to each other, the stronger the association among countries sharing these features, and the greater the distance of a modality from the intersection of the axes, the greater its significance in the interpretation of results. Note: the size of the circles reflects the relative weight of the regulatory modalities in defining axis 1. MCA of regulation of coverage of IVF: variance explained by axis 2. The strength of association between the modalities of variables is defined by their spatial closeness and their positions within the quadrants: the closer the modalities are to each other, the stronger the association among countries sharing these features, and the greater the distance of a modality from the intersection of the axes, the greater its significance in the interpretation of results. Note: the size of the circles reflects the relative weight of the regulatory modalities in defining axis 2. Regulation and utilization of IVF in Europe --- Discussion While there is a clear positive relationship between the rate of utilization and the breadth of general eligibility for IVF, no link was found between restrictions to publicly financed IVF and utilization. Nonetheless, our analysis underscored that the more generous the coverage in terms of number of cycles and share of cost publicly financed, the greater the restrictions with respect to who may be covered. Budgetary constraints likely play a role in this phenomenon. Two groups of countries with similar patterns of regulation of access to IVF were distinguished: one with more liberal regulations extending eligibility to single woman and lesbians and the other with more restrictive policies limiting access to couples with medical indications; there was no clear grouping of countries with regard to the regulations of public coverage and no relationship between the general access patterns and the coverage pattern. Other studies have explored the impact of various aspects of IVF regulation, particularly in terms of health economics or significant regulatory changes within a single country . However, to our knowledge, this is the first to examine the impact of a broad spectrum of regulations, including those establishing social eligibility criteria, on IVF utilization across Europe. Our findings are consistent with a recent study examining the reasons patients seek cross-border reproductive care, which found that most do so to avoid legal restrictions related to patient characteristics, such as age, sexual orientation and civil status, and to obtain specific services such as anonymous gamete donation . Our study has a few limitations. We faced the common challenge encountered in international comparisons: matching data across countries and across time. This is all the more difficult in this rapidly changing regulatory landscape. We chose to include the regulations in effect in 2009, while the latest utilization data is primarily from 2008. The temporal mismatch is exacerbated in the cases of Greece and Spain, for which the most recent utilization data are from 2003 owing to the low number of clinics reporting in later surveys. In terms of method, we selected MCA, a descriptive approach, to compare the countries' regulatory policies to their utilization rates. Normally an explanatory method, such as linear regression, would be used to measure the impact of these variables on the utilization rate. However, given the nature of the data, in particular the low number of countries, such a method could not be used. One option we considered was qualitative comparative analysis, which attempts to maximize the number of comparisons that can be made across cases in order to examine how configurations of conditions are linked to a certain outcome. We decided not to use this method because it is generally limited to dichotomous variables . Moreover, we believe that MCA is the best method for enriching the understanding of tabular data because of its graphical representation. While there is no fixed minimum number of variables or modalities in MCA, the limited number of countries in our analyis could lead to misinterpretation. In order to reduce this possibility, we limited the number of categorical variables to the extent possible and systematically verified that there were no fewer than two modalities per variable. In defining categorical modalities, certain nuances may have been lost or not captured. For example, the discretization of continuous variables, such as utilization and the number of covered cycles, entails some loss of information. The same is true with respect to the use of broad categories, such as full versus partial coverage. We tested various configurations of the categorical variables to ensure that the diversity of regulatory patterns across the countries studied remained apparent in the analyses while avoiding outliers that could dominate the interpretation of the axes. Nonetheless, the balance between parsimony and interpretability requires subjective choices. Similarly, the definition of a variable can lead to counterintuitive results. Indeed, while the association of strict age limits with more liberal regulatory regimes may seem anomalous, in fact, the strict age limits for general eligibility in the countries we examined were relatively high , corresponding to the natural limits of conception . Moreover, such active regulation may actually support higher utilization by sending clear signals to patients and healthcare providers regarding the contours of permissible treatment. We focused on utilization defined as cpm population as the outcome measure to assess the impact of regulations. We also examined the cpm women of reproductive age to determine whether age differences across the countries studied would affect the analysis. They did not. Nonetheless, utilization data alone do not permit us to draw conclusions regarding access to IVF. Indeed, it is debatable whether utilization is a good proxy for actual access to treatment, given that those with equal need and equal access to treatment may not make equal use of that opportunity because of individual preferences . Certainly with respect to IVF, factors beyond access, including religion and cultural norms, may come into play. Finally, the utilization data do not distinguish cycles for cross-border treatment as opposed to in-country care. While our study found that regulations related to public financing, including eligibility for coverage and level of financing, were not linked with utilization levels at the population level, it is important to underscore that we did not explore the effect of no public coverage on utilization, as all of the countries studied provided at least partial public coverage of IVF. Caution must be applied in drawing conclusions regarding the effect of public financing, particularly with respect to equity of access. Affordability has significant implications for equity, as only those individuals with the resources either to pay out of pocket and/or travel to a foreign jurisdiction will have access to treatment. This concern is illustrated by the significant drop-off in utilization that occurred in economically weak regions of Germany after IVF coverage was reduced to 50% in 2004 . Better data are needed on how much, and at what cost, treatment is privately financed versus publicly covered because the extent of IVF in the private sector implicates inequalities in access based on the ability to pay . In the UK, for example, nearly 60% of IVF treatments are privately financed . Cost, rather than restrictions based on social criteria, have been found to be a major motivation for British crossborder treatment seekers . Whether driven by cost or legal restrictions, regulatory policies lie at the heart of cross-border reproductive care. The actual level of cross-border IVF in Europe is unknown, although estimates suggest a minimum of 18 000-22 500 cross-border IVF cycles annually . In March 2011, the European Parliament adopted a directive regarding patients' rights in cross-border health care, delineating the circumstances under which a health system must finance treatment in another Member State . While coverage of treatments that are not included among the home country benefits is not required, the manner in which treatment is provided may differ and may have public health implications for the home country. ICMART, which collects worldwide data on IVF, has called for prospective monitoring of cross-border reproductive treatment, including the reasons for seeking cross-border care . At a minimum, collecting data on patients' countries of origin would provide a much clearer picture of this phenomenon and would give policy-makers an important data point regarding the extent to which their own citizens are seeking care elsewhere and therefore the degree to which there may be inequity of access to IVF in a given country. ---

How do the different forms of regulation and public financing of IVF affect utilization in otherwise similar European welfare state systems? summary answer: Countries with more liberal social eligibility regulations had higher levels of IVF utilization, which diminished as the countries' policies became more restrictive. what is known already: Europe is a world leader in the development and utilization of IVF, yet surveillance reveals significant differences in uptake among countries which have adopted different approaches to the regulation and and public financing of IVF. study design, size, duration: A descriptive and comparative analysis of legal restrictions on access to IVF in 13 of the EU15 countries that affirmatively regulate and publicly finance IVF. participants/materials, setting, methods: Using 2009 data from the European Society of Human Reproduction and Embryology study of regulatory frameworks in Europe and additional legislative research, we examined and described restrictions on access to IVF in terms of general eligibility, public financing and the scope of available services. Multiple correspondence analysis was used to identify patterns of regulation and groups of countries with similar regulatory patterns and to explore the effects on utilization of IVF, using data from the most recent European and international IVF monitoring reports. main results and the role of chance: Regulations based on social characteristics of treatment seekers who are not applicable to other medical treatments, including relationship status and sexual orientation, appear to have the greatest impact on utilization. Countries with the most generous public financing schemes tend to restrict access to covered IVF to a greater degree. However, no link could be established between IVF utilization and the manner in which coverage was regulated or the level of public financing. limitations, reasons for caution: Owing to the lack of data regarding the actual level of public versus private financing of IVF it is impossible to draw conclusions regarding equity of access. Moreover, the regulatory and utilization data were not completely temporally matched in what can be a quickly changing regulatory landscape. wider implications of the findings: Whether motivated by cost, eligility restrictions or the availability of particular services, cross-border treatment seeking is driven by regulatory policies, underscoring the extra-territorial implications of in-country political decisions regarding access to IVF. study funding/competing interest(s): There was no funding source for this study. The authors have no conflicts of interest to declare.

two strangers as between a husband and a wife . There is also no consensus about the degree to which such models can accurately capture preventative behaviour . Criticism of these approaches led to the development of alternatives that are network-based and assume interaction between fewer actors. Interestingly, these network-based models show more diverse patterns of epidemic spread. For instance, in more clustered networks, there is a faster drop-off in the initial transmission rate but a longer persistence of susceptible individuals in the population . Nevertheless, network models are limited by the degree to which they accurately reflect key features of real social networks . We argue that both the modelling of infectious disease and the public health response seeking to alter risk behaviour would be improved through greater consideration of the social psychology of the people in the networks that we seek to model, and whose behaviours leaders seek to change. In particular, we contend that peoples' engagement in behaviours that contribute to the spread of infectious disease is fundamentally shaped by their group memberships, most particularly by those groups which hold psychological meaning to themtheir social identities . We build on this fundamental proposition by outliningand providing evidence that speaks toa social identity model of health risk taking, in which we specify some of the key processes through which social identities shape health risk behaviours . At a time when the implications of such behaviours have rarely been more critical, we argue that being cognizant of these processes is paramount. --- A social identity model of health risk taking According to the social identity approach , people categorize themselves in terms of both their personal identity and their various social identities . In line with this, several thousand empirical studies have demonstrated that subjectively defining oneself in terms of a particular group membership affects thoughts, attitudes, perception, and behaviour . Through their capacity to transform psychology and behaviour, group processes fundamentally structure our perception of safety versus vulnerability. Central to our model is the proposition that potential threats arising from ingroup membersparticularly those with whom we share a strong social identitywill be perceived as less risky, and inspire greater risk taking behaviour, than potential threats arising from outgroup members. In our model, we also articulate two key processes through which we hypothesize that shared group membership affects risk: via increased trust, and via lowered disgust. In the sections that follow, we summarize key evidence for our model. Importantly, although this model focuses on the ways in which group processes can undermine the containment of infectious diseases such as COVID-19, we note that group processes also have a key role to play in an effective response. For instance, greater group identification is associated with increased conformity to group norms . For instance, depending on one's referent group, normative influence may prompt a person to get vaccinated , but could also prompt them to attend a rally protesting social distancing . Although norms are non-focal in the model we present here, we briefly return to the importance of normative influence in the Implications section below. --- The importance of group memberships in disease risk There is a long history of research examining the predictors of risk taking, with good evidence that various biological factors , individual factors , and social factors are associated with a greater likelihood of engaging in risky behaviour. There is also good evidence that people choose to engage in behaviours that pose a risk to their health because of concerns related to the way they will be perceived by others . It is likely that these factors are also relevant predictors of peoples' engagement in behaviours that confer the risk of disease contagion to themselves and others. Although important, these factors are not the focus of the present article. Rather, our interest centres on the role of shared group membership as a determinant of risk perception and behaviour. That is, a person may well be more likely to take the risk of shaking hands with another person during a pandemic if they are more impulsive, or if this is behaviour they, or people around them, normally engage in. However, we argue that they are also more likely to engage in this behaviour if they perceive the other person to be a fellow ingroup member with whom they share a valued social identity . As a corollary to this, we also propose that people more readily recognize the risk posed by strangers or outgroup members, and take appropriate precautions . This critical role of shared identity has been under-recognized in public health messaging that seeks to change individual behaviour. It is also a particularly insidious contributor to health risk behaviour because people may not even recognize the risk to begin with. That is, the danger posed by shared group membership is that people do not Figure 1. A social identity model of health risk taking. This conceptual framework summarizes key pathways through which social identities shape risk perception and engagement in risk behaviours that contribute to the spread of infectious disease. Adapted from Cruwys, Greenaway, et al. . even perceive a risk in the first place because ingroup members are considered 'safe'paving the way for risky behaviour to occur when one's guard is down. For example, speaking to the role of shared group membership as a predictor of disease risk perception, Hult Khazaie and Khan found that mass gatherings attendees felt less vulnerable to disease when they felt a strong sense of shared identity with fellow attendees. Loersch and Bartholow found that participants perceived beer consumption to be less dangerous when they were presented with beer cans that featured the colours of their university, rather than standard beer cans. Regarding risk behaviour, in a sample of 128 military and police officers, Firing and Ladberg found that shared identity with fellow officers significantly predicted the likelihood of participants taking the risk of jumping into a freezing ocean. Indeed, shared identity was a stronger predictor of risk taking than the personal characteristics of the participants. Most recently, across eight studies, Cruwys, Greenaway et al. found consistent evidence for a direct causal link between shared group membership and risk perception and behaviour, including in the context of disease. For instance, in an experiment utilizing the minimal group paradigm , a confederate ostensibly had a cold, and left used tissues in a shared workspace. Participants rated the health risk as greater when the confederate was an outgroup member compared to when they were an ingroup member. Another study explored health risk behaviour among 149 attendees at a controversial art festival. People who strongly identified with fellow attendees stayed for longer at a performance that posed a potential health risk . In a further experiment, participants were asked to imagine shaking hands with a colleague and later noticing the colleague had a rash on their hand. Participants perceived this to be less risky when their colleague was a member of their political party, rather than an opposing political party. There was evidence that these effects were not attributable to other plausible mechanisms, such as a generalized bias towards ingroup members . Ingroup favouritism did not mediate the effect of shared group membership on risk perception, and this relationship remained significant when ingroup favouritism was included as a covariate. --- The mediating role of trust A key mechanism through which we propose that shared group membership affects risk perception and behaviour is trust. A robust link has been established between shared group membership and trust, with much of this research using economic investment game paradigms. This research has found that participants are more likely to trust ingroup members to distribute monetary outcomes fairly , even when group membership is based on arbitrary criteria . Trust has primarily been studied as an adaptive cognition that facilitates effective social functioning for instance, by enhancing cooperation , and allowing for more effective communication . However, trust is also closely linked to risk takinga relationship that has also been indicated in controlled economic investment game contexts, where evidence suggests participants are more willing to risk gambling money during interactions with more trustworthy game partners . The link between trust and health-related risk taking is supported by evidence from diverse contexts. For instance, ethnographic studies on needle sharing indicate that this behaviour primarily occurs within close social networks and represents an expression of trust . Similarly, research in interpersonal contexts has found trust in one's sexual partner to be negatively associated with condom use . Applied to infectious disease, the capacity for trust to attenuate risk perception and risk avoidant behaviour has implications for contagion spread, because it may lead people to engage in behaviours they should avoid . In line with this, recent research has demonstrated the full mediational pathway of shared group membership on health risk perception and behaviour via trust . For example, in a study involving almost 1400 attendees of a school leaving festival, those who identified more strongly with their fellow attendees also trusted them more and, in turn, perceived activities such as accepting a drink from them as less risky. A more fine-grained analysis suggests that a specific form of trust drives this effect. In two further studies in the context of health and financial risk taking, integritybased trust consistently mediated the effect of shared group membership on risk perception and behaviour . This suggests that people tend to believe ingroup members pose less risk than outgroup members primarily because they assume ingroup members will behave in principled ways. In the context of COVID-19, this means that people may be more likely to assume that ingroup members are strictly adhering to good hygiene practices and are thus less likely to be carrying the disease. --- The mediating role of disgust The second pathway through which we propose that shared group membership can affect health risk is via disgust. Disgust can have substantial benefits in the context of Figure 2. People attended art events that posed a health risk for longer when they identified with fellow attendees . This is consistent with experimental evidence that suggests that we perceive health risks to be lower when they arise from ingroup, rather than outgroup, members. disease contagion. This is because it represents part of the body's behavioural immune system -'a collection of psychological mechanisms that enables individuals to detect pathogens in their environment and motivate behaviours that prevent these pathogens from entering the body' . A specific role that disgust plays in mitigating risk is to help distance people from others who could contaminate them . Research suggests that peoples' disgust response can even increase during periods when they are particularly vulnerable to infectious disease . Disgust also appears to be attenuated by shared group membership. In an early demonstration of the impact of social relationships on disgust, Case et al. found evidence across two studies that people are less disgusted by the smell of, and by changing, their own children's nappies compared to those of other children. Relatedly, experimental and questionnaire-based research indicates that people feel more comfortable being close to, and in a shared environment with, ingroup members . These findings align with the evolutionary perspective that we are 'hard-wired' to avoid, or at least be more cautious of, outgroup members because they are more likely to carry pathogens that we are not immune to . However, recent research suggests that these findings may be due to an ingroup attenuation effect on disgust, at least as much as an outgroup accentuation effect. Research by Reicher et al. speaks directly to the role of ingroup relations in attenuating disgust. In their first study, these researchers found that university students rated a sweaty t-shirt bearing the emblem of a different university as less disgusting after their identity as a university student had been made salient compared to when their identity as a member of their own university had been made salient , or when their personal identity had been made salient. In their second study, participants' disgust responseoperationalized as the time they took to walk to a bottle of hand sanitizer after touching a sweaty t-shirtwas lower when the t-shirt bore the emblem of their university, compared to both when it bore the emblem of a rival university and when the t-shirt was plain. Importantly, in both studies there was little difference in disgust ratings between the outgroup and personal identity conditions, suggesting that perceiving that someone is an outgroup member may not increase disgust towards them, so much as perceiving someone as an ingroup member decreases disgust towards them. In the COVID-19 context, this might mean that people experience an attenuated disgust reaction in situations that pose a risk for disease transmission specifically when the other people involved are perceived to be ingroup members. --- Implications for the COVID-19 response Our response to this emergency cannot afford to ignore social psychological factors that contribute to risk perception and risk behaviour. From the evidence reviewed above indicating that ingroup members pose a particular risk of transmitting infectious disease, we thus abstract three specific recommendations for the COVID-19 response . First, we propose that modelling of disease spread should weight interactions such that ingroup members both interact more often and have a higher risk of transmission with each interaction. Although it may not be practical to model the psychological perception of shared group membership in such models, network models typically include close contacts, which are an acceptable proxy for these purposes. Second, we propose that the elevated risk posed by ingroup members should be recognized in public health messaging, such that this explicitly highlights the risks associated with being in physical contact with others that we care the most about, and the need to be equally vigilant in the way we interact with these people as we are with strangers. Along these lines, we recognize that most people will find it easier to grasp the risk posed by a handshake with a stranger than by a hug with a relative. Thus, third and finally, we propose that public health messages should explicitly challenge this, and frame a lack of physical contact as an expression of care . More broadly, we urge political and health care leaders to work with, rather than against, group processes in their messaging. Indeed, we note that several group processes that are beyond the scope of this article have a critical role to play in optimizing the COVID-19 response. While people may find it difficult to perceive the risk of COVID-19 arising from those closest to them, they are also most motivated to protect their loved ones. Indeed, research indicates that if people perceive that failing to modify their behaviour will result in harm for others, or that modifying their behaviours may improve the health of others, then this positively influences their decisions to engage in protective behaviours such as vaccinations . Initial investigations in the context of COVID-19 support the idea that messages focusing on collective responsibility to loved ones increases compliance with public health directives . --- Conclusion Through their capacity to shape our perceptions of who we are, social identities structure our cognitive and emotional reactions towards others. Because of this, our evaluation of risk versus safety is inextricably tied to our group memberships. These processes are not, of course, limited to COVID-19. Indeed, prior research has suggested that our perceptions of, and engagement in, many kinds of risk are also structured by shared group membership . However, in the absence of biomedical solutions to vaccines), it is vital that we harness all the tools at our disposal in the fight against its deadly spread. We argue that applying our understanding of the social determinants of health risk perception and health risk behaviour is one such tool. --- Data availability statement Data sharing is not applicable to this article as no new data were created or analysed in this study. --- Conflicts of interest All authors declare no conflict of interest. --- Author contribution

In the face of a novel infectious disease, changing our collective behaviour is critical to saving lives. One determinant of risk perception and risk behaviour that is often overlooked is the degree to which we share psychological group membership with others. We outline, and summarize supporting evidence for, a theoretical model that articulates the role of shared group membership in attenuating health risk perception and increasing health risk behaviour. We emphasize the importance of attending to these processes in the context of the ongoing response to COVID-19 and conclude with three recommendations for how group processes can be harnessed to improve this response. As COVID-19 infection rates soar around the world, public health advice from epidemiologists, medical professionals, and political leaders is that the most powerful weapon we possess to combat its spread is to change our behaviours. Indeed, the measures that governments have imposedfrom closing bars to closing borders, from stopping mass gatherings to shutting down entire countriesall represent attempts to ensure widespread behaviour change. These broad evidence-based measures have been effective in helping to slow, and ultimately halt, the spread of infectious diseases such as SARS in the past (Wilder-Smith & Freedman, 2020), and have helped attenuate the spread of COVID-19 (Anderson, Heesterbeek, Klinkenberg, & Hollingsworth, 2020). However, individuals engage in dozens of micro-level behaviours every day that lead to infectious diseases being passed from one person to another. These behavioursgathering in close proximity, shaking hands, sharing drinks, choosing to visit one's ageing parentsare harder to prevent through broad policies, yet critically determine whether containment efforts succeed. Extant approaches to modelling the transmission of infectious diseases typically draw on economic models (e.g., game theory) and seek to predict the rate and distribution of disease spread through a population (for a review of such models that incorporate some psychological variables, see Weston, Hauck, & Amlot, 2018). Such models necessarily simplify human interaction. For instance, they assume that contact between individuals is random (called homogeneous mixing), such that transmission is just as likely between

Introduction Globalisation and advances in new communication technologies have shortened distances and changed the reality in which children and adolescents develop today. Information has become available more readily and social interaction has taken on radically different forms, changing both behaviours and expectations around communication, learning and exchange. However, it can be argued that as well as these micro-level changes, access to information and new technologies has also posed a new macro-level paradigm of inclusion/exclusion. This paradigm has been very influential in the Argentinean political agenda over the last decade, especially in terms of public policy demands and debates about the educational system. The three main outcomes were as follows. First, a national education law came into force in 2006, significantly changing the Argentine educational system. This law prescribed that, from 2009 in Buenos Aires City, all children attending public primary schools must learn a foreign language [2] from first grade. Second, from 2010, the same district implemented the Integral Plan of Digital Education [3], seeking digital inclusion of students and teachers of public primary schools through the delivery of laptops and digital literacy. Third, during the same year, the nationwide Connecting Equality Program [4] began operating; this programme had similar goals to the PIED, but with a nationwide reach, affecting adolescents attending secondary schools, special schools and public vocational schools. In this context, it can be inferred that from 2010, children and adolescents in public schools in Buenos Aires City have had access to foreign languages, laptops and digital literacy in their school curriculum. This article aims to describe the different ways that these children have come to connote value and represent the new educational tools they received in the framework of these policy changes. It asks three main questions: how do children and adolescents represent computer use and domination of a foreign language? How do they imagine that both tools modify people's course of life? What valuations are associated with the use of them? Also, given that the aim of such policies is mainly to reduce social gaps, the article explores the notion that social inequalities may be evident in the representations that children build around these tools. In order to answer these questions, it draws on data gathered from in-depth interviews conducted in 2012 with a group of students from primary and secondary public schools in Buenos Aires City, in which the researchers used a 'creativity method' as a means to determine students' perceptions about the new educational tools they had acquired. Specifically, subjects were asked to represent their answer to the following questions through a drawing: 'How do you imagine the life of a person is who cannot speak English or use a computer, and, in opposition, how do you imagine the life of a person is who can speak English and knows how to use a computer?' By being asked to imagine contrasting realities, the children were able express their representations of both. This methodology was based on the understanding that in drawings, an 'original' and 'spontaneous' process of denaturation operates . Selected cases from the results of these interviews were chosen for maximum variation sampling, using the following criterion for differentiating residential status: informal or formal . --- Background: public policies regarding foreign languages and digital inclusion A national education law was enacted in late 2006 and contained a renewed normative core, composed, among others, from the law of Technical Education Professional No. 26.058 and Education Financing Law No. 26.075. Its main feature was that it defined actions of compliance and the sovereign decision to generate a model of social and economic development, with education as both a priority and a requirement . --- About the Teaching of a Foreign Language Under this law , the 'Junior Languages' programme was born, proposing that a foreign language be taught from the first grade for all students in the city's public schools, so that they have the ability to access that knowledge for free. Thus, since 2009, in the same district, the teaching of a second language has been present, starting with first graders in all primary schools dependent on the Directorate of Primary Education ; by 2011 this measure had reached all third graders. Furthermore, this initiative was accompanied by an increase in language workload for fourth through seventh grades . --- About the Digital Inclusion Policies Furthermore, as mentioned in the introduction, National Education Law No. 20.206 gave rise to digital inclusion policies in Argentina, through the provisions of Article No. 27, which states that there is a 'need to generate pedagogical conditions for the use of new information and communication technologies, for students and teachers in public schools across the country'. Since 2006, then, policies for digital inclusion began to take shape at national and local level, which, in the case of Buenos Aires City, materialised into a plan with two main lines of action , according to the educational level of schools. On the one hand, a secondary-level national programme emerged as part of the Connect Equality Program. On the other hand, at the primary level, state intervention came from local government through the PIED. The implementation of this plan included delivering laptops to students , and also the distribution of laptops to teachers as well as other educational actors . The Connect Equality Program, unlike the PIED, was a policy of the national government created from the Decree 459/10, and implemented jointly by Office of the President, the National Social Security Administration , the Ministry of Education Office, the Cabinet of Ministers and the Ministry of Federal Planning Public Investment and Services. It began to be implemented in April 2010 through the distribution of netbooks at public secondary school level across the country, as well as to special education schools of all levels and vocational schools dependent on special education. By March 2012, both educational policies had achieved their implementation targets for children and adolescents in public schools in Buenos Aires City, the population from which respondents were selected for the investigation. --- The Research Approach The research approach was shaped by the understanding that imagination, creativity, expressiveness and experience are condensed as sensible practice in the gap between expressiveness and the subject's experiences. These are understood by Scribano as itinerant mediations that occur in the context of relations between positions and conditions of being-in-the-world subjects. This, in turn, implies that in experiences a crossing and re-assimilation of sensations arises, and this and the associated emotions give effect to the flow of everyday life. Indeed, the future of the social experience of living a narrative can be captured through the network of expressiveness . Furthermore, social experiences mediate the different positions of the subject as a 'being-inthe-world' from its life lived. Creativity allows subjects to enrol in various surface experiences while living, such as memory and projection, and the face they present to others when showing their emotions, feelings and meanings of the everyday world they inhabit. Thus, expressiveness has the purpose of unwrapping or unpacking what has been naturalised and placing it 'on the verge' of those experiences as class habitus, getting the wrapping and putting it in connection with what was involved, capturing the trajectory experience/story/experience . In this context, expression opens up paths to other ways of shaping the voice; the subject regains its silent identity that was hidden and re-takes its reality from his or her potential recognition. Thus, this practice is composed of symbols and meanings that are always beyond 'reality to the naked eye'. Pictures symbolise turning points in the 'naturalness' of the social world and re-map the meanings . In this context, an expressive methodological praxis was proposed for this inquiry, to make and remake the world of interpretations and representations of the subject in a creative way. Examples of this praxis include using aesthetic experience recording techniques -such as theatre, painting, photography, music, and many other ways of expression -that are presented as productive to inquire about the child and adolescent world . Plastic expression allows the subject to be brought into a game, in relation to the work and the senses/feelings. In the act of creating, horizons about 'here' -memory and projection, defining what, who, why and for whom it is produced -are created. Artistic expressions such as drawing and playing, but also watching, show the contents of the subject's experiences in connection with the individual-subjective-social, 'to shape and color the world' in his or her ability to create constructive ways of expressing it for the appropriation and interpretation of others . In this sense, drawing, as a tool in the creative process, allows us to investigate these ways of 'seeing' and 'feeling' in the world of these children and adolescents, and thus to deconstruct their representations and advance the valuations that they have on education and the educational tools analysed. For this article, the scheme for the interpretation of the drawings was based on three levels: The stage of expressiveness. The context of interaction in which creative work is performed and the imaginary synthesising of the experience of 'being-in-the-world' that subjects described takes place . Decomposition -recomposition of the drawing. Where the image is decomposed into its partswhich figures, which words, how they were made and how they were related to give authority to the whole, for the subsequent interpretation of expressive productions . To this end, Scribano's proposal is used via a table containing components, elements, details and organisation of the drawings, which allows disaggregation, reuniting and rejoining of the parts . Interpretation of drawings. Analysis of the different superimposed layers that account for the entire composition , paying attention to the analysis of parts, and interactions of the figures together . --- Results Using this scheme and interpretation of compositions by children/adolescents as respondents, this article will attempt to answer the research questions posed in the introduction. Each pattern can be interpreted as a discourse, by which the child expresses and communicates something; it is a tool to help solve problems, create new ideas and facilitate communication. For Moreno Sáez , for example, approaching the drawing of children involves contact with one of the most pure, spontaneous and sincere candidly sources of artistic expression. The essence of the design lies in the awareness that it is only a tool and is used to communicate with yourself and others; all drawing represents an abstraction -it is a language, as it handles what it perceives as real, which is not the same as reality . --- Expressiveness Scenario For the inquiry, students from public schools in Buenos Aires City were interviewed in depth using the frame of creativity outlined in the earlier sections to understand their perceptions of the educational policies in question. The cases were selected from the population reached by the children's module in samples 2010 and 2011 of the Survey of the Social Debt Argentina , through maximum variation sampling according to residential status, using the criterion informal or formal . Interviews were organised according to the respondents' educational level: for the primary level , four children were interviewed, two from the informal residential status and two from the formal residential status . In addition, four teenagers were interviewed from the secondary level according to residential status. The interviews were conducted individually with each of the boys/girls and adolescents in their home, and the dialogue was recorded, with prior permission from the parents and the boy/girl, during the months of March and April of 2012. The instructions given to respondents were that by means of a drawing, on one side they should express their ideas on how they imagine is the life of a person who cannot speak English or use the computer; and the other side, imagine how is the life of a person who can speak English and knows how to use the computer. Thus, by contrasting realities, they would have the opportunity to make representations about the symbolic value they have of the educational tools. At first sight, the production of each design was different, varying according to the age and type of neighbourhood the children came from. Also, an uneven use of colours was seen, no matter what the age group or area of origin was. However, further observation showed that those children who chose to use the colours provided did so throughout the whole composition. This use was observed in only two drawings, by a group of children at primary level. What was noted was that in both Drawing No. 5 and Drawing No. 8, teens chose to add just one detail of colour. In this article, we imagine that these subjects have mastered the tools offered by policies that are the subject of this research, whereas those people who have not mastered them only used black marker, for the same compositions. This was observed only in adolescents at the secondary level. Figure 1. Girl, 9 years old -informal residential status: 'She [left side] can be a computer teacher, and she [right side] cannot'. [5] Figure 2. Girl, 9 years old -formal residential status: 'This boy [left side] lives in a house, like this, very big and has an owl, and here has a door and a garage … and this one [right side] has nothing because he does not buy in Mercadolibre'. [6] Figure 3. Boy, 11 years old -formal residential status: 'He [right side] would be sleeping all day in his house and she [left side] is a teacher, she knows how to use a netbook … this is a piece of furniture [right side], this is a chair and this is another furniture'. Figure 5. Teenage girl, 13 years old -formal residential status: 'I imagine someone my age, attending a public school where they teacher her English and knows how to use a computer, and the other person, I imagine someone in not such good living conditions, that cannot but a computer or does not get one for example, because he does not go to school'. --- Composition: restructuring of the drawings One possibility for framing the analysis and interpretation of the compositions is to follow the path of 'break to re-compose' . The points for following this framework include: • Components: core elements of expression that focus attention at first glance; • Details: individual traits that the authors have given the drawing to show 'beyond' or 'here' component; • Organisation: the effect of composition, possibly in some detail, displays the representational map of cartoonists; • Relations: indicates how the components interact in the visual organization, revealing what the artist sees as a semantic connection . [7] Table I is a summary of this framework of composition analysis. --- Interpreting the Drawings The unbundled network layers reflected in Table I can be deconstructed by the different representations of children and adolescents in Buenos Aires City, in relation to the domain of the two educational tools analysed. Regarding the components, it can be seen that the drawings contained at least two people, one on each segment of the paper. In some cases, such as Drawing No. 1, it can be seen that the subject joined two others to the composition, linking each to the domain or not to the educational tools. Moreover, in Drawing No. 6, the teenager decided to represent the person who dominates the computer as a pair, as part of a dialogue in English, while the person who does not dominate is alone in the other segment of the composition. Also, as extraordinary components, an owl stands in Drawing No. 2: the girl decided to associate this with the person who mastered the use of computers, because you can buy one on Mercadolibre . In regard to details applied to the compositions in differentiate scenarios over and over, people drew personal characteristics, including a dress, a beard , and expressions of happiness and sadness seen in the faces of the people represented. If you emphasise the organisation of the compositions, there was a common factor in all representations, which is the horizontal division of the paper leaf, to separate each 'world' or reality. However, the organisation within each segment was varied in the compositions. In some cases, subjects drew a reference floor, while other components of the composition appear to be 'floating on air'. The most significant differences were observed in terms of the contexts that were decided on to accompany each of the situations, that they drew as domestic/labour, urban density, etc. Finally, the relations established by the children in the compositions can be interpreted as a closeness or distance to gender when choosing the sex shown in the drawing. The same gender was chosen to represent the person who mastered the tools, and the opposite gender was chosen to represent those who do not dominate either. Also, according to the account of the interviewees, they set different bridges of relationships that they associated with their families. An example is the case of Drawing No. 7, in which the person who dominates the tools perfectly matches the description the person doing the drawing made about his mother. For further analysis, we can also take into consideration some dimensions proposed by Huergo and Ibáñez : Absences: A constant seen in almost all the drawings was the absence of real content in the composition of the world of those who cannot handle the educational tools. In most cases these gaps, such as in Drawing No. 8, which contains only a man, or Drawing No. 2, in which the girl expresses the absence of an owl, are in the possession of the person speaking the foreign language and familiar with the use of the computer. Metaphors. A representation, linked to feelings, of the people who know how to or do not know how to use the tools in questions was observed; as evidenced in their representations of the drawings and stories, children associated knowledge to happiness and the lack thereof to sadness. This was expressed in both cases in drawn facial expressions, such as in Drawing No. 7, in which the person is lying on a bed. Metonymy. Other associations observed in both the drawings and stories were development/marginality. In some cases, such as in Drawing No. 5, it was seen that the lack of use of a computer or no foreign language proficiency was associated with rudimentary shelter, life in the trash , or life away from urbanisation, as evidenced by the last drawing. Almost in line with the association of urbanisation/marginalisation, a tendency to relate education to wealth, employment and economic success was observed, while those who were not educated were associated with unemployment and poverty. This was evidenced in more depth in the story in Drawing No. 4, which is expressed as a differentiation factor. Symbols. The symbols used in the drawings were mostly linked to the valuations that were associated with education, based on the two educational tools that were being investigated. This can be seen in the words that were used to demonstrate knowledge/ignorance of the provisions and in the use of colour in the representation of both worlds. In this sense, very distinct associations were observed in all cases related to the mastery of a foreign language or computer use, regardless of the sex or social stratum of the children and adolescents, in terms of success/failure. --- Final Thoughts Returning to the questions raised in the introduction to this article, it may be said, in general terms, that this analytical journey has shown that the value the subjects place on education is directly associated with economic success, participation in the labour market and access to commodities and services that promote the well-being of people. Representations of education by children and adolescents, in general, can be understood as a 'symbolic capital' that people can use in their future lives. In this context, the compositions showed that the subjects' imaginations reflect large asymmetries between the quality of life of people who dominate the tools and that of those who do not, symbolised through the feelings, attitudes and living conditions they drew. Furthermore, by focusing on partnerships within the drawings, not taking into account the vastly different strata from which respondents came, we saw that everyone seemed to reflect similar imaginations of these situations, prioritising one or another element, even combining features. However, differences were observed when the stratum of the children was taken into account, when education was viewed in terms of the dichotomy cause/consequence of poverty. It was observed that children from higher strata seem to infer that lack of education leads to poverty, while children from lower strata expressed ideas associated with poverty preventing people from accessing education. In this sense, the representations of the children and adolescents interviewed could relate to their current or potential social position, which, according to Bourdieu , is the subject of perceptions and assessments that depend on past history , and of possible reference frames. However, this conjecture is certainly only a first approximation to the perceptions of children and adolescents. It deserves to be deepened in later works. Finally, an overall observation is that it is important to consider the relationship of tension that exists between the drawing, the child's interpretation and the interpretation of the researcher. While the children and adolescents performed their artistic compositions with freedom, being asked to create in the context of an interview -with the presence of the researcher and being prompted by questions -may have influenced the scene of expressiveness. Then, in the context of the interview, the answers given by the subjects to questions may also have influenced their artistic interpretation. It is important to highlight the bias of the adult's view on the creativity of children, which takes its reading from a place of knowledge and asymmetry, and so can stain interpretations and reviews of subjects. --- Notes

This article aims to describe the different ways that children in public schools connote values and represent two new educational tools received in the framework of two educational policies, concerning access to foreign languages since first grade, and laptops and digital literacy. How do children and adolescents represent the use of computers and the dominance of a foreign language? How do they imagine that both tools modify people's course of life? What values are associated with the use of them? Also, given that the aim of such policies is mainly to reduce social gaps, the article explores the notion that social inequalities may be evident in the representations that children build around these tools. In order to answer these questions, it analyses data gathered from in-depth interviews conducted in 2012 with a group of students from primary and secondary public schools in Buenos Aires City, in which a 'creativity method' was used as a means to determine students' perceptions about the new educational tools they had acquired. The cases were selected with maximum variation sampling, using the following criterion for differentiating residential status: informal (slum or urban settlement) or formal (upper-middle socio-economic neighbourhood).

Introduction Child sexual abuse is widely recognized as a serious problem in sub-Saharan Africa and has been described as a "silent emergency" . While estimates vary by country, definition and study methodology, prevalence rates for some countries are alarmingly high, with one in three females in Swaziland, for example, reporting experiencing some form of sexual violence before age 18 . High HIV prevalence in much of the region adds risk of exposure to HIV to the consequences faced by victims of CSA . Additionally, high numbers of orphans and other vulnerable children and high levels of child poverty and displacement in SSA exacerbate vulnerability. Although all African countries with the exception of Somalia have ratified the United Nations Convention on the Rights of the Child , CSA continues to be shrouded in silence and shame , its victims too often failing to receive the care and support they need. CSA interventions in SSA are usually delivered at the local level by civil society organizations rather than by governments , 2011), and this limits their scope and integration; popular points of interventions are awareness-raising and increasing the availability of child helplines and safe houses for victims. Alongside interventions that are CSA-specific are approaches that situate CSA within the context of gender-based violence more generally, on grounds that the prevention of child rape cannot be dissociated from efforts to transform the social status of women and children . Across SSA, patriarchal and age-based hierarchies create opportunities for CSA and may also affect a community's perception of and response to CSA. It is, for example, customary for children to be raised to unquestioningly obey any adult's commands out of respect for their elders . In addition, regional myths may provide a pretext for CSA, for example, the myth that sex with a child has magical properties or is a way to cure a person from HIV . Reporting of abuse and use of available services is very low and may be influenced by embarrassment, fear of reprisal or incurring further shame on the family , and by poverty and distrust in the healthcare and legal institutions . Although an increasingly prevalent multi-level conceptualization of CSA in SSA supports holistic, integrated approaches to prevention and response, CSA efforts directed at parents and caregivers have been largely absent in SSA to date. Parents are well-placed to complement other sources of CSA prevention in the community due to their unique ability to engage their children in dialogues about sexuality-related issues early and on an ongoing, developmentally-and time-sensitive basis . In addition, because of their proximity to their children, parents may be more able to take swift action to stop CSA, address trauma, and minimize harm in the event that abuse does occur . Recent studies have documented both need and demand for parental education around CSA in SSAs , suggesting that an intervention directed at parents could be well received. --- Background The Families Matter! Program is an evidence-based intervention for parents of 9-12 year-olds that promotes positive parenting practices and effective parent-child communication about sex-related issues and sexual risk reduction . The US Centers for Disease Control and Prevention provides technical support for implementation of FMP in countries across SSA. FMP has been proposed as a valuable platform for addressing CSA because the program is widely accepted within communities and teaches parenting skills that closely map onto protective factors identified in previous studies on CSA . In response to the 2009 Violence Against Children in Tanzania report , a draft FMP session on CSA was developed as a supplement to the existing 5-session curriculum. Learning objectives and activities were developed within the context of the existing FMP curriculum and with reference to CSA literature . Initial drafts were refined following feedback from Tanzanian and international specialists in violence prevention and children's rights. The draft sixth session was field-tested in rural Tanzania in late 2012 with 8 parent groups who had recently completed FMP sessions 1-5. Field testing revealed that the session was well received and perceived to be an appropriate and valuable complement to the existing FMP sessions. However, it also identified the need to revise the session to incorporate more diverse and interactive learning approaches, including role-plays and audio narratives drawing on authentic scenarios. This paper describes the subsequent development of these resources and activities drawing on stories contributed by young people across Africa to the Global Dialogues/Scenarios from Africa scriptwriting competitions. Since 1997, the GD/SfA competitions have invited young Africans, up to age 24, to contribute scripts for short fiction films to educate their communities about HIV and AIDS . The winning ideas in each contest are selected by local juries and transformed into short fiction films by leading African directors . By 2011, the process had generated an archive of over 50,000 narratives written by young people from 47 countries. Analysis of samples from this archive provides insight into the context of HIV and related issues, including CSA, from the perspective of young Africans, thereby supplementing existing peer-reviewed and gray literature. While the youth-authored narratives do not present a comprehensive overview of CSA in SSA, they do have the advantage of providing access to authentic voices of young Africans on the subject in an emotionally-compelling and highly contextualized narrative form that can be readily adapted to provide powerful learning tools. The FMP curriculum makes use of narratives in an audio format that can easily be played on a battery-operated CD player in low-resource rural areas. It also incorporates role-play exercises which call on participants to improvise on a scenario and build specific skills. In light of the cultural diversity of SSA, and of individual sub-Saharan countries, our objective was not to create a definitive curriculum to be used across the continent, but rather to generate a reference curriculum that could be adapted -both culturally and linguistically -to specific regions following community needs assessment and pre-testing. In this paper, we describe the enhancements that were made to the FMP session on CSA following field-testing with a view to informing other programs working on CSA in SSA. One objective in enhancing the FMP session on CSA was to ensure that it resonated optimally with parents and was grounded in contextual examples that would be perceived as authentic. Given that FMP participants are parents or caregivers of a child aged 9-12, we felt their emotional investment in the CSA session could best be fostered if it focused on younger, pre-pubescent and pubescent children and if it stressed the short, medium and longterm consequences of CSA for the victim's physical and mental health and wellbeing. However, we also sought to work against the normalization of CSA involving postpubescent victims, by repeatedly stressing that all children -not only those within the FMP window of ages 9-12 -are at risk for CSA. In line with this strategic decision, this paper focuses on situations where a prepubescent or pubescent child is tricked and forced into a sexual activity in the home or community by an adult or older child. Earlier FMP sessions address in depth issues of sexual violence, coercion and pressure, including dating violence, faced by older children. --- Methods The narratives described in this paper were submitted to the GD/SfA contest, which was held continent-wide from 1 February to 15 April 2005. Over 63,000 young people from 35 African countries participated in this contest, submitting approximately 23,000 narratives. As part of a six-country study of young Africans' social representations of HIV and AIDS, a sample of 586 narratives from Senegal, Burkina Faso, South-East Nigeria, Kenya, Namibia and Swaziland was constructed . The de-identified data were transcribed verbatim in English or French and entered into MAXQDA 2007 qualitative data analysis software , where they were labeled with descriptive codes with reference to a detailed codebook covering a range of HIV-related themes, including "sexual violence/coercion/rape" . A summary was written for each story and these summaries were coded with up to 6 out of a possible 45 keywords, which included "sexual violence/coercion/rape" . These steps allowed us to isolate both individual text segments related to sexual violence and those narratives in which sexual violence was a central theme. We then isolated those narratives about sexual violence perpetrated against pre-pubescent and pubescent children and address those narratives here. Narratives about CSA were not elicited by the scriptwriting contest but were voluntarily contributed by young Africans in response to an invitation to submit ideas for a film about HIV and AIDS. It is impossible to know to what extent, if any, they draw on personal experience. We treat them as fictional stories which are nonetheless likely to draw on -and therefore provide insight into -situations of which the young authors are aware within their own communities, as well as normative understandings of CSA in these communities. The secondary analysis of the de-identified GD/SfA data was approved by [University name removed for blinding] Institutional Review Board. --- Results Eighteen narratives in the GD/SfA data set addressed CSA perpetrated against prepubescent and pubescent children. Authors came from Swaziland , Namibia , Kenya , Nigeria , and Senegal . Most of the narratives were by female authors; half by rural authors. Two of the narratives written by male authors, one from Nigeria and one from Senegal, relate stories of young boys being sexually abused. We focus on the following key thematic areas: known versus stranger perpetrator; progressive versus isolated abuse; secrets, gifts and threats as forms of manipulation or coercion; and disclosure, response and action. --- Known versus Stranger Perpetrators GD/SfA Data-The perpetrator was a relative or acquaintance of the victim in two-thirds of the narratives and a stranger to the victim in the remaining six narratives. Half of the CSA narratives, including five from Swaziland, recount stories about incest. In the three narratives where the perpetrator is known to the victim but not related, the assailant is a trusted individual, a person in a position of authority, or both. Three narratives -from Swaziland, Namibia and Kenya -recount child rapes motivated by a virgin cleansing myth and perpetrated by members of the extended family or the mother's boyfriend. Victims' respect for their elders increases their vulnerability: for example, a schoolgirl is the victim of sexual assault perpetrated by two of her teachers, who are able to use their authority and familiarity with the victim to isolate her from her peers . A Senegalese narrative relates the rape of a girl by a guest staying at the family house . In contrast to the detailed and often graphic descriptions of CSA by known assailants, the attacks by strangers are often brusque and opportunistic. These abusers often prey on the poverty-related vulnerabilities and innocent, trusting nature of the children they abuse, for example, offering a child a lift home or attacking a working child who is hawking fruit on the street. Application in FMP Curriculum-The narratives help us to pepper the curriculum with examples of risk situations that are concrete and contextualized, allowing us, for example, to draw parents' attention to the need to be aware of where their child sleeps at night and who else is in the house, of the importance of not accepting lifts, and of risks related to work or chores in the community. This concreteness is central to making the risk of abuse real and relevant for parents and helping them understand and communicate these risks to their children. A scenario about the risks of accepting a lift home from school appears in an audio narrative which models good parent-child communication around CSA. The audio demonstrates how key FMP parenting strategies -supervision, building a close relationship, and good communication skills -allow the mother to know quickly that something is troubling her child. These parenting strategies also allow the child to feel confident in communicating her concerns to her mother, making it easier for the mother to protect and guide her. Another audio recognizes the importance of parents communicating these skills to others who may be in charge of their children during their absence, for example, when a child is staying with relatives in the village during the harvest time. As the FMP curriculum cannot hope to address all potential risk situations children may face, it addresses those that are most salient across SSA. In addition, parents are encouraged to identify, plan for, and role-play other culturally relevant risk situations with their children. While the contextualized examples drawing on the GD/SfA data do not obviate the need for adaptation of the curriculum in specific settings, they do provide a culturally resonant point of departure for it. --- Isolated versus Progressive Abuse GD/SfA Data-Some of the CSA narratives recount only one isolated event, while others are a culmination of several years of progressive abuse at the hands of one or more perpetrators, often linked to incest and/or to the child's displacement as a result of poverty or orphanhood. Several narratives describe stratagems that are used by perpetrators to secure a victim's compliance or to take advantage of a mother's trust or her absence, for example, at a funeral. A narrative by a 12-year-old male from urban Swaziland recounts a girl's sexual abuse from the age of five, first by her grandfather , then by her two stepbrothers. In another Swazi narrative, by a 15-year-old female author, the main character discloses to a friend the progressive abuse she is suffering at the hands of her uncle: "my uncle tells me that I'm more beautiful and when telling me that he touches my body all over." Application in FMP Curriculum-Despite the lower likelihood of CSA by strangers than by known perpetrators, "stranger danger" and considerations of personal safety provide valuable entry points for parents to initiate the long-term process of talking about sexual abuse with their children. For example, reviewing personal safety rule, such as "avoiding places where you could be alone with someone," provides an opportunity to ask children why they think the rule exists and thereby gauge their understanding and readiness prior to providing additional information. This process is modeled in one of the audios, in which discussion of rules related to "stranger danger" transitions into discussion of the privacy of the genitalia. The narratives about progressive abuse translate into one of the most important and extensive audio narratives in the CSA session, designed to help parents understand the insidious nature of some forms of abuse. In the interactive discussion following the audio, parents are asked whether they think it is easy for the child to see right away that the uncle's behavior is not ok. It seeks to help them understand that, while a child might be able to understand the risk that she could face from a stranger when she is out alone after dark, it is much more difficult for her -based on her limited life experience -to understand risk within her own home or in other places where she normally feels safe. It drives home the message that children do not suddenly become adults when they reach puberty and that they need adult help in order to understand situations that can put them at risk. This narrative continues later in the session, illustrating one central prevention message: the "No Secrets, No Threats, No Gifts" rule. --- No Secrets, No Threats, No Gifts Rule GD/SfA Data-In the narratives, the perpetrators of CSA use three primary tactics to force or lure children into sexual acts and silence them after the abuse has taken place: threats, secrets and gifts . These often overlap as means for an abuser to exercise control over a victim. The perpetrator often concludes the act of rape by instructing the victim to keep the abuse secret, and this command is usually conveyed with an overlapping threat of further harm to the victim or his/her family if they reveal the abuse to anyone. Gifts are presented in the GD/SfA data in the form of sweets, money and free rides in motor vehicles as a means to lure a child or placate a child following abuse. One perpetrator in a Namibian narrative uses candy to lure the small child he plans to rape in hopes of being cured of HIV . Another narrative from Kenya relates a situation when two friends, Rose and Mary, accept a ride from two strangers. The two men take full advantage of the girls' naïveté. Rose and Mary are raped, and they later discover that they are infected with HIV. After the abuse has occurred, the men give Rose and Mary money to try to silence them from taking action against them. Despite the fact that the girls had not agreed on any monetary transaction before the rapes, they accept the money and tell no one what has happened to them . Application in FMP Curriculum-These descriptions inform the further elaboration of the audio about progressive abuse, which through interactive debriefing is crystallized into the "No Secrets, No Threats, No Gifts" rule that parents can share with their children. The audio also seeks to communicate to parents the strategies abusers may use to: manipulate their victims; shift the blame for the abuse to the victim; and manufacture a sense of complicity or responsibility on the part of the victim which militates against timely disclosure. This audio narrative also provides an opportunity to anchor the concepts that children should always know that their parents will believe them if they tell them about CSA, that parents will never be angry with them if they disclose, and that if a child is sexually abused it is never the child's fault. --- Disclosure, Response and Taking Action GD/SfA Data-The victim discloses the sexual abuse to a trusted person in eight of the eighteen GD/SfA narratives. One narratives describes how the child "kept the agony of the incident to herself because she did not know how to relate the incident" to her caregiver . The perpetrator of CSA is brought to justice in only five narratives, two from Namibia and three from Swaziland. One of the narratives from Namibia ends with imprisonment for the two perpetrators, a traditional healer and a man with HIV seeking a virgin cure who rapes his cousin's five-year-old daughter. In a Swazi narrative, the victim seeks support from a helpline after she is abused. However, in one case of abuse by an uncle in a narrative by a 16-year-old Swazi female author, nothing is done to bring the perpetrator to justice, primarily due to fear of the shame that this could be brought on the family. Application in FMP Curriculum-This data informs an audio narrative modeling a parent's response to a child's disclosure, which highlights the importance of maintaining calm, acknowledging and praising the child's courage, and reassuring your child that you love and believe her/him and that s/he is not to blame. It also identifies signs and symptoms that, in the absence of disclosure, could indicate CSA. The curriculum stresses the importance of prompt action to remove the child from the situation and seek medical care which -in the event of need for emergency contraception or post-exposure prophylaxis for HIV -is time-sensitive. It stresses that the child will need help from the parents and from service providers, if available, to come to terms emotionally with the abuse. A communityspecific handout shows where medical and mental health services can be sought locally and provides details of national CSA service providers. The CSA curriculum also stresses the importance of reporting the abuse to someone in the community, including community or religious leaders, medical professionals, the police or school principal to ensure that the abuse will not happen again to one's own or someone else's child. While acknowledging the challenges that reporting presents, particularly in situations where the family is financially dependent on the perpetrator, the FMP curriculum stresses that covering up the abuse or coming to an agreement with the abuser, will not stop the abuse and may put other children at risk. The curriculum encourages parents to use all the resources available within the community -traditional leadership, police, health workers and NGOs -to make sure that CSA is not accepted and that children who have been abused are supported. This section closes with advice on helping a child through the long-term healing process. --- Discussion As a parenting intervention that addresses CSA within the context of broader parent-child communication around sexuality and sexual risk-reduction, FMP has the potential to make a valuable contribution to multi-level efforts to prevent and respond to CSA in SSA. In addition, FMP seeks to interface with local community mechanisms and to promote linkages to local and national-level CSA-related services, including medical and mental health services. An outcome evaluation will assess the impact of the CSA-focused session on parents' awareness of CSA and its associated risks. The evaluation will also examine the intervention's impact on parents' self-efficacy to talk with their children about CSA, help protect their child from CSA, and respond in the event that their child experiences or has experienced CSA. An enhanced six-session FMP curriculum places particular emphasis on gender norms and gender-based violence throughout so that by the time participants have reached the sixth session on CSA, they are already well primed on the gender power dynamics and age-related hierarchies that increase opportunities for CSA in SSA. In addition, the curriculum encourages parents to engage their community around CSA. Hence, FMP is sensitive to the structural dynamics that underlie and perpetuate CSA in SSA and the broad-based mobilization that is necessary to its prevention. While poverty, patriarchal norms and impunity for perpetrators represent major obstacles to CSA prevention in SSA, parents can be empowered to protect their children and to respond in the event of abuse. When developing health education materials, it is customary to collect formative data by means of focus group discussions and in-depth interviews. However, transcripts from this kind of data collection are not always easily translated into curricular materials. The narrative data described here lend themselves more readily to adaptation into health education materials, which are well-suited for low-literacy adult learners. A recent comprehensive literature review on CSA in SSA called for greater attention to be paid to the voices of communities and of children themselves. We felt that narrative audios drawing on the GD/SfA stories would be a particularly valuable way to enhance the CSA curriculum. They are likely to resonate culturally, to be perceived as authentic, and to offer opportunities for richly contextualized and interactive learning in pursuit of the CSA goals and learning objectives. The GD/SfA data have clear limitations: our sample was small and unrepresentative; these are not necessarily personal accounts of CSA, but rather young Africans' unelicited stories about CSA and, as such, may be influenced by media representations. The narratives' concordance with the limited existing ethnographic literature on CSA in SSA suggests that they tap into deep-seated social understandings of CSA risk and dynamics in SSA. In addition, they provide plausible contextualized and emotionally-compelling accounts of CSA that, we hope, will allow the curriculum to resonate with participants. --- Conclusion FMP situates CSA prevention and response in SSA within the context of a holistic parenting skills program. It is conceived as a valuable component of comprehensive multi-sectoral response to CSA. The GD/SfA data allowed us to incorporate authentic voices of young Africans into our role-plays and audio narratives to create a diverse interactive curriculum around CSA that is adapted to the needs of low-literacy adult learners in SSA. --- Biography Kim S. Miller, PhD, is the Senior Advisor for Youth Prevention at the Centers for Disease Control and Prevention, Division of Global HIV/AIDS Prevention. She joined CDC 25 years ago after completing her doctoral studies at Emory University in Atlanta, Georgia. Her current international research and prevention activities focus on pre-risk prevention approaches to sexual risk reduction, understanding and reducing sexual risk among youth, youth development approaches to sexual risk prevention and the role of the family in the promotion of sexual risk reduction and protection of children. --- --- Table 1 Goals and Learning Objectives of FMP Session 6 -Understanding Child Sexual Abuse Goals: • To raise parents' and caregivers' awareness about child sexual abuse • To increase parents' and caregivers' understanding of their role in preventing child sexual abuse • To increase parents' and caregivers' awareness of their role in protecting and supporting their children when responding to child sexual abuse --- Learning Objectives After completing this session, participants will be able to: • Explain what child sexual abuse is and why children need to be protected • Explain why parents and caregivers play an important role in protecting their children from child sexual abuse and responding to child sexual abuse • Apply the main parenting strategies of FMP to protecting their children from child sexual abuse • Talk with their child about child sexual abuse and situations that could put them at risk for child sexual abuse • Identify actions that can be taken within the family or household and within the community to protect children from child sexual abuse • Recognize the signs of potential or actual sexual abuse

Despite widespread recognition of child sexual abuse (CSA) as a serious problem in sub-Saharan Africa (SSA), few far-reaching programmatic interventions addressing CSA in this setting are currently available, and those interventions that do exist tend to focus on response rather than prevention. Parents are in a unique position to engage their children in dialogues about sexualityrelated issues and thereby both help prevent CSA and take swift action to stop CSA, address trauma, and minimize harm, in the event CSA does occur. The Families Matter! Program (FMP) is an evidence-based intervention for parents and caregivers of 9-12 year-olds in sub-Saharan African countries that promotes positive parenting practices and effective parent-child communication about sex-related issues and sexual risk reduction. This paper describes the enhancement of a new FMP session on CSA, drawing on authentic narratives contributed by young people to the Global Dialogues/Scenarios from Africa (GD/SfA) youth scriptwriting competition. The GD/SfA data permitted incorporation of young Africans' voices and resulted in an interactive curriculum that is grounded in contextually-relevant and emotionally-compelling scenarios and adapted to the needs of low-literacy adult learners in SSA. Experiences are shared with a view to informing the development of interventions addressing CSA in SSA.

Introduction Female genital mutilation is a significant global health concern and is likely to become an increasingly important healthcare challenge in destination countries such as the UK owing to rising levels of migration from FGM-affected countries. Currently, there is no consensus on the optimal timing of deinfibulation surgery for women who have experienced type 3 FGM and care provision remains suboptimal in the UK. This qualitative study aims to explore the views of survivors, male partners and healthcare professionals on the timing of deinfibulation and delivery of NHS FGM services. Methods and analysis A qualitative study, informed by the Sound of Silence conceptual framework, will be undertaken via two work packages . WP1 will explore views on timing preferences for deinfibulation and NHS FGM services through interviews and discussion groups with FGM survivors , male partners and HCPs . WP2 will use established techniques via two workshops and national stakeholder ) to synthesise qualitative research findings and inform best practice and policy recommendations around the timing of deinfibulation and NHS FGM care provision. Supported by trained interpreters, data collection will be audio recorded and transcribed. Data will be analysed using the framework method to facilitate a systematic mapping and exploration of qualitative data from multiple sources. --- Ethics and dissemination The study has received ethical approval from the North West Greater Manchester East Research Ethics Committee . The outputs for this study will be recommendations for best practice and policy around FGM care provision that reflects the views and preferences of key stakeholders. The findings will be disseminated via conference presentations, peer-reviewed publications, patient groups, third sector organisations and social media. trial registration number ISRCTN 14710507. --- IntroduCtIon Female genital mutilation or cutting is an increasingly global issue owing to international mobility and migration. 1 The practice of FGM has been performed for millennia 2 and continues to be prevalent in some countries in Sub-Saharan Africa, Asia and the Middle East. 3 An estimated 200 million women and girls live with FGM globally. 1 In the UK, FGM is identified increasingly among migrants from FGM-affected countries, with 137 000 women and girls reported as currently living with the consequences. 4 Between April 2015 and December 2017, 15 390 women and girls presented to NHS services where FGM was identified as a relevant condition and/or Open access treated. 5 The annual cost of NHS care for FGM survivors is estimated at £100 million. 6 FGM involves the partial or complete removal of, or injury to, the external female genitalia for non-medical reasons. 7 The WHO joint statement has classified FGM into four main types , with a further seven subtypes identified to capture more closely the variation in practices. 8 9 Generally, the extent of genital tissue cut increases from type 1 to type 3, with type 3 being the most extensive and often requiring deinfibulation 7 . There are immediate and lifelong health, obstetric, sexual functioning, psychosocial and economic impacts associated with FGM. 4 10-17 The risks of adverse outcomes appears to be greater the more extensive the FGM, 15 with 9 in 10 type 3 FGM survivors reporting complications. 18 The consequences of type 3 FGM may lead to loss of life and reduced quality of life. 10 19 The WHO reports 20 that deinfibulation is associated with improved health and well-being, as well as, allowing sexual intercourse and childbirth, although there is limited direct evidence to support this statement. A recent systematic review found no evidence that deinfibulation improved urological complications. 21 There is however a case to be made that deinfibulation is associated with improved gynaecologic and obstetric outcomes, although this is based on low-quality observational evidence. 22 There has been slow progress in the development of evidence-based care to improve health outcomes for FGM survivors, in particular, around the optimal timing of deinfibulation. 20 23 24 Deinfibulation can be undertaken outside of or during pregnancy 24 ; however, there is considerable variation within and between clinical recommendations for when deinfibulation should occur. 23 For example, Royal College of Obstetricians and Gynaecologists guidelines 25 recommend that deinfibulation should be offered prior to pregnancy and preferably before first sexual intercourse. These guidelines also state that deinfibulation can be performed antenatally, in the first stage of labour, at delivery or perioperatively after a caesarean section. Royal College of Nursing FGM guidance 26 does not provide a clear indication on the optimal timing of deinfibulation, with one statement indicating that the procedure is best performed when not pregnant and another that deinfibulation is best undertaken before or at least within the second trimester of pregnancy. WHO guidelines on the management of FGM 20 recommend either antepartum or intrapartum deinfibulation with a suggestion that timing should be based on wider contextual factors including patient preference, access to healthcare facilities, place of delivery and the skill level of the healthcare professional . In addition to a lack of consensus about when deinfibulation should be performed, there is also debate about whether timing affects outcomes, with some studies suggesting that obstetric risks increase the later deinfibulation is undertaken. 27 28 However, these findings were not substantiated in a recent systemic review of low-quality observational evidence comparing childbirth outcomes between antepartum and intrapartum deinfibulation. 24 A gap analysis of current FGM research has suggested that there is an urgent need for well-designed research to inform evidence-based guidelines and to improve the healthcare of women and girls with FGM. 23 A recent qualitative systematic review highlighted that it is important for future research to focus on care-seeking and decision making around the timing of deinfibulation. 29 To date, in the UK, there has been one qualitative study that directly explored women's experiences of deinfibulation 30 and no studies exploring the views of men and HCPs. With no clear consensus on the optimal timing of deinfibulation for type 3 FGM survivors, 20 23 24 there is a specific need to focus on exploring preferences, involving a diverse range of FGM survivors, their male partners and HCPs, across multiple centres. 23 31 Such research may help to inform the strategic planning and development of cost-effective, culturally acceptable NHS services, leading to improved outcomes for women and their families and help other organisations plan for improved support for FGM survivors and their families. This paper presents the protocol for the FGM Sister Study. --- Aims and objectives The aim of the FGM Sister Study is to explore and understand the views of FGM survivors, male partners and HCPs on the timing of deinfibulation and how NHS services can best be delivered to meet the needs of FGM survivors and their families. We will address this overarching aim via two work packages . The aims and objectives of each WP are presented in table 1. --- MEthodS And AnAlySIS Study design and conceptual framework The FGM Sister Study is a qualitative study, informed by the Sound of 'Silence' conceptual framework. 32 The 'Silences' framework is underpinned by broader theoretical approaches with a worldview that accepts that reality is not objective, rather the social world is influenced by people in a particular society at a particular point in time. 33 'Silences' define areas of research and experiences that are little researched, understood or not heard 34 and are useful for researching sensitive issues and/or the healthcare needs and perspectives of marginalised populations. 32 Within the context of this study, although FGM is a contemporary issue that has increasingly become the subject of political and media interest, it remains a sensitive issue prevalent among marginalised populations and one that is under-researched. 'Silences' elucidates and underpins the research using a four-stage approach: working in Silences, hearing Silences, voicing Silences and working with Silences. There is an additional fifth stage, planning for Silences, that is not incorporated into the core four-stage model but will be used to help inform service delivery action planning Open access Aim: To qualitatively explore and understand the timing preferences for deinfibulation and how NHS FGM services could be improved for type 3 FGM survivors , their male partners and HCPs . Aim: To use established techniques with survivors and stakeholders to synthesise the qualitative research findings, inform best practice and policy recommendations around the timing of deinfibulation and FGM care provision and identify future actions. Objectives: 1. to explore knowledge, awareness and understanding of FGM and deinfibulation 2. to elicit views on preferences for the timing of deinfibulation and the rationale for these 3. to explore perspectives on the decision making process around deinfibulation 4. to explore knowledge, awareness and experiences of FGM services and support 5. to understand the enablers, motivators and barriers to FGM care seeking behaviours 6. to explore how HCPs describe, explain and reason about their care provision for FGM survivors and their families 7. to understand how FGM care provision could be improved to best meet the needs of FGM survivors, their families and HCPs who support them in their local context Objectives: 1. to explore views and reflections on the trustworthiness of our interpretation of the data and the conclusions drawn 2. to establish if there is consensus about the optimal timing of deinfibulation 3. to identify the key recommendations to inform NHS FGM care provision 4. to explore the facilitators and barriers to implementation of changes to NHS FGM care provision 5. to explore views on the requirements for future FGM research FGM, female genital mutilation; HCPs, healthcare professionals; RCT, randomised clinical trial; WP, work package. and future recommendations for research, policy and practice --- Study setting and timing The study will be undertaken with FGM survivors and male partners in three high FGM prevalence areas of England and with HCPs and key stakeholders across the UK including areas of high and low FGM prevalence. Data collection for WP1 commenced in September 2018, and initial analyses are planned to be completed ahead of the start of WP2 in November 2019. Synthesis of the results of WP1 and WP2 and final recommendations are planned to be completed by May 2020. --- Eligibility --- WPs 1a, 1b and 2a Women who have experienced FGM and male partners of women who have experienced FGM will be eligible to participate if they are aged 18 years or over, are resident in the UK, speak fluent English, Somali, Arabic or French and are willing and able to provide voluntary informed consent . We will exclude anyone where their clinician, support worker or the researcher judges that their distress relating to FGM may influence their ability to provide voluntary informed consent. Additionally, we will not snowball recruitment of a male partner where the participating FGM survivor does not consent to their participation. --- Work package 1c HCPs will be eligible to participate if they are aged 18 years or over, speak fluent English, are willing and able to provide written, electronically completed or verbal informed consent and are currently or have recently been involved in the delivery of care to FGM survivors and their families in the UK. --- Work package 2b Key FGM stakeholders including HCPs , policy makers, FGM specialist researchers/academics, health economists, commissioners and representatives from third sector organisations will be eligible to participate if they are aged 18 years and over, speak fluent English, are willing and able to provide written, electronically completed or verbal informed consent and are currently or have recently been involved in some aspect of service and/or care provision to FGM survivors and their families in the UK. --- Sampling and recruitment --- Work package 1a Four groups of pregnant and non-pregnant FGM survivors will be purposively sampled 35 including those: Open access who have not had deinfibulation, who have had deinfibulation for health and/or personal reasons, who had deinfibulation antenatally and who had deinfibulation during labour/at the point of delivery . Within these groups, we will try to ensure that we have maximum variation and diversity of views by including women from a range of FGM-affected communities , locations, ages and education levels. 36 Women will be recruited via multiple pathways including, for example, their HCP; advertising within FGM clinics, community settings and on social media, culturally sensitive snowballing 37 from women approached to participate and FGM community groups/third sector organisations. Multiple NHS trusts and third sector organisations will support the study. Relevant staff at participant identification sites will be trained by the research team to approach potential FGM survivors. Women will be approached, in the first instance, by a member of their usual care team or by a trusted advocate in a third sector organisation. Recruiters will be asked to briefly screen for eligibility and introduce the study. All documents to support participant identification and recruitment will be available in English, French, Somali and Arabic. If the woman responds positively, she will be asked to complete and sign a contact details form giving permission to be contacted by the research team. A researcher will then contact the woman to check her eligibility, discuss the study further, answer any questions she might have and arrange a mutually convenient time and location for the interview or to let her know the times and locations of discussion groups. At the end of the interview/discussion group, women will be asked discretely if their partner may wish to participate in the study. If she responds positively, then contact details of the research team will be left or if the man is present, then the recruitment and consent process will be initiated. --- Work package 1b Men will be identified via participants from WP1a, via support of local community groups and third sector organisations and via social media. --- Work package 1c HCPs will be purposively recruited 35 from across a range of groups including general practitioners, practice nurses, midwives, obstetrics and gynaecology clinicians, genitourinary clinicians, health visitors and sexual health specialists. HCPs across the UK will be identified via multiple pathways including, for example via: FGM service listings; contacting NHS trusts with maternity services in low FGM prevalence areas directly; the study team's FGM networks; advertising the study via electronic communications , professional bodies and membership societies; and snowballing from HCPs approached to participate. HCPs who express an interest in participating will be screened for eligibility by the research team, provided with further study information, given time to answer questions and then a mutually convenient time and location for the interview arranged, or they will be told about the times and locations of discussion groups. --- Work package 2a FGM survivors who take part in WP1a will be invited to participate. If required, recruitment will be supplemented via the same pathways as identified in WP1a above. --- Work package 2b HCPs interviewed in WP1c will be invited to participate. Other key stakeholders will be identified via the research team's networks and collaborators, social media and knowledge of FGM services acquired during the study. --- Anticipated sample sizes WP1 We will seek to recruit: ► up to 50 women who are FGM survivors, ► up to 10 male partners, ► up to 50 HCPs. --- WP2 We will seek to recruit: ► 20-25 FGM survivors for the community engagement event, ► 30-35 stakeholders for the national stakeholder event. Numbers will however remain flexible to ensure that we collect sufficiently rich data to answer the research questions and achieve core analytic saturation. 38 data collection Work package 1 Semistructured interviews have been identified as an appropriate data collection method, given that they can facilitate an in-depth exploration of participants' 'Silent' views 39 and are particularly useful in discussions of sensitive or traumatic experiences such as FGM. In addition, discussion groups will be used as an alternative data collection tool for FGM survivors and male partners, as Patient and Public Involvement representatives felt that women and men in some communities may be more willing, given the nature of the topic, to participate in a group rather than an individual discussion. Interviews and discussion groups will be conducted by a trained qualitative researcher and two researchers will be present in all discussion groups. Independent, professional interpreters will be employed to provide real-time oral translation during the interviews/discussion groups where there is a language barrier between researcher and participant. Debriefs between researcher and interpreter will be held after interviews/discussion groups if interpretation issues are identified. In addition, Open access researchers will keep a research journal throughout the data collection period to help to provide reference points in the journey to expose 'Silences'. 32 Participants will be given the choice as to whether they wish to take part in an interview or a discussion group. If they choose to take part in an interview, they will be given a choice as to where it takes place, for example, in their own home, in a clinic room where they were recruited/ work or via telephone . Discussion groups will take place in an appropriate location, for example, a community venue or clinic. Aligned with 'Hearing Silences' , 32 discussion guides for WP1a-c will be informed by a critical reflection of the FGM evidence base . These 'Silences' will then be heard by and discussed within the research team and with the survivor group. Semistructured interviews and discussion groups will be conducted in a participant-focused manner, allowing experiences and views important to participants to develop naturally. 40 The composition of the participants in each discussion group will be carefully considered, taking into account language, the community from which they are from, their deinfibulation experience and wider demographic characteristics. Discussion guides will be refined iteratively to ensure that all views are captured. Data collection and analysis will take place concurrently 39 and will continue until the research team judge that the data and sample have sufficient depth and breadth to address the research objectives. 38 At each interview/discussion group, participants will be asked to complete a short demographic questionnaire to facilitate maximum variation sampling and provide a description of the sample characteristics. --- Work package 2 A community engagement event and a national stakeholder event will be run by FGM experts at Barnardo's, with support from the wider research team and the survivor group. Participants will be shown a plain English summary in advance. At the start of each event, a study overview will be presented. Participants may be split into smaller discussion groups facilitated by a member of Barnardo's and supported by a member of the research team and/or the survivor group. Discussion groups were identified as an appropriate data collection method, given that they provide an opportunity for interaction and communication between participants in order to generate data and can provide a permissive and empowering environment where participants feel comfortable enough to share their views and question those of others. [41][42][43] Discussion will focus on participants' reflections of the trustworthiness of our interpretation of the data and the conclusions drawn; an exploration of 'what has or can change as a result of this study' in terms of NHS policy and practice and identifying future research to address other identified 'unheard Silences'. 32 Recommendations from each group will be shared and discussed within the whole group to establish if a consensus on the timing of deinfibulation can be reached and to identify the next steps following study completion. --- data analysis Interviews, discussion groups and events will be audio recorded and transcribed using an intelligent transcription system by a specialist company and subsequently checked for quality and anonymised by the research team. Up to six transcripts of interpreted interviews will have both the English and the second language translated and transcribed. This is to enable the research team to check that interpreters are employing content interpretation rather than word-for-word translation and to ensure that interpreted responses are conducive to and reflect the depth of response provided by the participant. Data analysis will be informed by framework analysis, 44 which provides a systematic and flexible model for managing and mapping qualitative data from multiple sources. The research team will use a five-stage framework approach 45 that captures, but also condenses, other framework approaches involving a greater number of stages. These five stages include compiling , disassembling , reassembling , interpreting and concluding . Each stage of the framework approach can be aligned with the stages of the Sound of 'Silence' conceptual framework to ensure relevant synthesis of qualitative methodology and philosophical/conceptual underpinnings that is sensitive to the research topic and is appropriate for the research study . --- Patient and public involvement In addition to our co-applicant and Study Steering Group FGM survivors, a lay group of type 3 FGM survivors, with different experiences of deinfibulation, has been established. Discussions, where possible, are co-facilitated by the chief investigator and our FGM survivor co-applicant. The survivor group and co-applicant have been actively involved in the design of the study and writing of the study documentation. For example, the group chose the logo and name of the study, have provided critical review of all participant facing materials and helped to co-produce the discussion guides. In addition, they will provide on-going problem solving support, critical review of interview transcripts and the research team's interpretation of data and support dissemination plans. --- Open access EthICS And dISSEMInAtIon data management, monitoring and oversight The University of Birmingham is the nominated sponsor and data controller for the study. Data management and storage will be subject to the UK Data Protection Act 2018 and will follow relevant University of Birmingham policy and procedures. Identifiable data will be securely stored and then safely destroyed within 12 months of publication of the main results of the study. Anonymised data will be securely stored for a minimum of 10 years after the publication of the main study results. An independently chaired SSG has been convened to oversee the study and includes an FGM survivor, academics, FGM-specialist HCPs and third sector representatives. The SSG has agreed the study protocol and will agree any subsequent amendments. This group will monitor adherence to protocol and participant safety and will ensure that the study runs in accordance with the principles of good clinical practice and relevant regulations. --- outputs and dissemination Dissemination is likely to focus on: the findings of the qualitative research with FGM survivors and their male partners; the qualitative research with HCPs working with FGM survivors and their families and the overarching policy and practice implications and recommendations of the research. The study final report will be available through the National Institute for Health Research website, published in open access peer-reviewed journals and presented at relevant conferences and events. --- Ethics approval The study has received a favourable opinion from the North West-Greater Manchester East Research Ethics Committee and approval from the Health Research Authority. Voluntary informed consent to participate and be audio-recorded will be gained from all participants. Provenance and peer review Not commissioned; externally peer reviewed. open access This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https:// creativecommons. org/ licenses/ by/ 4. 0/. orCId id Laura Jones http:// orcid. org/ 0000-0002-4018-3855 rEFErEnCES

► This is one of the largest qualitative studies in the UK to explore views on the timing of deinfibulation and NHS female genital mutilation (FGM) care provision with a range of key stakeholders including survivors, male partners and healthcare professionals (HCPs). ► This cross-cultural and cross-language study will allow us to hear narratives from a range of FGMaffected communities across the UK. ► This qualitative study is informed by the Sound of Silence conceptual framework. ► The study is supported by an active FGM survivor group, a charity partner, a number of NHS trusts and third sector organisations. ► Recruitment, data collection and interpretation may be challenging due to the cross-cultural and cross-language nature of the study.

Introduction It is morning and children are arriving at kindergarten. Carriages are lined up in the corner of a "play kitchen". Two-year-old Matti arrives at kindergarten. He wanders around, observes the children arriving, and starts playing in the play kitchen. Nearby, other children are reading books with the teacher. After a moment, Matti grabs a wooden carriage and starts moving around the kindergarten . This small story of everyday life in early childhood education above invites the reader to follow what happens to a child with a doll carriage from the viewpoint of child participation. The research material comprises video recordings obtained over two days at a Finnish kindergarten. Our research is inspired by Löw's and Massey's theories of space and new materialist aspects . Drawing from these theoretical ideas, we approach child participation as a relational, rather than individual, phenomenon. We understand child participation in ECE as a space that emerges continuously in various combinations of social, material and cultural relations. Our interest is not only in people's actions, but also 'in the diverse weave mass between people, material and structures' . Based on recent narrative childhood studies , we employ a holistic view that emphasizes the active and embodied nature of young children's narration. We consider children as already complete and being, and also becoming with their environment . Along similar lines, Sumsion and colleagues have described how children are doing space in their everyday activities by being, doing, and interacting with their human and non-human environments . Further, the material environment and artefacts play a vital role in children's lives . Based on this notion, in our study, the doll carriage represents the lens through which we view how children do space of participation in the ECE setting. Our aim is to deepen understanding of child participation in ECE by exploring child participation as a space in a child's everyday life. The role of the teacher is significant for child participation as a space, and therefore, further exploration of child participation from the viewpoint of children's other relationships in ECE is important . Here, we focus on children's relations with their environment by exploring the small stories of children under three years of age, guided by the research question 'How do children do space of participation with material items of everyday life in ECE?'. --- Child Participation as a Space in ECE Child participation has been a topical theme in societal and educational discussions internationally for decades . Similar to some other countries, child participation is a core principle of the ECE curriculum in Finland . However, despite its topicality, child participation often remains at the level of rhetoric in ECE practices , and research on participation from the child's perspective has been scarce . Based on Löw's and Massey's theories, we approach child participation as a space that emerges in various relationships within ECE. The space is open-ended and always in process, and there exist, simultaneously, multiple spaces . Human action is vital for the creation of a space, but relationships are not limited to humans; rather, spaces emerge in the web of social, material and cultural relationships . Further, the space is shaped through everyday practices . When people are doing space, they organize different elements, such as social relations, material items, places, cultural rules and routines, values, and curriculum, both consciously and unconsciously . Similarly, children have an influence in and are influenced by their environment . Children do space of participation through 'entangled flows between people and things' in everyday activities, such as playing and communicating with peers , and they try actively to control their lives and share this control among peers . The space can be seen as particular moments of relations . The space of child participation can take many forms, and children experience the opportunities and constrains for participation from their own positions . Therefore, a child's own sense of participation is crucial . The everyday life of ECE can provide favourable spaces for child participation, for example, by responding to children's Fig. 1 The carriages initiatives, enabling opportunities for agency and influencing in their environment, and enhancing their engagement in joint action , which are regarded as crucial cues for child participation in this study. However, although children have the potential for active agency, they are vulnerable, dependent on adults and need protection . Therefore, ECE is also embedded with values and ideologies that frame the children's doing space of participation . Moreover, teachers affect the space, even if they are not present physically, for example, through material organizing and pedagogical planning . Similar to adults, children accumulate understanding of the rules and resources available in ECE . In addition, materiality plays a significant role in children's lives , as children seek to establish relationships not only with the people around them, but also with material and cultural objects . --- Methodology In recent narrative studies with young children , children's narratives have been approached from a holistic viewpoint, where embodiment and multimodality in children's narration are emphasized. Accordingly, our study is based on the understanding of children making sense of the world in a continuous process through narratives . Juutinen describes 'children's narration as something that, in a sense, is already "ready", not something that is going to develop under adult guidance ' . This is an alternative approach to considering children's narratives from the predominant perspectives of developmental psychology or children`s language development . Accordingly, we do not search for children's narratives that are produced for research purposes and have a linear and logical form; rather, we assume that children narrate spontaneously through their everyday activities, play, bodies, and gestures . We apply the concept of 'small story', which represents an interactive everyday moment where children's experiences, emotions and motivations entangle in multiple ways . Short, even minimal, moments can include rich narration . Thus, listening to children's multimodal narration requires researchers to go beyond children's verbal communication and settle into relationships with participants . We rely on Viljamaa's notion of dialogical re-telling, which goes beyond the idea of outsider researchers analysing narratives as research material. Rather, the idea of narrative penetrates the entire research process, which is based on a dialogue between researchers, participants' narratives, and previous research literature. Thus, researchers not only listen to children's narratives, but also engage in narration themselves. --- Research Context, Participants, and Research Material This study was conducted at a Finnish open ECE centre. Open ECE, which is regulated by national policies, differs from full-day ECE with regard to day length, as children do not have lunch or nap time. The study group included eight children and two teachers with a group assistant. Enni-Mari participated in the study, first, as a teacher and, later, as a researcher. The children gathered 2 times a week for 3 h each. Data collection occurred in early autumn, and some guardians were present with their children to familiarize themselves with the kindergarten. A two-year-old child named Matti is featured in this article: It was his first time in the ECE setting and he was not familiar with its people, material environment or practices. In the reporting of this research, we have used pseudonyms to protect the identity of the children. The research material includes video recordings of the group gathered over 2 days from 1 week that were part of a larger set of recordings previously used by other researchers, including Enni-Mari and Satu. We have not influenced which situations have been recorded, because the material was gathered by our colleague as part of a previous research project. Thus, some important situations may not have been captured because they were recorded at our colleague's discretion . The research process posed several ethical questions , and we considered questions about narrative childhood studies, our positions, power relations, and our views of the child. Our guiding ethical principle has been to respect the rights of children and other participants . Consent for the research was provided by the concerned municipality, ECE staff, children, and their guardians. The children's guardians were informed on many occasions about the research, for instance, during parent's evenings. The researchers were also available for discussions when guardians arrived with their children. The teachers and researchers co-operatively opened the research process up to children, for instance, by introducing the video cameras and discussing the study. Throughout our process, we aimed to be sensitive, and continuously assessed whether children wanted to share their life. --- 'In-Between-Out': Different Angles of Entry to the Analysis Our analysis involved multi-layered dialogical interpretation by the three authors, who have different relations to the research material: Enni-Mari had an inside position as a teacher of the children's group; Anna-Maija looked at the material from an outsider's position; Satu had a position somewhere between the two, as she knew the children and had utilized video recordings from this setting in her earlier studies. Due to COVID-19 restrictions, we mainly used Microsoft Teams for our analysis, with which we could record our analytical discussions and produce text. Discussion recordings emerged as an important in-between space where we were able to reflect on our analytical discussions. The multifaceted and dynamic nature of child participation challenged us throughout the research process and required us to come to a consensus about what we are trying to capture. Relying on our theoretical understanding of child participation, we zoomed our focus on children's multiple relationships with peers, adults and their ECE environment. In these relationships, children influence their everyday environment: they choose how and where to act, and with whom and what material. In addition, children make, receive and respond to initiatives with peers in multiple ways. We interpreted these moments as potential spaces of child participation . Our process started with Enni-Mari, as the first author, watching all the video recordings to get an overview and making notes on interesting moments from the viewpoint of child participation. From this first viewing, Enni-Mari noted that children often 'acted with' the carriages. Based on her notes, we held the first analytical discussion and decided to focus on the children with the carriages. With this lens, Enni-Mari watched the videos and took notes again. We continued our discussion with Enni-Mari's notes, most of which concerned moments involving Matti. We decided to continue by further analysing moments in which Matti was acting with the carriage. However, other children were involved in many situations with Matti and displayed similar patterns of behaviour in our material. Thus, our findings are not only about Matti but reflect the group of children more broadly . At this point, Enni-Mari wrote small stories of the chosen moments. In our discussions, Enni-Mari shared her memories as a teacher, and Satu shared her memories as a researcher. Anna-Maija, in turn, asked questions that were meaningful for the interpretation of the small stories. For instance, her question below reminded other researchers about how Matti had a younger sibling and usually came to kindergarten accompanied by the sibling in the carriage. Enni-Mari: 'I remember the carriages were more of a disturbance. As teachers, we controlled the children's play with those. Actually, the carriages were at some point in a "repair shop" .' Satu: 'There were not many rules, but I remember one; it was not allowed to go to the stage with carriages.' Anna-Maija: 'Is the carriage somehow familiar to Matti? Maybe from home?' Our analysis proceeded as a continuous movement between the research material and theory . We watched the video recordings together and captured moments of the small stories as screenshots. We had a special ethical consideration about the use of images as part of the research reporting and aimed to strengthen the anonymity of the children by drawing the images into sketches . Enni-Mari took a final look at the videos to ensure that the moments were described in detail in the small stories. This was important, because some new observations emerged, for example, the sound of the carriage catching Matti's attention in one story. --- Findings We identified three distinct, though intertwining, perspectives about children doing space of participation with the doll carriage: exploring the conditions for participation, doing space of participation in peer relationships, and shaping the borders of participation as a space. Below, the small stories about children`s acting with the carriage reveal how tightly material items are involved in their doing space of participation. --- Exploring the Conditions for Participation with a Doll Carriage Matti moves with the carriage from the play kitchen to the other side of the kindergarten. He heads to the stall bars, in front of which are placed gymnastic mats. He lifts the carriage from the front bow to get on top of the mats, turns, and leans on the stall bars. Constantly holding onto the carriage, he looks around at the kindergarten, other children and adults. At times, Matti takes a spin and returns to the same place . The doing with the doll carriage provided opportunities for Matti to create a space of participation in a new environment. Matti starts formulating his space of participation with the carriage by familiarizing himself with the social, cultural and material environment . The gymnastic mats seemed to be a significant place for Matti's doing: in several stories, Matti is or finds his way there. Almost the entire kindergarten can be viewed from this place. Thus, by utilizing a favourable and safe spot for observation, Matti aims to understand what happens in the kindergarten. By observing from a distance, Matti can sense his position in relation to others, their activities and the environment as a whole . The small story 2 illustrates how Matti repeats his doing with the carriage. The repetition might deepen Matti's sense of familiarity, trust and security and enable him to find his own means to operate in the space . That is, Matti familiarizes himself with the kindergarten's practices, norms, and prevailing rules related to the use of the carriage and, simultaneously, shapes his own space of participation . The repetition is not a mechanical playback of the prior doing, but rather, is tuned by the contextual cues in the everyday life of ECE . These actions are also crucial for the child's sense of belonging, which emerges from familiarity, ownership and continuity . Through repetition, the doing becomes more predictable and supports child participation, and ownership of the space is possible to emerge. Matti is walking around with the carriage. He crosses the dinosaurs on the floor, turns, and crosses them again. He wanders in different places with the carriage . The carriage, as a moving matter, provides Matti with expanded opportunities to explore the different physical places of the kindergarten. Matti's active, autonomous doing fits the expectations of a 'proper kindergarten child' . The story also illustrates the embodiment in a child's doing with the material . Matti pauses in his journey to touch various material objects; for instance, he presses the keys on the teacher's laptop and moves the mouse so that the screen light turns on. Matti chose the road where he had to work hard, zigzag around the furniture and, sometimes, Fig. 2 On the gymnastic mats even lift the carriage . Matti probes how the strollers of the carriage and the toys on the floor react to each other. The carriage seems like an extension of the child in his doing and significant for his participation. Matti heads to the table, parks the carriage nearby and starts playing with play dough. When he plays, he glances a few times at the carriage. After a while, he grabs the carriage and starts moving again with it. The previous story demonstrates the child's strong relationship with the material: The relationship with the doll carriage extends beyond the 'actual doing' with it, and this is interpretable in Matti's glances at the carriage even when he is occupied with other things. In addition, the significance of the carriage is perceptible when Matti continues his doing with it after finishing other activities. Occasionally, Matti seems to be on the borderline between familiarity or unawareness-belonging to a group or being an outsider or being together with others or alone with the carriage. Through moving, observing, and repetitive doing, Matti creates the rhythm and borders for his space. Matti takes control of the ownership of his space of participation from nearby, but also from a distance, in relation to others. --- Doing Space of Participation with a Doll Carriage in Peer Relationships Matti plays in the play kitchen and raises his head; he seems to hear something. The sound gets louder and closer. It comes from the doll carriage pushed by Niilo. Matti stands up, walks to Niilo and grabs another doll carriage. Fig. 3 Exploring with the carriage In the small story above, Matti's attention is suddenly drawn from his ongoing play to the approaching sound of the carriage. He seems to recognize the sound, probably from his earlier activities with the carriage . This situation is, however, different from the ones discussed earlier as now there is another child, Niilo, pushing the carriage. Niilo's doing with the carriage serves as an invitation for Matti to join: Matti grabs a similar doll carriage, and the children begin to act together with the carriages. It is meaningful that there are at least two similar carriages available for the children, as the material environment and items in educational settings are influential: they both invite and inspire children . Here, the carriages provide a space of participation by enabling the children to construct peer relationships and create joint action. Niilo and Matti start walking with the carriages in the same direction. At times, they walk side by side, and occasionally, they walk in line. At times, they stop and look at each other. They circle the tables and walk over the tiny cars on the floor. First, Niilo leads, and Matti follows. Then, Matti leads, and Niilo follows; they imitate each other's doing. Matti walks ahead to the mats and looks back at Niilo. Niilo looks at Matti but walks past him. Matti stays on the mats for a while and walks in a different direction. Niilo stops in his way, turns around and walks in the same direction with Matti. Again, Matti heads to the mats and looks over his shoulder to where Niilo is. Matti turns around and leans on the stall bars. Niilo spurts to the mats but is soon on his way somewhere else. Matti looks at Niilo, shakes his head, raises his hand and says, 'Bye!' Niilo comes back to Matti. Matti moves a little closer to Niilo; it looks like he makes room for Niilo. Niilo glances at Matti and walks away again. Matti waves to Niilo saying 'Bye!' Matti does not move away from the mats. Niilo turns to Matti; they look at each other. Matti uses his hand to gesture to the spot next to him. Niilo walks to Matti. For a while, they stand next to each other with the carriages on the mats . This small story demonstrates how Matti's and Niilo's joint play emerges with the carriages and highlights the significance of repetitive actions, reciprocity and embodiment in children's doing of participation. The children communicate through their bodies: 'Let's do this with the carriages!' They begin to create common rules and routines for their joint doing by staying close to each other, observing and acting in a similar way . The children's shared space of participation appears, also, in how they walk side by side or in line. Matti also invites Niilo back to the mats by gesturing with his hand. The exchanged gazes and gestures exemplify the embodied communication between children . Despite not having a common spoken language, they have embodied negotiations on how to continue the joint activities . By negotiating, they seem to find together a common rhythm for doing their space of participation. Neither Matti nor Niilo takes power over the other to determine the joint play; rather, both children make and respond to the other's initiatives. Sharing power is regarded as a meaningful aspect of participation . Moreover, the familiar and individually meaningful place for Matti's participation becomes a shared space for Matti and Niilo. While Matti stays on the mats, Niilo repeatedly goes back and forth between the mats and other places. Moreover, Matti's staying on the mats seems to be a call for Niilo to return. When Niilo walks away again, Matti takes a few steps forward but stops at the edge of the mats. Matti observes what Niilo is doing: Niilo puts the carriage aside and runs to start a play with dinosaurs. Matti lifts his hand to Niilo with a 'Bye!' and smiles. Soon, Niilo grabs the carriage and turns to Matti, who goes down on his knees. After taking some steps with the Fig. 4 Children doing shared space of participation carriage, Niilo's attention is drawn to an assistant who brings the 'firefighter' play material for the children. Again, Niilo puts the carriage aside and starts exploring the firefighter material. Matti stays on his knees on the mats and 'knocks' the carriage against the floor. Niilo does not pay any attention to this. Matti stands up and moves back and forth with the carriage. Every time, Matti stops at the edge. After a moment, Tommi, who is also playing with the firefighter material, walks to Matti. Matti goes down on his knees, still holding the carriage. Tommi points a 'water tube' at Matti. The firefighters head to the play kitchen, and Matti cannot see Tommi. Matti walks to the water tube left on the floor. He touches the tube with his foot. Then, he crosses over the tube with the carriage, turns, and crosses over the tube again. Matti tries to take the tube but notices that Tommi is back and pointing at him with the tube again. Matti runs back to the mats and keeps checking if Tommi is still after him. Tommi follows him while pointing with the tube and then goes in a different direction. Matti stays on the mats with the carriage . The former short story portrays children's doing space of participation as a continuous process. While the children's joint activities with the carriages maintain their shared space of participation, Niilo's shifting interest in other material items changes the situation. In line with Nordtømme's study, the children seem to move between different positions regarding the joint play, either changing their focus or trying to protect the ongoing play. It is obvious that Matti would like to continue playing with Niilo, and his appearance expresses disappointment at Niilo's leaving. This is demonstrated when Matti calls for Niilo by knocking the carriages on to the floor. Matti seems to be more committed in doing with the carriages than Niilo, and Niilo's own space of participation has already emerged elsewhere. At this point, Matti appears to be balancing on the border of participation in joint play. In addition, Matti becomes an outsider in relation with the firefighter play. While the other children engage in the shared play, Matti shows both hesitation and interest in joining the play. It is apparent that Matti wants to explore the water tube and even touches it. However, Tommi's actions make him return to the mats with the carriage. It is perceptible that Matti does not take Tommi's physical closeness and his actions as an invitation to a joint play, when Matti moves as if he wants to hide between the carriage and the wall . Even though Matti does not participate in play with peers, the carriages and the familiar place of the mats play a significant role in Matti's sense of ownership, familiarity and security and are regarded as significant indicators of the space . --- Shaping the Borders of Participation as a Space with a Doll Carriage Matti heads with the carriage to the stage. Before getting up, he stops for the blink of an eye and turns his head as if to make sure that no one can see. Matti goes on top of the stage and then back down, turns and heads back up. On top of the stage, he makes a couple of circles with the carriage before coming back down, but slower than the last time. When Matti heads to the stage again, an assistant says 'Matti!' Matti turns the other way. Later, a teacher is playing with other children nearby and notices Matti directing the carriage to the stage. 'Matti come down, thank you!' says the teacher. Matti continues his way in another direction . The doll carriage appeared to be meaningful for shaping the borders of Matti's space of participation. As mentioned before, going on to the stage with the carriages was not permitted. The story above demonstrates not only children's and adults' shared cultural knowledge about the rules , but also how Matti starts testing the borders of his space. Matti's gazing around before walking to the stage illustrates that Matti already knows the rules about playing with the carriage. This is also demonstrated when he changes direction on hearing his name called out by the assistant. The repetitive action in Matti's doing occurs again. Matti seems to find a way to take over his own agency as a border maker . Tommi heads to the stage with the carriage, where Matti is playing. Matti stands up and spreads his hands in front of Tommi while shaking his head. Tommi stops. Matti walks down from the stage and starts running, with Tommi by his side, to get one of the carriages for himself. Together, they start walking with the carriages and head to the stage. 'Matti and Tommi, please. There is plenty of room on the floor', the teacher says. The two children go elsewhere with the carriages . The story illustrates that children test the ECE rules also in peer relationships. First, Matti shows Tommi with an embodied expression what is not acceptable doing with the carriage. Later, the children together head with the carriages to the stage. The small story illustrates the embodied doing and the ease in coordinating the joint actions with the carriages . Additionally, the story uncovers the prevailing contradictions between educator's and children's perspectives: the spaces favourable for children's participation can be interpreted differently from educator's and children's viewpoints. This raises the question of who has the expertise when it comes to child participation. Matti stands with the carriage on the mats. Suddenly, Janne grabs Matti's carriage and pulls it towards himself. Matti tries to push Janne's hand away and moves forward with the carriage. Janne holds on tightly. Matti starts knocking the carriage on to the floor and looks at the adults. Matti moves backwards on the mats with Janne still holding on to the carriage. Janne's mother guides Janne elsewhere. Matti stays and watches Janne leaving . The previous story portrays how Matti defends the borders of his participation as a space. Even young children set boundaries for peers and defended their own space, possessed materials, and their bodies . Here, Matti protested and sought the adults' attention in the conflict situation. Thus, Matti appears to gain more agency with the doll carriage, and losing the carriage reduces his space of participation. That is, Matti and the doll carriage are together more than separate . --- Concluding Discussion The present research provides remarkable insights into children doing participation as a space with the material items in ECE by analysing the small stories of everyday life in ECE. We found that children do their participation with the doll carriage by exploring the conditions for participation, doing space in peer relationships, and shaping the borders of the space. Thus, children's doing space of participation is a dynamic process between children, educators, material items, and cultural rules and routines in ECE . Neither the spatial nor the narrative approach is common in ECE studies on child participation, so our study makes an important theoretical and methodological contribution to the topic. Our research strategy was fruitful, as systematically following children's actions with the doll carriages opened insights into the entangled relations meaningful for children's space of participation. While previous research considers child participation primarily as a social phenomenon, our study shows how tightly the material environment and culture of the ECE setting are involved with the social space in a child's doing of participation. The study, thus, contributes to the theoretical understanding of child participation by uncovering the three-fold nature of the space, that is, the social, material and cultural dimensions. The findings illustrate how material items, such as doll carriages, can provide children with a means for familiarizing themselves with the ECE environment, creating peer relationships, joint play and shared spaces of participation. The material item also played a significant role for children when exploring, testing, expanding, and defending the borders of their participation. The present findings provide empirical evidence about how Matti and other children do space of participation. Children's doing of participation emerges as a tension-filled phenomenon from the viewpoint of children's active agency. On the one hand, children play an active role in doing participation by actively exploring the material and cultural environment of the ECE setting and keenly attempting to make connections with their peers. Embodiment, continuity, similarity and repetitive actions appear to be crucial for young children's doing space of participation. Hence, children's doing participation as a space exceeds the language Fig. 7 Testing the rules with a peer Fig. 8 Defending the borders boundaries . On the other hand, children cannot freely do their participation, as their doing is partly dependent on other people's actions and the cultural rules of the setting, which have also been highlighted in previous research . Thus, doing space of participation requires children to balance between the active doing of participation and adapting to the expectations of other people and the environment. For example, the ECE rules and curricula can enable, but they also restrict children's active doing of participation. Prior ECE research has largely focused on adults' perspective of child participation and included children who can linguistically express themselves. Our study contributes to the field of ECE by bringing in perspectives not only on children's multiple ways of narration but also on their ways of doing participation. Similar to the findings of narrative childhood studies, children's embodied ways of expressing themselves emerged central, while their verbal narration remained secondary. Our findings might help educators recognize the important aspects of child participation in ECE. Enhancing every child's sense of participation requires that the educator has awareness not only about the relational nature of child participation but also about what is meaningful for each child individually. Enni-Mari's memory of not recognizing the meaningfulness of a carriage for child participation is an excellent example of everyday moments that could be overlooked by an educator. In line with previous studies , we suggest that the first step in developing child participation in ECE is to acknowledge and listen to children's embodied narration. Secondly, it is necessary to pay increased attention to the matters and relationships meaningful for children. Recognizing these emerging childlooking spaces could open new doors for enhancing child participation and developing practices in educators' work. We humbly would like to acknowledge the limitations of our interpretations. For one, we are unable to label analysed moments as 'here is where the space of participation truly stands'; rather, we understand them as fruitful potentials for participation to flourish . Nonetheless, the dialogue between us researchers enriched the research process and strengthened our study's credibility. We found the different perspectives on the interpretation of child participation as fruitful; together, we constructed new knowledge that we might not have been able to do individually. We are aware that our personal histories as ECE teachers and researchers affect in many ways how we read the material . Nonetheless, we can conclude that the everyday action in ECE already has potential for young children to do spaces of participation, and the spatial approach can provide tools for identifying these everyday meanings for children. However, further research is needed for deepening understanding of child participation. In particular, there is a need to study child participation in ECE practices from a wider relational point of view, for instance, by focusing on the societal, cultural, or discursive dimensions in doing of child participation as a space. --- Author Contributions All authors contributed to the study conception and design. Data collection were performed by EMY and SK. All authors contributed to the analysis and writing of the manuscript. However, EM was most responsible for the writing process and finalizing the manuscript. All authors read and approved the final manuscript. Funding Open Access funding provided by University of Oulu including Oulu University Hospital. No funds, grants, or other support was received. ---

This article focuses on child participation in early childhood education (ECE). The theoretical basis for this study is a relational approach that highlights the significance of spatiality and materiality. Drawing from these theoretical perspectives, child participation is understood as a space that children actively do in various combinations of social, material and cultural relationships. The study relies on a relational understanding of the child as not only being, but also becoming by doing space within one's environment. The methodological framework draws on narrative childhood studies that emphasize the multimodality of children's narration. The research material consists of small stories from a Finnish ECE centre. The analysis focused on children under 3 years of age doing space of participation with material items in ECE. The study shows that space of child participation is dynamic in the everyday relationships. The findings illustrate children's doing space of participation as a tension-filled phenomenon which requires children to balance between the active agency and adapting to the expectations of the environment. The findings uncover the three-fold nature of the space-namely, the social, material and cultural dimensions-that is meaningful for child participation. The doll carriage as a material item played a crucial role in children's doing space of participation: it provided opportunities for children to formulate the ownership of the space, to construct peer relationships, and to shape and defend the borders of one's space. The study contributes theoretically and methodologically to ECE research by opening new perspectives on the participation of young children.

Introduction In this paper we study the impact of European intermarriage 1 , transnational social relations and experiences on the emergence of European identity on the basis of original data from a survey among couples in Zurich, Switzerland. Sociologists have only recently begun to take a serious interest in processes of Europeanization and the formation of a European society . Hitherto, sociology has tended to regard society as an all-embracing and self-sufficient social system within the territory of a nation state. This assumption has led to the problem of methodological nationalism and the neglect of transnational forms of social integration and sociation . This problem is heavily discussed in migration research on the one hand and in more general research about internationalization and transnational relations on the other hand. The dominant perspective in migration research focuses on the process of migration from one national society to another and the ensuing dynamics of integration into the host society, usually conceived of as taking place in several dimensions, such as acculturation , structural assimilation , social integration and identificational assimilation . This nation-based paradigm in migration research has been challenged by the transnationalism concept. Several authors have argued that contemporary migration is not a one-way process of migration from a country of origin to a country of destination and a subsequent process of integration . Instead, migration today is depicted as a process characterized by several steps, including travelling to the destination country, commuting and sometimes finally returning to the country of origin. These flows of people are taking place in transnational social spaces without reference to political borders, often connecting geographically distant regions . Thus, it is claimed, integration into the country of destination is replaced by integration into transnational social fields, which connect the countries of origin and destination and are characterized by different forms of economic, political and social transactions . Subsequent empirical research has shown, however, that transnational relations are not as prevalent among migrants as originally envisaged, and that the national society of the destination country is still the most important framework for analyzing processes of integration . Prior to this discussion in migration research, empirical studies on more general processes of internationalization and transnationalization have been conducted in the social sciences . They started from the assumption that national societies exhibit only a relative degree of closure with respect to different forms of interaction and exchange, be they economic, political or social . Spatial mobility, social networks, transactions and organizations have always crossed the borders of societies and nation states, and the necessity to take this into account has become more pressing with the onset of Europeanization and globalization processes . Current social science research does not only study Europeanization on an institutional level, but also on the level of systemic exchanges , and social exchanges like migration, social networks and binational marriages. Research is also undertaken on the emergence of a European outlook and identity . In this paper, we build on these discussions by studying the relationship between intermarriage, transnational social relations and experiences on the one hand and European identity on the other hand. We rely on Recchi's structuralist explanation, which focuses on European social relations and experiences as main determinants of European identification. Kuhn and Delhey have shown that it is the interpersonal exchanges between persons from different European nations in sociable gatherings, in friendships and in marriages that are decisive for the development of trust, identity and pro-European attitudes. So along with the effect of intermarriage, we study the degree to which transnational social relations and experiences contribute to the more encompassing, supranational identity of feeling European for people in different types of partnerships. With our data, we are able to compare respondents from different origins in either mono-national or binational partnerships . In our analyses we will firstly analyze their degree of identification with Europe. Secondly, we will investigate how people construct the meaning of Europe which should help us to detect whether individuals in the different types of partnerships differ with respect to the meaning they attach to Europe. Thirdly, we will analyze which factors contribute to a European identity and which might explain existing differences between the groups. --- Theory "Identity" is a contested concept in the social sciences . Identity can refer to both what distinguishes us from others and what we share with others. The identity of every individual is made up of aspects that make her different and of aspects that he shares with others . Brubaker and Cooper distinguish between the mere membership in a group, the characteristics of a group and the individual degree of identification. This is very similar to the discussion by Abdelal et al. , who differentiate between the content of a collective identity and its degree of contestation. The first refers to the meanings and characteristics attached to the collective and its collective identity, the second refers to the degree of support a certain notion of identity has in the group. In our paper we study both the degree of identification with Europe and the characteristics and meanings that are attached to it. European identity is a special case of collective identity, since although it can denote identification with a group , in contrast to other group identities like gender or class it is mainly a territorial identification. In general, social or collective identities have to be conceptualized as plural phenomena, since people usually belong to different categories or groups , to which they feel a certain degree of attachment and identification. However, in the case of territorial identification, they are not only plural but are often nested within each other. Being an ardent Bavarian does not rule out being a strong German patriot or supporter of Europe. The latter may simply encompass the former . However, the extent to which this is true usually depends on the meaning of the respective type of identity. If for example the meaning of national identity is constructed as being in opposition to regional identity, there should be a negative correlation between the degree of national and of regional identification. The literature on the meaning of European identity often assumes a distinction between a civic and an ethnic/cultural dimension . The focus of the ethnic/cultural dimension is on common values, language, religion and mythsa kind of common national or European culturewhereas the civic concept emphasizes the common rights and duties that go along with the citizenship of a certain polity. Since the respective meaning of European identity may be important for the attachment to Europe in relation to the identification with the nation, we take it into account in our study. Especially the more civic aspects of a European identity may be present to a higher degree for EU-citizens in contrast to Europeans from non-EU countries like Switzerland. Since territorial identities can be weak or strong on different levels of territorial identification, in our empirical analysis we follow Delhey et al. in using a relational measurement of identification. For instance, to justify the thesis of a rise in European identification, not only does the absolute level of European identification have to increase but so too does the level of European identification in relation to the level of national identification. In addition to the absolute level of Europeanization, Delhey et al. . According to the Eurobarometer survey little more than 10% of respondents only feel European, whereas roughly 50% feel both European and national. Thus, only a small fraction of Europeans exhibit a mainly European identity. How can we explain the emergence of an individual identification with Europe? Based on survey evidence, we know that highly educated persons, males, professionals and managers have a stronger European orientation than other social groups . There are two main explanations for these results: firstly, these groups may profit more than others from the emergence of a European transnational field, which offers labor market and educational opportunities, the chance to travel and move freely within Europe and to make friends and acquaintances all over Europe . Thus, this is mainly an interestbased explanation. Fligstein has forcefully put forward this thesis, but because of data limitations he was only able to show that the groups that exhibit a stronger European identification are also the groups which travel more, know more languages and move more within Europe. This brings us to the second explanation, which Recchi calls the structuralist theory of identity formation . This structuralist perspective takes up the Deutschian thesis that the emergence of social relations in a society leads to more communication and thus to a feeling of belonging to a community . This thesis is clearly applicable to the emerging European field, where there is unquestionably a variety of social relations and experiences that are likely to facilitate the formation of a European identity . Based on statistical results from an analysis of survey data, Díez Medrano et al. claim that such transnational relations and experiences do not explain the socio-demographic distribution of European identity, whereas Recchi argues that such relations and experiences are especially important for explaining the emergence of such identities among less well educated persons. We are able to contribute to this debate on the basis of high-quality data, which covers a broad array of European experiences and social relations among our respondents. We expect especially strong impacts of intermarriage on the formation of a European identity. Intermarriage is usually taken as a key indicator of social integration of immigrants into a host society, since marriage is a strong, intimate and durable social relation, the impact of which is felt beyond the couple because it brings together the family and friends of both partners. We depart from and extend the classical migration view of intermarriage and identification as core indicators of social integration within a national society and study them instead as indicators of integration into transnational fields, in this case the European society . Moreover, intermarriage can be seen as a source of solidarity and an indicator of the social distance between groups . We assume that partnerships between persons from different European countries have a strong impact on the enhancement of European identity. Partners, as significant others, contribute to how individuals identify and perceive themselves and others. The idea that identities emerge as the result of social interactions and social relations is a very basic assumption of identity theories in sociology . This does not only apply to intermarriage, but also to other social relations like friendships or family relationships, thus theoretically supporting Recchi's structuralist perspective. However, this also implies that different relationship patterns lead to different identifications depending on the origin of the partners. Thus, for a Swiss married to a person from another European country, we expect a higher degree of identification with Europe and the world, however, for a Swiss married to a non-European person we just expect a higher degree of identification with the world. Identity theory generally expects a homology between the pattern of relations and experiences on the one hand and the pattern of identifications on the other hand. Building on this discussion, we will study the following hypotheses on European identification. Since our survey mainly contains European respondents from EU countries, we had to exclude the few Europeans from non-EU countries, and thus refer in the analysis only to persons from EU countries. For the reasons explicated above, we expect intermarriage to have a positive effect on identifying as European and as world citizen. H 1 : People in intermarriages, i.e. with partners from another country, exhibit a higher degree of identification with supranational entities than people with partners from the same countries as themselves. H 1a : More specifically, Swiss with a partner from an EU country exhibit a higher degree of --- European identity than Swiss with Swiss partners or partners from non-European countries. H 1b : Swiss with a partner from a non-European country exhibit a lower degree of European identity than Swiss with Swiss partners or partners from European countries. The effect of intermarriage should also apply to people who grew up in intermarriages. Again, we hypothesize that these individuals tend to attach more relevance to a supranational category, as their identity will have been shaped by the experience of having two nationally diverse parents. H 1c : Individuals whose parents are intermarried , exhibit a higher degree of a European identity. Based on identity theory we will take account of other, in comparison to intermarriage possibly weaker, forms of integration into transnational social fields. Friendship relations, for example, are normally another indicator of social integration in migration research, therefore we also hypothesize that a network of friends within Europe strengthens European identity via the same mechanisms as intermarriage. At the same time, people can have social relations with non-European countries. Hereas with a non-European partnerwe expect a reverse effect on European identification vis-à-vis the global reference frame. Another negative effect on European identity should result from close friendship networks in Switzerland. The more someone is embedded in social relations with people from the same country of origin, the more important the national reference frame is likely to remain. In order to take account of this, we also analyze the effect of having friends in the country of residence. H 2c : The higher the share of friends from the same country of origin, the lower is the identification as European with respect to national openness and relative Europeanization. The hypotheses are formulated without referring to individuals of one particular nationality. We assume that transnational experiences should have the same effect on Europeanization for all people living in Switzerland. Again, reverse effects might be possible due to trips to, or stays in non-European countries. Experiences outside of Europe should foster a broader supranational identity and reduce the effect on European identification. Particularly more extended stays abroad, when reflected in actual contact with the people and culture abroad, should have the most pronounced effect. But also travelling to various world regions should promote an inclusive reference frame for identity formation . --- Data Our empirical analysis is based on data from an online survey which ran 2012 in Zurich as part of the project "Toward a European Society: Single Market, Binational Marriages, and Social Group Formation in Europe " . The inquiry was addressed to persons in mono-and binational partnerships . The sample included individuals living together with their partners and holding citizenships from Switzerland, the EU27 countries or other European and non-European countries. The couples consisted either of two Swiss partners, one Swiss partner and a partner from one of the countries of the aforementioned foreign groups, or two EU27 citizens from different countries. The basic sample was drawn randomly from several predefined strata of persons from the population register of the city of Zurich. Within each couple, one partner was chosen randomly. All selected persons were contacted by mail in German and English and invited to participate in the online survey . At increasing time intervals we sent three reminders to the sampled individuals. By following this procedure we were able to achieve a response rate of about 40 percent . In order to test the effect of intermarriage on European identification, we run our analyses on couples in different partnership combinations. The country of birth of the respondent and of his/her partner is the main criterion for assignment to one of the groups as it does not change throughout the life course . We differentiate six groups: Swiss-born with a Swiss-born partner , Swiss-born with a partner from one of the EU27 countries , Swiss-born with a partner born in a non-European country , individuals born in an EU27 country with a partner either born in another EU27 country or in Switzerland , and individuals born in a non-European country with a Swiss-born partner . Furthermore, we restrict the analyses to those respondents who possess the citizenship of their country of birth and indicate it as their most important citizenship in case of double citizenship. This decision was taken because for non-Swiss we only surveyed identification with the country of citizenship, not with the country of birth. As the country of birth did not always correspond with the citizenship, the procedure led to the exclusion of almost 13% of the cases. The item on European identification was embedded in a battery of corresponding items. On a scale from zero to 10 , individuals were asked to indicate how much they would agree or disagree with certain statements on identification, e.g. "I feel European." The list of statements ranged from smaller political and regional units to larger ones. We included "I feel Swiss" as well as the respective national category for foreigners and individuals with double citizenship. Besides feeling European, the item battery also comprised feeling as a citizen of the world. These items are used to construct the dependent variables, i.e. the four indices of Europeanization applied to identification as proposed by Delhey et al. . Furthermore, respondents were asked to indicate how much they would associate various terms with Europe. The list of terms reads as follows: common history, geographic region/continent, European Union, Christian religion, cultural diversity, dominance of economic interests, political cooperation, tolerance, the welfare state/a strong social network, separation of church and state, and loss of our own national identity. The scale ranges from one to seven . The terms were used for measuring the meanings our respondents associate with Europe. The explanatory variables were constructed as follows: born in intermarriage is coded one if the countries of birth of the respondent's parents vary, zero otherwise. The two variables on friendship networks abroad have three categories. They are coded one if respondents indicate that they have friends a) in a country belonging to the European Union or b) outside the EU and coded two if they have regular personal contact with them , otherwise zero. The share of friends from the same country of birth represents the percentage of friends born in the same country among the closest friends in Switzerland . As we partly expect different effects for the different nationality groups, a net interaction effect is used to account for the citizenship of the respondent . The number of languages refers to all languages respondents reported as those they speak fluently. A value of "1" indicates that a person only speaks his mother tongue and no foreign languages. The maximum of the variable is five or more languages. The two variables on short trips refer to short term mobility lasting a minimum of one overnight stay and up to three months. The number of European countries visited in short trips ranges from zero to 34. The number of continents visited in short trips is restricted to six, including Europe, North America, South America, Africa, Oceania, and Asia. Stays abroad refer to long term mobility, i.e. stays that lasted at least three months. Depending on whether it was in a European or non-European country, the stay is counted within number of stays in European countries or number of stays in non-European countries. Education is a categorical variable with four values. As the majority of the respondents are highly educated, we distinguish between secondary education or less, postsecondary, nontertiary education, and two levels of tertiary education. Higher tertiary education applies to persons holding a PhD or an equivalent degree. The size of our sample in the analyses is 1,918. Different case numbers result from list wise deletion of missing values in the dependent variable. --- Results Table 1 shows an overview of the average absolute level of identification with the different reference frames: Europe, the nation and the world. We find that in the sample the level of feeling as a European is relatively high for all individuals born in Europe. European migrants show the highest level of identification with Europe, even more so when their partner is from a different EU27 country. Respondents born in non-European countries show a markedly lower level of European identification. However, Swiss with a Swiss partner do not significantly differ from Swiss with a partner from the EU, although the latter show a slightly higher level of European identity. As far as the nation is concerned, we find the highest level of identification within Swiss-Swiss couples. But here too, the difference between these and Swiss with a European partner is negligible. With regard to identification as a citizen of the world, we do find rather high mean values between 6 and 7 and no significant variation over the groups. In sum, we find that Swiss and non-Europeans rate identification with their own nation higher than identification with Europe, while European migrants put more emphasis on identification with Europe. -TABLE 1 ABOUT HERE -Table 2 presents the means of the four indices of European identification for the various partnership combinations. With respect to national openness we find the highest salience of Europe in relation to the nation for EU citizens. However, there is no significant difference between those in partnerships with other EU citizens and those with a Swiss partner, the exception being the absolute level of Europeanization. For the Swiss, we do find a slightly significant effect of European intermarriage only in respect to national openness. But even intermarried Swiss rate their national identity more highly than their European identity . Furthermore, no difference is discernible as regards European closure and relative Europeanization. So far, the data hardly supports hypothesis 1a. Yet, for hypothesis 1b, we find partial empirical evidence: Swiss intermarried with non-Europeans score markedly lower on European closure than Swiss who do not have a partner from another country. In spite of the fact that we did not ask for identification with the European Union but with Europe, EU citizens show a significantly higher degree of identification with Europe than Swiss . --- -TABLE 2 ABOUT HERE - The relative measures of European identity demonstrate that Europe is a more meaningful category for EU citizens than the respective national category. On average, the EU citizens in our sample have a value higher than 0.5 on national openness. It seems that, at least for EU citizens who live in a European country other than their country of birth, national identity loses relevance as the ultimate point of reference, which points to the importance of migration experiences. Nevertheless, the high mean of European closure demonstrates that the boundary against "the other" remains important. It should be noted that the vast majority of EU citizens in our sample were born in Germany . They are followed by Italians , Spaniards , and French, Austrians and Britons . This implies that many of the European migrants in our sample come from countries that are founding members of today`s European Union and as such probably more familiar with the European reference frame and the idea of a united Europe. However, we find that a non-European partner has a pronounced weakening effect on European closure and the absolute level of Europeanization; so far, we can conclude that the foreign partner of a Swiss does not necessarily enhance or reduce identification with Europe. Still, feeling as a European might mean different things to different people, depending on what their concept of Europe is. We will therefore take a look at the meaning of Europe for persons in the different types of partnerships. We exclude the non-Europeans from the analyses because they are not our primary interest and their case numbers are on the low side. As described above, the survey included 11 questions on the meaning of Europe. Exploiting that some related meanings go together more often than others a factor model with just six latent variables proved sufficient to represent the data structure . The dimensions can be subsumed under the headings: "Egalitarian Values," "Common Ground," "Politics," "Christianity," "Loss of National Identity," and "Dominance of Economic Interests". Figure 1 shows the latent mean values of the factors, i.e. the levels on those factors relative to Swiss-Swiss couples. Positive deviations indicate that a dimension is more important for a group compared to Swiss with a Swiss partner, and negative deviations indicate that a dimension is less important. -FIGURE 1 ABOUT HERE -With regard to the meaning of Europe, too, we do not find a significant difference between Swiss that are intermarried and those that are not. EU citizens living in Switzerland have a different image of Europe. Compared to the Swiss, they are more strongly geared toward "Common Ground," and put more emphasis on the political dimension of Europe. This is even stronger for EU citizens with a partner from a different EU country. Europe as a threat to national identity is less of an issue for EU respondents than for Swiss. Though both Swiss with an EU partner and Swiss with a non-European partner do not differ significantly from the Swiss-Swiss couples, those two groups do differ significantly from each other, i.e. Swiss with a partner from outside Europe are more afraid that Europe means a loss of national identity than Swiss with an EU partner. If we compare the correlations of the dimensions of meaning with the four indices of European identification , we find rather similar patterns for all indices except for European closure . Whereas "egalitarian values," "common ground," and "political cooperation" work as attractors for European identification, loss of national identity is a clear repeller. The dimensions "dominance of economic interests" and "Christianity" are neutral with respect to the level of European identity. This analysis shows quite clearly that identification with Europe depends strongly on the meanings attached to the nested national and European levels and their interrelation . Our next step is to test the explanatory hypotheses. Table 3 presents linear regression models for national openness and European closure . Apart from the main effects of type of partnership, models 1 and 3 include only the socio-economic control variables. The variables on transnational networks and experiences are added in models 2 and 4. For national openness we firstly find a positive and slightly significant effect for Swiss with an EU partner controlling for socio-demographic variables . We also find empirical evidence for hypothesis 1c: having intermarried parents enhances national openness. The social networks do not seem to be important for identification with Europe in relation to the nation . The variables on transnational experience show mixed results. The number of languages has the predicted effect: the more languages someone speaks, the more salient becomes the European reference frame relative to the national . For the most part, the predictions concerning the influence of mobility on European identity do not seem to hold. There is, however, one exception: every stay abroad in another European country increases the level of national openness . -TABLE 3 ABOUT HERE -For the explanation of European closure other variables prove to be important. Here, we do find an effect of the social networks, at least with respect to friends outside the EU: respondents who have regular personal contacts to friends in non-EU countries show a significantly lower degree of European identification in relation to the global reference frame . Similarly, the number of continents visited and the number of stays in non-European countries both have a considerable negative effect on European closure, indicating that global experiences and networks lead to more global identifications. An interesting result is the significant, positive effect for the share of friends of the same country for EU citizens. The more the circle of closest friends consists of same nationals, the higher is the distinction of Europe vis-à-vis the world. Table 4 shows the results for relative Europeanization and absolute Europeanization . Again, apart from the main effects of type of partnership, models 5 and 7 include only the socio-economic control variables, while the variables on transnational experience are added in models 6 and 8. With respect to relative Europeanization, we find only two of the hypothesized effects. While the effects all point in the expected direction, the number of stays in European countries turns out to have a small, though significant effect . For absolute Europeanization, i.e. the only non-relational measure of identification, the total of languages and the number of European countries visited in short trips also have a positive effect. With respect to education we find in general rather small effects, with less well educated persons exhibiting a somewhat lower degree of European identification . The inclusion of variables, which measure transnational relations and experiences, decreases the impact of education somewhat, but they do not fully mediate the relevance of education. --- -TABLE 4 ABOUT HERE - The models with controls for the socio-economic variables reveal a slight difference between Swiss-Swiss couples and Swiss with a European partner, indicating that these marital unions are socially selective, such that the effect was not visible in tables 1 and 2 . What we also find is that nationals from countries that are members of the EU show markedly higher levels of identification as Europeans. The reason for this could, of course, be that they associate Europe much more strongly with the EU than do the Swiss, as demonstrated in the factor analysis . Furthermore, EU citizens, unlike Swiss citizens, are much less likely to construct European identity as conflicting with national identity. Transnational networks and experiences have much less influence on the four indices of European identity than expected. A plausible explanation is that these variables have been conflated with the types of partnerships, resulting in some issues of endogeneity. --- Summary and Discussion We analyzed the impact of European intermarriage and transnational social relations and experiences on European identity. It is an empirically well-founded result from migration research that intermarriage is an especially important indicator of social integration and often also leads to identificative integration . We studied this idea from a transnational perspective and discussed whether European intermarriage, i.e. binational unions between individuals from different European countries, also leads to a stronger identification with Europe. According to the structuralist explanation of identification and identity theory , this should be expected on the grounds that intermarriage is an especially strong, durable and intimate social relation. In addition, the theories also predict that different patterns of social relations and experiences predict different types of identification. We based our study on survey data from Zurich, Switzerland. The structure of the meaning of Europe was similar for Swiss and EU citizens and thus for individuals in different partnerships. However, the importance of the dimensions differed between the groups. Respondents from EU countries placed more emphasis on common ground and politics in contrast to Swiss respondents; this was even stronger for EU citizens with a partner from a different EU country. Furthermore, as suggested in the discussion about nested identities , Swiss respondents constructed the meaning of Europe as more in conflict with national identity, whereas EU citizens saw less of a conflict here and thus identified more strongly with Europe. This was also evident in our statistical analysis, where EU citizens showed much higher levels of identification with Europe than all other groups. In contrast to the considerable importance of category membership our measures of transnational social relations and experiences had only a small number of significant statistical effects. Intermarriage between Swiss persons and individuals from the EU led to slightly higher levels of European identification, but only after controlling for socio-demographic variables, indicating that these partnerships have a rather selective social profile. Swiss that are married to non-Europeans exhibit clearly lower levels of European identification. Furthermore, having parents who are intermarried also leads to a somewhat stronger European identification. Thus, there is some support for the impact of European intermarriage on European identification. With respect to other forms of European social relations and experiences there are only some significant effects. The exception was that having non-European friends leads to a lower level of European closure and for EU-citizens having more co-national friends increases the level of European closure. With respect to other forms of transnational experiences we find that proficiency in different languages does indeed increase identification with Europe. The same is also true for longer stays in different countries, whereas stays in different continents tend to decrease European identification. Thus, the pattern of transnational relations and experiences are homologous to the respective patterns of identification. In previous publications there has been discussion whether transnational relations and experiences also explain socio-demographic differences in identification . Recchi's assumption was that more highly educated persons show a stronger European identification because they have more transnational relations and experiences. We studied this question with respect to education and our results did not support this conclusion, since the already rather weak effects of education were not statistically mediated by the indicators measuring transnational social relations and experiences. Thus, overall, transnational relations and experiences do shape the degree of identification with Europe, but not to the extent expected by the structuralist approach. This approach seems to underestimate the importance of category membership and the construction of the meaning of Europe. However, our study is limited insofar as the sample consists of rather highly educated people. As the structuralist approach particularly makes assumptions on the effects of transnational social relations and experiences for the less educated, further research is needed to substantiate our conclusion. Nonetheless, especially the empirical differences between persons of different origin point towards other explanations. We found stronger European and lower national identification only for migrants from EU countries. Migrants from EU countries, in contrast to migrants from non-European countries clearly profit from their EU citizenship, because mobility for them is not only easier within EU countries but also to and within Switzerland. Thus their citizenship gives them privileged access to the European, including the Swiss, labor market and educational system. Thus, they have interest-based reasons to identify more strongly with Europe and less with their country of origin. This lends support to Fligstein's ideas about the roots of European identity. European identity seems to be more about interests and less about social relations and experiences. However, this explanation has to be tested in a much stricter design, e.g. comparing EU citizens with and without migration experience within Europe. A further important point that must be taken into account is the fact that our conclusions are based on cross-sectional data. Thus, we are able to determine if our statistical results are in accordance with our hypotheses, but we cannot infer the direction of causality. Intermarriage, transnational relations and experiences may be endogenous and may themselves be the result of more open and supranational outlooks. Thus, we are only able to conclude that identities and social relations to a certain degree form a coherent pattern, but not the causal relations. --- Notes 1 Intermarriages in the following encompass marital and non-marital partnerships alike. We expect the same mechanisms to be at work because both forms of relationships have a similar legal standing in Switzerland . The notion Europe refers to Europe as a whole and not only to the European Union. 2 If respondents indicate "zero" for both identification with Europe and the nation, national openness is set on 0.5 as both reference frame are weighted equally. The same procedure is done for European closure and relative Europeanization if both the enumerator and the sum in the denominator are "zero" due to non-identification. --- European Identity in Switzerland: The Role of Intermarriage, Transnational Social --- Relations and Experiences --- Tables & Figures

We analyze the impact of intermarriage, and transnational social relations and experiences on the emergence of European identity. According to the structuralist theory of identification, European social relations, with European intermarriage as an especially important relation, and experiences should explain European identifications. Our analysis is based on a survey in Zurich, Switzerland, providing a broad array of data that allow testing the impact of a European partner on European identification for Swiss and how transnational social relations and experiences contribute to both Swiss and non-Swiss feeling European. Overall, we find that a partner from another European country (for Swiss natives) and transnational social relations and experiences have an important role in explaining European identification. The most important differences are between Swiss and EU citizens living in Switzerland where, for the latter, the meaning of Europe is differently constructed. Specifically, EU citizens see less conflict between national and European identification.

Introduction The Joint United Nations Programme on HIV/AIDS [1] estimated the number of youth living with HIV as of 2017 to be 3.9 million worldwide. Among the global population of youth between the ages 15-24, there were an estimated 610,000 new cases in 2016 according to the United Nations Children's Fund [2]. Approximately 84% of these new cases occurred in Sub-Saharan Africa [2]. These numbers suggest a need to further investigate HIV prevention efforts among youth, especially in parts of the world with high HIV prevalence rates. Sub-Saharan Africa accounts for approximately 64% of all new HIV diagnoses worldwide [3]. Of the 630,000 children being administered antiretroviral therapy in countries designated as low-and middle-income, 544,000 live in Africa according to the World Health Organization [4]. This population faces many challenges associated with health services, including stigma, lack of prevention education, and lack of support services and treatment options [1]. --- HIV Testing as An HIV Prevention Strategy HIV testing is a vital step for both treatment and prevention [5]. Once aware of HIV infection, patients can begin antiretroviral therapy , which is designed to reduce viral load. Current research suggests people with HIV who have undetectable viral loads are not likely to transmit the virus to others. Thus, identifying cases of HIV through testing and linking newly diagnosed patients to, and retaining them in, care has been shown to be effective for preventing the progression of the virus and eventual spread to others. According to the Kenya HIV Testing Services Guidelines from the National AIDS and Sexually Transmitted Infections Control Programme [6], all healthcare facilities should offer HIV testing and counseling to adults and adolescents 15 years or older. It is recommended that children 14 years and younger be tested if living in an area identified as having a high incidence of HIV or if the child has been exposed to high-risk situations or environments . Nonetheless, recent estimates suggest only 46% of women and 53% of men in Kenya's general population had been tested for HIV, suggesting a need to better understand the facilitators and barriers to HIV testing [7]. --- Facilitators and Barriers to HIV Testing Barriers to HIV testing exist particularly in areas of high incidence, such as Sub-Saharan Africa. Existing research suggests facilitators to HIV testing may include high levels of knowledge, social support, subjective well-being, and mental health symptoms , while barriers may include high levels of perceived stigma and substance use. In the sections to follow, both facilitators and barriers to HIV testing are explored. --- HIV Knowledge Behavioral intentions related to HIV testing may be affected by misconceptions about HIV and how the virus is transmitted [8]. Understanding an individual's HIV knowledge-level is tantamount to understanding how much they perceive risk when engaging in behaviors that could lead to transmission of the virus, such as unprotected sex [9]. Glick and Sahn [9] reported a positive relationship between education and HIV testing in nine African countries. Other research indicates a positive relationship between HIV knowledge and testing [10]. Findings from a meta-analysis of 60 studies revealed that HIV knowledge ranked among the most common factors selected by researchers when studying HIV testing behaviors, and that HIV knowledge was positively correlated with HIV testing [11]. --- Stigma HIV-related stigma has been established as a key barrier to HIV testing and treatment [8]. HIV-related stigma can manifest in various forms and be either perceived, anticipated, or believed to be true by an individual or group, notwithstanding HIV status [12]. Simply put, regardless of whether or not an individual has been tested and is aware of their HIV status, one may hold certain beliefs or preconceptions about individuals with HIV. These beliefs often serve as a barrier to both HIV testing and treatment. One method being considered as a means to lower stigma related to HIV and HIV testing is to normalize screening for the virus by making screening routine in healthcare settings. In interviews with men aged 18-24, Knight, Small, and Shoveller [13] found that while, overall, participants viewed routine testing as more convenient and less stress-inducing than seeking out testing as a result of high-risk behavior, others felt targeted by routine testing. The latter group noted that appearing to be homosexual, a drug user, or impoverished may have led the clinician to recommend testing. The authors concluded that while routine testing may break down certain barriers, making those more at risk feel targeted may be a drawback. In a re-analysis of a 2009 study from South Africa, Maughan-Brown and Nyblade [14] found that women were influenced by stigma when it came to HIV testing. Further analysis of this dataset showed that women were more likely to have been tested for HIV if they had stigmatizing attitudes toward others . Conversely, women who reported more perceived stigma, or had personally observed someone with HIV being treated poorly, were less likely to seek testing. This research highlights the importance of better understanding the influence of stigma on HIV testing behaviors. --- Social Support Social support has also been identified as a key factor in HIV test-seeking and treatment adherence [8]. Social support can be conceptualized as the notion of having other individuals to provide reinforcement and emotional support [15]. Among nearly 5000 adults in South Africa voluntarily seeking HIV testing and counseling, low social support and poor mental health were associated with testing positively for HIV [16]. When examining correlates of social support, researchers found that low social support was most strongly correlated with having no prior HIV testing [15]. Lypen, Lockwood, Shalabi, Harper and Ngugi [17] studied type and source of social support among Kenyans aged 18-27. Their research indicated the important role of social support in populations with high risk of HIV in seeking testing and treatment. However, they stress that many pathways of social support have not yet been explored in HIV intervention studies. Other studies also suggest social support is important for HIV testing and that low levels of social support may serve as a barrier to testing among numerous populations of men who have sex with men in the United Kingdom and United States [18,19]. Social support also has been found to be an important factor in disclosing one's status to others. In their meta-analysis of 21 studies on social support, HIV stigma, and disclosure, Smith, Rossetto, and Peterson [15] found that disclosing one's positive HIV diagnosis was positively correlated with having some form of social support. --- Subjective Well-being Subjective well-being refers to one's assessment of tangible and intangible evidence of health or quality of life [20]. Among South Africans, subjective well-being was associated with income and social support from family and religious affiliations. Compared to people without HIV, subjective well-being was significantly lower among people living with AIDS in Australia and the United States [21]. These authors reported a negative correlation between stigma and subjective well-being among people living with HIV. Currently there is no research on the relationship between subjective well-being and HIV testing. However, findings from a study in Portugal suggested subjective well-being is correlated with adherence to antiretroviral medication, indicating its importance in the scope of treatment and living with HIV [22]. In their study, Reis, Guerra, and Lencastre [22] found that patients with greater subjective well-being were more likely to be asymptomatic. Given its impact on how people living with HIV cope and adhere to medication regimens, subjective well-being is a factor of interest among the HIV research community. --- Mental Health Mental health symptoms such as depression, anxiety, and stress can affect HIV risk exposure as well as HIV testing and treatment [8]. Among a sample of 325 men and women in Uganda, depression was correlated with alcohol consumption and intimate partner violence-both risk factors for exposure to HIV [23]. In South Africa, researchers conducted interviews based on the Diagnostic and Statistical Manual of Mental Disorders with patients at HIV testing sites and found that 14% of those seeking testing experienced depression, 5% experienced anxiety, and nearly 5% experienced posttraumatic stress disorder [24]. In India, those seeking repeat voluntary testing were found to have significant levels of depression and anxiety [25]. In a Taiwanese study, researchers found that nearly 40% of their sample of men seeking HIV testing reported depression [26]. The relationship between HIV testing and mental health symptoms may be bidirectional. That is, individuals with mental health symptoms may be at high risk for exposure to HIV and, therefore, more likely to seek testing at the same time that HIV testing may produce symptoms of anxiety, depression, or stress. For example, Worthington and Meyers [27] identified four themes related to anxiety among individuals seeking HIV testing in Canada: risk to health, stigma, the power dynamic of the patient and test provider, and techniques employed to improve patient control when interacting with test providers. In another study, mothers in rural Kenya reported they may be less likely to take measures to prevent transferring HIV to their infants due to stigma and stress associated with HIV, testing their infants, and seeking drugs to prevent transmission [28]. Yehia and colleagues [29] found that individuals with mental illness in the United States were more likely to seek HIV testing than those without a diagnosed mental illness. This research included disorders such as schizophrenia, bipolar disorder, depression, and anxiety, and participants within each category were found more likely to have been tested for HIV than the general population. --- Substance Use Substance use contributes to HIV on multiple levels; it may affect HIV testing, put individuals at higher risk of contracting the disease, and lower the ability to adhere to treatment regimens [8]. Substance use has been established as a factor that increases HIV risk through engagement in high-risk sexual behavior [30,31]. UNAIDS [32] estimates that individuals who use injection drugs are 28 times more likely to be living with HIV. Substance users in Indonesia noted stigma and fear of test results as the major barriers to their seeking HIV testing [33]. --- Social Cognitive Theory Applied to HIV Bandura's [34] social cognitive theory posits that the person, the environment, and one's behavior equally share multidirectional influence over one's learning. SCT has often been applied to health behaviors since these are typically learned throughout an individual's lifetime [35]. Bandura [35] himself contributed to the HIV/AIDS literature by applying SCT to HIV/AIDS education and prevention. Based on a thorough review of the HIV/AIDS literature, Bandura posited that personal factors , one's environment, and one's behavior are constantly interfacing in a way that influences an individual's decision to engage in risky sexual behavior or to consider HIV testing. Bandura stressed that while understanding HIV risk behaviors is important, the impact of one's environment and behavioral changes involved in making sound decisions about HIV prevention are also important. Since Bandura's theory emphasizes a multidirectional interaction between factors, it cannot be assumed knowledge alone will influence health behaviors. Researchers in China applied SCT to HIV education in high schools and found that students who were exposed to an intervention program encompassing self-efficacy, one's environment, and behavior significantly differed from students in a control group [36] The intervention program focused on HIV knowledge, perceptions, stigma toward those with HIV/AIDS, and behavioral intentions related to testing. --- Purpose of This Study Guided by SCT, this study, which was part of a larger research program on HIV-stigma reduction, aimed to examine the relative impact of the following factors, which can be conceptualized as facilitators and barriers, on behavioral intentions related to HIV testing: HIV knowledge, stigma, social support, subjective well-being, mental health, and substance use. Specifically, this study aimed to answer the following question: What is the relative contribution of various factors on behavioral intentions related to HIV testing among youth in Kenya? We hypothesized that participants who reported greater HIV knowledge, social support, subjective well-being and depression/anxiety/stress would be more likely to seek HIV testing and participants who reported greater HIV projected stigma and substance use would be less likely to seek HIV testing . --- Materials and Methods --- Study Design A one group correlational research design was employed in this study using existing pre-test data from the HIV SEERs Project, a community-based participatory research project involving four components that was delivered to 1526 participants aged 12-36 in Kenya [37][38][39]. A correlational design is advantageous in the current study because it can help elucidate the strength and direction of relationships between personal/environmental factors and an individual's intent to receive HIV testing in advance of any intervention. --- Research Variables Framed by SCT, this study examined the relative contribution of personal, behavioral, and environmental factors affecting HIV testing among youth. Personal factors included HIV knowledge, HIV projected stigma, subjective well-being, mental health and substance use. The environmental factor was social support, and the behavioral factor was HIV testing intent. These factors were conceptualized as either facilitators or barriers to HIV testing. --- Participants and Procedure The entire HIV SEERs Project was evaluated using pre, post, and three-month follow-up questionnaires, which were translated using a back-translation method and offered to participants in either English or Swahili, the native language of Kenya. All participants elected to complete the questionnaires in English, which is the language of instruction in Kenya [40]. This study used existing pre-test data only and included 1007 Kenyan youth aged 13-24 who were not HIV-positive. However, 356 participants did not provide complete data on demographic or study-related variables and were therefore not included in the primary analyses. Results henceforth are based on the 651 participants who provided complete data. Participants who were included were not different than participants who were excluded on a number of demographic characteristics: age = -1.20, p = 0.23), gender = 0.763 p = 0.38), sexual orientation = 2.19, p = 0.334), ever having been tested for HIV = 0.510, p = 0.775), or sexually active status = 2.883, p = 0.237). The Institutional Review Board at the University of South Florida reviewed this project and deemed it exempt given its use of existing pre-test data from the HIV SEERs Project. --- Measures A post-hoc analysis of reading level revealed that the Flesch-Kincaid reading level of the measures described below ranged from 4.7 to 7.3 with the exception of the CRAFFT Screening Tool for Adolescent Substance Use, which had a reading level of 9.1. --- Behavioral Intentions This scale was developed by the HIV SEERs Project investigators to collect information about HIV treatment-seeking behaviors. Only one item addressing HIV testing behavioral intentions was used in this study: "If you are not HIV-positive, do you plan to be tested for HIV?" Response options included "Yes," "No," and, "I am HIV-positive." Participants who responded "I am HIV-positive" were excluded from the analysis. Since this measure consisted of a single item, Cronbach's alpha was not calculated. --- Brief HIV Knowledge Questionnaire The HIV KQ-18 [41] is an 18-item instrument that uses true/false prompts to measure an individual's understanding of HIV and methods of transmission. "A person will NOT get HIV if she or he is taking antibiotics" is an example of the type of items on this measure. Carey and Schroder [41] reported a Cronbach's α = 0.75 to 0.89 across multiple samples, demonstrating the internal consistency of this measure. In the current sample, α = 0.39 was found. Missing data on this measure was scored as an incorrect. --- AIDS-Related Stigma Scale The ARSS [42] is a nine-item instrument that assesses an individual's beliefs and projected stigma toward people living with HIV/AIDS. Response choices include Agree or Disagree. Examples of items on this measure are "People who have AIDSs are dirty," and "People who have HIV should be isolated." Kalichman and colleagues [42] reported a Cronbach's alpha of α = 0.75, demonstrating the internal consistency of this measure across five African communities, and also used test-retest reliability estimates to demonstrate the stability of this measure over a three-month period, r = 0.67. In the current sample, α = 0.59 was found. --- Social Provision Scale The original 24-item measure [43] was adapted to a 12-item measure, which assesses social support using six provisions [44]. Participants respond to items such as "There are people I can depend on to help me if I really need it" using a five-point Likert scale with 1 = Strongly Disagree and 5 = Strongly Agree. In the original sample, α = 0.92 was reported for the entire scale, demonstrating good internal reliability. In the current sample, α = 0.62 was found. --- Subjective Well-being Scale The Subjective Well-being Scale [45] gauges well-being on a ten-point Likert scale with 1 = Extremely Unhappy and 10 = Extremely Happy using a single question, "Taking all things together, how happy are you?". Since this measure consisted of a single item, Cronbach's alpha was not calculated. --- Depression, Anxiety, and Stress Scale The DASS-21 [46] is a 21-item measure that assesses depression, anxiety, and stress . Examples of items from this measure include "I felt scared without any good reason" and "I felt I was close to panic." Participants respond using a four-point Likert scale where 0 = Did not apply to me at all and 3 = Applied to me very much or most of the time. Cronbach's alphas ranging from α = 0.87 to α = 0.94 across the three subscales have been reported, demonstrating good internal reliability [47]. In the current sample, α = 0.86 for the depression scale, α = 0.83 for the anxiety, and α = 0.87 for the stress scale. --- CRAFFT Screening Tool for Adolescent Substance Abuse The CRAFFT [48] is a nine-item, yes/no measure of an individual's experiences with drugs and alcohol. The assessment is divided into two sections: three behaviors within the last 12 months, and six behaviors over the lifetime. The six lifetime items are used for scoring purposes. Example items from this measure are "During the past 12 months, did you smoke any marijuana or hashish?" and "Do you ever forget things you did while using alcohol or drugs?" Knight and colleagues [48] reported α = 0.79, demonstrating good overall reliability of this measure. In the current sample, α = 0.63 was found for the last six items. --- Date Analysis Descriptive statistics were used to assess the sample and performance on measures. Cronbach's alpha was computed to determine the reliability of measures as reported above. Pearson's r was used to examine the bivariate relationship between two continuous variables and point-biserial correlations were used to examine bivariate relationships between a continuous and dichotomous variable. Logistic regression was used to determine the predictive effect of each of the independent variables on HIV testing intent . Forward selection was used to identify the best combination of independent variables likely to predict HIV testing intentions. We followed suggestions by Lee and Koval [49] and set our alpha level to 0.15 for entry into the analysis. Sexual activity was included as a covariate and tested as a moderator in the analysis. SPSS version 25 was utilized to complete all analyses. --- Results --- Participant Demographics Among the 651 participants aged 13-24, the mean age was 16.7 . The sample was 53.5% female and 46.5% male. Only 3.5% identified as gay or lesbian while 92.5% did not identify as gay or lesbian and 4.0% wished not to answer. The majority of participants identified their religion as Christianity while 2.9% identified as Islam and 1.4% combined identified as Hindu, Buddhist, or Other. Less than 1% selected No Religion or wished not to answer. Most said they were not sexually active while 20.8% said they were sexually active and 7.2% wished not to answer. A slight majority of the sample reported having been tested for HIV while 40.1% had never been tested and less than 1% wished not to answer. --- Descriptive Statistics --- Behavioral Intentions Approximately two-thirds of participants reported they intended to be tested for HIV, while 33.0% said they did not. --- Brief HIV Knowledge Questionnaire Scores on this measure ranged from 4-18 with a mean score of 13.65 , suggesting a moderate level of HIV knowledge among participants in this sample. Some common incorrect items on this measure were "Coughing and sneezing DO NOT spread HIV" and "A person can get HIV by sharing a glass of water with someone who has HIV." --- AIDS-Related Stigma Scale Scores on the ARSS ranged from 0-10 with a mean score of 1.32 , indicating low levels of projected HIV stigma among participants. --- Social Provision Scale Scores on this scale ranged from 6-60 with a mean score of 34.00 , indicating a low to moderate level of perceived social support among the sample. --- Subjective Well-being Scale Scores on this scale ranged from 1-10 with a mean score of 7.92 indicating moderate to high levels of reported well-being among the sample. --- Depression, Anxiety, and Stress Scale This scale was divided into the three mental health subscales of depression, anxiety, and stress for more valuable interpretation. Depression scores ranged from 0-38 points with a mean of 7.57 , indicating low to moderate levels of depression among the sample. Anxiety scores ranged from 0-34 points with a mean of 6.80 , indicating low to moderate levels of anxiety reported by the sample. Stress scores ranged from 0-38 points with a mean of 8.08 , indicating low to moderate levels of stress reported by participants. --- CRAFFT Screening Tool for Adolescent Substance Abuse Participants reported total scores ranging from 0-6 with a mean score of 1.39 ; 30.4% of participants scored two or higher on this measure, which is the recommended threshold for intervention [45]. --- Impact of Independent Variables on Intent to Test for HIV We first examined zero-order correlations among all of the study variables. As expected, HIV knowledge was significantly positively related with an intent to seek HIV testing and projected stigma was significantly negatively related to an intent to seek HIV testing. Contrary to expectations, substance use was positively-and social support negatively-correlated with an intent to seek HIV testing. We next evaluated our primary hypotheses with a logistic regression model which included intent to be tested for HIV as the outcome variable and HIV knowledge, stigma, social support, subjective well-being, depression, anxiety, stress, and substance use included as potential predictor variables. Sexual activity was included as a covariate. The logistic regression model was statistically significant, χ 2 = 24.02, p < 0.001. The model explained 5.7% of the variance in HIV testing intentions and correctly classified 66% of cases. Results from the logistic regression analysis are presented in Table 1. Four variables emerged as significant or marginally significant predictors of HIV intentions: HIV knowledge, substance use, depression, and social support. HIV knowledge was significantly associated with a higher likelihood of intent to seek testing, b = 0.087, χ 2 = 6.19, p = 0.013. The odds ratio was 1.09, suggesting that every one-unit increase in HIV knowledge resulted in an 9% increase in HIV testing intentions, reflecting a small effect. Substance use also predicted intent to test for HIV, b = 0.346, χ 2 = 7.33, p = 0.007. The odds ratio was 1.41, indicating that a one-unit increase in CRAFFT scores increased intent to test by 41%, reflecting a medium effect. Depression was marginally associated with a lower intent to seek testing, b = -0.019, χ 2 = 3.72, p = 0.054, reflecting a small effect. Finally, social support was marginally negatively associated with intent to seek HIV testing, b = -0.024, χ 2 = 3.68, p = 0.055, reflecting a small effect. Projected stigma, subjective well-being, anxiety, stress, and sexual activity were not associated with HIV testing intentions and were therefore not included in the final model. Sexual activity did not moderate any of the associations between the predictor variables and intent to test. --- Discussion Framed within the context of social cognitive theory , this study used existing data from the HIV SEERS Project, a community-based participatory research project designed to increase HIV knowledge and decrease HIV stigma, to examine facilitators and barriers to HIV testing among youth in Kenya. Understanding what might motivate or prevent Kenyan youth from seeking HIV testing is key in advancing effective education programs for this high-risk population. In this study, HIV knowledge and substance use were found to be significant facilitators to HIV testing. The association between HIV knowledge and intent to be tested for HIV is consistent with existing studies [8][9][10][11]. The association between substance use and HIV testing is consistent with some studies, but not others. Similar to findings in our study, Luseno & Wechsberg [50] reported an association between substance use and HIV testing, which may be a function of those using substances perceiving themselves to be at higher risk for HIV. Inconsistent with our findings, Altice, Kamarulzaman, Soriano, Schechter, and Freidland [51] reported that people who used drugs of any kind were less likely to seek HIV testing early on and, for this reason, were more likely to be diagnosed with advanced stages of HIV. Also among their findings were other important health indicators, such as increased risk of HIV exposure and suboptimal healthcare. Given the relationship between HIV testing and substance use, some current research suggests carrying out HIV testing at substance use treatment facilities [52]. Substance use is understood to increase risk-taking behaviors such as unprotected sex, needle sharing, and one-night stands, which may increase the chances of individuals contracting the disease [30,31]. Regarding mental health, our results suggested that depression was marginally significantly predictive of intent to test while anxiety and stress did not significantly predict HIV testing. Rueda and colleagues [8] found that depression, anxiety, and stress were all contributing factors to risk-taking behaviors that may increase the need for HIV testing. These same mental health symptoms were present among a sample of individuals seeking HIV testing [24]. Also, those who were previously diagnosed with a mental illness were found more likely to have had prior HIV testing [29]. It is important to note that many studies in this area have been conducted with adult populations in the United States and Canada. Age and context may contribute to current findings. That is, mental health symptoms may be viewed differently by youth and/or by other individuals in low-to-middle income countries. Social acceptability of admitting to mental health concerns or even differing thresholds of tolerance may have had an impact on the significance of these factors in the current analysis. The marginally significant negative predictive effect of social support on intent to be tested for HIV found in this study is inconsistent with what is understood about the importance of social interaction in maintaining physical and mental health [15,17,19]. Cultural factors and stigma may help explain current findings. Given the stigma associated with HIV in Kenya, it is possible that fear of rejection by social groups may have served as a barrier to HIV testing intentions for participants in this sample [53]. That is, youth with a strong social support network may fear losing support if others find out they were tested for HIV given assumptions that might be made about promiscuity or other high risk behaviors. While several of the hypothesized facilitators and barriers to HIV testing were not found to be significant in the current study , findings do provide some support for the use of social cognitive theory to explain HIV testing behaviors among youth [34]. Specifically, social support-an environmental factor-and substance use, HIV knowledge, and depression-personal factors-were found to be predictors of HIV testing-a behavioral factor. See Figure 1. However, several of the personal factors hypothesized for the model were not significant among the study sample as described above. Specifically, projected stigma, anxiety, stress, and subjective well-being did not predict HIV testing in the current sample. With regard to stigma, it is important to note that HIV stigma was associated with HIV testing intentions in a simple regression, but lost significance when other variables were taken into account. This is likely due to multicollinearity among the predictors used in this study, which suggests that stigma does not uniquely explain HIV testing intentions when more powerful predictors are taken into consideration. There were several limitations to this study. First, while the measures used in this study were established as reliable measures in the literature, estimates of internal consistency among our sample were moderate for the social support and substance use measures and low for the HIV knowledge and projected stigma measures. One explanation for this relates to the homogeneity of the sample, since all participants were part of the same culture and fell within a restricted age range. In addition, there are other measure-related factors that can affect alphas-the length of the measures, use of subscales, interrelatedness, and construct heterogeneity [54]. Alphas should not be considered the only indicator of reliability, since low alphas do not always indicate poor measures, and high alphas are not necessarily proof of a reliable measure [55]. Second, cultural differences and language barriers may have been limitations in this study. While questionnaires were translated into Swahili and offered in both English and Swahili, all participants chose to complete the English version; this is likely due to the fact that English is taught and used in educational settings in the region [40]. Thus, youth might be more comfortable reading and writing in English while they are more comfortable speaking in their native language of Swahili [40]. Nonetheless, it is possible that individual items on measures were misinterpreted or misunderstood due to the cultural context. Finally, the research design posed limitations. The use of existing program evaluation data resulted in limited control over the variables of interest in this study. For example, intent to test for HIV was the main outcome in this study. While several theories suggest a relationship between intentions and behavior [56] and it is not uncommon for researchers to assess intentions as opposed to actual behavior in studies related to HIV testing [57,58], we do not know if intent to test resulted in future HIV testing among our participants. Also, the use of a correlational design is limiting with regard to the conclusions that can be made about relationships between variables. Further research implementing experimental designs is needed to identify causal relationships between personal and environmental factors, intent to test, and actual HIV testing. --- Conclusions This research relied on existing data from a community-based participatory research project. Despite limitations, this study contributes to our knowledge regarding factors related to HIV testing intent among youth in Kenya. There is still much to be learned about youth at risk for HIV in Kenya There were several limitations to this study. First, while the measures used in this study were established as reliable measures in the literature, estimates of internal consistency among our sample were moderate for the social support and substance use measures and low for the HIV knowledge and projected stigma measures. One explanation for this relates to the homogeneity of the sample, since all participants were part of the same culture and fell within a restricted age range. In addition, there are other measure-related factors that can affect alphas-the length of the measures, use of subscales, interrelatedness, and construct heterogeneity [54]. Alphas should not be considered the only indicator of reliability, since low alphas do not always indicate poor measures, and high alphas are not necessarily proof of a reliable measure [55]. Second, cultural differences and language barriers may have been limitations in this study. While questionnaires were translated into Swahili and offered in both English and Swahili, all participants chose to complete the English version; this is likely due to the fact that English is taught and used in educational settings in the region [40]. Thus, youth might be more comfortable reading and writing in English while they are more comfortable speaking in their native language of Swahili [40]. Nonetheless, it is possible that individual items on measures were misinterpreted or misunderstood due to the cultural context. Finally, the research design posed limitations. The use of existing program evaluation data resulted in limited control over the variables of interest in this study. For example, intent to test for HIV was the main outcome in this study. While several theories suggest a relationship between intentions and behavior [56] and it is not uncommon for researchers to assess intentions as opposed to actual behavior in studies related to HIV testing [57,58], we do not know if intent to test resulted in future HIV testing among our participants. Also, the use of a correlational design is limiting with regard to the conclusions that can be made about relationships between variables. Further research implementing experimental designs is needed to identify causal relationships between personal and environmental factors, intent to test, and actual HIV testing. --- Conclusions This research relied on existing data from a community-based participatory research project. Despite limitations, this study contributes to our knowledge regarding factors related to HIV testing intent among youth in Kenya. There is still much to be learned about youth at risk for HIV in Kenya and the behavioral, personal and environmental factors that may impact their healthcare. This study serves as a building block for future studies. Additional research is needed to explore factors affecting HIV testing among youth in low-to-middle income countries . Post and follow-up data from the HIV SEERs Project may yield important information about the impact of a community-based participatory research project on HIV testing and treatment. Beyond SEERs, it is imperative that further research include larger and more representative samples and take into consideration language and cultural barriers. Additionally, intervention studies designed to improve rates of HIV testing among youth in LMICs will build upon the information collected in this and similar studies. ---

With the high prevalence of HIV among youth in sub-Saharan Africa, it is vital to better understand factors affecting HIV testing among this population; this is the first step in the HIV treatment cascade. The purpose of this study was to examine factors related to behavioral intentions regarding HIV testing using existing pre-test data from the HIV SEERs (Stigma-reduction via Education, Empowerment, and Research) Project, a community-based participatory research program targeting 13-24-year-olds in Kenya. It was hypothesized that HIV knowledge, social support, subjective well-being, and mental health (depression, anxiety, and stress) would serve as facilitators to HIV testing while projected stigma and substance use would serve as barriers to HIV testing. In partial support of our hypotheses, findings from logistic regression analyses revealed that HIV knowledge, substance use, depression, and social support were significant predictors of HIV testing intentions. However, HIV knowledge and substance use served as facilitators while depression and social support served as barriers. While projected stigma was correlated with HIV testing intentions, it was not a significant predictor in the regression analysis. Subjective well-being, anxiety, and stress were not significant predictors in the regression analysis. These findings have important implications for HIV testing initiatives designed for youth in Kenya as well as future research on HIV testing with this population.

Introduction Emerging evidence in the UK suggests that some ethnic minorities have better psychological well-being in early adolescence than their White peers , and that markers of the quality of family interactions and connectedness, such as shared activities and perceived parenting, are key correlates of psychological well-being regardless of socio-economic circumstances . Adolescence is, however, a formative period when family influences may decrease and characteristics of the wider context, neighbourhoods and schools in particular, could become critical influences . There is a significant body of literature on the harmful effect of racism on physical health and psychological well-being . Some evidence suggests that individuals spending much of their time among people of the same ethnicity are less likely to report racism . The ethnic density hypothesis posits that common social norms and support networks in ethnically dense areas may be protective of health . Longitudinal studies of these issues in ethnically diverse cohorts are sparse. In this article, we explore the extent to which trajectories in psychological well-being between early and late adolescence are influenced by the experience of racism, whether the prevalence of racism varies from one context to another, and whether ethnic density and deprivation play a significant role in shaping psychological well-being through this early stage of the life course. Evidence in support of the ethnic density hypothesis in relation to psychological well-being, mainly from cross-sectional studies, has been equivocal to date. Beneficial effects on health of more ethnically dense contexts , a mix of beneficial and negative associations depending on the ethnic group , negative effects at higher levels of exposure , and no effect of ethnic density on psychological well-being have all been reported. Only a few of these studies have considered psychological wellbeing in childhood and adolescence . The historic differences in migration between the USA and the UK relates to contrasts in settlement patterns between the two countries. For example, in the 2000 US census, while 60 per cent of the Black American population in Chicago lived in neighbourhoods where they formed between 90 and 100% of the population, there are no single ethnic minority neighbourhoods in the UK . Neighbourhoods thought to be ethnically dense in the UK are ethnically and socially diverse although, overall, ethnic minorities are more likely to live in deprived areas than their White peers. Ethnic diversity may also be an important influence on psychological well-being. Blumer's group position theory of prejudice and the conflict hypothesis contend that individuals within ethnically-diverse contexts have less trust and fewer ties with people of other ethnic groups, focusing interactions more towards those of the same ethnicity . The conflict and ethnic density hypotheses both imply that ethnic diversity increases the risk of tension between ethnic groups. Putman's constrict hypothesis goes further in contending that greater ethnic diversity results in withdrawal from collective life and the reduction of both in-group and out-group solidarity; that is, in Putnam's terms, less 'bonding' and 'bridging' social capital . Not only may the risk of experiencing racism be greater, but ethnic diversity may also have an independently negative effect on health through the absence of local social support. By contrast, Allport's contact hypothesis proposes that positive relations between ethnic groups and social solidarity are encouraged in more ethnically diverse settings, where the opportunity for contact with people from a different ethnic group is greater . With more contact, people are said to overcome initial reservations and prejudices, while developing trust and solidarity. Unlike the ethnic density hypothesis which implies benefits of being around people similar to oneself, the contact hypothesis suggests that prejudicial beliefs may be reinforced when individuals are surrounded by more of their own group and with fewer opportunities to meet other ethnic groups on a regular basis, thereby increasing the risk of experiencing racism when interactions do occur. In comparison to the rapidly expanding literature on ethnic density, however, these theories of ethnic diversity have received less attention in research on ethnic inequalities in health. In this analysis, we use UK-based longitudinal data at 11Á16 years on the main ethnic groups in the Determinants of Adolescent Social well-being and Health study to address the following questions: What is the effect of racism on ethnic-specific age trajectories of psychological well-being? Is ethnic density or deprivation related to perceived racism across different ethnic groups? Does the ethnic density of neighbourhoods or schools modify the effect of racism on psychological well-being, independent of deprivation or family life? --- Methods Detailed information on the DASH study is available elsewhere . In brief, DASH is a longitudinal study of 6645 adolescents from 51 schools in 10 of London's 32 boroughs. These 10 boroughs were selected as having high proportions and numbers of people from ethnic minority groups. Ethical approval was obtained from the Multi-centre Research Ethics Committee and from Local Education Authorities. The baseline survey was conducted in 2003/2004 when participants were aged 11Á13 years, with 4782 adolescents participating in a follow-up in 2005/2006 at age 14Á16 years. In both surveys, each individual self-completed a structured questionnaire on their health and social circumstances. The sample included White UK, Indian, Pakistani and Bangladeshi, Black Caribbean, Nigerian and Ghanaian, and Other African boys and girls. For reporting convenience we refer to the White UK in the text as 'White'; Nigerians and Ghanaians, Other Africans and Black Caribbeans are referred to as 'Black African origin' where similar patterns are reported. Data for the remaining DASH participants of other ethnicities are not presented in this article as they are of insufficient number to draw reliable statistical inference. --- Outcome measure Psychological well-being was measured using Goodman's 25-item Strengths and Difficulties Questionnaire , a behavioural screening tool providing coverage of children's behaviour, emotions and peer relations , validated for use in multi-cultural settings . The SDQ comprises five scales: emotional symptoms; conduct problems; hyperactivity; peer problems; and prosocial behaviour. The first four of these scales are summed to calculate a total difficulties score ranging from 0 to 40, with higher scores representing increasing difficulties . Differences in scores between groups are within the 'normal' range and are therefore not indicative of clinical morbidity, per se. Each 1-point increase in child-reported SDQ, however, corresponds to an increased probability of clinician-assigned mental disorder . --- Individual level explanatory or confounding variables Experience of racism was the key variable of interest, assessed with the question: 'Has anyone made you feel bad or hassled you because of your race, skin colour or where you were born?' Separate responses were requested for the settings 'at home', 'at school' and 'where you live' . A single dichotomous outcome variable was constructed with affirmative responses in any or all settings as the 'yes' category. Ethnicity was primarily self-reported; individuals identifying as 'Black British', 'Asian British' or no ethnic group were re-classified according to parental ethnicity and parental/grandparental country of birth. Family type was defined as two-parent, reconstructed or cohabiting, lone parent and other . Family activities were assessed using a cumulative frequency of activities, including watching television, eating meals together, indoor and outdoor games, visiting family and friends. Perceived parental care and control were measured using an eight-item scale; higher scores denote more care and control respectively. Individual socio-economic circumstances were measured using tertiles of a score based on 17 standard of living items. Generation status was defined as UK-born or overseas-born. --- Contextual explanatory variables School characteristics were derived from the 2003 and 2006 school censuses and neighbourhood characteristics from the 2001 census . We calculated the neighbourhood measures based on output areas , the smallest level at which census data are accessible, which contain about 297 individuals on average . Own-group ethnic density was calculated as the percentage of the same ethnic group within the residential neighbourhood or the school attended. For example, Black Caribbean adolescents were assigned the percentage of the neighbourhood population that identified as Black Caribbean in the 2001 Census, where a higher percentage denotes greater own-group ethnic density. Nigerians, Ghanaians and Other Africans were each assigned 'Black African' density in neighbourhoods as census data was not available for each group individually. Own-group ethnic density varied, with any ethnic minority group being far less likely to be the majority group than Whites. Mean ethnic minority density in schools varied between 3.1% for Other Africans to 26.8% for Nigerians and Ghanaians; in neighbourhoods it varied between 13.0% for Black Caribbeans to 18.2% for Indians. The corresponding figures for Whites were 33.3% in schools and 60.1% in neighbourhoods. Ethnic diversity was calculated using the Herfindahl index , which is the sum of the squared proportions of each ethnic group within a neighbourhood or school. A score equal to the proportion of all possible ethnic groups reflects an ethnically diverse context. For example, 10 ethnic groups equally proportionally represented within a neighbourhood would give a score on the Herfindahl index of 0.1. Higher scores denote contexts where the proportional representation of some ethnic groups are greater than for others, while a score of one indicates the presence of one ethnic group only. Deprivation measures were derived separately for neighbourhoods and schools. In neighbourhoods, we used the 2001 census to calculate the Carstairs index based on overcrowding, male unemployment, low social class and car ownership. In schools, we calculated the percentage of the school population eligible for free school meals, which individuals were entitled to receive if their families claimed UK welfare benefits . Higher Carstairs scores and percentages of free school meals denote more deprived circumstances. --- Statistical analysis Repeated measures were obtained from the same pupils at 11Á13 and 14Á16 years. Although pupils were nested within schools and neighbourhoods, preliminary analyses showed clustering at each level was not statistically significant. A two-level random slopes model with measurements nested within pupils was used, with a continuous measure of TDS as the outcome. All variables were considered as time-dependant, except generational status, gender and ethnicity. The log-likelihood ratio test was used to identify statistically significant effects . To investigate ethnic-specific age trajectories of psychological well-being and the effect of racism on these trajectories, gender-specific models were run, adjusting for racism and other individual level explanatory variables. Age was linearly associated with TDS through adolescence . We fitted interaction terms to explore how TDS trajectories varied between ethnic groups ethnicity), the effect of racism on TDS varied through adolescence racism) and for differential effects of racism across ethnic groups ethnicity). We used information on all individuals to estimate mean TDS between 12 and 16 years old, accounting for 98% of the sample. Ethnic-specific cross-tabulations and logit regression were used to explore the extent that ethnic density or deprivation was related to the reporting of racism for different ethnic groups. Using ethnic-specific models, we explored to what extent the ethnic density and deprivation of neighbourhoods and schools modified the effect of racism on TDS, independent of individual explanatory variables. Contextual variables were added to the baseline model sequentially. To maintain statistical power, gender was included as an explanatory variable, controlling for gender-specific age trajectories within ethnic groups using interaction terms. We explored modification of the racism effect using interaction terms between contextual variables and racism. Results from linear mixed models are presented as predicted mean values with 95% confidence intervals. Approximately 20% of answers to the racism question at 11Á13 years were absent, with some variation by ethnic group. We used Multiple Imputation by Chained Equations ). Models were run in Stata v.10. --- Results Table 1 shows that by age 14Á16 years, all ethnic minority groups reported significantly more experiences of racism than their White peers. The reporting of racism significantly increased for Nigerian/Ghanaian boys and girls, while a decrease was noted among Pakistani/Bangladeshi boys. The increased prevalence of racism reported among Nigerian/Ghanaian boys coincided with the lowest age-adjusted mean TDS at baseline and followup. By 14Á16 years, significantly lower mean TDS scores were observed among Pakistani/Bangladeshi and Black African origin boys, and Indian girls compared to their White peers. Mean TDS scores among boys were significantly lower at 14Á16 years compared to 11Á13 years. Compared to their White peers, all ethnic minorities were more likely to attend schools or live in neighbourhoods that were classified as deprived. White boys and girls were significantly more likely to attend schools and live in neighbourhoods with high own-group ethnic density than their peers in other ethnic groups . Other Africans attended schools with the lowest levels of own-group density. Whites were also more likely to be situated in the schools and neighbourhoods with low ethnic diversity than any other ethnic group. In baseline models adjusted for individual-level characteristics only , the psychological well-being advantage for the ethnic minority groups noted above remained. Mean differences in TDS for these ethnic minority groups compared to Whites among boys were for Black Caribbeans; Nigerians/ Ghanaians; Pakistanis/Bangladeshis; Other Africans; and for Indian girls. In these adjusted models, psychological well-being scores improved through adolescence among both boys and girls, and this effect was constant by ethnicity. Figure 1 shows the effect of racism on psychological well-being through adolescence, adjusting for individuallevel characteristics only. Those who reported racism had higher mean TDS compared to those who did not, in each ethnic group. The improvement in psychological well-being was greater for those who reported racism than those who did not, indicating that the impact of racism on psychological well-being decreased with age. Age-specific mean difference in TDS between those who experienced racism and those who did not at age 12 years 0'1.88 ;'1.19 at age 16 years for boys;'2.29 at age 12 years;'1.43 at age 16 years for girls. This effect was constant across ethnic groups. Additionally the psychological well-being advantage for the ethnic minority groups detailed above remained with the adjustment for racism. Table 2 shows the proportions that reported racism within each ethnic group by school and neighbourhood deprivation/ ethnic density measures. Among Whites, more racism was reported in more deprived schools and neighbourhoods. A similar pattern was seen for Pakistanis/Bangladeshis, and Nigerians/Ghanaians in relation to neighbourhood deprivation. There was a broad pattern of more racism reported in schools and neighbourhoods with low than high own-group density. For Other Africans there was no difference by either school or neighbourhood own densities, and for Indians and Black Caribbeans, there was no difference by neighbourhood own density. Only for Whites did the prevalence of racism significantly vary according to the level of ethnic diversity, with more racism reported in more diverse schools and neighbourhoods. Table 3 shows the effect of school deprivation and ethnic density measures, adjusted for all individual-level variables, on psychological well-being. School contextual measures were not associated with TDS. There was one exception Á Other Africans in schools with high own-group density now showed a higher TDS than those in low own group density. Table 4 shows the corresponding results for neighbourhood deprivation and density measures, adjusted for all individual level variables. Neighbourhood deprivation was associated with increased mean TDS for Whites, but not for the ethnic minority groups, remaining after controlling for own-group density and diversity . Neither own-group density nor ethnic diversity was associated with TDS. After controlling for diversity, the effect of deprivation on psychological well-being among Black Caribbeans became significant. No interaction was found between deprivation and the effect of racism on mean TDS among Whites or Nigerians and Ghanaians. The simultaneous adjustment of all school and neighbourhood characteristics within each model did not materially alter these results. To check the sensitivity of our results to definition, we re-ran all analyses using neighbourhood variables measured for lower super output areas , which contain a mean of 1500 individuals, and also for separated responses for the racism variable . These sensitivity checks did not change the overall findings. --- Discussion --- Principal findings After taking into account individual characteristics, psychological well-being improved over time for all adolescents irrespective of ethnicity. Ethnic minority boys , and Indian girls reported better psychological well-being throughout adolescence compared to their White peers, consistent with findings from previous cross-sectional research which indicated an advantage in psychological well-being among ethnic minority adolescents . The significant psychological well-being advantage reported for Indian girls, but not for Indian boys, may be explained by potentially fewer externalising problems among girls. Further longitudinal research to investigate ethnic and gender differences in TDS subscales is warranted. 4 27.8 35.0 29.6 30.9 29.3 Higher 22.9** 34.0 27.3 24.2 29.7 27.0 Ethnic density Lower 23.7 36.1 39.0 31.7 36.9 Ethnicity & Health 81 --- Ethnicity & Health 83 In line with a systematic review of the effect of racism on health by Paradies , we found that racism was associated with poorer psychological well-being throughout adolescence for all ethnic groups. The effect was constant across ethnic groups despite greater reporting of racism among minority groups. We also found a decrease in the size of the effect of racism on psychological well-being with age, which has not previously been reported. This increasing resilience with age may potentially be explained by coping strategies developed through experience and increasing maturity . The reporting of racism varied by context and ethnic group and, consistent with the conflict and constrict hypotheses, Whites in more ethnically diverse contexts were more likely to report racism. The reporting of racism was generally lower in neighbourhoods and schools with higher than lower owngroup ethnic densities. However, we found little evidence of an effect of own group ethnic density or diversity on adolescent psychological well-being, in contrast to previous studies among adult populations . Previous studies have reported increasing ethnic density associated with externalising problems among Dutch minority youth , and high neighbourhood own-group ethnic density associated with attenuation of the psychological well-being advantage among UK 'South Asian' adolescents . Thus, to date, there is little support for the ethnic density hypothesis in relation to adolescent psychological well-being. Xue et al. found that neighbourhood deprivation was associated with increased mental health problems among US children in different ethnic groups. We found neighbourhood deprivation was associated with poorer psychological well-being among Whites, and for Black Caribbeans after adjustments for neighbourhood density or diversity. All other ethnic minority groups appeared resilient to the influence of neighbourhood deprivation on psychological well-being. Overall, we report little evidence of contextual effects on psychological well-being in these analyses. Two definitions of neighbourhood of contrasting sizes were compared and similar results were found for both. Nevertheless, we acknowledge the possibility that each of these definitions may mask real neighbourhood effects and may poorly reflect what 'neighbourhood' really means for individuals . --- Strengths and weaknesses The key strength of the DASH study is that it is longitudinal with large samples of the main ethnic groups in the UK. For adolescents, both the neighbourhood and school contexts may be simultaneously important for their psychological well-being. We took into account the cross-classification of contextual exposure in neighbourhoods and schools. We used both own-group ethnic density and ethnic diversity, in recognition of the contemporary context of children straddling different symbolic worlds. Most of the DASH respondents were born in the UK, but their parents are likely to have been born in home countries. Children may return home to fairly culturally concentrated environments but they attend schools with diverse ethnic populations. There are limitations to the study that warrant mentioning. Missing data was a significant issue for the measure of racism at 11Á13 years, however, we found no material difference in the effect sizes, their direction or significance, between imputed and non-imputed data. Additionally, although longitudinal, the follow-up period of our study may not be long enough to observe the effects of neighbourhood and school characteristics on psychological well-being, which could manifest later in the life course. Arguably, the temporal mismatch between the ethnic density data and the DASH data could have contributed to a non-effect but the extent of ethnic clustering is not thought to have changed substantially . As in all observational studies, reverse causality cannot be discounted, since it is plausible that adolescents with poorer well-being may report more racism. As DASH is a Londonbased study, it is possible that our findings are contextually specific and contrasting results may be found among studies set in different geographical areas. Some suggest that the UK housing allocation system promoted the concentration of ethnic minorities in some of the poorest neighbourhoods . Whether by choice or discriminatory forces, a failure to account for the non-random nature of neighbourhood or school exposure limits the validity of any observational study inferring contextual effects . --- Conclusion Racism adversely influenced psychological well-being throughout adolescence, while the ethnic density of schools and neighbourhoods had little significant impact. The effect of racism on psychological well-being was significantly greater in early compared to late adolescence. White adolescents in more deprived neighbourhoods had poorer psychological well-being, while their ethnic minority counterparts appeared resilient to a potential effect of neighbourhood deprivation. Despite being more likely to experience racism, the psychological well-being of ethnic minorities was not significantly poorer than their White peers. Additional longitudinal studies are required to monitor trajectories in psychological well-being through adolescence into young adulthood, at which time in the life course any ethnic minority advantage may be eroded by prolonged exposure to racism in neighbourhoods and places of study and work. --- Key messages What is known: Racism adversely affects mental health in adult populations. There is a paucity of longitudinal research, exploring relationships between contextual factors and psychological well-being among adolescents from different ethnic groups. What this study adds: Racism was significantly associated with psychological well-being throughout adolescence and the effect was greater in early adolescence. Despite minority groups being more likely to report racism than their White peers, the effect of racism on psychological well-being was similar across ethnic groups. Own-group density, ethnic diversity and deprivation of schools and neighbourhoods were generally not associated with the psychological well-being of ethnic minority groups. Policy/wider implications: Efforts to reduce exposure to racism will benefit the psychological wellbeing of all adolescents, regardless of their ethnic group. Further longitudinal studies are required to investigate the role of context and racism on ethnic differences in psychological well-being during the transition from late adolescence to young adulthood.

Objective. To investigate the effect of racism, own-group ethnic density, diversity and deprivation on adolescent trajectories in psychological well-being. Design. Multilevel models were used in longitudinal analysis of psychological well-being (total difficulties score (TDS) from Goodman's Strengths and Difficulties Questionnaire, higher scores correspond to greater difficulties) for 4782 adolescents aged 11Á16 years in 51 London (UK) schools. Individual level variables included ethnicity, racism, gender, age, migrant generation, socioeconomic circumstances, family type and indicators of family interactions (shared activities, perceived parenting). Contextual variables were per cent eligible for free school-meals, neighbourhood deprivation, per cent own-group ethnic density, and ethnic diversity. Results. Ethnic minorities were more likely to report racism than Whites. Ethnic minority boys (except Indian boys) and Indian girls reported better psychological well-being throughout adolescence compared to their White peers. Notably, lowest mean TDS scores were observed for Nigerian/Ghanaian boys, among whom the reporting of racism increased with age. Adjusted for individual characteristics, psychological well-being improved with age across all ethnic groups. Racism was associated with poorer psychological well-being trajectories for all ethnic groups (p B0.001), reducing with age. For example, mean difference in TDS (95% confidence interval) between boys who experienced racism and those who did not at age 12 years 01.88 ('1.75 to'2.01); at 16 years 0'1.19 ('1.07 to'1.31). Less racism was generally reported in schools and neighbourhoods with high than low own-group density. Own ethnic density and diversity were not consistently associated with TDS for any ethnic group. Living in more deprived neighbourhoods was associated with poorer psychological well-being for Whites and Black Caribbeans (p B0.05). Conclusion. Racism, but not ethnic density and deprivation in schools or neighbourhoods, was an important influence on psychological well-being. However, exposure to racism did not explain the advantage in psychological well-being of ethnic minority groups over Whites.

Introduction Access to safe water, sanitation, and hygiene facilities is considered a basic human necessity for survival and well-being [1], without these basic needs, the health conditions of millions of people especially children are at risk [2]. However, 2.3 billion and 844 million people across the globe lack access to basic drinking water and sanitation facilities, respectively [3], causing 842,000 deaths every year [4], which is undoubtedly a major public health concern. WASH services are considered means of contacting and at the same time preventing diseases [5]. It has been estimated that overall 9% of the global burden of disease could be prevented through improvement in adequate WASH facilities [6,7]. Children are one of the most vulnerable groups affected by lack of water, sanitation, and hygiene facilities [2]. In developing countries, the high mortality rate resulting from diarrheal among children under the age of five was majorly due to WASH challenges [2,5]. e provision of safe WASH facilities has been greatly influential on people's health status and livelihood; however, the availability of these facilities remains critical in Nigeria especially in the rural areas [8][9][10][11][12]; a large percentage of rural communities in Nigeria live without access to safe WASH facilities [13]. e situation has thus subjected the communities to the utilization of water from rivers, ponds, and streams for drinking and domestic activities [14] and to the practice of open defecation [15] which consequently has often led to deaths, illnesses, and spread of waterborne diseases [16][17][18]. e few improved water facilities from boreholes and wells with hand pumps available are largely insufficient; women and children mostly travel far distances to access water, which is energy and time consuming, thus affecting children's education and women's household and economic productivity [19]. On the other hand, hygiene facilities such as excreta disposal have also been inadequate for usage at community households and public places such as schools [20], markets [21], and even hospitals [22], which left people with no alternative but to defecate openly and sometimes in and around water sources [23] with no use of soap or any cleaning agents for protection [24]. Moreover, toilet facilities available were poorly maintained and mostly shared among numerous people with no consideration of gender segregation and women integrity [21]. However, government failures have led to the intervention by organizations such as UNICEF and WHO for aid through their programs known as WASH. UNICEF's WASH team works in over 100 countries globally to provide water and sanitation facilities [25]. Knowledge, attitudes, and practices associated with WASH are of pertinent concern towards sustainable and effective implementation of WASH programs in communities [26]. KAP regarding WASH are contributing factors to waterborne disease prevalence in communities; poor WASH knowledge leads to unhygienic practices and poor attitudes which pollute water and spread illness [27,28]. Such inadequate WASH knowledge leads to wrong perception of quality of water resulting in large dependence on surface waters for drinking [14,29], open defection practices being perceived normal and commonly practiced, minimal household water purification practices to prevent diseases [30,31], and poor water collection and storage behaviors contaminating water and causing illnesses [32][33][34]. Household and environmental hygiene also tend to be poor, and children stool is often overlooked and perceived harmless in sanitation programs, hence increasing the risk of disease transmission [35][36][37], all due to limited WASH understanding and poor attitudes and practices towards WASH. erefore, there is a need to provide hygiene education programs and increased awareness towards promoting good WASH practices and ensuring good public health in the communities. In Nigeria, it is expected that there is currently a dearth of data on the status of WASH; thus, it is increasingly becoming difficult to plan any meaningful WASH program to improve health and well-being. erefore, this study aims at assessing the knowledge, attitudes, and practices related to WASH in 5 local government areas in Kaduna state, Nigeria, with a view to providing reliable and sufficient data for effective implementation of WASH programs and ensuring good public health. e KAP survey is a follow-up of an earlier nationwide survey on water, sanitation, and hygiene carried out by UNICEF, Nigeria, during the years 2007-2009. e study objectives are to assess the demographic characteristics of the study areas relevant to the survey, the various water sources for drinking and domestic activities, home water treatment methods and practices, water collection and storage attitudes and practices, knowledge of quality drinking water, excreta disposal systems used and related perceptions, personal and environmental hygiene, and water related diseases relevant to the study. LGAs were selected from 23 LGAs . From each LGA, one community was picked up randomly for the KAP survey. e communities are Gwagwada , M/Kajuru , Soba , Kachia , and Gukwu . --- Materials and Methods --- Sample Size and Survey Procedure. Scientific steps were followed to ensure the technical appropriateness of the survey sample size and sampling procedure. e following formula was used to determine the sample size [38]: n � D * • Z 2 /2 pN d 2 + Z 2 /2 p , (1 ) where n is the sample size estimate, D is the design effect, p is the proportion practicing hand washing as a hygiene practice, Z 1-α/2 is the standard score corresponding to 95% confidence level, N is the total population, and d is the degree of precision. Using hand washing as a proxy for the indicators, p was chosen to be 50% . e design effect was estimated at 2 because of the clustering in the target population, and the level of precision was set to 0.05. e total population of Kaduna state was obtained from the official gazette of the Federal Republic of Nigeria for the 2016 population census and used as N. After the application of the above formula, the estimate obtained for Kaduna state was adjusted for a nonresponse rate of 10%. A three-stage sampling procedure was adopted for selection of respondents. In Kaduna state, the IYS LGAs were 5 . e first stage was a random selection of 40% of the IYS LGAs in each of the state LGAs. e NPC population estimates of the selected LGAs were obtained from the 2016 census [39]. e proportion of the total population constituted by each LGA was computed. Sample sizes for each LGA were then proportionately allocated to the selected communities. e second stage involved the random selection of 40% of the IYS communities in each of the selected LGA. e sample sizes allocated to each LGA were equally allocated to the selected communities. Table 1 shows the breakdown of sample sizes for the state and selected LGAs and communities. e last stage was the random selection of street blocks based on the map of the selected community. Household heads were interviewed in selected clusters. --- Data Collection and Analysis. e study is a crosssectional field survey involving the use of structured questionnaire and field observation. A total of 854 questionnaires were administered and retrieved. e number was distributed across the LGAs as follows: Chikun: 236 ; Kachia: 156 ; Kajuru: 71 ; Soba: 188 ; and Zango/K: 203 . e study variables include sociodemographic characteristics, water sources, and KAP regarding household water treatment, water collection and storage systems, excreta disposal, and household and environmental hygiene. In order to support the questionnaire data, observational checklists were used to capture and ensure the practices of household compound cleanliness, excreta disposal, and storage systems. Data collection was performed and supervised by 3 trained and experienced environmental and public public health professionals with assistant from various trained field staff members and researchers in Kaduna state. e survey was closely monitored for data quality assurance. Results of the surveys were entered into EpiData and analyzed. --- Results --- Background Characteristics of the Study Population. Table 2 shows the background characteristics of the survey respondents of which the majority were males. About 81.8% of them have been residents for more than 10 years. Some 47.6% were aged 30-50 years, and about 88.8% were married. Some 80.2% reported that they have been to school. However, 36.1% of them had Quranic education, 20.4% primary schooling, and 24.4% secondary schooling, while 14.0% had postsecondary education. Hausa accounted for 39.7% of the ethnic groups; others accounted for 58.4% in the selected LGAs. Christianity is the most common religion of the respondents followed by Islam . Major occupations were farming , civil service , and self-employed , and unemployment stood at 3.4%. --- Sources of Water for Drinking and Other Activities in the Communities. As shown in Table 3, across the 5 LGAs, sources of drinking water include surface waters , unprotected hand dug wells , and protected hand dug wells . Rain water harvesting was practiced by 52.5% of the communities. Kachia and Zango/K collect rain water to a large extent, 85.8% and 63.9%, respectively. ese communities also utilize surface waters to a great extent, 89.0% and 95.1%, respectively. During dry season, unprotected hand dug wells and surface waters were the main sources. However, some still sought additional water from vendors. e sources of water for other domestic purposes were as follows: surface waters , unprotected hand dug wells , and protected hand dug wells . --- Water Treatment Methods Used and Communities Level of Practice. About 46.2% of the respondents claimed to have treated their water supply . e most common method was filtration through cloth . Other methods mentioned were boiling and chlorination, though very few as shown in Figure 2. ere is frequency of household water treatment in all the LGAs as 43.3% indicated that they treated water that day, 22.6% the previous day, 15.8% less than one week ago, 3.3% less than a month ago, and 5.7% more than a month ago, and very few could not remember the last time they treated water. --- Household Drinking Water Storage Practices, Handling Attitudes, and Knowledge of Quality Drinking Water. e facilities used for storing drinking water were mostly covered clay pots and covered plastic containers . Open containers were also common among 15.1% of the respondents. Plastic buckets with taps were used only in Kachia by 24.5% of the respondents. Items used in fetching drinking water from the storage facility included cup with handle , cup without handle , calabash , and bowl . ese items were common in all the LGAs studied. ese items were either kept on the storage container , kept in a basket or shelf , or hung . Cleaning of the storage facilities was done daily by 72.6%, weekly by 15.0%, and when dirty by 11.5%. Again, the frequency of cleaning varied widely across the LGAs. When respondents were asked about the qualities of safe drinking water, there were responses such as visually clear , free from germs , odourless , and sweet taste . ese views were consistent among the LGAs. In the communities, it was common for under-5 children to defecate around the house , in the toilet , and in the potty/chamber pot . Defecation around the house is most common in Zango . e use of potty/ chamber pot was more frequent in Kajuru and Chikun . After children defecation, the feces were dropped into a toilet facility , thrown into the bush , or eaten by dogs . Dropping of children feces into toilet facility was generally a common practice across the LGAs as shown in Table 7. --- Perception of a Good Toilet, Type of Toilet, and Preferred Ownership by the Population. e information on how the respondents perceive a good toilet, the type of toilet, 9. When asked about when is it important to wash hands, there was a consensus among the respondents from the LGAs that hands should be washed before meal , after meal , after defecation , and after cleaning children feces . After defecation, hand washing is practiced the most in Kachia followed by Zango . Items used for hand washing include soap with water , water only , and sand and water . A sizable number understood that personal hygiene means bathing , washing of clothes , cutting of hair , and cutting of nails . e knowledge was comparable among the communities. --- Household and Environmental Hygiene. Sweeping of the house , cleaning of kitchen , proper disposal of waste water , cleaning of toilets , and proper disposal of solid waste regularly are practices referred to as household and environmental hygiene by the respondents. is understanding of household/environmental hygiene is fairly and the lowest in Chikun . In order to prevent water stagnation, 81.0% said they would clear it, while 3.0% said they would divert it to farms. ere was however a consensus that community members should be responsible for prevention of stagnated water. --- Water Related Diseases Perceived in the Households and Communities. ere were reports of epidemics of water related diseases within the last year by 40.9% of all respondents. e common diseases are malaria , LGAs. Diseases of great concern are typhoid and diarrhea which were relatively more in Kachia and Zango LGAs. Yellow fever was reported to be high in Kachia , and no Guinea worm was reported in any of the LGAs. --- Structured Observations Findings. e results of the spot checks are summarized in Table 11. Presence of feces was reported around the house , inside the house , and near the water source . Kachia showed the highest presence of feces around the households , and all the LGAs showed high presence inside the households. Cow dung and animal excreta and children's feces were also seen around the premises. However, water sources were kept fairly free from the feces. Traditional pit toilets were the most observed. e observed features of the toilets are as follows: small enough hole , adequate privacy , safe floor , presence of slab , and having superstructure . Locations of the toilets were mostly outside the compound . In terms of the indicators for the current usage of the toilet, the following results were obtained: clear paths leading to it , cleanliness , being free of smell , and being free of flies . Hand wash facilities were located inside the house or within walking distance and next to the toilet . --- Discussion 4.1. Sources of Water for Drinking. Access to safe water supply is integral to health and survival [1]. In the study area, there is virtually total absence of improved source of water; thus, communities extensively utilize surface water and unprotected wells for drinking which can be infectious [12]; this is similar to many KAP surveys in developing areas [32,40] including a recent counterpart study in Kaduna [41], Nigeria, where the majority of surveyed communities utilized polluted water sources due to lack of adequate clean water sources. However KAP studies by Pradhan et al. [42] and Hothur et al. [20] depict otherwise where almost 73.6% of the households were consuming water from improved source of water; this disparity is not unconnected to lack of effective water policy and governmental commitments in the respective study locations; efficient water policy can be helpful extensively towards mitigating the high WASH related mortality and morbidity rate in Nigeria [13]. --- Home Water Treatment Methods and Practices. Household water treatment practices can improve dramatically microbial drinking water quality and prevent diseases [43]; it is found to reduce rate of diarrhea infection among children [31] and shown to decrease level of cholera outbreak and disease transmission among people [44]. In the study area communities' level of home water treatment practices was considerably low as 54.8% do not treat water particularly using efficient methods such as chlorination. e absence of home treatment practices is often consistent in many communities across developing countries in the world as shown by many studies [28,40,[45][46][47], being more pathetic in the rural areas [48,49]; it is perceived unimportant by the rural dwellers due to lack of education and awareness on WASH, which has continued to impact negatively on their health status. --- Storage System Practices, Handling, and Knowledge of Quality Drinking Water. Appropriate use of storage vessels and handling attitudes are vital to maintaining quality drinking water and preventing waterborne diseases at households [50][51][52]; uncovered water storage containers and those with wide openings make water susceptible to contamination; frequent cleaning of containers is essential in mitigating household water bacterial recontamination [33], and it is highly recommended as part of sanity that cups used to retrieve water from storage containers should have ladles or handles [46], to avoid unclean hands dipped into water in the process of fetching, and be kept on clean surface [40] or hung after usage. In the study area, different storage systems used are safe for storing water and are fairly covered and cleaned periodically; this is similar to findings by Reddy et al. [46], Pradhan et al. [42], and Ssemugabo et al. [53]. However the study population lacks knowledge of quality drinking water as many responded that quality water means visually clear water. is understanding affects home water treatment practices, consequently subjecting communities to diarrhea and waterborne diseases [31]. Contrarily, some similar studies show that water consumers are knowledgeable about safe drinking water [53]. --- Excreta Disposal Practices. e lack of improved toilets facilities was a huge concern in the study areas, resulting in open defecation being a common practice especially by children at households, which is a major health and social burden for the community at large [54]; this corresponds to various findings in developing countries [30,46]. Nonetheless, the respondents have positive perception regarding good toilet facilities and are willing to stop open defecation; however, it was reported that unavailability of improved toilets was due to lack of financial capability. Toyobo et al. [55] and Miner et al. [15] reported similar scenarios, where communities suffer WASH challenges due to lack of fund. Livelihood empowerment and poverty eradication in the rural areas are vital to sustainable toilet facilities in the area. --- Personal and Environmental Hygiene. In the study area adequate knowledge of personal hygiene led many to the good practices of using water and soap for cleaning hands after defecation, which is essential in preventing diseases [31], similar to studies by Orimoleye et al. [30] in Ibadan and Miner et al. [15] in Jos, Nigeria; practices of environmental hygiene were also fairly good as children feces were majorly perceived as harmful and were disposed in a toilet facility as reported by 63% of the respondents. However increased education is essential at critical times as children feces were observed in many households during inspection. 4.6. Waterborne and Water Related Diseases. Relating to the KAP on WASH in the study area, waterborne diseases of diarrhea, malaria, and dysentery seem to be prominent in the area; dysentery and diarrhea were more common among children under 5, relating to the findings by Yaya et al. [2] and Prüss-Ustün et al. [5] where leading cause of death under the age 5 in developing countries was diarrhea. Generally, the disease prevalence was low among the communities. --- Constraints, Challenges, and Strength of the Research. During the research exercise, we were faced with some constraints and challenges which can affect the quality of the exercise which are lack of efficient means of transportation, poor accessibility to some communities, noncooperative attitude of some few respondents due to bulky nature of the questionnaire, and inadequate time for the exercise; however, the strength is the general cooperation of the community and stakeholders. --- Conclusion and Recommendations e deficiency in knowledge, poor attitudes, and lack of practices of WASH, particularly with regard to home water treatment, use of unsafe water sources, and open defecation, are routes of exposure to waterborne infections associated with the study area. Effectiveness of WASH does not depend on facilities provision alone. erefore, WASH education is fundamental for promoting good practices and behavior towards WASH in order to protect public health. e study further suggests investigation regarding personal and environmental hygiene practices and the related disease implication in the study area. --- Data Availability e data used to support the findings of this study are available from the corresponding author upon request. --- Conflicts of Interest e authors declare that there are no conflicts of interest regarding the publication of this paper.

Access to safe water, sanitation, and hygiene (WASH) facilities is a basic necessity for human livelihood, survival, and well-being. Adequate WASH facilities provision is a critical issue to most developing countries around the world including Nigeria. Knowledge, attitudes, and practices regarding WASH are integral to effective and sustainable WASH facilities provision. is study assessed the level of knowledge, behavior, and practices towards water, sanitation, and hygiene in Kaduna state, Nigeria, with a view to ensuring sustainable WASH facilities intervention in the region. Data collection tools included spot check observation and questionnaire involving 854 participants, selected from five local government areas (LGAs): Chikun, Kajuru, Soba, Kachia, and Zango Kataf. From the results, major drinking water sources were surface waters (52.5%) and unprotected hand dug wells (44.8%); only 46.2% treated their water supply and few (16.6%) used chlorination method. Pit latrine toilets were the major (76.5%) excreta disposal means, and open defecation practices were widespread (41.4%). Level of personal and environmental hygiene understanding was fairly good in all the local government areas, and 65.4% claimed to use water and soap for washing hands after defecation. Incidence of water related diseases is generally low in the area. Despite the commendable findings in the study areas, communities are still at risk due to lack of safe water supply and poor practices of home treatment and excreta disposal. erefore provision of WASH facilities and WASH education is fundamental for ensuring public health in the study area.

Introduction Ours is a digital world. As such, we are almost entirely dependent on digital technologies for news, entertainment, communication, and commerce. On one hand, life in a digital world offers unparalleled convenience, connectivity, and access to information, potentially empowering us in unprecedented ways. On the other hand, it means that screens increasingly mediate social interaction, algorithms influence the nature of our experiences, and digital environments blur the boundary between the material world and a virtual one. In short, digital technologies pervade and shape nearly every aspect of our social lives. Thus, it comes as no surprise that digital technologies are transforming the nature of work. They are, for instance, encouraging entrepreneurialism, eliminating or streamlining repetitive tasks, enhancing productivity and efficiency, and enabling remote work, flexible schedules, and cross-cultural and interorganizational collaborations. But, perhaps less obviously, digital technologies are also giving rise to new models of leadership. For one thing, digital technologies make leadership-like virtually everything today-decidedly more public. Increasingly, leaders at every level-from heads of state to corporate CEOs to the line manager at a local Starbucks-may choose to conduct business and "lead" in full view of the public, largely on social media platforms. During Donald Trump's presidency, for instance, one rarely had to wonder what the president was thinking about or what his approach to a particular issue might be, as he consistently broadcast both on Twitter [1]. Similarly, one need not imagine what it might be like to work for Elon Musk, as his management of Twitter, which he recently rebranded as "X", has unfolded one tweet at a time right before our eyes. Donald Trump and Elon Musk, one a politician and the other the world's wealthiest person, are not random examples. As Robert Reich observes, "both represent the emergence of a particular . . . personality in the early decades of the 21st century: the wildly disruptive narcissist" [2] . But they share, we argue, more than a particular personality; they share an evolving style of management that is spreading rapidly in politics and business, as well as in educational and religious contexts. Our central goal in this study is twofold: to chart the contours of that style, which we have dubbed digital authoritarianism, and to illustrate how it operates through a critical case study. In terms of its contours, we specifically argue that digital authoritarianism is a unique combination of: the destructive behaviors and dysfunctional psychological traits of toxic leaders; the ideology of authoritarianism; and the underlying structural logics of digital technologies. In terms of our supporting case study, we have selected Elon's Musk takeover and leadership of Twitter. We made this decision because Elon Musk offers a particularly clear example of public corporate leadership, regularly making and communicating management decisions on his recently acquired social media platform. We refer to that platform throughout the essay as Twitter since we had mostly completed our study by the time Musk rebranded it as X. To facilitate these goals, our essay unfolds in four stages. First, we review and reflect on the relevant literature regarding toxic leadership, authoritarianism, and media ecology. Second, we propose a critical approach uniquely suited for investigating digital authoritarianism. Third, we analyze Elon Musk's leadership both of and on Twitter as an example of digital authoritarianism, highlighting the ways it remakes toxic leadership. Fourth, we discuss the broader implications of the spread of this management style and consider its damaging personal, organizational, and social consequences. Before proceeding, we wish to acknowledge that not everyone regards authoritarianism as damaging or dangerous. Both Donald Trump and Elon Musk have amassed vast followings of individuals who fervently believe that their style of management has led to predominantly positive outcomes. Here, we invoke the observation first made by Theodor Adorno et al. [3], who argued that a certain personality type exists that will find authoritarianism not only acceptable, but preferable for addressing societal problems. Bill Jones cautions that "there are foreboding signs of Adorno's warnings coming to pass in the US", as it increasingly abandons democratic norms [4] . Again, we recognize not everyone views this "authoritarian slide" as problematic or novel, and, in fact, one of the reviewers of this essay suggested that democracy itself may be a historical "aberration". Our own view is that-aberrant or not-democratic values are worth defending, and that to do so we must have a thorough understanding of how the features of our digital world amplify authoritarian strategies. --- Literature Review and Reflection Toxic leadership is new neither in practice nor in theory [5]. But it is, we contend, evolving in practice, which invites a rethinking of the theory. That evolution, we further argue, is a product largely of changes in our communication environment and the technologies that animate it. Before looking at how toxic leadership is changing in a digital world, we briefly review the history of the concept. As Steven Walker and Daryl Watkins recount, the political scientist Marcia Wicker introduced the phrase toxic leader in 1996 [6], though Jean Lipman-Blumen renewed interest in the topic and heightened its popularity with her 2004 book, The Allure of Toxic Leaders. In her book and elsewhere, Lipman-Blumen defines toxic leadership as, "a process in which leaders, by dint of their destructive behavior and/or dysfunctional personal[ity] characteristics inflict serious and enduring harm on their followers, their organizations, and non-followers, alike" [3] . Expanding on this definition, Asha Bhandarker and Snigdha Rai observed, a "leader can be considered toxic if [their followers are] physically or psychologically harmed by the leader's actions and it creates long-lasting impairment in the subordinates" [7] . As these definitions suggest, the scholarship on toxic leadership has tended to view it through the lens of its effects or consequences on followers. Similarly, the research has sought to understand why toxic leadership is compelling to many followers. As Jean Lipman-Blumen has noted, "followers not only tolerate, but . . . often prefer, and sometimes even create toxic leaders-in for-profit corporations, non-profits, government, even educational and religious institutions" [8] . People are attracted to toxic leaders, according to Lipman-Blumen, for six primary reasons; they appeal to deep psychological needs, ease existential anxiety, provide order in a chaotic world, and foster a sense of belonging, belief, and purpose. Charlice Hurst et al. proposed a seventh reason some employees follow toxic leaders, which is that they themselves show signs of psychopathy, and employees with high primary psychopathy are more likely to flourish than their peers under toxic leaders [9]. While much research focuses on the motivations of followers, scholars have identified a series of destructive behaviors and dysfunctional personality traits typical of toxic leaders [10,11]. Some of the key destructive behaviors in which toxic leaders engage include misleading, lying, undermining, stifling, silencing, demeaning, demoralizing, bullying, intimidating, coercing, marginalizing, scapegoating, disenfranchising, and favoring. In addition to these behaviors, toxic leaders also exhibit a series of related dysfunctional behavioral traits, including insatiable ambition, narcissism, self-aggrandizement, arrogance, and a lack of honesty, integrity, trustworthiness, transparency, empathy, and self-reflection. Summarizing these behaviors and personality traits, Birol Başkan writes, "There are three critical elements of this destructive style of leadership: an apparent lack of concern for the well-being of subordinates, a personality or interpersonal technique that negatively affects the organizational climate, and a belief by subordinates that the superior is motivated primarily by self-interest" [12] . Inasmuch as all these elements run counter to prevailing understandings of effective leadership, toxic leadership can be understood as incompetent leadership combined with abuse. Put another way, toxic leadership is the exercise of abusive incompetence in a leadership position, and thus, at least in the extreme, not really leadership at all. Due to the deficit of leadership behaviors associated with toxic leadership, toxic leaders lean heavily on authority to achieve their aims, which partially explains why toxic leaders are prone to authoritarianism. While some scholars regard authoritarianism as its own style of leadership [13,14], we posit that it is better understood as a mode of ruling rather than leading. Authoritarians, by definition, exert absolute authority and control and demand unconditional obedience [15]. Leaders , by contrast, do not exercise absolute authority. Indeed, toxic leaders typically do not embody all the destructive behaviors and dysfunctional personality traits of toxic leadership or enact them in all contexts [8], though the more of these traits a leader possesses and the more routinely they are enacted, the more authoritarian that leader is likely to be. In the political sphere, the spread of authoritarianism around the globe has accelerated in recent years, capturing the attention of commentators and scholars alike. In one worldwide assessment, Sarah Repucci and Amy Slipowitz note that the number of countries with declines in democratic indicators has far outpaced the number of countries with democratic improvements over the past 16 years [16]. The systematic violation of democratic norms in favor of authoritarian rule has dire consequences for citizens, societies, and international relations, and understanding these consequences will become increasingly important if current trends continue [17]. While authoritarianism shares commonalities with toxic leadership, as well as other recognized styles of destructive leadership [18], its central concern with rulership rather than leadership makes it distinctive. Because authoritarianism is premised on centralized power, authoritarian rulers often have limited or no accountability. In short, they are neither responsible for their decisions nor accountable for their actions. They also exercise far greater control over the flow of information, often eliminating any possibility for discussion, let alone dissent. Indeed, one of the key differences between toxic leaders and authoritarian rulers is that authoritarians do not, properly speaking, have followers; they have subjects, and those subjects have limited rights. Subjects may, of course, openly support an authoritarian ruler, even a brutal one, though this is typically done out of fear or favor and is tied to a cult of personality. Thus, while authoritarianism and toxic leadership are distinctive, we maintain that toxic leadership is evolving in the direction of authoritarianism. To understand why and how this is happening, we need to turn to the subject of media ecology. The central premise of this essay, as well as the field of media ecology generally, is that communication technologies are not merely tools in our social environment, but constitute the very social environment we inhabit, thereby conditioning our habits of mind [19]. Alternatively stated, media ecologists posit that the prevailing technologies of communication-the media forms that dominate at a given historical moment-shape and influence not so much what we know , but how we know . With respect to the topic of leadership, we are interested in two questions. First, how has the development and widespread use of digital technologies altered thinking ? In answering this question, we suggest that-among other habits of mind-some workplace leaders increasingly view employees as subjects to be ruled rather than followers to be led. Second, how have digital modes of consciousness or habits of mind altered the character and conduct of leadership? To answer the first of these questions, we review the relevant literature on the structural logics of digital media. Answering the second question is the central aim of our analysis. According to media ecologists [20], all media forms have relatively fixed and distinct structural traits. In other words, they are designed and engineered to operate in particular ways. These structural traits or biases mediate human experience through repeated use, effectively training us to process the world in certain ways. Scholars generally divide human history into three major periods [21,22]: primary orality, literacy , and the electronic era . Based on the structural biases of the prevailing media forms, each one of these eras gave rise to different modes of thought and expression. For instance, we know that the experience of our social environment and, subsequently, thought in the period of primary orality, which existed before the invention of writing, was aggregative, immediate, participatory, and situational, while during the print era it was more analytical, interior, isolated, and distanced [22]. Today, in a digital world, experience and thought tend to be more affective, connected, interactive, and simulated. Brian Ott argues that digital media has three defining structural traits: digitality, algorithmic execution, and efficiency. These traits, in turn, foster the logics of intransigence, impertinence, and impulsivity [23] . Intransigence, which arises from digital media's basis in binary code, trains us to see the world in simple, dichotomous, and dogmatic ways. Impertinence, which arises from the programmed nature of computers, conditions us to be insensitive and unresponsive to others, while impulsivity, which is related to the efficiency of microprocessors, invites us to act affectively rather than analytically. To these, we would add a fourth logic, publicity, as digital media ensure that we are chronically online. We offer a more detailed discussion of these traits in our analysis. In sum, Table 1 highlights the key traits of toxic leadership, authoritarianism, and digital technology characteristic of digital authoritarianism. --- Critical Approach Because we are operating within a critical research paradigm, we adopt a broad approach or "view" rather than a rigid "method". In this section, we outline our unique approach to studying digital authoritarianism. But before doing so, we need to review briefly why a novel approach is warranted. Traditionally, research on leadership in organizational communication and management has studied it independent of the technologies of communication that create the very social environment we inhabit. This oversight blinds us to the habits of mind specific to the social environment at a given historical juncture. In other words, while scholars have explored how technologies are used by leaders, they have generally ignored how oral, chirographic, typographic, analog, and digital cultures all foster different models or styles of leadership. But a digital culture is fundamentally different from an oral or print culture, and it produces different ways of thinking both by and about leaders. Take toxic leadership, which existed long before digital technologies dominated all aspects of our social lives. In a pre-digital world, an employee with a toxic boss had to endure their boss's destructive behaviors and dysfunctional personality traits at work, but those behaviors and traits were not on full display for the public. Nor were online trolls or sycophants-a few of whom perhaps work at the same organization-liking, praising, and encouraging those toxic behaviors publicly. A digital culture has changed this. A toxic boss today, who perhaps is particularly given to the dysfunctional personality trait of narcissism, enthusiastically posts on social media about the restructuring of their organization. Users -most of whom do not work at this organization and who have no way of knowing this decision was made in an entirely top-down manner that silenced and disenfranchised employees-praise the realignment. In this instance, the symbolic violence visited upon employees by the manner of the decision is revisited upon the employees by the social media post and subsequent responses. This phenomenon is evident, for instance, in the reorganization that took place at Twitter following Elon Musk's takeover. Moreover, the toxic leader, seeing and feeling affirmed by the public praise for their destructive behavior, is encouraged to engage in similar behavior again. But perhaps the most significant point about the preceding example is that it focuses mainly on the logic of publicity and the way digital technologies have made leadership more public in the digital world. As we highlighted previously in the literature review, the logics of intransigence, impertinence, and impulsivity, in addition to the logic of publicity, are all remaking the character of toxic leadership in the digital era. It is crucial, therefore, that scholars who study leadership attend carefully to the structural features of digital technologies and highlight the styles of leadership they foster. In response to these issues, our approach to studying digital authoritarianism is twofold: first, we examine how the structural logics of digital communication technologies influence the thinking and, consequently, the behaviors of toxic leaders; and second, we demonstrate the subsequent style of leadership through a critical case study, namely Elon Musk's takeover and leadership of Twitter. Musk is hardly the only example of this evolving style of leadership. But he is, we think, a particularly high profile and clear example of how digital media remakes toxic leadership. Given the extraordinarily public character of his leadership, we conducted our analysis based on publicly available data, attending to public accounts of Elon Musk's leadership as reported in news media and examining several of Musk's actual tweets as evidence of leadership practices and behavior. These data are attached as supplementary materials under the title, "Elon Musk Leadership Action Chart". We recognize that some readers may object to our approach on the grounds that it conflates an analysis of media coverage of Elon's Musk's leadership with an analysis of Musk's actual leadership. But from our perspective, such a critique fails to understand how digital media technologies have obliterated that distinction. Digital authoritarianism as a style of leadership is significant and warrants our attention precisely because it is mediated by digital technologies. Thus, any serious attempt to give an account of it would, out of necessity, require attention to its fundamentally mediated character. To put it another way, Elon Musk's "public" management of Twitter is his management of Twitter. One of our central arguments is that there is no longer a meaningful distinction between the two. --- Critical Analysis The core argument of this essay is that our digital world and, more precisely, the communication technologies that have given rise to that world are producing and proliferating a new style of leadership known as digital authoritarianism. This evolving style of leadership encompasses the destructive behaviors and dysfunctional personality traits typical of toxic leadership. But it subsumes those behaviors and traits to the prevailing logics of digital media, which collectively function to make toxic leadership more authoritarian. Basically, the inescapability of digital media and its structural biases are transforming toxic leadership into digital authoritarianism by enlisting its central features in service of publicity, intransigence, impertinence, and impulsivity. Given this transformation, we have two primary goals in this portion of the essay. Our first goal is to clarify the structural biases and attendant logics of digital media, paying particular attention to the ways those logics incorporate and remake the destructive behaviors and dysfunctional traits of toxic leadership. There is not, however, a simple oneto-one relationship between the four digital logics and various features of toxic leadership. On the contrary, each of the digital logics enlists a constellation of toxic behaviors and traits, and several of the logics enlist the same behaviors and traits. For instance, the toxic leadership trait of arrogance serves all four digital logics. Our second goal is to illustrate the operation of these logics in Elon Musk's leadership of Twitter by drawing on news media accounts of his management practices and analysis of his tweets. --- Publicity: Self-Promotion and Self-Interest in an Attention Economy If our digital world has a master logic, a logic through which all other logics pass, it is publicity. The logic of publicity combines self-promotion and self-interest with a near continuous desire to remain in the public eye. Like all digital logics, publicity is a consequence of the inherent structural biases of digital communication technologies, though it receives its fullest expression on social media platforms like Facebook, Instagram, YouTube, LinkedIn, TikTok, and Twitter. The development and widespread adoption of digital communication technologies have eroded the traditional distinctions between our personal, professional, and public lives. Increasingly, thoughts and actions exercised in private and professional settings are shared publicly, sometimes by choice and sometimes not. Either way, the increasingly public nature of our personal and professional lives is a consequence of two key structuring principles associated with digital media: easy access to the modes of production and digital tracking. First, unlike legacy media, whose ownership and control were concentrated in the hands of a select few, the ubiquity of digital media, especially mobile telephony, make anyone a potential purveyor of mass information. By decentering the modes of production, digital media turn users into producers, democratize the dissemination of information, and invite "sharing." Second, digital media, which "track" every keystroke, website, and user preference, condition us to unreflexively accept the reality of continuous surveillance, making us even less likely to police traditional boundaries between our personal, professional, and public lives. Simply put, the desire to be seen is deeply embedded in the structural logic of digital media and social media in particular. On these platforms, people curate and share carefully crafted versions of themselves and their lives, often highlighting achievements and successes. The culture of following, liking, and commenting on social media platforms, which is engineered to produce regular hits of dopamine, further incentivizes people to behave in self-promotional and self-interested ways. This raises the question of what happens when the digital logic of publicity infects a leader's habits of mind and, consequently, management practices? Because publicity reflects the master logic of the digital age, it tends to be expressed more clearly in the general management philosophy of a leader than in specific leadership actions, though there are instances when it is acutely evident in those as well. As such, it is not so much what Elon Musk tweets that exemplifies the logic of publicity but how and how much he tweets . As Linda Chong et al. wrote in The Washington Post, "Elon Musk is a Twitter superuser. He has tweeted more than 19,000 times since joining the platform 13 years ago. This year, he has tweeted an average of six times a day. At least 150 of those posts are about Twitter itself" [24] . Similarly, Kari Paul, writing in The Guardian, described Musk as a "prolific user" [25]. While the specific content of Elon Musk's tweets since taking over as CEO regularly introduced policy changes at Twitter, the need to announce them publicly on the platform appears to reflect a deeper psychological impulse. As Jesse Lehrich, co-founder of the advocacy group Accountable Tech, told Billy Perrigo at Time in April 2023, "Throughout his career, Musk has had an almost pathological need to promise grand visions and make himself the center of attention. He's very Trumpian in his need to capture media attention with constantly-shifting promises, which everyone in the media covers" [26] . Elon Musk not only routinely promises grand visions such as rebranding Twitter as "X", telling the world it will become the "everything app", but he also often makes those pronouncements with little warning or context, which heightens their surprise or shock value and ensures wider circulation. After all, the algorithms that govern social media platforms prioritize content that garners more attention, which encourages individuals to act in ways that attract a larger audience, and what better way to gain attention than to violate norms and expectations. Tomas Chamorro-Premuzic refers to this practice as "grandiose exhibitionism" in his 2023 book, I, Human, and suggests it is, "One of the key facets of narcissism . . . which is characterized by self-absorption, vanity, and self-promotional impulses and is especially well-suited to a world in which human relations have been transferred almost entirely to digital environments" [27] . Narcissism is, of course, one of the primary dysfunctional personality traits of toxic leaders, and "More than anyone else, narcissistic individuals feel the constant need to be the center of attention, even if the means to achieving this is to engage in inappropriate, awkward, or eccentric interpersonal behaviors" [27]. The impulse to be the center of attention animates virtually everything Elon Musk does. Indeed, one could say that publicity is Musk's leadership philosophy. Comparing Musk to Donald Trump, Robert Reich offered the following assessment in The Guardian, "Both are indefatigable self-promoters. Both are billionaires, but they are not motivated primarily by money. Nor are they fueled by any larger purpose, principle or ideology. Their singular goal is to imprint their giant egos on everyone else" [2] . In fact, many commentators believe that ego and self-interest are what drove Musk to purchase Twitter in the first place. "Elon Musk", Tomas Chamorro-Premuzic writes, "wasn't content with monopolizing so much attention on Twitter, so he offered $44 billion to buy the entire business . Adopting the logic of publicity as a guiding leadership philosophy naturally influences individual decisions, of course. One of the clearest examples of this with regard to Elon Musk's management of Twitter was his response "when his Super Bowl tweet performed worse than . . . president [Biden]'s" [28] . Writing for Vox, Shirin Ghaffary reported, "Musk, apparently livid because his tweets about the Super Bowl were getting fewer views than President Joe Biden's, flew to Twitter's headquarters and ordered engineers to change the algorithm underlying Twitters main product to boost his own tweets above everyone else's" [29] . While Elon Musk's embodiment of the digital logic of publicity may seem victimless, his alteration of Twitter's boosting algorithm as a result of that logic suggests otherwise. It highlights that Musk is motivated not by what is in the best interest of his employees and even less so by what is in the best interest of Twitter, but in his own self-interest. Leaders who make decisions based on self-interest make bad business decisions that harm both employees, often lowering morale, and the organization they serve, often hurting the bottom line. Musk's decision to purchase Twitter to feed his own ego caused him to grossly overpay for the platform and his self-interested business decisions after becoming CEO chased away many advertisers on Twitter and undermined stakeholder confidence. --- Impertinence: Dichotomous and Dogmatic Thinking The second logic associated with digital media is intransigence, which describes a habit of mind involving simple, short-sighted, either-or, inflexible thinking. Basically, impertinent thinkers lack cognitive complexity, which in the case of leaders prompts them to arrogantly propose reductionistic solutions to complex problems and to eschew responsibility for consequences by blaming and scapegoating others when things go awry. This logic arises out of digital media's most basic structural feature: binary code. In contrast to humans, who have traditionally processed their environment and everything in it as a continuous and interconnected stream of sensations, images, and words [30], "A computer thinks-if thinking is the right word for it-in tiny [discrete] pulses of electricity, representing either an 'on' or an 'off' state, a zero or a one" [31] . Repeated exposure to binary code, according to Brian Ott, "unconsciously urges us to divide our otherwise contiguous world into discrete units, units that are not only separate but also fundamentally opposed. In binary code, one and zero are opposing states; when the human mind tries to make sense of our analog world in this manner, it contributes to dichotomous, absolutist, and inflexible thinking" [23] . Just as the logic of publicity conditions us to want to be seen, the logic of intransigence conditions us to see in blackand-white. The prevalence of intransigent thought in digital environments is evidenced by the speed and ease with which issues-both serious and seemingly innocuous-become irrevocably polarized. The logic of intransigence is also evident in a wide range of Elon Musk's management decisions at Twitter, including his advocacy of free speech absolutism, his subsequent devaluing of content moderation on the platform, his reinstatement of Donald Trump's Twitter account based on an online poll, his removal of the legacy blue verified checkmark system [32], and his hostility toward research on the platform [33]. Indeed, the shortsightedness of these and other decisions led to a wide array of problems at Twitter after Elon Musk took over, not the least of which was advertisers abandoning the platform. As Clare Duffy reported for CNN in February 2023, "More than half of Twitter's top 1000 advertisers in September were no longer spending on the platform in the first weeks of January . . . After Musk completed his takeover of the company in late October, advertisers began to worry about the safety and stability of the platform given his plans to cut staff and relax content moderation policies" [34] . The decline in advertising revenue, which "Musk has admitted . . . is down by 50%" [35] , is significant because historically advertising "has made up 90% of Twitter's revenue" [36] . The clearest example of intransigent thinking by Elon Musk involves his advocacy of free speech absolutism and subsequent relaxing of content moderation. In a 5 March 2022, tweet, Musk proclaimed himself "a free speech absolutist", a statement the news media has circulated widely without objection from Musk ever since. In an interview for Time, Jason Goldman, a member of Twitter's corporate board from 2007 to 2010, described Musk's free speech absolutism as "naïve" for failing to understand the political and legal complexities of the issue [37] . But Musk did more than articulate this simple-minded position and dogmatically refuse to moderate it, he acted upon it. Specifically, "Musk", wrote Billy Perrigo, "fired many members of Twitter's platform safety team just days before the U.S. midterm elections, . . . removed bans on dozens of accounts including Neo-Nazis, and disbanded the platform's already-existing Trust and Safety Council" [26] . Even before implementing these changes, Elon Musk's declaration of being a free speech absolutist had prompted a proliferation of hate speech on the platform. As The Guardian reported in October 2022, "many began testing the limits of the site just hours after the billionaire took the helm. . . . dozens of extremist profiles-some newly created-circulated racial slurs and Nazi imagery while expressing gratitude to Musk. And researchers found a surge in new followers flocking to the accounts of high-profile rightwing figures in the 24 h after Musk took over" [38] . To combat the loss of advertising revenue created by his absolutist position on free speech, weakening of moderation standards, and subsequent proliferation of hate speech, Elon Musk decided to replace the legacy blue checkmark system, which functioned to verify users were who they claimed to be, with a paid checkmark system. But Musk, who had fired nearly half of the staff at Twitter shortly after being hired, had no plan to verify users who paid for the checkmark, which led to a host of difficulties. In his reporting for CNN, Brian Fung captured the chaos that ensued: Twitter users awoke Friday morning to even more chaos on the platform than they had become accustomed to in recent months under CEO Elon Musk after a wideranging rollback of blue check marks from celebrities, journalists and government agencies. The end of traditional verification marked the beginning of a radically different information regime on Twitter, one highlighted by almost immediate impersonations of government accounts; the removal of labels previously used to identify Chinese and Russian propaganda; and a scramble by the company to individually re-verify certain high-profile figures such as Pope Francis. [39] Elon Musk's management of Twitter consistently suggests that he sees the world in very simplistic, black-and-white terms. As such, he proposes preposterously reductionistic solutions to complex problems like content management. But he also lacks the self-awareness and reflexivity to take responsibility for his colossal missteps, choosing like many toxic leaders to blame others. In an interview for CNBC Make It, Harvard leadership author and expert Bill George told the outlet, "If you had to write a case study on an example of a really poor takeover of an organization, Elon Musk's takeover of Twitter would fit that perfectly well. . . . I don't think he understands social media" [40] . --- Impertinence: Insensitive and Inhumane Behavior The third logic of digital media is impertinence; it reflects a habit of mind that favors callousness over compassion and cruelty over caring. The logic of impertinence derives from the essential nature and function of digital media. Digital media technologies are unaware, unreflexive, and dispassionate machines . Their operating logic-their mode of information processing-is cold and calculated . As such, according to Henry Perkinson, digital communication technologies are incapable of ethics or morality [41]. These machines are not immoral; they are literally a-moral, meaning they are entirely without morality. In contrast to digital media, human beings are self-aware, reflexive, and qualitative decision-makers, equipping them with the unique capacity for ethics and morality. "So", queries Brian Ott, "what happens when the human mind tries to 'process' [and act on] information like a computer? If digitization and the logic of binary code foster an intransigent mind, one that is both dichotomous and dogmatic, algorithmic execution within a closed system habituates an impertinent mind, one that is increasingly insensitive and unresponsive" to the thoughts, feelings, and needs of others [23] . In short, digital media invite persons to become more machine-like. In the case of leaders, the logic of impertinence urges decision making that shows little concern or regard for what others think and feel. Much like authoritarianism, it rarely involves the solicitation of input and even more rarely alters course in response to feedback. It is rooted in an attitude of "I know best" and, consequently, does not hold itself as answerable or accountable to anyone. In keeping with this uncaring and unresponsive attitude, digital authoritarians enact a wide range of toxic behaviors in positions of leadership, including but not limited to stifling, silencing, bullying, intimidating, demeaning, coercing, marginalizing, and disenfranchising. These behaviors, along with others that demonstrate a lack of human empathy, manifest widely in Elon Musk's management of Twitter. The logic of impertinence was evident from the start of Musk's leadership both in the act of mass layoffs and in the manner in which those layoffs were carried out. As Johana Bhuiyan reported for The Guardian, "In the second week, nearly half of the company's workforce were laid off with little notice, prompting some to . . . file a class-action lawsuit alleging Elon Musk violated California labor law" [42] . Those who remained were, according to Shana Lebowitz at Business Insider, not treated much better: "Shortly after Musk took the helm, some employees received instructions to work 12-hour shifts, seven days a week, without being told whether they would receive overtime pay or time off, CNBC reported. At the same time, Musk started ranking employees against one another" [43] . Writing for Forbes, Bryan Robinson offered this assessment: "Experts on workplace leadership assert that so far Musk's leadership style is headed in the wrong direction. . . . Musk is treating people like collateral damage instead of human beings, forgetting basic human decency in the way he's handling the layoffs" [44] . Just a few weeks later, Eli Sopow rendered a similar judgment at The Conversation, noting that, "Musk's cold, impersonal approach to management and leadership is antithetical to . . . kinder, more humanistic approaches to work. Management approaches like Musk's threaten current business management practices that advocate for healthy, happy and engaged workplaces. Musk adheres to a mechanistic style of management that treats employees like cogs in a machine, rather than human beings" [45] (paras. [5][6]. While the improper handling of mass layoffs, setting of unreasonable work expectations, and pitting of employees against one another subsume a series of toxic leadership behaviors under a general management attitude of impertinence, Elon Musk is not above targeting individual employees with the same degree of insensitiveness and cruelty. Johana Bhuiyan reported that, "Musk publicly announced the termination of an engineer named Eric Frohnhoefer, tweeting 'he's fired' in response to Frohnhoefer's tweet correcting an assessment Musk made about why the site was so slow" [42] . Elon Musk also mocked a worker with a disability , tweeting on 7 March 2023, "The reality is that this guy did no actual work, claimed as his excuse that he had a disability that prevented him from typing, yet was simultaneously tweeting up a storm. Can't say I have a lot of respect for that". Likely trying to avoid a defamation lawsuit, Musk later deleted that tweet. These examples suggest that Elon Musk, who is "notoriously thin-skinned" [46], despite claiming to be a free-speech absolutist is, like Donald Trump, intolerant of any kind of criticism. Specifically, Elon Musk sued the Center for Countering Digital Hate , a nonprofit anti-hate research group that found hate speech had proliferated on the platform since he took over [47], banned journalists from Twitter who were critical of him [48], fired several employees who tweeted corrections to or countered things he has said on Twitter [42], and "in one case publicly called out a former employee's tweets about him saying that they were the result of 'a tragic case of adult onset Tourette's'" [49] . As Ellen Pao wrote in The Washington Post, "Musk . . . often punches down in his tweets, displaying very little empathy. He called a British caver who helped to rescue trapped young Thai divers 'a pedo guy' " [50] . While Elon Musk appears willing to bully, intimidate, and potentially fire anyone who is critical of him, he has demonstrated a particular insensitivity on matters of gender and diversity [51]. On 21 June 2023, Musk tweeted, "The words 'cis' or 'cisgender' are considered slurs on this platform." Two months earlier, Twitter had removed protections for transgender people from its hateful conduct policy. As Clare Duffy reported at CNN: Twitter appears to have quietly rolled back a portion of its hateful conduct policy that included specific protections for transgender people. . . . Twitter also removed a line from the policy detailing certain groups of people often subject to disproportionate abuse online, including "women, people of color, lesbian, gay, bisexual, transgender, queer, intersex, asexual individuals, and marginalized and historically underrepresented communities". [52] , and white heterosexual patriarchy. In fact, perceived threats to the hegemony of "whiteness" and "masculinity" in some spheres is likely a key contributing factor to the evolution of toxic leadership in the direction of digital authoritarianism. That perception was certainly central, as Brian Ott and Greg Dickinson noted, to Donald Trump's presidential victory in 2016. In their view, "The commonality between communication practices and communication platform [struck] a powerful emotive chord with [Trump's] followers, who [felt] aggrieved at the decentering of white masculinity" [1]. Basically, authoritarians' followers are drawn to speech and platforms where they can say anything they like without consequence . --- Impulsivity: Erratic and Unpredictable Decision Making The fourth and final digital logic is impulsivity. It is a habit of mind that favors the rash over the reasoned and the affective over the rational. The logic of impulsivity leads to erratic, irregular, unpredictable, and unreliable decision making. Paradoxically, this habit of mind arises from humans' desire-but inability-to replicate the guiding principle of digital media, namely ever-increasing computational power and efficiency. Digital communication technologies, which are powered by microprocessors, are engineered to be as efficient as possible. Simply put, they are designed to perform as many mathematical calculations as rapidly as they can. Each generation of microprocessor possesses greater computational power, meaning it is more efficient. While computers are highly efficient at receiving information, processing it, and executing commands, humans are comparatively less so. "That difference matters", explains Ott: because repeated exposure to the logic of efficiency formally invites humans to strive for greater efficiency. As humans attempt to mimic computer efficiency, they rely more heavily on instinct and affect. In short, as humans try to speed up their information processing and decision-making capabilities, they are less careful and rational and more impulsive and affective, which . . . undermines the quality of their decision making. [23] In management and leadership contexts, this habit of mind routinely results in impulsive and ill-considered decisions. Moreover, it regularly enlists the destructive behaviors of misleading, lying, and a lack of transparency along with the dysfunctional personality traits of arrogance, deceitfulness, and untrustworthiness to justify those decisions. After all, digital authoritarians do not see themselves as responsible for their decisions thanks to the logic of impertinence. Further, the decisions arising from the logic of impulsivity are not only rash and reckless, but they are also simple-minded and reductionistic thanks to the logic of intransigence. Collectively, the digital logics of intransigence, impertinence, and impulsivity lead to poor decision making, which leaders, paradoxically, regard as effective thanks to the positive user feedback that often accompanies the logic of publicity. The logic of impulsivity is a hallmark of Elon Musk's management of Twitter. As Walter Isaacson shared with The Wall Street Journal in an excerpt from his new book on the mercurial leader, "The way Musk blustered into buying Twitter and renaming it X was a harbinger of the way he now runs it: impulsively and irreverently" [53]. This habit of mind is evident in everything from his decision to purchase Twitter and subsequent attempt to back out of the deal [27]; to his ending of the legacy blue checkmark system and multiple restarts of a new paid checkmark system [32]; to his banning of linking to external social media sites and reversal of that decision [54]; to his limiting of how many tweets users can view and changing of that limit multiple times in a few hours [55]. The near instantaneous reversal and/or revision to these decisions highlights their impulsivity, leading journalists to routinely describe the situation at Twitter as "chaotic" [43], "chaos" [44,56], "chaos and confusion" [57], and "widespread chaos and turmoil" [45]. Since Elon Musk's leadership of Twitter is a product not simply of his decisions, but also how the media covers his decisions, we share an extended quote from Ryan Mac et al.'s assessment of Elon Musk's purchase of Twitter in The New York Times: To a degree unseen in any other mogul, the world's richest man acts on impulse and the belief that he is absolutely right. ... As Twitter negotiated a sale to Elon Musk last month, the social media company pulled out a corporate takeover playbook. Mr. Musk, the world's richest man, did the opposite. He had no plan for how to finance or manage Twitter, Mr. Musk told a close associate. . . . And when Twitter resisted his overtures, Mr. Musk pressured the company with a string of tweets-some mischievous, some barbed and all impulsive. [58] The preceding assessment is intriguing because it highlights that not only is Elon Musk's management style and decision making impulsive, but so, too, are his tweets. This is consistent with research that suggests "impulsivity" is a defining structural logic of both digital media generally and the platform of Twitter specifically. As Brian Ott explains, "Tweeting . . . requires almost no effort . . . It is ridiculously easy. Thanks to wireless technology, one can tweet from virtually anywhere at any time. Since tweeting requires little effort, it requires little forethought, reflection, or consideration of consequences. Tweeting is, in short, a highly impulsive activity" [59] . Because tweeting is impulsive, it invites ill-advised tweets such as Elon Musk's tweet perpetuating a "fringe conspiracy theory about the violent attack on Paul Pelosi" [60] . The danger of this tweet is not simply in its recirculation of misinformation, though it accumulated more than 28,000 retweets and 100,000 likes before Musk deleted it. It is also dangerous-from a leadership perspective-because Elon Musk has chosen to manage Twitter on/through Twitter, making it impossible to meaningfully separate Twitter CEO Musk from private citizen Musk. Once a leader engages in management behaviors on social media, all their behavior on social media becomes part of their public record of management. The implosion of management and nonmanagement spheres can, in light of the intersecting logics of publicity and impulsivity, become especially messy and fraught, as it did on 7 November 2022, when Elon Musk tweeted: "My commitment to free speech extends even to not banning the account following my plane, even though that is a direct personal safety risk". Musk's tweet referred to @ElonJet, a popular Twitter account that tracked and reported the movements of his private Gulfstream G700 jet in real time. Since the account regularly tracked short, 40-mile flights from San Franscisco to San Jose, it undermined Musk's "environmentally-friendly image" as CEO of an electric car company [26], which our previous analysis would suggest likely did not sit well with him. About four weeks following Musk's tweet making "a big deal about how he wouldn't ban @ElonJet because of his 'commitment to free speech'" [61] , Jack Sweeney, the account's owner, shared that an anonymous Twitter employee told him that Twitter had been shadowbanning @ElonJet, which severely limited its visibility. After Sweeney made this news public, Twitter "reversed course and reinstated @ElonJet's visibility" only to reverse the decision again and take the far more drastic step several days later of suspending the account altogether. This series of events captures well, we maintain, the dangers associated with the intersecting logics of digital authoritarianism. Elon Musk publicly and perhaps impulsively tweets something that he believes makes him look good: his supposed staunch commitment to free speech in a context involving him personally. But privately, he engages in management behavior that punishes a private citizen and non-employee for causing personal offense. When a whistleblower at Twitter exposes this behavior, Twitter temporarily reverses the shadowban to avoid the appearance of Musk targeting a private citizen. Then, after a few days have passed and people have hopefully moved on, Twitter suspends the account permanently. Authoritarians do not recognize any limits or bounds to their authority. Because they exercise their authority publicly, they regard the entire public as subject to their authority. This is what it means to say, as we argue in the implications, that the harms of digital authoritarianism are social as well individual and institutional . --- Critical Implications As our analysis of Elon Musk's management of Twitter illustrates, the destructive behaviors and dysfunctional personality traits of toxic leadership coalesce around the four digital logics of publicity, intransigence, impertinence, and impulsivity, thereby transforming toxic leadership from a mode of abusive incompetence into outright authoritarianism. In the final section of our essay, we reflect on the implications of that evolution, highlighting what it means for the perceived stability of traditional categories of leadership, as well as their treatment in academic literatures; the value and importance of attending to the changing nature of communication technologies in leadership research; and the individual, institutional, and social harms that accompany the spread of digital authoritarianism. --- On the Evolution of Toxic Leadership As our analysis demonstrates, styles of leadership are not static. Rather, they evolve over time to address our changing social environment. In this essay, we examined how toxic leadership is being remade by the transition from an analog age to a digital one. In an analog era, the impact of toxic leadership was limited largely to the organizations and employees who had toxic bosses. In that context, it made sense to theorize toxic leadership chiefly in terms of followership. But in a digital world, which obeys the controlling logic of publicity, toxic leaders increasingly enlist "publics" in their cause. As such, it no longer makes sense to approach toxic leadership, or any style of leadership for that matter, primarily through the lens of followership, as least not as that category has traditionally been understood since "followers" no longer refers exclusively or even principally to employees. Moreover, at the same time that the logic of publicity has collapsed the personal, professional, and public, the related digital logics of intransigence, impertinence, and impulsivity have gradually cultivated authoritarian tendencies in those who occupy positions of leadership. As P. D. Harms and Karen Landay have concluded, "Whether conceptualized as a set of values, attitudes, or a personality trait, authoritarianism is associated with a variety of characteristics, including dogmatic, inflexible thinking [,] . . . intolerance of and lack of empathy for [others, and] . . . lower cognitive abilities" [62] . Apart from impulsivity, it is striking how closely these authoritarian characteristics map onto digital logics. Additionally, just as it no longer makes sense to approach toxic leadership through the frame of traditional followership, we question the wisdom and appropriateness of continuing to treat authoritarianism as a valid and legitimate category of leadership in academic literature. Authoritarianism is obviously an important social phenomenon, and therefore worthy of academic study. But given that it does not involve anything resembling recognizable and agreed upon leadership practices, treating it as a valid style of leadership legitimizes its abusive management practices in business and nonbusiness contexts. At best, the concept of authoritarian leadership is oxymoronic, and, at worst, it is unethical. We want to acknowledge that there is a body of literature that argues "authoritarian leadership" is effective in some contexts. For example, scholars have documented its positive effect on employee motivation in some contexts and have argued the case of an "authoritarian advantage" in crises response [63,64]. Our concern is not with the conclusions of this scholarship, but with the terminology being employed. There are of course some contexts where authoritarian practices and behaviors are sensible and warranted. But referring to these contexts in terms of leadership suggests a level of follower choice and power-sharing that simply does not exist. Scholars across many disciplines recognize that words matter, and the words we use in our scholarship matter as well. In that spirit, we would slightly revise our own claim in this essay. Digital authoritarianism is not so much a new style of leadership as it is the transformation of toxic leadership into a form of rulership by persons in positions of authority. If we truly wish to take a stand against the growing authoritarianism in our world not just in political contexts but also in business, educational, and religious contexts, then we can start by passionately insisting that there is no context in which authoritarianism constitutes leadership. --- On the Importance of Media Ecology In her research on toxic leadership, Jean Lipman-Blumen stresses the significance of social change and the uncertainty and unease it creates as important factors contributing to the followership of toxic leaders [8]. However, she stops short of identifying the drivers of social change and the implications of those drivers, stating simply, "As humans, we are constantly bombarded by uncertainty, change, turbulence and crises" [8] . Social change and its consequences, however, are not unmappable. In this essay, consistent with media ecology, we suggested that prevailing media forms or technologies of communication are the primary drivers of social change because they create the environment that we inhabit. As Lance Strate explains, "our . . . modes of communication, our technologies and techniques, and inventions and innovations . . . have had a profound impact on human affairs, arguably greater than any other factor [,] . . . media . . . constitute the conditions we live under, the conditions we live in [,] . . . conditions that in turn condition us" [65] . Another media ecologist, Jack Goody, made a similar point decades earlier: "Systems of communication are clearly related to what [humans] can make of [their] worlds both internally in terms of thought and externally in terms of . . . social and cultural organization. So changes in the means of communication are linked in direct as well as indirect ways to changes in the patterns of human interaction" [66] . Consequently, scholarship on leadership needs to take seriously the insights of media ecology. While the focus of the present essay was on identifying and analyzing the influence of digital logics on toxic leaders and leadership, future research could employ this perspective to better understand the shifting terrain of followership noted earlier. After all, the digital moment and its attendant logics have created a dangerous and deleterious situation in which social media users, who often know nothing about how an organizational decision was made or executed, publicly celebrate and endorse it. In the case of digital authoritarianism, this problem is greatly magnified by the cult of personality surrounding figures like Elon Musk and Donald Trump, who can do no wrong in the eyes of their followers. In 2018, The Verge published an exposé on "Muskateers", which Bijan Stephen described as, "a vast, global community of people who revere the 46-year-old entrepreneur with a passion better suited to a megachurch pastor than a tech mogul" [67] . Muskateers are not "followers" in the academic sense of the term, or, at least, not as used in leadership and management contexts. They are, for the most part, not employees. Muskateers are followers in the literal sense of the term in that they "follow" Elon Musk on various social media platforms; but they are also followers in an ideological sense, meaning they are loyal to Musk in much the same way Trump's supporters are loyal to him: unconditionally. As such, they are always "at the ready" not only to defend him, but also to attack those who are critical of him. Wrote Bijan Stephen: The most vocal of those fans have an impact: they're an army of irregulars waiting to be marshaled via a tweet and sent on the digital warpath against anything Musk decides he doesn't like, the iron fist in Musk's velvet glove. They've become known for haranguing people they believe have crossed him, journalists especially, with relentless fervor. The attacks are standard social mediaera fare: free-for-all bombardment across social platforms by people who are not always vitriolic but who nevertheless barrage the perceived enemy with bad-faith questions. As is often the case on these platforms, if you're not a cis white man, the harassment scales proportionally based on how far you deviate from that perceived norm. [67] and white masculinity deserves closer scrutiny. As critical management scholars investigate that intersection, they would do well to be mindful of prevailing communication technologies and their attendant habits of mind. --- On the Individual, Institutional, and Social Harms of Digital Authoritarianism Digital authoritarianism reflects a uniquely dangerous evolution in toxic leadership, subsuming the destructive behaviors and dysfunctional personality traits of toxic leadership to the structural logics of digital media . This constellation is dangerous because it moves toxic leadership in the direction of authoritarianism, which inflicts serious and enduring harms not only upon persons and organizations as toxic leadership does [10], but also upon society in general. Given the breadth and depth of these harms, it is worth reflecting upon them at length. At an individual level, digital authoritarians heighten the "conflicts and emotional damage to their subordinates" created by toxic leaders [7] . Such damage takes a debilitating toll, not only on the careers of said subordinates, but also on their physical and mental wellbeing. Scholars have documented such psychological distress in the form of agitation, withdrawal, and loss of self-worth [7], as well as anxiety, fear, and depression [68]. As Birol Başkan indicated, "individuals within the organization [begin] to be treated as objects needed for the goal [. . .] instead of assets that facilitate the achievement of the organizational goal" [12] . One need not have more than a basic understanding of human needs to understand the harms of being treated as an object. At an institutional level, digital authoritarians negatively impact a number of key organizational outcomes. For example, authoritarian leadership negatively affects overall organizational performance [69], employee creativity [70], and turnover [71]. Perhaps more relevant to our arguments here, authoritarian leadership may also lead to counterproductive behaviors among subordinates. As Sunita Mehta and G. C. Maheshwari explained, "counterproductive behaviors tend to be attributed to perceived injustice by employees who retaliate by inflicting harm and producing systemic damage in an organization like sabotaging operations, providing inaccurate information, and being uncooperative to co-workers" [72] . At a societal level, digital authoritarians normalize incivility, undermine democratic norms and institutions, and even provoke violence. We want to concentrate on the last harm, which suggests potential links between leadership and violence. As the science fiction writer Isaac Asimov once observed, "Violence is the last refuge of the incompetent" [73]. Given that digital authoritarians are incompetent and abusive leaders who are interested only in power, they will do nearly anything to remain in power. Donald Trump's fomenting of an insurrection at the US Capitol following his defeat in the 2020 US election is a prime example. Importantly, while Trump did not engage in violence himself, he created a context in which his followers regarded violence as an acceptable and reasonable response. The link between a leader and violence need not be nearly so explicit and direct, however. Some followers, especially those in the public sphere who were not subject to the abusive behaviors of a toxic leader , may take it upon themselves to respond to critics of a leader in a variety of abusive and violent ways including online harassment, mobbing, and doxing. But there is yet another potential link between digital authoritarianism and violence, which is attitudinal mimicry. We worry that employees who engage in workplace violence may, in some instances, be taking their cues from years of toxic and abusive leadership in their organizations. If the leader of an organization consistently models impertinent behavior that suggests the ideas and opinions of others, but especially subordinates, do not matter, will employees begin to internalize attitudes that similarly show little regard for others? Moreover, if "attitudes" are, as Kenneth Burke, argues, "incipient acts" or "the first step towards an act" [74] , does an impertinent attitude increase the likelihood of workplace violence? More research on these potential links is desperately needed. --- Closing Thoughts As the preceding discussion stresses, the implications of digital authoritarianism could scarcely be more serious. Media ecologists have long recognized that our prevailing communication technologies shape and condition our habits of mind. That insight has historically been used to understand the broad social differences between various eras such as orality and literacy. But in this essay, we suggested that organizational and management scholars can also benefit from the insights of media ecologists by studying how our digital communication environment privileges the logics of publicity, intransigence, impertinence, and impulsivity. Specifically, we argued that these logics transform toxic leadership into a dangerous form of authoritarianism that places the narcissistic needs of a leader ahead of their organization, favors simple and inflexible decision making, demonstrates little regard for the wellbeing of employees, and produces erratic and ill-considered decisions. Only by understanding the structural biases of media, and then making critically conscious decisions regarding habits of mind, can we begin to envision ethical models of leadership in our digital world. --- Data Availability Statement: Not applicable. ---

Employing a critical approach typical of humanities-based research, this article investigates the changing nature of toxic leadership in our digital world. Drawing on the perspective of media ecology, which asserts that the prevailing communication technologies at a given moment create the social conditions that, in turn, condition us, the authors illustrate how the digital logics of publicity, intransigence, impertinence, and impulsivity remake the contours of leadership. Based on a critical case study of Elon Musk's public management of Twitter, which has subsequently been rebranded as "X", it is argued that the four digital logics transform toxic leadership into digital authoritarianism, an unabashed form of authoritarian rule. A concluding section of the essay explores the implications of this evolution for traditional categories of leadership; the importance of attending to communication technologies in leadership research; and the individual, institutional, and social harms of digital authoritarianism.

Introduction The article focuses on the relationship between women's impoverishment and the democratic process, comparing care relationships and the removal of women from decision-making power. The central objective is to research, using the deductive method, the imbalance in the dimensions of power in the face of the invisibility of unpaid domestic work, as well as the consequent lack of legal recognition. REVISTA JUSTIÇA DO DIREITO DOI 10.5335/rjd.v37i2.15275 JUSTIÇA DO DIREITO v. 37, n. 2, p. 158-177, Mai./Ago. 2023. 160 Thus, by critically analyzing it, it is recognized that among the methodologies for the nationalization of heteropatriarchal power is the foundation of this unequal situation and lack of rights in the duty of care, sometimes based on the discourse of love and sometimes on women's natural instinct. It should be noted that when it comes to women in the labor market, in slow and costly steps, the recognition of the need for legal protection was driven by the Consolidation of Labor Laws itself, which earmarks a specific chapter to deal with the Protection of Women's Work. Therefore, socially visible and legally important. However, the normalization of inequality and the endless accumulation of work is justified and convinced by strategies of need and care. However, this unpaid work does not generate direct financial resources, which justifies women's exclusion from choices and maintains the imbalance in power relations between the sexes. In this case, the Consolidation of Labor Laws is silent, i.e. the law does not regulate, consequently it does not discipline, it does not protect and it effectively contributes to invisibility and helplessness. However, in today's socio-political climate, it is well known that women's work is essential to the family income in most families. However, even in these cases, responsibility for domestic work rests with women. In this way, the growing need for income from women's paid work and their consequent purchasing power is leading to a historic moment of reflection on the legitimacy of care relationships from the point of view of decision-making power. The study proposed here is unavoidable, since it will analyze unpaid domestic work and its repercussions in terms of social structure and maintaining the imbalance in gender relations, as well as women's health and rights. At this point, the central objective is to critically reflect on the discourses that legitimize care, institutions and the order of power hierarchies. --- The invisibility of women's domestic work: power asymmetries and difficulty in accessing democratic processes To begin with, it is important to emphasize that the linguistic choice already reflects a change in social perception, since it was chosen to talk about --- 161 work and not domestic activities, housework or any other euphemism. When the terminology work is chosen, it immediately warns the most inattentive, whether culpably or maliciously, that it is something exhausting, with responsibilities, demands and burdens. In the inhospitable scenario of overlapping inequalities that affect women, another choice made here is precisely to use the term women, in the plural, because there are many of them, each with their own stories of oppression and overcoming, since they occupy different positions on the scale of inequalities. Therefore, the myth of the universal woman who was relegated to domestic work in the confines of the home is unacceptable, while it was up to men to do external, recognized, regulated and paid work. This discourse falls to white and middleclass women, while bourgeois women were responsible for managing the paid domestic work carried out by socially and economically vulnerable women, especially black women, who, in addition to the incessant tasks of the home and family, performed the same duties in other people's homes. 5 As a final point of clarification before going any further, it's worth noting that when we talk about women's domestic work, we'll simply use the term domestic work, because, as we know, domestic work is not identified as a social problem for men. In fact, taking out the garbage, washing and drying the dishes, correcting a school lesson are all said to be help by men themselves. It should be said that for many decades, with special reinforcement since the First World War, it has been established that women are born knowing how to take care of things, or even that it is an instinctive and natural condition, or that, in fact, women do it out of unconditional care and love. 6 Finally, there is something that women are the best at, in the meritocratic race of patriarchal convenience. By instinct, nature and experience, women are the best servants and caretakers of the human species and should therefore continue to be so alone. But why is this social position not sought after and contested, why is it removed from the gender struggle? --- The capitalist mode of production and domestic work In this sense, one of the most persistent problems affecting women's access to democratic decisions is the fact that domestic work is not considered work, at least not with the legal requirements and consequences that the market determines, generating the well-known multiple working hours. Women in all times and places have contributed to the subsistence of their families and also to creating social wealth. However, the emergence of capitalism takes place under conditions that are adverse to women. In the complex process of individualization inaugurated by the capitalist mode of production, they have absorbed social disadvantages of a double dimension, because at the superstructural level there was, and still is, an undervaluing of women's abilities, presented through myths such as male supremacy and female fragility. On the structural level, as the productive forces developed, women were progressively marginalized from productive and managerial functions, in other words, they were placed on the periphery of the production system. 7 The transition from one system of production to another involves the unequal distribution of burdens between sectors of the population. However, the capitalist mode of production not only makes the social division and hierarchy of power explicit, it is also very successful in justifying the actual or potential marginalization of certain sectors of society. This has led to the formation of social markers of difference and exclusion, with gender being a factor that has long been selected as a source of women's socio-political inferiority, interfering in the established competitive mode. The social elaboration of the natural sex/gender factor by capitalism creates the heteropatriarchal pattern and with it an unprecedented success on the legitimacy scale, since it is apparently the biological deficiencies of the female being that determine the imperfection in productivity. 8 v. 37, n. 2, p. 158-177, Mai./Ago. 2023. --- 163 Women thus take on the role of an obstacle to economic development, when in fact it is the capitalist system itself that puts obstacles in the way of women's full realization. The discourse of women's inferiority was and still is based on unquestionable grounds, given that nature is seen as the first and last source of all marginalization, and in order to avoid the confrontations that science would inevitably bring to prove the natural inferiority discourse wrong, the title of male guardianship and protection is added, without which the feminine would not be able to resist. Heteropatriarchalism/heterocapitalism is therefore established9 , which, based on the intersection between the social, the political and the economic, engenders the triad of male power that elides women from the processes of discussion and deliberation.10 This creates a feedback paradox, in which heteropatriarchalism constructs the false discourse of the inferior nature of women in order to legitimize its strategies of domination/submission, simultaneously placing itself in the position of protector, creating supply and demand like the market. At this point, it is important to note that multiple working hours affect all women, regardless of their social class, but to varying degrees. Unpaid work is for everyone, disabling discourse is for everyone and so is the difficulty of accessing spaces of power. Thus, there are the so-called matrifocal households, which are run solely by women without partners or whose responsibility is placed exclusively on them; women who occupy the labor market and who take on domestic work at the same time, and even those who outsource this mission to domestic workers, the latter being the last category and the target of stark contradictions.11 All three types of work have their main roles, and in all of them there is a combination of roles and responsibilities. However, domestic workers, who are the result of the needs of other working v. 37, n. 2, p. 158-177, Mai./Ago. 2023. 164 women, who hire them to replace all or part of their domestic duties, represent the worst case scenario of overt capitalist plundering, as they are not yet recognized as a profession and do not receive the recognition they deserve. The position of queen of the home12 does not represent decision-making power or domination in the socio-economic structure. This queen is a translation of the eternal Cinderella, but with the demands of youth, beauty, affection, cheerfulness and contentment. Domestic work has the magic of perfect invisibility, in other words, it's only noticeable when it's not being done. Thus, absolute success is most often identified as nothing accomplished and, in this sense, how can the "queen" be tired, sad and unhappy if she doesn't work? As Silvia Federici says: 13 It is precisely this particular combination of physical, emotional and sexual services that is involved in the role that women must play so that capital can create the specific character of the maid who is the housewife, making her work so heavy and, at the same time, so invisible. Domestic work plays a fundamental role in the functioning of society, as well as guaranteeing the well-being of the family and sustaining and ensuring traditional economic development. It involves a wide and diverse range of functions, such as cleaning, cooking, caring for children, adults and the elderly, household administration and shopping, all of which are essential for the healthy development and emotional well-being of a family and society as a whole. This work requires particular skills, knowledge and means of production for the realization of each of its products. The mission of motherhood includes: giving birth, breastfeeding, preparing food for different times of the day and according to the possibilities and peculiarities of each, hygiene, health maintenance, encouragement and participation in the development of the body, intellect, socialization and quality time to formally educate, play, and establish bonds of affection and protection, in order to enable the future with emancipated and secure adults. 14 With such high social and market expectations, women develop performativities that subdivide them into different characters15 , which are: the mother, the wife/partner, the daughter, the domestic worker, the market worker. All of them overlapping and with different demands and no breaks. These obligations can extend beyond regular working hours, without a clear break or adequate financial compensation. All of this is the responsibility of women. And why don't they denounce it? Why don't they object? Out of love, fear of blame and social criticism, fear of abandonment, but this reality is changing. Women have played a significant and challenging role at work throughout history. --- The sexual and social division of labor and the imbalance in power relations Women's efforts to enter the job market and gain financial independence mean that they have to take on a number of responsibilities. Hirata and Kergoat16 state that "the sexual division of labor is the form of the social division of labor that stems from the social relations of sex; this form is modulated historically and socially". It is important to note that there are inequalities in this division, both in relation to productive and domestic work. There is a hierarchy and valuation of each of these jobs, with domestic work being mostly carried out by women, which has been devalued and made invisible. However, without it, paid work would be impossible. There is also the racial dimension of labour relations which cannot be overlooked, since slave labour was heavily used in Brazil, assigning unequal and racialized roles in the sphere of productive and reproductive work. In this context, the spheres of production and reproduction were differentiated, with the work of social reproduction being assigned to women, which involves a wide range of tasks in the domestic world, and paid work, in productive life, to men. Thus, caring for children, the sick and the elderly, cleaning the house, preparing food, shopping, washing and ironing clothes and many other domestic chores became purely a family responsibility, and were carried out mainly by women without pay 17 . The sexual division of labour arose at the same time as capitalism and the organization of paid work would not have been possible without the existence of domestic work. Although there was a division of roles between the sexes before the emergence of the capitalist system, it was at the birth of capitalism that the productive and reproductive spheres were separated for the first time 18 . The inequalities to which women are subjected, who have to work more, unpaid, significantly increase female poverty. Worldwide, women account for less than 40% of total employment, but make up 57% of people working part-time, which reveals the intermittency of women's work. In addition, they earn 77% of what men earn and are still subject to interruptions in the world of work due to maternity, which, due to the informality characteristic of their participation in the productive sphere, leaves them more vulnerable and poorer. 19 The ILO states that in 2015, the gender gap in the employment rate reached 25.5 percentage points to the detriment of women, only 0.6 points less than in 1995 and that the percentage of opportunities for women to participate in the labor market in the world remains almost 27 points below the opportunities for men. 20 According to the Brazilian Institute of Geography and Statistics , the occupation level of women in formal or informal employment, in the 25-49 age group, living with children up to 3 years old, was 54.6%, while that of men was 89.2%. Those without young children, on the other hand, had an employment 17 BARCELOS, Mariana Silveira. A incorporação da família nos serviços de saúde: um debate a partir das concepções dos profissionais num hospital de alta complexidade. 2011. 120 f. Dissertação -Universidade Federal de Santa Catarina, Florianópolis, 2011. Women also have more schooling: in the population aged 25 and over, 15.1% of men and 19.4% of women had completed higher education in 2019. However, it is perplexing to note that this majority is not seen in the staff of the country's Higher Education Institutions , as only 46.8% of HEI professors are women. Furthermore, despite studying more, women occupied 37.4% of managerial positions and earned 77.7% of men's income. While the average monthly income for men was R$2,555, the average monthly income for women was R$1,985. 21 It is important to consider that the Maria da Penha Law itself -Law No. 11.340/2006 -when providing instruments to protect women in situations of domestic and/or family violence, provides in article 9, second paragraph, item II, for the maintenance of the employment relationship of the victim of violence, and does so by guaranteeing the possibility of the worker's absence from her work environment for a period of up to six months. Although it does not clarify how this leave of absence will occur, or rather does not present the regulations for the effective exercise of this right, the provision does exist. However, women who make care and unpaid domestic work their main occupation are not included in the protection mechanisms. Therefore, the main Brazilian legislation, a legal framework for the protection of women, fails to identify and recognize such frequent characters in the country's reality. In this context of unpaid domestic work, the patriarchy manifested in the capitalist structure has played a fundamental role in assigning and perpetuating domestic work to women, who have been socialized to take responsibility for maintaining the home and caring for the family, while men are encouraged to engage in lucrative and leadership activities in the external market environment. v. 37, n. 2, p. 158-177, Mai./Ago. 2023. --- 168 This model of the sexual division of labor reinforced the gender hierarchy, assigning a lower and less valued status to domestic work. 22 This historical relationship between heteropatriarchy and domestic work has contributed to the invisibility and devaluation of women's work, which is understood and exposed by the market as an activity of lesser importance, often neglected and not recognized as legitimate work and, consequently, not recognized as an activity worthy of remuneration or social status. This has led to the invisibility and overburdening of women and, above all, to an asymmetry of power, perpetuating gender inequality and limiting women's opportunities in other spheres of life, such as the democratic process. --- Care relationships and the removal of decision-making power: Women's political poverty and the urgent revision of the democratic deficit The political and economic relationship between domestic work and women's work represents the traditional sexual division of labor, in which women take on the role of caretakers of the home, children, the elderly, dependents and partners. Over time, social movements and legal progress have sought to make women's demands for rights visible and recognized in the challenge of material survival. All of the above demonstrates the many facets of the perversity of inequality. Devaluation and invisibility, the disproportionate overload of domestic work, the addition of paid work, as well as professional or academic responsibilities, result in an excessive workload, which can lead to stress, exhaustion and a lack of time for self-care. As a result, women are kept in the position of supporting actors in their own lives, but with the workload of protagonists and the social recognition of extras. The problem is of global magnitude and for this proportion to be transformed requires a joint commitment from women, social movements, governments and international organizations. The oppressed cannot be held solely responsible for oppression. --- Care and decision-making power: women's political poverty and reversing the democratic déficit The idea that women are naturally better suited to domestic work has been widely accepted and perpetuated over the centuries. However, the agenda of economic development, the goal of profit and the confusion between happiness and consumption, causes the fragility of social ties and the fluidity of relationships. This situation inaugurates and maintains the path of satisfaction until the next purchase and the programmed obsolescence to things and people. The bourgeois revolution ushered in a new relationship between time and space, where organizations are increasingly demanding greater productivity from their employees, forcing them to work at an intolerant pace. 23 In order to achieve the growing goals of the market and make unattainable profits in the enslaving ring of competitiveness, workers are identified as collaborators, the same dynamic of holding the oppressed accountable that is adopted in domestic work. --- As Federici points out: This fraud that is hidden under the name of "love" and "marriage" affects all of us, even if we are not married, because once domestic work is totally naturalized and sexualized, once it becomes a female attribute, all of us, as women, are characterized by it . 24 The maintenance of multiple working hours is legitimized in the discourse of care and under this essentialist mantle, naturalized and personified in the figure of the woman, it spreads throughout society as a safe and immutable space. "Nor can we deny the values that women have shown to make this care possible, but this must be recognized and universalized." 25 In other words, the ethic of care is part of that set of values massacred by the goal of economic development, which we must rescue and incorporate into social relations in every measure. However, we must reinterpret it in such a way that it becomes a model for relationships and the formation of bonds between human beings and between human beings and nature. It is not just a question of reclaiming the ethical value of care, but above all of recognizing the excessive demands that have fallen on women to date and better distributing this responsibility. Care needs to be re-signified, because in the model of domination imposed by patriarchal power, it represents the massacre of women's citizenship, since the Federal Constitution expressly includes the principle of the dignity of the human person as the center of radiation of the normative system, a legal provision which is in line with the Universal Declaration of Human Rights of 1948 and the Convention on the Elimination of All Forms of Discrimination against Women of 1979. Thus, by assuming sole responsibility for multiple working hours, women are primarily fighting for survival in material terms, being forcibly deprived of political decision-making power, consigned to the status of a mass of maneuverers and living under layers of multiple dependencies and without building their own life project. This situation Demo26 points to women's political poverty, since poverty is not only linked to the deprivation of material goods, but also to the deeper condition of the destruction of the subject of rights, i.e. the removal of decision-making power and thus the ability to participate in democratic processes, which leads over time to the permanence of the current conditions of submission and socio-political atomization. "As people are unable to manage their own destiny, they are confined to the will of others, and this is all the harder as the final regulator of everything becomes the capitalist market." 27 The phenomenon of material poverty does indeed fall more heavily on women. In addition to the reasons already given, it is well known that despite higher levels of schooling, women do not enjoy the same proportion of the best opportunities to enter, remain in and move up in the job market. This is due to the avalanche of responsibilities imposed on them, which make it difficult for them to access qualified time for dedication, mental and emotional health and also the lack of an equitable distribution of tasks and, above all, the power to decide their own destiny. Thus, when gender inequality and inequality of access to material goods and social rights overlap, there is the complexity of women's poverty, the direct result of the sum of political and economic deficiencies. As Demo says: "Being a woman is a brutal 'hardship'"28 . The fact is that much progress has been made in uncovering domestic violence against women and the Maria da Penha Law has played a central role in recognizing gender inequality and not accepting the normalization of this type of aggression. However, structural violence against women is still maintained as normal and inevitable. In this sense, it is important to say that this violence is normalized in society due to the culture that for years held that men were superior to women. Given this context, it is necessary to understand what this violation of rights actually is. The Federal Constitution, in its article 7, item XX29 , defines the "protection of the through specific incentives, under the terms of the law". In addition, as already mentioned, the Consolidation of Labor Laws, in chapter III, starting with article 372, regulates the protection of women's work in detail. It is important to note that article 373 defines that women's working hours are identical to men's, with a weekly limit on working hours, based on article 7, XIII, of the Constitution. In this regard, the labor reform prevented employers from removing the guarantee of higher pay for night work than for day work, in article 611-B, VI, of the CLT 30 , preventing the customary practice and dictating that negotiation cannot be above legislation. The legal basis described above reinforces the conclusion that paid work is the object of state protection, but it does not shed any light on the reproduction of concepts, conclusions and principles for unpaid work. Labor and constitutional legislation remains restricted to the labor market, failing to recognize the importance and necessity of domestic work, maintaining it as a territory of contemporary enslavement and violation of the precepts of the Democratic Rule of Law. It is clear that implementing public policies to redistribute income is fundamental to rebalancing power relations between the sexes. There can be no talk of political participation, decision-making power and social transformation without respect for fundamental social rights. However, economic balance is not enough when it comes to women. The intensity of inequality is much more serious, complex and sarcastic than that. It necessarily involves social relations of care, which are a fundamental ingredient in the desired social transformation, paradigmatic in the task of building and strengthening bonds shaped by empathy, solidarity and balance in power relations. "It is essential to find new values, principles and legal elements to reevaluate care, without essentialist connotations, and to place the person at the center of rights and policies." 31 In no way are we trying to deny the values of care, but rather to universalize responsibility for it. It is impossible to deny the high social demand for care imposed on women and to recognize that this phenomenon implies reviewing the logic of the market and the distribution of power, especially the possibility of making decisions. There will necessarily be a reallocation of resources, because it is necessary to build and implement different public policies, especially in terms of rights and education, but not only that, it is also essential to incorporate a feminist perspective that values the importance of reproductive values. 32 Today we know that mere economic growth is not enough to combat poverty, as a lack of material resources,33 the system of domination of capitalism, identified in the figure of the white, Western man, is restricted to organizing and reorganizing budget leftovers for policies that silence social struggles. The path to transforming power relations and changing women's democratic deficit, however, lies through truly redistributive public policies, which promote a reckoning in the social markers of inequality and allow women to be political and able to debate and discuss. Political debate requires willingness and qualified time, and is necessary for the development of political awareness and the construction of decisionmaking power to move the paths of destiny. Reversing the democratic deficit is an indispensable strategic fact for equal rights, the real exercise of decisionmaking power and the transformation of society. It is therefore necessary to think systemically, making the construction of a new constitutional pact a priority, starting with the re-signification of care to enable participation in democratic processes in conditions that are feasible for exercising decision-making power. --- Final considerations To the extent that patriarchy alone regulates society, the sociopolitical impasse worsens and shows that the autophagic structure that remains regulates an impossible socio-political model, since exclusion and inequality become absolute rules. Growing material poverty is real. However, the most acute problem lies in the feeling of powerlessness. A rebalancing of power relations, where the socio-political determines the economic and human rights takes precedence over the market, seems impossible in the eyes of everyone, especially everyone. This feeling of atomization is due to the removal of democratic processes and, consequently, of decision-making power. The death of the political subject due to the asphyxiation of rights viscerally affects women, who are suffocated in the duties and responsibilities of care with perverse exclusivity. The invisibility of the violence resulting from the omnipresence of the social markers of inequality as a structuring methodology of socio-political relations is striking and has long defined the structures of power and domination. The moment calls for reflection, the identification of the problem of women's material and consequently political poverty and the necessary and urgent denunciation. The unveiling of gender inequality comes in the concomitance of paid and unpaid work relationships developed by women and the inescapable conclusion is that the maintenance of heteropatriarchy is largely due to the naturalization of domestic work, which is always unfinished and invisible, as well as unpaid, clad in the ethical discourse of care. The indignity of simultaneous working hours leads to women's material and political poverty, as well as their distancing from instances of power and democratic processes of discussion and deliberation. This absence of women's politicality is a successful strategy for maintaining the paradigms of inequality. Therefore, it is essential to reverse women's democratic deficit and build their political protagonism in order to create a new constitutional pact based on a new ethic of care. It should be noted that the research does not deny the value of care, just the opposite, because in a society marked by the predominance of economic power, it is essential to strengthen the social bonds of solidarity and empathy. However, care also needs to be redistributed and recognized not as an instinct or determined by biology, but as something learned and developed socially and as something that can be reorganized. Material poverty is a reality in the contemporary world, where the poor are exponentially more people than the rich, and economic and social inequality is a notorious fact. However, it is still worth noting that among the poor, women are the protagonists and therefore the supporting and peripheral actors in REVISTA JUSTIÇA DO DIREITO DOI 10.5335/rjd.v37i2.15275 JUSTIÇA DO DIREITO v. 37, n. 2, p. 158-177, Mai./Ago. 2023. 175 democratic participation and political decision-making. This situation needs to be reversed so that democracy ceases to be a discourse and becomes a daily practice. The path to be followed involves recognizing gender inequality as a preponderant factor for democratic transformation. In other words, it implies recognizing that women's political poverty stems from material poverty and social inequality. Next, the socio-political visibility of women's multiple working hours, the redistribution of responsibilities and, as a matter of urgency, the legal protection and regulation of unpaid domestic work. Legislation does not have the power to automatically transform lives, but the visibility, regulation and legal protection of unpaid domestic work will be decisive in rebalancing power and breaking down the social model imposed on the basis of gender inequality. Standardization will therefore reveal that care is indispensable, as one of the core values of the Democratic Rule of Law, and as such needs to be exercised by all people and not just, or mainly, women. The resignification of care will have repercussions on the quality and diversity of democratic participation and, consequently, better and more appropriate access to decision-making power, with easier identification of demands and redistribution of public policies.

O artigo apresenta como problemática central o impacto do empobrecimento feminino no processo democrático, analisando criticamente as relações de cuidado e o afastamento do poder decisório. Objetiva, por meio do método dedutivo, pesquisar o desequilíbrio de poder e a invisibilidade do trabalho doméstico não remunerado, bem como seu consequente não reconhecimento por parte do Direito, para tanto delimita o estudo na invisibilidade do trabalho doméstico feminino; as assimetrias do poder e a dificuldade de participação na vida democrática; o cuidado e o afastamento do poder decisório e, por fim a pobreza política das mulheres e necessidade de revisão do déficit de

Introduction Since emerging in China in December 2019, the beta coronavirus SARS-CoV-2 has been expanding rapidly throughout the world [1]. On 30 January 2020, the World Health Organisation declared coronavirus COVID-19 to be a Public Health Emergency of International Concern [2]. Thereafter, in March 2020, the WHO declared COVID-19 a pandemic due to the identification of more than 118,000 cases in 114 countries [3]. Millions of people worldwide have been infected, and hundreds of thousands of people have lost their lives due to COVID-19 [4]. Different regional approaches related to the use of face masks to mitigate the transmission of COVID-19 have been developed. In East Asian countries, for example, wearing masks was ubiquitous and was performed as a hygienic habit due to past positive outcomes in 2003 during SARS. On the contrary, in Europe and North America, the population was informed that masks were not recommended for general use [5]. The WHO states that masks can be used either for protection of healthy persons or for infection control [6]. Ma et al. showed that N95 masks, medical masks, and even home-made masks could block at least 90% of the virus in aerosols [7]. From the perspective of disease spread, at the population level, wearing masks by infected individuals may be important in helping retain contagious droplets, aerosols, and particles that can infect others and contaminate surfaces [5]. The WHO notes that health workers and caregivers in clinical areas must continuously wear medical masks where there is known or suspected community transmission [3]. However, due to the lack of robust evidence in clinical trials, the WHO's recommendations about wearing masks by the general population has been ambiguous. In its interim guidance issued on 5 June 2020, the WHO advises that to effectively prevent COVID-19 transmission in areas of community transmission, governments should encourage the public to wear masks. Universal masking, as a public health intervention, would probably intercept the transmission of COVID-19. This could especially be the case for asymptomatic infected individuals with high viral load at the early stage of the disease [8]. Therefore, community-wide mask usage irrespective of symptoms may reduce the infectivity of silent asymptomatic individuals. Masks are helpful for source control of asymptomatic infectious persons but also for protecting healthy people [9]. Universal masking may become the default solution in high-risk areas with a large number of patients and without sufficient testing, where everyone can only be seen as potentially infected [10]. High compliance in wearing a mask is a crucial factor for stopping transmission. This is similar to vaccines: the more people that are vaccinated, the higher the benefit to the whole population, including those who cannot be vaccinated, such as infants or immune-compromised people [11]. Moreover, the gain is greater the earlier masks are adopted and when face masks are used to complement other measures such as social distancing [12]. Hand hygiene is a discontinuous process and sometimes difficult to practice in the community. However, wearing a mask is a continuous form of protection to stop respiratory droplets to and from others [13]. Thus, controlling harms at the source by wearing masks is at least as important as other mitigation actions such as handwashing. Universal masking would also help in removing stigmatization that could discourage symptomatic patients to wear a mask in many places, preventing any discrimination that might arise. However, various authors have justified not wearing masks for different reasons. One of the key reasons is that there is limited evidence supported by clinical trials on the effectiveness of masks [14]. Secondly, it is claimed that prevention depends on an individual's behaviour and compliance, which has been shown to be inconsistent or inappropriate in trials, for example, people may repeatedly touch their mask [6]. It is also claimed that wearing a mask might make people feel safe and hence reduce adherence to other nonpharmaceutical measures such as hand washing and social distancing [5]. Moreover, at one stage it was also argued that because of the shortage of masks, the public should not wear them because healthcare workers would need them more and that the public buying masks could lead to major supply chain problems along with an increase in prices [15]. In this context, the aim of this study was to look at potential for Twitter to highlight public views towards universal masking. Social media is a useful platform for raising awareness of various issues and Twitter is a valuable platform for listening to public views and opinions on a range of topics in real-time. Moreover, from a public health standpoint, it is important to develop an understanding of the drivers of the discussion around masks and to gain insight into key topics of discussion. More specifically, the research questions of this study were as follows: 1. What was the overall network shape of the discussion on Twitter? 2. What were the key hashtags? --- 3. Who were the most influential users? 4. Who were the most mentioned users? 5. What were the key themes of discussion that were taking place? --- Methods --- Tweet Sampling Twitter data were collected using the keyword "mask" from 27 June 2020 21:21:21 to 4 July 2020 19:50:40 to provide coverage of about a week of data in a period when this topic was highly present in social media. The total number of tweets which were gathered were n = 452,430. By using the keyword "mask", tweets including words such as "masks" or "#mask" were also retrieved and included. A systematic random sample of 1% of the tweets was extracted and analyzed using social network analysis, as described in the next section. --- Social Network Analysis The software NodeXL was used to conduct a social network analysis of the data [16]. In understanding the network graph, the results of this study build upon previous research [17,18], which has highlighted that Twitter topics may follow six network shapes and structures [19]: broadcast networks, polarized crowds, brand clusters, tight crowds, community clusters, and support networks. In the network graph provided in Figure 1, circles represent individual Twitter users and the lines between them represent connections such as mention and reply. The network graph was laid out using the Harel-Koren Fast Multiscale layout algorithm which is built into NodeXL. NodeXL uses the Search Application Processing Interface . Influential users were identified using NodeXL and were anonymized by providing a description of the account in line with previous research [18]. A specific subanalysis was performed for tweets that originated solely from the USA. Regional information was extracted as follows: a total of n = 13,265 tweets were extracted where users had included "USA" in their user bios and a 5% sample of tweets was extracted and analyzed in NodeXL. Individual users and/or organisations that were deemed to be not sufficiently in the "public domain" were anonymized. 5. What were the key themes of discussion that were taking place? --- Methods --- Tweet Sampling Twitter data were collected using the keyword "mask" from 27 June 2020 21:21:21 to 4 July 2020 19:50:40 to provide coverage of about a week of data in a period when this topic was highly present in social media. The total number of tweets which were gathered were n = 452,430. By using the keyword "mask", tweets including words such as "masks" or "#mask" were also retrieved and included. A systematic random sample of 1% of the tweets was extracted and analyzed using social network analysis, as described in the next section. --- Social Network Analysis The software NodeXL was used to conduct a social network analysis of the data [16]. In understanding the network graph, the results of this study build upon previous research [17,18], which has highlighted that Twitter topics may follow six network shapes and structures [19]: broadcast networks, polarized crowds, brand clusters, tight crowds, community clusters, and support networks. In the network graph provided in Figure 1, circles represent individual Twitter users and the lines between them represent connections such as mention and reply. The network graph was laid out using the Harel-Koren Fast Multiscale layout algorithm which is built into NodeXL. NodeXL uses the Search Application Processing Interface . Influential users were identified using NodeXL and were anonymized by providing a description of the account in line with previous research [18]. A specific subanalysis was performed for tweets that originated solely from the USA. Regional information was extracted as follows: a total of n = 13,265 tweets were extracted where users had included "USA" in their user bios and a 5% sample of tweets was extracted and analyzed in NodeXL. Individual users and/or organisations that were deemed to be not sufficiently in the "public domain" were anonymized. --- Results --- Top 10 Hashtags Used Table 1 below provides an overview of the most used hashtags during this time, showing that "wearamask" and "maskssavelives" appeared among the most used hashtags. The hashtag "kidlitformasks" referred to authors of children's books who used this hashtag to highlight the importance of wearing masks. Twitter users also shared images from a Japanese manga series "haikyuu" and at the same time encouraged others to wear a mask using the "#haikyuu" and 'hq' hashtags, and these appeared as the 7th and 8th most popular hashtags that were used. As the discussion during this time was global in nature, there also appeared to be relevant hashtags from around the world such as "マスク" which refers to "mask" in Japanese. The word "ビール" also appeared, which is the Japanese word for "beer". This was because users conversed about a humorous mask invention which allows a user to store beer inside the mask which can be consumed. The hashtag "andhrapradesh" appeared as a popular hashtag as this referred to a state in the south-eastern coastal region of India and news reports were shared at this time related to reports about the death of a young person who was allegedly "beaten to death" by police for not wearing a mask. --- Overview of Network Structure Figure 1 is a social network graph of the discussion taking place during this time. The largest group is an isolates group which contained users who sent tweets that did not contain mentions. Overall, the group resembled a community network shape because there were many groups of users conversing about this topic. NodeXL clustered users into different groups based on mentions. A community cluster indicated that many users were talking to each other across several groups. It was also interesting to see influential users scattered around the network, which indicated that the topic brought in a wide range of influential actors. --- Top 10 Users Ranked by Betweeness Centrality Table 2 highlights Twitter users that were influential during this time. The users in the network were ranked by betweenness centrality, which is a measure of centrality. These users would have acted as important bridges within the network. The follower's column refers to the amount of followers each user has. Many of the influential nodes within the network derived from ordinary citizens who became important bridges in the network. A number of popular culture figures also appeared among the most influential users within the network. --- Top 10 Users Most Mentioned Table 3 provides an overview of users that were most mentioned during this time which ranged from political figures, organizational accounts, and popular culture related accounts. Certain users may not have tweeted about masks but were mentioned frequently by other users. Examining the influential users from this time highlighted that the discussion had been focused in and around the United States. However, it must be noted that not all account mentions may have been relevant to medical masks, as the band Slipknot are known to wear non-medical masks. --- Content Analysis NodeXL was used to identify the most frequently occurring co-words, as shown in Table 4. These co-related keywords provided insight into the types of discussions that may have been taking place on Twitter. Word pairs containing mentions of Twitter user handles were removed. This occurred on three occasions in group 3. 'Wear, mask' 'see, someone' 'wearing, mask' * asterisk placed by authors in expletives. The most frequent co-related keywords were "wear" and "mask" to form the sentence "wear a mask" and, in some cases, also involved an expletive. There were also tweets related to humor. There were also co-occurring words such as "stay" and "home". Other co-occurring words which appeared across the clusters included "breathing" and "problem", which revolved around the debate of whether face masks should be mandatory as they may lead to breathing problems. In regard to this debate, Twitter users provided evidence to highlight that face masks did not cause breathing problems, whereas other users highlighted the potential for masks to cause breathing issues. --- Regional Analysis of the USA In this part of the analysis, tweets were extracted to only focus on the USA. Data were filtered by users who noted in their bios that they were from the USA. --- Top Word Pairs Table 5 provides an overview of the 5 most popular words that were used from users from the USA. The most frequent words used together included "wear mask", "wearing mask", "mask, public", "face, mask", and "breathing problem". The first two co-words appeared to be encouraging the use of face masks and the third and fourth most used co-words appeared to be centred around general discussions around the use of masks. The fifth most used co-word appeared to relate to discussions around whether masks could cause breathing problems. Overall, there appeared to be overlaps between phrases and words used in tweets when filtering specifically for the USA, compared to analyzing tweets overall. Furthermore, content such as top hashtags and users appeared to be similar to that of the results of the analysis overall. --- Discussion The overall shape of the network resembled a community as there were a range of users conversing amongst each other in different clusters. When examining the most frequently used hashtags, it appeared that the most popular hashtags encouraged mask wearing among the public. It was found that a range of accounts were influential and/or mentioned, ranging from ordinary citizens, politicians, and popular culture figures. The discussion had been politicized by some users on Twitter, which led to politicians appearing as influential users within the network. Japanese hashtags also appeared within the network, which highlights the global nature of the discussions around this issue. It is important to note that some of the accounts such as the official Twitter account of YouTube might not have tweeted using the word 'mask', however, users may have shared content which contained the word 'mask' and also used the mention "@YouTube". The most common theme to emerge was the encouragement for the public to wear masks and discussions around this. Other themes were related to jokes and discussions related to whether face masks were safe for those who may have breathing problems. Our study also examined Twitter data emerging solely from the United States, which demonstrated that there was overlap in content. This could have occurred because the Twitter has the most active users in the United States. It must be noted that Twitter discussions are constantly evolving and potentially alter on a weekly basis. A limitation of our study is that it focused specifically on the Twitter network from 27 June to the 4 July, hence our findings may not be applicable to other time periods. Future research could seek to expand time periods and examine Twitter discussions based on other locations. A further limitation is that with the 1% random sample we extracted, generic keyword and 7-day approach, we have captured some tweets coming from "temporary" discussions . If the study had examined a longer time period, i.e., 6 months, and taken a 0.005% sample, the study would have been able to perform a much wider analysis . At the same time, however, the approach adopted in this study had a better chance of capturing other issues such as conspiracy theories and/or short-lived time-based discussions. Future research could combine both approaches in the analysis. Further research could also seek to conduct a sentiment analysis of the data. A wider limitation related to research on Twitter is the potential of 'off-topic' discussions surrounding a particular keyword or hashtag to take place. Future research could seek to eliminate irrelevant content prior to analyzing data. Twitter data can be used to study a wide range of public health topics and was recently used to study views into personal health records [20], which utilises a similar methodology to this present study. Other research has also examined disclosure of patient information [21] and COVID-19 conspiracies [22]. Future research could also seek to examine the role of influential accounts during this time. --- Conclusions Overall, it was found that the shape of the network resembled a community as there were a range of users conversing amongst each other in different clusters. It was found that a range of accounts were influential and/or mentioned ranging from ordinary citizens, politicians, and popular culture figures. The most common theme and popular hashtags to emerge from the data encouraged the public to wear masks. Public health authorities and influential accounts could continue to utilize social media platforms to encourage users to wear masks.

Background: High compliance in wearing a mask is a crucial factor for stopping the transmission of COVID-19. Since the beginning of the pandemic, social media has been a key communication channel for citizens. This study focused on analyzing content from Twitter related to masks during the COVID-19 pandemic. Methods: Twitter data were collected using the keyword "mask" from 27 June 2020 to 4 July 2020. The total number of tweets gathered were n = 452,430. A systematic random sample of 1% (n = 4525) of tweets was analyzed using social network analysis. NodeXL (Social Media Research Foundation, California, CA, USA) was used to identify users ranked influential by betweenness centrality and was used to identify key hashtags and content. Results: The overall shape of the network resembled a community network because there was a range of users conversing amongst each other in different clusters. It was found that a range of accounts were influential and/or mentioned within the network. These ranged from ordinary citizens, politicians, and popular culture figures. The most common theme and popular hashtags to emerge from the data encouraged the public to wear masks. Conclusion: Towards the end of June 2020, Twitter was utilized by the public to encourage others to wear masks and discussions around masks included a wide range of users.

Introduction Child health and development has been one of the biggest issues in the World Health Organization. In recent years, Chinese president Xi Jinping has put public health at the center of the country's policy-making agenda, clarifying the need to include public health in official government policy. "The Healthy China 2030 Planning Outline", issued by the Chinese State Council, is the first long-term strategic plan of public health developed at the national level in China [1]. One of the aims in the plan is to enhance children's health and reduce children's mortality by the construction of pediatrics and the prevention of pediatric critical diseases [2]. As an important area of study in public health, children health advocates the prioritization of children healthcare in public health community as a basis for the improvement of national health [3]. Concerning this great emphasis on children's health, sustainable investment and efforts have been put into relevant fields, especially online. Chinese Premier Li Keqiang once put forward a guideline named "Internet+," aiming to integrate online resources with other domains including education, logistics, and health care. Social media is of particular importance in disseminating health information and promoting health communication as China records the world's largest number of registered social media users [4][5][6]. Of all the social media in China, Sina-microblog is one of the most popular platforms for health communication, with 516 million active online users at the end of 2019 [4]. In recent years, the number of doctors' 2 of 23 accounts in Weibo has been booming. It is estimated that Weibo has more than 50,000 registered doctors and generated 70 million followers in total, which facilitates the dissemination of high-quality health information [7]. These online doctors are coined as social media influencers in our study. SMI refers to those who are well connected with social network users and persuasive in the sphere in which they exert opinion leadership to affect users' decision-making process [8,9]. They play a crucial role in forming public views and impacting the spread of information in social media [8]. It is their expertise, ease of access, and social ties on social media platforms that enable them to influence followers [9]. In this regard, social media influencers in pediatrics are more convenient for consultations, as they use social media to address parents' concerns. This includes opening online clinics, spreading children's health knowledge online, responding to questions online, sharing thoughts of children's health issues in blogs, and more. Likewise, owing to their expertise and the ease of access for their followers, SMIP play an essential role in the circulation of children's health knowledge, formation of public opinion, and the behavioral effect of online followers [10]. Given that acquiring children healthcare information from, interacting with, and following SMIP online is becoming more commonplace, attention should be paid to the quality of health information and the effectiveness of SMIP's health communication. --- Previous Studies on Health Communication via Chinese Social Media In recent years, health communication via social media has received extensive attention from international academic communities and health practitioners. Moorhead et al. [6] reviewed 98 worldwide research studies to identify the benefits, limitations, and uses of social media for health communication among the general public, which provides insights for future health communication research. Given that social media have been widely used in China, scholars began to place more emphasis on health communication with the public via Chinese social media. Previous studies have identified possible factors affecting online health communication in China. These studies mainly focus on the analysis of non-content factors. For instance, Gao et al. [11] argued that Chinese users' perceived credibility of health information related to participants' involvement level and prior knowledge. Jiang [12] illustrated that the presence of interactive functions in Chinese social media contributed to productive communication between health care organizations and the public. Moreover, prior study found that social networks and social support affected how Chinese people viewed the health issue and disseminated specific information [13]. A few studies looked into the content of health information on social media. Jiang and Beaudoin [14] studied the smoking information posted from an organization's account on Weibo. The results uncovered that perceived risk and self-efficacy content positively influenced audience engagement. Liu et al. [15] illustrated that messages emphasizing the benefits of recommended healthy practices and harmful outcomes of unhealthy practices were associated with a Chinese user's behavior, such as shares and comments. Tian et al. [16] analyzed online messages indicating depression, which provided insights for medical practitioners to better communicate with people suffering from depression in China. To sum up, we found that most of the studies fell into the following three categories: the effectiveness of health information provided by Chinese organizations and non-professional online users [13]; typical health problems, such as depression, smoking addiction, and HIV in China [16]; positive effects of Chinese social media on health communication [6]. However, few empirical studies focused on health information provided by Chinese doctors via social media, especially pediatricians. Likewise, the effectiveness of driving factors in Chinese SMIP's communication on public engagement has not been sufficiently discussed. Public engagement is important because it reflects the public's attitude and affects their trust and relationship with involved members [17,18]. When there is a high level of public engagement, the public's panic, anxiety, and uncertainty will be minimized and the trust on the SMI will be improved [19,20]. Therefore, the study of the effectiveness of driving factors in Chinese SMIP's online communication on public engagement is highly warranted. The study can help pediatricians, and other doctors, better understand the constituent parts of online messages to be employed when they attempt to engage the public in health conversations through social media. Moreover, a better understanding of how people view doctors' online health communication would strengthen the value of the principles that guide good communication. --- The Importance of Genuineness and Its Role in Health Communication Among all the driving factors in health communication, genuineness has been suggested as a contributing factor in effective medical communication, especially in patient-centered psychotherapy [21]. Norcross and Newman [22] pointed out that health practitioners considered doctors' genuineness as "important for significant progress in psychotherapy, and, in fact, fundamental in the formation of a working alliance" . Likewise, therapists' characteristics, especially genuineness, authenticity and honesty can enhance their credibility which was essential for promoting therapeutic alliance and patients' trust [23][24][25]. Genuineness has been widely studied in face-to-face communication between doctors and patients. Little attention has been paid on the importance of genuineness on health communication in the context of social media. Due to the absence of concrete operational dimensions in studying genuineness, we developed an integrated framework that included four dimensions-"self-disclosure", "genuine response", "functional interactivity" and "genuineness in Chinese culture", for examining genuineness in social media communication based on previous studies in health communication, dialogic communication, and the study of Chinese culture. The first three are universal dimensions that occur regardless of the cultural context while the fourth dimension is a cultural determined dimension which is essential to Chinese communication. --- Self-Disclosure There is no universally agreed-upon definition of genuineness. However, the common features of genuineness focus on "self-dimension," referring to transparency, realness, and the authenticity of one's mind and behavior. In the medical field, Landreth stated, "the most significant resource the therapist brings to therapy relationship is the dimension of self. Skills and techniques are useful tools, but therapist's use of their personalities is their greatest asset" . Egan also specified that genuineness is "beyond professionalism and phoniness" . It refers to an attitude or behavior that can only be expressed if the doctor is self-aware [28]. Similarly, studies have noted that doctors' genuineness could be conceptualized as being real, being their true authentic self, and getting rid of dishonest and false behavior [29]. Nevertheless, how to concretely perceive genuineness via "self" has been understudied. Previous studies once demystified the idea that "self-dimension" of doctors' genuineness could be identified by self-disclosure during the health care process [30][31][32]. Self-disclosure is defined as being willing to consciously and intentionally reveal personal feelings, life experiences, and professional knowledge in the process of communication to establish a positive relationship [30]. Self-disclosure has received extensive attention in medical research because of its benefits to patients' positive health practices and doctor-patient relationships [31]. Previous research [30][31][32] found three main types of self-disclosure being preferred by the doctors, namely the disclosure of personal thoughts/feelings, disclosure of personal life, and disclosure of personal expertise . A study reported [31] that patients liked their doctors more when doctors disclose personal feelings and thoughts. Patients viewed an act of expressing feelings and thoughts from doctors as friendly and helpful because it encouraged patients to participate in a dialogue and enhance patient's self-exploration [31]. Another study suggested that when a doctor disclosed his/her own lifestyle , patients considered the doctor to be more credible and approachable [32]. Likewise, patients particularly valued when doctors disclosed the accumulated skills, experience, and specific expertise in the field [28][29][30]. Expanded on the previous studies, our study aims at investigating these three types of self-disclosure exhibited in the SMIP communication, and how public responded to different types of disclosure. --- Genuine Response In addition to self-disclosure, prior studies [33][34][35][36] confirmed that genuineness could be manifested when healthcare workers communicate consistently and provide expertise and emotional support to patients. In the health communication, a consistent response from doctor matters because it reflects "the degree to which one person is functionally integrated in the context of the relationship with another, such that there is an absence of conflict or inconsistency between their total experience, their awareness, and their overt communication in their congruence in the relationship" . A genuine response is not a response that simply expressed Yes or No answer or a simple act of reaction . It emphasizes on the recognition of interlocutors' concerns, thereby providing professional and emotional support to address their problems [33][34][35][36]. A genuine response to a patient's question or concern, is useful for building a positive therapeutic relationship [35]. Yet, an absence of an analytical framework for examining genuine response was noted. As such, we have modified frameworks from previous studies [33,36] on health communication studies and proposed three main sub-dimensions to measure genuine response: consistency, knowledge, skill, experience and treatment advice, and facilitation of hopefulness. Consistency emphasizes on whether patients' concerns are well understood, and the response is on the right track [33,36]. Van et al. [36] noted that healthcare workers often rephrase or repeat the patient's questions or concerns before providing follow-up treatment and explanation to demonstrate their understanding on patients' needs. Bottorff et al. [33] pointed out that nurses who responded with expert knowledge, such as treatment and medical advice were able to reduce patients' anxiety and uncertainty. They suggested that such responses enable patients to make informed decision-making and be more actively involved in a dialogue [33]. Moreover, previous research [33,35,36] indicated that nurses usually communicated emotional support through facilitation of hopefulness with patients during a therapeutic process. Bottorff et al. [33] and Van et al. [36] found that facilitation of hopefulness that nurses adopted in interactions contributed to reassuring patients and avoiding escalation of emotional instability, thereby leading to positive outcomes of treatment. In view of these, we intend to investigate and reveal genuine responses in SMIP messages by examining the three sub-dimensions adapted from previous studies on health communication [33,36]. --- Functional Interactivity In addition to health communication studies, this study drew on insights from dialogic communication theory in public relation and communication studies where functional interactivity serves as one of the principle elements in creating a genuine and dialogic communication online [37]. Functional interactivity refers to the interface's elements that allow an online user to interact with someone/an organization and build a dialogue between interlocutors [38]. Such elements include hyperlinks, multimedia, live-chat rooms, and questions [39]. Functional interactivity is of particular importance in social media communication where dynamic, two-way interactive communication is advanced by the proliferation of social media [40,41]. For a genuine and dialogic communication to emerge [37,40], interactive functions including "generation of return visits," "conservation of visitors," and "dialogic loop" were deemed necessary. The "generation of return visits" emphasizes on the return visit of the public while the "conservation of visitors" highlights the importance of connecting the public to the SMI. Both "generation of return visits" and "conservation of visitors" could be achieved by providing external links and hashtags to engage the public [41,42]. "Dialogic loop" placed much attention on promoting dialogue between SMI and the public where strategies including providing frequent responses, asking questions, and using multimedia are most employed [41,42]. Subsequently, this study examines the use of interactive functions for building genuine dialogue in SMIP communication and their association with public engagement. --- Genuineness in Chinese Culture If the first three dimensions of genuineness are universal dimensions that occur regardless of the cultural context, the fourth dimension can be identified as a cultural determined aspect which is essential in Chinese communication. In Chinese culture, honesty and kindness are viewed as necessary components for developing genuine dialogue [43][44][45][46], and therefore, are of particular value to the Chinese audience. Honesty is the essence of Confucianism and has a deep impact on the moral personality development [47]. Kindness, along with compassion, care, friendliness, righteousness, and affection, is one of the Confucian values about a "good person" in Chinese culture [48]. A kind individual is positively related to excellent job performance [49]. Zhang et al. [50] also illustrated that the kinder a nurse is, the more satisfied patients are. The genuineness in SMIP communication that attributes to the portrayal of a positive personality trait [46,47], could be measured by the use of lexical indicators for the expression of honesty and kindness [43][44][45][46]. Honesty in the Chinese culture is denoted as the moral quality of being consistent in words and deeds; opposite to hypocrisy; loyalty and open-mindedness; no lying, no fraud, no exaggeration, no distortion of facts [51]. As such, we postulate that lexical indicators related to reasoning and explanation , personal sharing and views , and truth/facts are important in expressing honesty. Kindness in the Chinese culture denotes personalities of being friendly, harmonious, kind-hearted and nice, and behaviors of altruistic, affectionate, righteous, and caring [52,53]. In SMIP communication, we expected kindness to be expressed through the use of lexical indicators related to caring , friendliness , gratitude , blessing , and compliment . --- Public Engagement in Social Media In our study, public engagement refers to the public's responses to the content communicated via social media which reflects the public's cognition and attitude on a particular issue [54]. Different level of public engagement on social media reveals their trust and relationship with involved members [20,55]. Previous research has studied public engagement in different contexts with varied definitions. In corporate-stakeholder communication, Bruce and Shelley defined public engagement as "the interaction between an organization and those individuals and groups that are impacted by, or influence, the organization" . In CEO communication, Men et al. conceptualize public engagement "as a behavioral construct focusing on publics' interactions with CEOs" . In government communication, public engagement refers to the involvement of citizens in public affairs [58]. In this aspect, public engagement aims to boost mutual understanding and build up a good relationship between the local government and the public [58]. In recent years, scholars have started to study public engagement and perception in an online context due to the arrival of global social media platforms [59,60], such as Weibo, YouTube, Twitter, and Facebook, which all include the common feature of real-time public interaction. Social media includes a variety of functions to engage with the public , and offers the ability to express attitudes via reaction buttons, appearing at the bottom of the relevant content: Like, Share, and Comment [60]. "Like" is an indicator to express awareness and interest, which can be used to identify the popularity of messages [61,62]. "Share" provides the opportunity to connect the organizational message to one's social group, and "Comment" enables direct dialogue with organizations [61]. These engagement indicators fall into different engagement levels. Like is the lowest level of engagement as it requires less cognitive effort and commitment than other indicators [62]. Share has a higher engagement level [63], as it can be viewed not only as an important indicator of user recognition but as user recommendation. This indicates that sharing requires certain time to evaluate the post's value [64]. A comment is the highest level of public engagement, as it requires more effort by the public to figure out the meaning of posts and directly respond to the messages with words or descriptors [61]. The number of likes and shares may indicate an overall positive effect but analyzing constituting parts embedded in comments helps estimate outcomes more concretely and accurately [65]. For instance, Fan [66] argued that how people perceive products could be revealed in the comments thread. By studying the comments, the organization will know the weaknesses and affordances of products. Public perceptions towards the content can also be amplified or constricted by reviewing other users' comments [67]. Therefore, comments can be quite persuasive on affecting public opinions [68]. The present research categorizes comments as the high, shares as the intermediate, and likes as the low level of public engagement indicators. Beyond the three engagement indicators, we paid particular attention to the valence of positive user comments. Kim and Yang [63] found that positive comments towards an organization are more likely to affect how people remember the organization, and further influence the organization's reputation. In tourism, for example, positive e-comments on businesses strongly influence travelers who read e-comments when they decide to select a hotel [69]. In the health field, positive online comments were positively associated with the effectiveness of anti-smoking persuasion on the public's attitudes [70]. --- Development of Research Questions Given the impact of SMIP on public views and boosting children's health, we aim to identify the effectiveness of genuineness, one of the most influential driving factors in health communication, in SMIP's online communication. The paper employs the coding framework of four genuineness dimensions generated from dialogic communication, health communication and Chinese cultural studies to examine the association between genuineness and public engagement. Further, it provides an in-depth understanding on the relationships between genuineness and public reception indicators . For the first research question , we aim to investigate the association between the four dimensions of genuineness and public engagement, therefore the following research question is put forward: RQ1: What is the association between the four dimensions of genuineness and public engagement? To fully understand the relationship within sub-dimensions in each dimension and public engagement, our second set of research questions is formulated as follows: RQ2a: What are the associations within the sub-dimensions of "self-disclosure" and public engagement? RQ2b: What are the associations within the sub-dimensions of "genuine response" and public engagement? RQ2c: What are the associations within the sub-dimensions of "functional interactivity" and public engagement? RQ2d: What are the associations within the sub-dimensions of "genuineness in Chinese culture" and public engagement? --- Materials and Methods --- Sample Selection First, we employed a self-developed Python program programmed by our research assistant with a postgraduate degree in computational science to identify the top pediatricians based on their number of followers in Weibo, one of the largest microblogging sites in China. The crawler is designed to search and identify verified pediatrician using the keywords "pediatrician" and the label "V-users". "V-users" referred to verified users where doctors need to submit their medical certificates to Weibo to prove their authenticity. Once approved, the letter "V" with a yellow badge will be assigned to these doctors' profile pictures. Verified pediatricians are preferred in our study, as they are much more influential in the social media community than non-verified ones [71]. The identified pediatricians with the highest number of followers in March are recognized as social media influencers in pediatric in our study as they are well connected and persuasive in their field. --- Data Collection As an exploratory study, we scrutinized the number of posts published by these top 10 SMIP for six months , to ensure that they are active communicators online. Subsequently, we replaced two inactive users who published fewer than two posts/day on average with the next two SMIP on the list. Table 1 presents the final list of Top 10 SMIP. Unlike Twitter, Weibo tends to change its open API at times for the purpose of data security and timely technical updates. Moreover, Weibo has a strict "restrictions on the API usage rate and unsolicited data requests" . Therefore, we had to manually collect the SMIP posts, record the number of comments, likes and shares and analyze positive comments for our study. Due to the complexity of the ten sub-dimensions embedded in the four dimensions of genuineness, we decided to code the sub-dimensions manually to ensure an accurate interpretation [54] on the use of genuineness in the SMIP posts. Taking all these into considerations, we decided to harvest a sample size of 300 posts to represent the target population. We have employed the sample size calculator developed by the Australian Statistics Bureau [73] to estimate a sample size of 300, giving a confidence level of 95%, a confident interval of 0.056, and standard error of 0.029. Through systematic random sampling, we randomly sampled 30 posts from each SMIP's Weibo account between March 1 and August 31 in 2019 for content analysis. --- Coding Scheme and Procedure Content analysis was employed to examine the four dimensions of genuineness adopted in the 300 posts of the top 10 SMIP on Weibo. Content analysis is a widely employed method in the study of media communication [74] and can be applied to "virtually any form of linguistic communication to answer the classic questions of who says what to whom, why, how, and with what effect" . It is concerned with the context where the occurrences of words, signs, and sentences are examined to provide in-depth understanding [74,76]. Researchers could adapt and integrate framework from previous research for conducting coding in content analysis [74,76]. In this study, we have drawn insights from health communication, dialogic communication and Chinese cultural studies to develop a four-dimension framework in genuine communication for pediatricians. A code book that includes the four dimensions of genuineness, the ten sub-dimensions, and descriptors to investigate the genuineness in SMIP's communication has been developed . The coding procedure of each dimension is listed below: For the dimension of "Self-disclosure", we coded the pediatrician's willingness to disclosure information related to his/her: personal life, personal feelings and thoughts, and personal expertise in pediatrics [30][31][32] in the post on sentence basis. For the dimension of "Genuine response", we coded to reveal if the pediatrician demonstrates: consistency, knowledge, skill, experience and treatment advice, and facilitation of hopefulness in his/her response in comment thread on sentence basis [33][34][35][36]. For the dimension of "Functional interactivity", we coded the number of interactive elements used to facilitate: "the generation of return visits and conservation of visitors," and creation of "dialogic loop". Refs. [37,39,41,42] in the post and comment thread. For the dimension of "Genuineness in Chinese culture", we coded the number of lexical indicators that demonstrates pediatrician's personality: of honesty, and kindness [43,45,46]; Refs. [50][51][52][53] in the post and comment thread. To ensure a high accuracy of analysis, a face-to-face meeting was held by the first author and the second author before the coding exercise. The authors identified the related descriptors, including lexical indicators and features in each dimension. Relevant examples were retrieved from the database collected to guide the coders in the process of coding. The first author and a well-trained research assistant who possesses a MA in communication conducted the coding in this study. Table 2 presents the four dimensions, ten sub-dimensions, and descriptors of the code book. The related examples extracted from the database could be found in Appendix A. For the evaluation of public engagement, the number of shares, likes, comments, and positive comments were identified. Beyond three engagement indicators, we paid particular attention to the valence of positive user comments, as positive comments can contribute to the excellent reputation of social media influencers and enhance public trust [65,77]. Online positive comments are characterized by the expression of compliment and affirmation, admiration and gratitude, usefulness and goodness [78]. In the corpus of this study, the comments, such as "thank you doctor," "beneficial advice," "great," and "feel the same way" were recorded. --- Interrater Reliability The coding was conducted by the first author, the primary coder, and a well-trained research assistant who possesses a MA in communication. To ensure inter-rater reliability on the coding of "self-disclosure", "genuine response", "functional interactivity", "genuineness in Chinese culture", and public engagement, the coders were highly trained on the coding scheme. Any disagreement between the two coders was discussed in the coding process until the agreement was achieved. The measure of interrater reliability was based on the co-coding of 60 posts from the top two SMIP . For all categories, the average agreement was higher than 0.95, and the average Cohen's Kappa was greater than 0.9, indicating an almost perfect agreement [79]. Please refer to Table 3 for the interrater reliability of all categories. --- Data Standardization and Statistical Analysis Since the content in posts and responses in comment threads varied from two words to 140 words, all coded data have been standardized, especially the coding done on sentence basis. As such we have standardized the coding data of the sub-dimensions in "self-disclosure" and "genuine response" by dividing the number of sentences yielded in each sub-dimension in every post by the overall number of sentences in each post. As for the sub-dimensions of "functional interactivity" and "genuineness in Chinese culture", we standardized the data by dividing the number of features harvested in each sub-dimension in every post by the total count of features in each post. As likes, shares, comments, and positive comments are count outcomes, Poisson regression was employed for statistics analysis. However, we found overdispersion exhibited when testing for assumptions in Poisson regression. Then we decided to employ negative binomial regression to replace Poisson regression as suggested in previous research [80,81]. Negative binomial regression fits various types of data arising in communication research [81], and "the negative binomial model is a more general model compared with the Poisson regression model that relaxes the strong assumption that the underlying rate of the outcome is the same for each included participant" [82] . Moreover, negative binomial regression allows various information to be included [82]; it is appropriate for the data in this study, especially in the presence of overdispersion. Thus, RQ1 and RQ2 were examined via negative binomial regression in which likes, shares, comments, and positive comments were taken as dependent variables. For the examination of associations between the four dimensions and public engagement in RQ1, standardized data in the sub-dimensions were summed up in the related dimension. For instance, the data in "disclosure of personal life", "disclosure of personal feelings and thoughts", and "disclosure of personal expertise" were combined to form the "Self-disclosure". As for RQ2, we used the standardized data in the sub-dimensions to examine if there was a significant association between sub-dimensions in each genuineness dimension and public engagement. --- Results In this section, we aim to reveal the association between the four dimensions of genuineness and public engagement and then identify different levels of impact of sub-dimensions in each genuineness dimension on public engagement. In response to RQ1, the NB2 findings indicated the number of "genuine response" was positively associated with the number of comments and positive comments, but negatively related to number of shares. For every extra sentence on "genuine response", 1.344 times more comments were generated, a statistically significant result . Similarly, there was a 16% increase in the number of positive comments for each extra sentence on "genuine response" . Likewise, a positive association was found between the occurrence of "functional interactivity" and shares, whereas there was a negative correlation with comment and positive comments. A 19.8% increase in the number of shares is expected for every extra feature in "functional interactivity" found . In addition, the frequency of "genuineness in Chinese culture" was positively related to the number of shares. For every extra lexical indicator in "genuineness in Chinese culture", 1.122 times more shares were generated . Table 4 summarizes the negative binomial regression results on the four dimensions of genuineness and public engagement. The results above show that three sub-dimensions of genuineness, namely "genuine response", "functional interactivity" and "genuineness in Chinese culture", have significant associations with public engagement on social media. Therefore, we intend to further examine the association between the sub-dimensions in the four genuineness dimensions and public engagement. Table 5 summarizes the negative binomial regression results on the sub-dimensions of "self-disclosure", "genuine response", "functional interactivity", "genuineness in Chinese culture" and the number of shares, likes, comments and positive comments. In response to RQ2a, we found that in the dimension of "self-discourse", "disclosure of personal life" had positive effects on the number of user shares and likes, while "disclosure of personal expertise in pediatrics" is positively associated with number of shares. For every extra sentence in "disclosure of personal life" and "disclosure of personal expertise in pediatrics", 1.23 and 1.13 times more shares were generated. For every extra sentence in "disclosure of personal life", 1.26 times more likes were expected. However, the "disclosure of personal thoughts and feelings" was negatively associated with the number of comments. For every extra sentence in "disclosure of personal thoughts and feelings", 0.89 times fewer comments were expected . Regarding the sub-dimensions in "genuine response" , our findings revealed that "consistency" had positive effect on the total number of likes, comments, and positive comments. 1.402 times more likes , 1.581 times more comments and 1.347 times more positive comments were witnessed for every extra sentence on "consistency" provided. Similarly, the sub-dimension of "knowledge, skill, experience and treatment advice" was positively associated with the number of comments and positive comments. For each extra sentence on "knowledge, skill, experience and treatment advice", 1.342 times more comments and 1.166 times more positive comments were yielded. However, the sub-dimension of "facilitation of hopefulness" was negatively associated with the number of likes and shares. For every extra sentence on "facilitation of hopefulness", 0.55 times fewer shares and 0.742 times fewer likes were generated, as presented in Table 5. For the dimension of "functional interactivity" , the "generation of return visits and conservation of visitors" had positive effects on the total number of shares and likes, whereas "dialogic loop" had a negative association with the number of comments and positive comments. For every additional "conservation of visitors and the generation of return visits" included, the shares and likes increased by 43% and 22% respectively while comments and positive comments decreased by 84% and 88% for every extra feature of "dialogic loop" provided, as shown in Table 5. Last but not least, "honesty" in the dimension of "genuineness in Chinese culture" had a positive association with the number of shares, likes, and positive comments in contrast to kindness, which showed no significant association. For every extra lexical indicator on "honesty", 1.158 times more shares , 1.106 times more likes , and 1.07 times more positive comments were generated, as presented in Table 5. --- Discussion Our results revealed that a variety of genuineness dimensions was employed by the SMIP to communicate with the public on social media. The findings yielded insights into how the "genuine response" alongside "functional interactivity" and "genuineness in Chinese culture" played an active role in engaging the public. --- Strong Effect of Genuine Response on Public Shares and Comments Corroborated with previous studies [33,36,83], our findings revealed that responses with high level of consistency and expert knowledge were positively associated with public engagement . A doctor responded by acknowledging the public's need helps develop a trustful relationship [84], even in online doctor-public communication. Furthermore, response with medical knowledge and treatment advice indicates the doctor's understanding of patient's concerns and his/her intention to address the issues [85]. This could be the reasons attributing to the positive association between "genuine response" and the number of likes and comments, especially the positive comments. "Genuine response" that aimed to address patients' concerns created more opportunities for the public to express feelings in the comment threads, and allowed them to continually ask questions if their concerns were not fully addressed. The phrases "thanks, doctor," "beneficial advice," "learn a lot," "really appreciate your patient guidance" and "what I need to do in the next step" were frequently unveiled under the comment threads. --- Interactive Features Promote Public Shares In line with previous studies [37,40,42,57], our results also revealed the strong effect of "functional interactivity" on public shares . We found a range of interactive features, in the form of links/hashtags, such as "#SIMP name+topic#", "@+other online users name" and "link to other Weibo pages", employed on the SMIP posts which foster the public's access to various and detailed information. Hashtags lead users to daily hot topics where users can make synchronous conversations, discuss relevant issues with others, and share insightful ideas [57]. Links enable users to return the site and increase the time of stay when reading messages [42]. Owing to word limit on Weibo, the SMIP messages may not explain the ins and outs of a health problem thoroughly. The offering of external links expands messages in greater detail and strengthens the usefulness of corresponding posts, thereby fostering information sharing. Given that the act of sharing can potentially reach out to a large audience, online doctors can adopt interactive elements to express genuineness and extend their influences. --- Honesty is Highly Valued in Chinese Culture Noticeably, the sub-dimension "honesty" positively engendered public engagement of likes, shares and positive comments . The expressions in honesty mainly involve verbs and adverbs related to explanation, personal views, and facts, such as "for instance," "include," "I think," "I suggest," "according to," and "the document shows." Tuckett [86] specified that "honesty" is "perceived as truth-telling" , and the extent to which truth-telling is preferred is highly related to culture and context. As noted in previous study [86], "honesty" is a fundamentally ethical principle in doctor-patient relationships. The majority of patients in China demonstrate that they want truthfulness and authenticity about their illness, which could enable them to manage uncertainty and make decisions independently [50]. This might also explain the negative association between sub-dimension of "facilitation of hopefulness" in the "genuine response" and shares and likes. To some extent, expression of hopefulness is intended to comfort patients instead of telling the whole truth [33,36], and the truthfulness of such expressions often arises suspicion. Also, the shared post represents the user [62]. A previous study [87] found that online self-presentation was a crucial part of impression management, in which the public carefully evaluated someone by how he presented himself. This suggests that sharing requires more cognitive effort [64]. Given that "honesty is the traditional morality of Chinese nationalities and is regarded as the basis of the making of a man" , it is not surprising to see the public's willingness in sharing "honest" so as to promote positive personality traits on social media. --- Self-Disclosure as a Controversial Communication Behavior Despite previous studies suggested doctor's "self-disclosure" may have a positive impact on patients' reactions and foster a stronger therapeutic relationship, our results reveal that "self-disclosure" has no significant association with any level of public engagement . Beach et al. [89] argued that doctors' personal disclosure to patients have sometimes been regarded as a boundary transgression. Doctors should be more careful about disclosing personal information [86]. "Self-disclosure" has been viewed as a positive intervention in doctor-patient communication but it could also hinder effective communication and lead to negativity [90]. Kelly and Achter [91] found that patients concerned the helpfulness and benefits of the disclosure information from doctors for their decision on engagement. If they think the information would be useful for their situations, they are willing to further interact with doctors and listen to their suggestions [92]. However, given that "self-disclosure" messages in this study mainly involve personal life, opinions, and feelings that may not be relevant to the public's concerns and problems, a lower level of public engagement is expected. Forest and Wood [90] commented that it is not surprising that people may disapprove or doubt the information provided by therapists who share personal opinions and experience frequently. Likewise, disclosure of personal expertise may involve the discouraging expressions [92], which makes the public feel sad and stressful. In addition, McDaniel et al. [93] found that the frequent statements about the doctor's personal life and professional information are, occasionally, of little value to impair the doctor-patient relationship because they may result in fewer opportunities for patients to express themselves. General information is prevalent in SMIP's posts on Weibo as each post is limited to 140 words, but there are a variety of followers with different needs. In other words, the posts cannot meet everyone's demands even though SMIP want to provide detailed information. Therefore, the public may not react to some information that is not tailored to their problems. --- Conclusions Academically, this study contributes to the research of health communication in the following aspects: developed an integrated framework to conceptualize and measure genuineness in social media communication and shed lights in the understanding of effect of genuineness on Chinese public engagement in SMIP online communication. In terms of practical implications, this study provides insights to health information providers such as pediatricians in engaging public on social media communication. For instance, the use of "genuine response" could raise public awareness which in turn facilitates the fostering of a healthy lifestyle. This study also has strong social implications. In recent years, the Chinese government has placed public health at the center of the country's entire policy-making agenda and initiated a national long-term strategic public health plan. One of the missions is to improve the well-being of citizens by spreading health knowledge and promoting quality medical care. Sustainable investment and efforts have been put into relevant fields, especially online, because in China, the Internet is becoming increasingly important in health care and offers a great number of platforms for patients to seek health information and conduct medical consultations. Therefore, this research provides insights into the formation of high-quality health information and productive medical services. Given the ongoing interaction between the public and SMIP on Weibo, further scrutiny on SMIP communication is highly warranted. Future studies could expand the database and include multiple social media outlets such as Wechat, TouTiao, and Baidu to fully examine the associations between different dimensions of genuineness and public engagement. With regard to the coding procedure, the data were not exhaustively coded as double coding within dimension was not allowed in this study. Lastly,"genuineness in Chinese culture"in this study was measured based on the lexical markers derived from the Chinese dictionary and relevant studies. To ensure a higher level of objectivity, we could harvest the most used words on content and style of genuineness by interviewing/surveying social media users that have interacted with the SMIP on Weibo. Literal translation: Hello doctor, my baby is 9 months old. He can laugh, make the voice of "mommy", grasp toys, and stand by armpit, but can not sit alone for more than 7 or 8 seconds. He often comes to cry for a few seconds when he is prostrating. I wonder whether he would recover from this situation? Pediatrician: "9个月，能逗笑，会发妈妈音，能抓玩具，能扶着腋窝站立，不能独 坐超过7.8秒，趴几秒也要哭就把两只手放两边。"据叙述，大运动发育落后。"能 扶着腋窝站立"这一定是家长扶着孩子站，不利于大运动发育，反而会有消极作 用。建议看神经康复科医生，是发育问题，还是家长养育问题。 Literal translation: "Nine months, can make you laugh, can make your mother's voice, can grasp toys, can stand by your armpit, cannot sit alone for more than 7 or 8 seconds, can cry even if you lie down for a few seconds." According to your illustration, it could be said that the development of large motor skills is backward. "Can support the armpit to stand" indicates parents was helping the child to stand, which is not conducive to the development of large motor skills and will have a negative effect. I suggest you should consult a neurologist to see if it is a developmental or a parenting problem. Literal translation: As long as the mother has no discomfort after perming and dyeing her hair, she would not affect her children through breastfeeding. Just keep in mind that do not let the child lick his/her mother's hair. Everyone has a desire for beauty. After giving birth to the baby, most mothers want to recover their body shape and bright skin as soon as possible. Thus, hair dyeing, perm, fingernail dyeing can be carried out. Mothers will take care of themselves. Literal translation: Is it necessary for lactation mothers to endure illness with silence? Can they still provide breastfeeding? Many mothers fear that they may not be able to breastfeed when they are ill in case of affecting the growth of their babies. Some concern that the child's grandparents will blame them for being sick and infecting their children. For this concern, click "my clinic" to learn more No.8: June.26, 14:33 @pediatrician: 哈哈，太可爱了//@User：我闺女也这样，笑得我呀！ Literal translation: @pediatrician: ha ha, it is so cute//@User: my daughter was just like this. How amusing! Reply by the Pediatrician to a user's comment in a post/comment thread No.8: June.26, 14:33 @pediatrician: 哈哈，太可爱了 ： // @User 我我我也这我，笑笑我笑！ Literal translation: @pediatrician: ha ha, it is so cute//@User: my daughter was just like this. How --- Appendix A Exemplification of four genuineness dimensions and relevant expression identified in the corpus. --- Coding items Examples Self-disclosure 1. Disclosure of personal life No. 7: May.12, 20:36 Literal translation: In the morning, I took Xiaoqi to Xianlin Golden Eagle Square to watch the launching ceremony of national nutrition week. The ceremony was great. Xiaoqi performed well. It was scorching, but Xiaoqi watched the whole process carefully. I ate cantaloupe, chicken wings, spaghetti, and half of the dessert at lunch. I knew that she would be allergic, so I did not give the food to her. However, she insisted on eating them, and I had not found any obvious allergic symptoms with her after eating. --- Disclosure of personal thoughts and feelings No. 5: Aug.4, 19:36 谢谢信任和认可。不过我们作为管理者知道，目前还是有蛮多可以改进和提高的地 方。 Literal translation: Thank you for your trust and recognition. However, as a manager, we know that there are still many things that can be improved. --- Disclosure of personal expertise in pediatrics No. 3 --- Coding items Examples Expression of compliment No 1. Apr. 9, 06:13 User one: 那需不需要给孩子补充维生素k呢? Literal translation: Do we need vitamin K for the children? Pediatrician: 问的很好！首先要保证孩子青菜等深色蔬菜的摄入。青菜含有维生 素K，在肠道正常菌群作用下，产生维生素K1和K2。不光是孩子，成人也是如 此，进食绿叶菜好处多多。现在已有维生素D3和维生素K2的混合制剂了。 Literal translation: Good question! First of all, we should ensure the intake of vegetables for the kids. Vegetables contain vitamin K, which is produced by healthy intestinal flora. Eating green leafy vegetables has many advantages for both children and adults. There is a mixture of vitamin D3 and vitamin K2 for now.

There is a growing need for the public to interact with pediatricians through social media in China, and genuineness is a crucial factor contributing to effective communication, but few studies have examined the relationship between genuineness and its effect on public engagement. This study developed a four-dimension framework including self-disclosure, genuine response, functional interactivity, and genuineness in Chinese culture to investigate the effect of genuineness in the communication of Chinese social media influencers in pediatrics on public engagement. Content analysis was employed to examine these dimensions and the related public engagement in 300 social media posts on the largest microblogging site in China. The findings indicate that genuine response was positively associated with the number of comments and positive comments, while negatively related to the number of shares. Functional interactivity made the site more appealing, resulting in likes and shares. Genuineness in Chinese culture was reflected in engagement through sharing posts by the public. This study is the first to develop an integrated framework to measure genuineness in online health communication and contributes to the understanding of the effect of genuineness on Chinese public engagement in social media.

Introduction The textile and apparel industry makes a major contribution to the global economy via the trade, income and the employment it generates; however, it also heavily challenges sustainability . It is estimated that it leaves the third largest global greenhouse gas emissions footprint and also that the textile sector was the third largest contributor of ocean plastic waste in 2019 . Simultaneously, textile and apparel manufacturing is highly labor-intensive, and the industry is known for labor law violations, especially in developing economies . Social and environmental negative impacts of the industry have prompted the inclusion of sustainability programs and agendas worldwide at institutional as well as firm levels. And at no surprise, consumer behavior research on sustainable apparel has increased each year since 2009 , since sustainability programs and agendas desperately necessitate consumer engagement and predisposition to sustainability as catalysts. Given the relevance and urgency to adhere to more sustainable lifestyles with the purpose of slowing the deterioration of the environment and society, it becomes critical to identify agents of change that increase the demand for and mainstream use of sustainable apparel. Cosmopolitan consumers might be an idoneous type of consumer for this role, since it is argued that a cosmopolitan approach incorporates a more ethical citizenship perspective, cultivating a more responsible contribution not limited by country borders . This study, in tandem, responds to a recent call for cosmopolitan research to broaden the range of product categories studied in relation to cosmopolitanism , since it is a challenge for scientists and marketers to comprehend how cosmopolitan consumer orientation drives consumer behavior toward specific product categories . Thus, the purpose of this study is to empirically examine the impact of CCO on sustainable apparel consumer behavior. Examining the link between cosmopolitanism and sustainable apparel consumer behavior is significant to explain, predict and improve the diffusion of sustainable apparel. Guided by identity and attitudinal theories , this study hypothesizes that aspects of the cosmopolitan selfconception affect beliefs toward sustainable apparel and apparel sustainability knowledge of consumers to pursue lines of conduct compatible with their self-structure. This is a relevant pursuit for scholars that also holds much importance for managers. A heightened understanding of the influence of CCO in furthering behavioral beliefs , normative beliefs , apparel sustainability knowledge and intentions to purchase sustainable apparel should provide insights about how CCO drives marketplace behaviors. Further, better understanding of how CCO shapes factors such as behavioral beliefs, normative beliefs, apparel sustainability knowledge and intentions to purchase sustainable apparel should also enhance management strategizing. For example, whether and the extent to which the model factors predict marketplace behaviors can assist managers in segmenting markets, selecting particular target groups to focus on and providing direction on how to better communicate their products positioning to their target groups. Considering the rapid growth of the cosmopolitan consumer segment, the globalization of the apparel industry and the worldwide acceptance of sustainable lifestyles, the value of this study lies in the advancement of knowledge in identifying the influence of CCO on attitudinal, normative, knowledge perception and purchasing behaviors toward sustainable apparel, while concurrently identifying an identity-based viable market for sustainable apparel, as well as providing direction for marketing practitioners to build stronger relationships and engagement with consumers. In the following sections, we review the pertinent literature, describe the research methodology and results and discuss the implications of our findings. --- Literature review --- CCO in consumer behavior and apparel research CCO literature frequently explores and seeks to explain discourses of consumer tensions within cultural authenticity, diversity and modernity ). Cosmopolitan consumers are omnivorous . They are agents of cultural change and transmission . CCO has been applied in consumer behavior and international marketing to provide insights concerning when and where marketing strategies could be standardized across countries or when and where strategies should be customized , thus, the relevance in identifying consumers with favorable predisposition toward sustainable apparel in the world and engaging them in the consumption of sustainable apparel. The impact of CCO on consumer behavior varies by consumption contexts . Cleveland et al. 's landmark study suggested that CCO can positively predict consumption of products of higher social value or aspirational products and, hence, the motivation to conduct research in the context of sustainable apparel cosmopolitan consumption. Literature suggests that consumers use apparel to communicate their cosmopolitan identity . For example, Chakraborty and Sadachar found that CCO, Western acculturation and consumer ethnocentrism predict attitude and purchase intention toward Western apparel brands among Indian consumers. Khare studied the incidence of CCO on the fashion involvement of Indian consumers. The author implied that cosmopolitan values influence consumers' lifestyles and views about the world, and Indian consumers are willing to imbibe global fashion brands that communicate their distinct global identity. Cosmopolitan consumers tend to emotionally attach to brands high in ideal self-congruity . Literature sustains that cosmopolitan consumption is "a symbol of social status and of one's moral worthiness" . Aspirational products represent "the modern lifestyles or an association with the global elite that cosmopolitanism symbolizes" . Consumers communicate their self-concept through the acquisition of aspirational products because these products represent new cultural ideals and are considered better signals of cultural capital and differentiation from the masses . Sustainable products are being recognized as an aspirational product category that embodies contemporary cultural and social values. Intellectual, value-laden and expertise-related product categories such as sustainable apparel are to be perceived by cosmopolitans as effective in differentiating themselves from the masses and signaling their aspirational self-concepts, social and cultural capital, as well as their moral worthiness. Thus, this study is instrumental in examining the acquisition of social capital, cultural capital and moral worthiness of the cosmopolitan consumer self-concept activated across sustainable apparel-linked choices. --- Development of research hypotheses Identity theory and the theory of reasoned action offer theoretical support for the study. Identity theory assumes that the self consists of set of identities . Individuals internalize the meanings that they apply to themselves when they are occupants of positions in the social structure. According to identity theory, individuals pursue lines of conduct compatible with their self-structure to the extent that the meanings are salient . In other words, aspects of the self will be activated across a variety of situations and will influence the role choices made by the individual . The TRA distinguishes two kinds of beliefs to guide the decision to perform or not perform a behavior: behavioral beliefs and normative beliefs . These beliefs then are expected to affect purchase behaviors . This study hypothesizes that aspects of the cosmopolitan self-conception affect beliefs , apparel sustainability knowledge, and intention to purchase sustainable apparel to pursue behavior compatible with self-structure. While identity theory provides a solid theoretical ground to explain the role of identities in the activation of sustainable apparel behaviors, the TRA guides relationships between beliefs and purchase behaviors. --- Effect of CCO on apparel sustainability knowledge In the context of this study, apparel sustainability knowledge refers to the consumers' perceived knowledge about social equity and child labor/sweatshop issues in fashion apparel manufacturing, as well as environmental issues during manufacturing including environmental impact across the supply chain. Cultural openness and global awareness are essential for the cosmopolitan consumer self-concept . Cosmopolitan literature has related cultural openness to connoisseurship and global awareness to global responsibility , which implies distinctive as well as selective avidity for knowledge and competency. In the context of coffee consumption, Bookman argues that the cosmopolitan attains knowledge of coffee and engages with coffee by being open to the exoticness and romance stories from different cultures that the coffee origins transmit. In addition, Bookman implies that coffee consumption communicates care for communities and the environments it is cultivated in. In terms of specific sociodemographic characteristics of cosmopolitan consumers, the literature suggests cosmopolitan consumers are better educated and have a higher financial status , which in turn increases their awareness of environmental and social issues and the associated knowledge of sustainability issues . It is expected that consumers with stronger CCO are more apt to develop connoisseurship and global responsibility regarding apparel sustainability since they have more access to education, resources and experiences. Additionally, literature portrays cosmopolitan consumers as avid travelers and consumers of global media, which reinforces their CCO . Cosmopolitan consumers' cross-border consumption lifestyle enhances their awareness of environmental degradation and protection and likely motivates them to actively search and obtain apparel sustainability knowledge. Thus, it is expected that CCO has a positive impact on consumer's apparel sustainability knowledge. --- H1. CCO positively impacts consumers' apparel sustainability knowledge. --- Effect of CCO on consumer attitude toward purchasing sustainable apparel Contemporary consumer markets worldwide are characterized by both a rapidly growing need for sustainability and an increasingly cosmopolitan lifestyle . Grinstein and Riefler provided evidence that CCO relates positively to the environmental concern of ecological issues . Cosmopolitan consumers are characterized by openmindedness and core values that speak for a heightened environmental concern and positive disposition toward sustainable behavior . Since cosmopolitan consumers tend to score high on values related to universalism, benevolence and egalitarianism , they tend to be tolerant, have an appreciation for all people and regard the protection of the environment as well as justice and equality as important . Values are guiding norms in life universally recognized within and across cultures . Value orientations are the most broadly defined identity facets on which consumers can be compared, regardless of cultural background, and are relatively invariant across situations . They have been found to have a positive effect on consumers' attitudes toward purchasing sustainable apparel . This coincides with Dickson and Littrell's and Kim et al.'s perspective on global values, which embodies that consumers guided by global values show concern for people, and this concern influences their support for sustainable apparel businesses. Thus, it is expected that cosmopolitan consumers hold positive attitudes toward purchasing sustainable apparel. --- H2. CCO positively impacts consumers' attitudes toward purchasing sustainable apparel. --- Effect of CCO on perceived norm Cosmopolitan consumers do not appear to deliberately try to win others' approval to enhance their self-image, gain rewards and/or avoid punishments ; however, sustainable apparel helps them express their personality , enables them to show their affiliation to certain social groups and communicates self-construal . The focus theory of normative conduct posits that norms cause conformity when they are salient, and the relevance of descriptive aspects and/or injunctive aspects of norms determine their saliency . Recent literature suggests loyalty, as a moral base for purchase dispositions, positively drives CCO . This attests to cosmopolitan consumers' desire for intergroup cooperation or desire to work for a common purpose with others. Since sustainable behavior is an emerging norm and sustainable apparel is an aspirational product category with high social value , it is expected that sustainable apparel allows cosmopolitan consumers to display their global benevolent, egalitarian and universal identity, which demonstrates increased authentic affiliation to as well as the agreement with social group norms. Thus, the following hypothesis was proposed. --- H3. CCO positively impacts perceived norm. --- Effect of apparel sustainability knowledge on attitude toward purchasing sustainable apparel Literature supports that knowledge of apparel sustainability is frequently a prerequisite to engaging in apparel sustainable behaviors and that a lack of knowledge is a barrier . Consumers who are more knowledgeable about apparel sustainability are more likely to show concerns for the negative impacts of global apparel production and consumption, as well as support for more sustainable apparel production and consumption, thus develop positive attitudes toward sustainable apparel . Given that the cosmopolitan consumer appreciates diversity, is innovative and is benevolent, the apparel sustainability knowledge that cosmopolitan consumers expose themselves to is likely to resonate and become highly accessible in memory. Consequently, it is expected that individuals who are more knowledgeable about apparel sustainability are more likely to have a positive attitude toward purchasing sustainable apparel. H4. Consumer apparel sustainability knowledge positively impacts attitude toward purchasing sustainable apparel. --- Effects of attitude toward purchasing sustainable apparel and perceived norm on purchase intention of sustainable apparel Literature provides evidence of a positive effect of attitude toward sustainable apparel on sustainable apparel consumer behavior . Thus, the more positive the attitude toward purchasing sustainable apparel the consumers possess, the stronger their intention to purchase sustainable apparel. H5. Attitude toward purchasing sustainable apparel positively impacts sustainable apparel purchase intention. Perceived norm refers to the perceived social pressure to engage in a certain behavior . That is, the more cosmopolitan consumers perceive that their important referents think they should purchase sustainable apparel and these referents also purchase sustainable apparel, then the more favorable or stronger their purchase intention toward sustainable apparel is likely to be. For example, Hyllegard et al. and Hyllegard et al. empirically found that the inclusion of subjective norm in their model improved the model's explanatory power regarding US consumers' purchase intentions of sustainable apparel with hangtags featuring prosocial marketing claims. Similarly, Kang et al. also found that subjective norm positively impacted college students' intention to purchase organic cotton from data collected from the US, South Korea and China. Cosmopolitan consumers' loyalty -as a moral base for purchase dispositions -attests to cosmopolitan's desire for intergroup cooperation , which likely enhances cosmopolitans' social pressure to purchase apparel that fits with social norms and group acceptability. Since sustainable apparel implies a good fit with social norms and group acceptability, it is expected that cosmopolitan consumers' perceived norm positively influences their purchase intention of sustainable apparel. --- H6. Perceived norm positively impacts sustainable apparel purchase intention. Lee et al. studied the direct impact of CCO on intention to purchase fair trade coffee and found a positive relationship. Also, Grinstein and Riefler found evidence to support the positive effect of CCO on environmentally friendly behavior . Cleveland et al. , a landmark study for CCO, shows that cosmopolitanism predicts the consumption of symbolic products and several types of apparel which are products of higher social value or aspirational products. Literature sustains that cosmopolitan consumption is "a symbol of social status and of one's moral worthiness" , and the cosmopolitan aspires to acquire social and cultural capital as well as moral worthiness . --- Effect of CCO on intention to purchase sustainable apparel As a new type of aspirational product , sustainable apparel is expected to satisfy the needs of cosmopolitan consumers due to its rich social, cultural and ethical values. Thus, it is expected that the cosmopolitan consumer aspires to purchase sustainable apparel. --- H7. CCO positively impacts consumers' sustainable apparel purchase intention. Figure 1 illustrates the conceptual model that shows how the seven hypotheses tie together. The model builds from the rationale that CCO influences consumer apparel sustainability knowledge, attitude toward purchasing sustainable apparel and perceived norm. Then CCO, as well as the three factors mentioned above, cause an effect on the consumer intention to purchase sustainable apparel. --- Research Method A structured questionnaire was designed based on a careful review of the literature. The measurement scales selected for this study are established in academic research, and most of them have been applied in multiple contexts and cultures. CCO in participants was measured by eight items using a seven-point Likert scale ranging from 1 to 7 ; the items were adapted from Cleveland and Laroche . Consumer's apparel sustainability knowledge was measured by six items of a seven-point Likert scale ranging from 1 to 7 , which was adapted from Shen et al. . Attitude toward purchasing sustainable apparel was measured by six items using a seven-point semantic differential scale adapted from De Lenne and Vandenbosch . The scale of the construct of intention to purchase sustainable apparel, adapted from Putrevu and Lord , was measured by three items using a seven-point Likert scale ranging from 1 to 7 . The perceived norm construct was measured by four items using a seven-point Likert scale ranging from 1 to 7 and was adapted from Fishbein and Ajzen and De Lenne and Vandenbosch . Table I shows the scales and scale items included in the survey instrument. The items for perceived norm, apparel sustainability knowledge, attitude toward purchasing sustainable apparel and sustainable apparel purchase intention were modified to the specific research context. Seven-point Likert scales were selected for the study to give the respondents enough natural points of discrimination without complicating the selection with too many response options. Psychometric literature suggests there is a diminishing return after 11-point scale ranges . The survey instrument contained an introduction section, a main questionnaire section and an ending. The survey introduction included the definition of sustainable apparel to avoid misunderstandings, ambiguity or confusion. The definition of sustainable apparel read as follows: --- What is sustainable apparel? Sustainable apparel refers to garments developed in a way that decrease or minimize negative impacts on the environment and/or the society . Sustainable apparel includes ethical apparel, green apparel, socially responsible apparel, fair trade apparel, organic cotton apparel, eco-conscious apparel, environmentally friendly apparel, environmentally responsible apparel, etc. The survey was distributed online using Qualtrics, a reputable web-survey platform. Data were collected from a US sample using Amazon Mechanical Turk, which is a crowdsourcing website usually used by academics to request crowd workers to perform on-demand tasks such as filling surveys. An US consumer sample makes up an appropriate starting point to study the effect of cosmopolitanism on purchase intention of sustainable apparel because the market is diverse and ample. Additionally, the US economy is advanced , with consumers who are optimistic about spending and shopping across channels . Thus, these consumers are potentially in a better position than consumers in developing economies to be economically able to spend in apparel that they are interested in. Also, US consumers shop in a variety of retail channels and retailers, where they can probably find an appropriate selection of sustainable apparel should they be interested in it. Prior to data collection, Institutional Review Board approval was obtained. Also, the instrument was pretested among a small group of 10 people to gather impressions and feedback. A few minor format modifications were performed to improve the clarity and readability of the survey items. Also, the order of appearance of constructs in the survey was slightly altered from the original set up to improve the flow of the survey. --- Results A total of 612 US responses were collected using MTurk. From the 612 responses, 469 were retained for the analysis after screening for unengaged responses. One hundred and thirty-seven cases were deleted because subjects failed to pass an attention check question included in the questionnaire or completed the questionnaire in less than 3.5 min. Additionally, six extra cases that showed extreme response patterns were deleted. --- Sample demographic characteristics The sample is fairly balanced between male and female participants. Thirty-three per cent of the respondents were 18-30 years old, and 42 per cent were 31-50 years old. Fifty-one per cent of the respondents were married. About 69 per cent of the respondents have bachelor's, graduate or professional degrees. Seventy-three per cent of the respondents identified themselves as Caucasian, followed by Asians . Sample characteristics are shown in Table II. When asked about past purchase experience with sustainable apparel, 46 per cent of the sample were aware of having purchased sustainable apparel in the past 3 years, whereas 23 per cent admitted that they had not purchased sustainable apparel and approximately 31 per cent were not sure. Of the 216 participants that who aware of having purchased sustainable apparel in the past 3 years, 51 per cent declared that they purchased sustainable apparel because of its sustainable features, whereas 26 per cent purchased it because it was fashionable and 19 per cent bought it because it fit well. The information about past purchase experience is shown in Table III. --- Exploratory factor analysis Exploratory factor analysis was conducted to evaluate the factor structure of the variable indicators in the model. All scales were entered as reported on the questionnaire using principal component analysis extraction method and Varimax rotation method with Kaiser Normalization; 27 Likert-scale items were subjected to EFA. Two items with poor psychometric properties loading on more than one factor were removed after establishing that their removal would not compromise the content validity of the constructs. The retained solution contained 25 items in five factors , accounting for 73.17 per cent of the total variance. The five factors are apparel sustainability knowledge , attitude toward purchasing sustainable apparel , intention to purchase sustainable apparel , perceived norm and consumer cosmopolitan orientation . The sampling adequacy measure Kaiser-Meyer-Olkin of 0.93 and the significant Bartlett's test of sphericity suggest that the factor model is appropriate. All communalities are above 0.63. The results of the EFA suggest that all 25 items load significantly on the proper theoretical dimensions proposed in the conceptual model and factors are internally reliable. In addition, there is no indication of normality problems in the data as per visual inspection of skewness and kurtosis statistics in variable indicators. Item means ranged between 4.06 and 5.92. --- Structural equation modeling A two-step structural equation modeling approach was conducted using IBM SPSS AMOS 26. First, the measurement model was evaluated using confirmatory factor analysis to determine the adequate model fit and establish a satisfactory level of scale reliability and validity for the model variables. Second, the structural model was tested to examine the seven research hypotheses. --- The measurement model Following the two-step SEM method described by Anderson and Gerbing , first CFA was used to test the characteristics of the measurement model in which each item loaded in its respective factor and all the factors were correlated. The CFA suggested that the measurement model exhibited a good fit: χ2/df = 2.35; root-mean-squared-error-of-approximation = 0.05; comparative-fit-index = 0.96; adjusted-goodness-of-fit-index = 0.88. Table IV depicts indicator variables and their underlying factors. The standardized loadings were all significant and ranged from 0.69 to 0.87. The average variance extracted for all constructs exceeded 0.50, ranging from 0.58 to 0.68, and the construct reliability estimates are uniformly high , providing support for convergent validity. The square root of AVE for each latent factor exceeded the respective inter-construct correlations between the factors , providing support for discriminant validity . --- The structural model After the validity and good fit of the measurement model was established, the structural model was analyzed. A reasonable fit was achieved for the structural model χ2/df = 2.98; RMSEA = 0.07; CFI = 0.94 and AGFI = 0.85. The results from the evaluation of the structural equation model are shown in Figure 1. With 99 per cent confidence, the hypothesized relationships in H1, H2, H3, H5, H6 and H7 are supported by the structural equation model, except H4. CCO has a positive and significant effect on apparel sustainability knowledge , attitude toward purchasing sustainable apparel and perceived norm . Therefore, CCO impacts consumers' attitudes toward purchasing sustainable apparel and perceived norm, as well as their apparel sustainability knowledge. Results suggest that the impact of CCO on the perceived norm is stronger than the impact of CCO on apparel sustainability knowledge or attitude toward purchasing sustainable apparel. This implies that when considering sustainability knowledge, attitudes and norms, pressure from what other people do and what is appropriate is affected by CCO the most. Also, attitude, perceived norm and CCO have a positive and significant effect on purchase intention . The effect of the perceived norm on purchase intention appears stronger and more influential than the effect of attitude on purchase intention. In addition, the results suggest that the hypothesized effect of apparel sustainability knowledge on attitude is not supported at the 95 per cent level of confidence. --- Profiling consumers To explore consumer profiles and past behavior, a two-step cluster analysis using log-likelihood distance measure and Akaike's Information Criterionwas conducted. Fair fits for two and three cluster solutions were obtained, and the two-cluster solution was retained because it allows better interpretability. Significant differences were detected between the two clusters for all predictors. The solution is provided in Table VI. Cluster 1 represents the group of consumers who admittedly have not purchased or are not aware of purchasing sustainable apparel in the past 3 years, so they were denominated "Unengaged". Consumers in Cluster 2 were denominated "Engaged" consumers, and they represent the group of consumers that declared to be aware of purchasing sustainable apparel in the past 3 years. In line with expectations, these consumers know more about apparel sustainability, have a stronger CCO, stronger intention to purchase sustainable apparel, show a higher tendency to follow social norms, tend to live more in metropolitan areas and tend to be slightly younger than the consumers in Cluster 1. --- Discussion and implications This study explores a structural model that examines the relationships among CCO, apparel sustainability knowledge, attitude toward purchasing sustainable apparel, perceived norm and intention to purchase sustainable apparel among consumers in the US. As expected, the structural results imply that consumers with stronger CCO are more predisposed to purchase sustainable apparel. This notion is also empirically reflected in our cluster analysis results, which shows that the segment of consumers that have purchased sustainable apparel in the past 3 years scores higher in CCO. The main contribution of this study is the identification of the link between two major trends in society and industry, cosmopolitanism and sustainable apparel behavior. Yet, the findings identify and paint a clear picture of how CCO functions as a driver of sustainable apparel consumer behavior. Considering the rapid growth of the cosmopolitan consumer segment, the globalization of the apparel industry, the worldwide acceptance of sustainable lifestyles and simultaneously the dearth in adopting sustainable lifestyles, this study is instrumental in providing theoretical and practical implications. --- Theoretical implications Guided by identity theory and the TRA and the structural model findings, our study implies that consumers are activating aspects of their cosmopolitan self when making choices regarding sustainable apparel. Consumers with stronger CCO embed in their identities particular aspects that prompt them to show stronger apparel sustainability knowledge, as well as more favorable attitudes toward purchasing sustainable apparel, perceive sustainable apparel as a social norm and show a stronger intention to purchase sustainable apparel. Also, our findings imply that favorable attitudes toward purchasing sustainable apparel and perceived norms influence intentions to purchase sustainable apparel positively. This is a significant pursuit because it allows for a better understanding of consumer behavior toward sustainable apparel in a structural model that identifies a network of impactful connections between model constructs. These impacts allow ultimately for better understanding and prediction of consumers' behavior toward sustainable apparel. The results reveal that consumer cosmopolitanism influences consumers' attitude toward purchasing sustainable apparel and intentions to purchase sustainable apparel. Thus, implying that cosmopolitan global values of universalism and egalitarianism are likely activating consumers' behavioral beliefs toward purchasing sustainable favorably, as well as their behavioral intentions. These cosmopolitan global values also impact their perceived knowledge about social equity and child labor/sweatshop issues in fashion apparel manufacturing, in addition to their knowledge of environmental issues during manufacturing including environmental impact across the supply chain . Interestingly, although CCO has the strength to influence upward apparel sustainability knowledge, this knowledge is not influential enough to impact consumers' attitudes toward purchasing sustainable apparel. Early research claims that consumer knowledge on apparel sustainability is frequently a prerequisite for consumers to engage in sustainable consumption and that a lack of sustainability knowledge is a constraint . It is possible that, although our results show that apparel sustainability knowledge in consumers is moderately high and current literature claims that consumers are becoming more aware of sustainability issues in apparel , there are supplemental factors affecting consumer attitudes toward purchasing sustainable apparel negatively. For example, some of these factors can be perceived aesthetic risks , greenwashing concerns , lack of appropriate/effective labeling or perceived behavioral control. Our findings highlight the importance of CCO impacting apparel sustainability knowledge; however, it is worth recognizing that apparel sustainability knowledge might not be as relevant as other factors in driving consumer behavior. Two particularly noteworthy findings are the observed positive impacts of CCO on perceived norm and perceived norm on the intention to purchase sustainable apparel. Perceived norm was the factor most influenced by CCO and the factor that affects the intention to purchase sustainable apparel the most. The finding not only implies that cosmopolitan consumers' purchase intentions toward sustainable apparel are more susceptible to social influences than their attitudes but also that sustainable apparel allows cosmopolitan consumers to considerably fit with social norms and show their affiliation with aspired social groups. This is an important finding because it implies that when cosmopolitans consider purchases of sustainable apparel, they heavily rely on their referents' opinions and behaviors, which is not necessarily the modus operandis of cosmopolitans since they tend to be unlikely to conform to expectations or uniformity when considering other type of purchase decisions . The theoretical significance of this finding therefore lies in the unexpected relationship direction and strength between CCO and perceived norm observed in the context of sustainable apparel and the relevant influence on intention to purchase sustainable apparel. This study reveals that it is, therefore, compulsory to consider perceived norm when considering/hypothesizing cosmopolitan sustainable apparel consumption. Our study results support Lee et al. 's study, which found that the relationship between perceived norm and purchase intention of fair trade coffee in South Korea is stronger for people with high CCO than for people with moderate or low CCO. Our study also supports Khare's findings that cosmopolitan consumers balance global values and lifestyle with group conformity. Considering that fair trade coffee in South Korea, as well as fashion clothing in India, are aspirational products, our findings contribute evidence to support the confirmation of previous research. However, our findings also extend the understanding because they imply that CCO activates the notion of what is appropriate and what others do more strongly than it activates knowledge, attitudes and even purchase intention directly. This insight is a powerful theoretical standpoint in the study of CCO and the consumption of sustainable apparel, not only because it identifies the importance of including referents when studying sustainable apparel but also because it establishes normative beliefs' relative importance among other factors and presents evidence that this behavior extends to the consumption of aspirational products. --- Practical implications To be successful in the complex and constantly changing global marketplace of apparel, managers of multinational corporations as much as managers of local firms need to know how to effectively address consumers to make their products and brands attractive. Our results indicate that consumers with strong CCO have a predisposition to purchase sustainable apparel and that their CCO also influences their apparel sustainability knowledge, their attitudes toward sustainable apparel, as well as their perceived norms, indicating that these factors are important to use in segmentation, marketing/advertising studies and even product design. Our study findings imply that the availability of apparel sustainability knowledge consumed by the public may not be enough or effective to affect attitudes. Thus, marketing practitioners should enhance cosmopolitan consumers' curiosity and emotional attachment with the brand and should highlight the ways that their sustainable apparel allows them to be more ecologically and socially responsible in line with global cosmopolitan values. Also, since cosmopolitan consumers have a favorable predisposition for foreign products, we anticipate that segments of cosmopolitan consumers make a good target market for the international expansion of firms offering sustainable apparel. In particular, managers are advised to focus on the influences of CCO on perceived norm and perceived norm on the intention to purchase sustainable apparel. They should pay attention to the followers of a brand who show cosmopolitan orientation, since marketing strategies appealing to global values and good fit with social norms would particularly affect their disposition toward sustainable apparel products. While consumers with strong CCO will be inclined toward sustainable apparel that fits with their global values of universalism, benevolence and egalitarism, it will also be crucial that their referents are also on board with the brand of sustainable apparel. For example, we advice brands of sustainable apparel interested in capturing the promising cosmopolitan consumer segment to act transparently because they would probably be more susceptible to worldly publicized brand transgressions and negative information that their referents criticize/judge. Concurrently favorable opinions, approval and consumption of the brand would improve the likelihood of brand purchases. While in recent years there seems to be a surge in sustainable apparel options in the US market, our cluster analysis results reveal that there might be a sizable group consumers that has not consciously participated in the sustainable apparel bandwagon. Almost 54 per cent of this study sample reported that they have not consciously purchased sustainable apparel in the past 3 years. Therefore, the study implies that sustainable apparel ownership in the US is moderate. The engaged group of consumers , who consciously consumed sustainable apparel in the past 3 years, represents less than half of the sample . This makes a call for attention to society, including managers of sustainability programs to focus on engagement of consumers with sustainable apparel products. Consumer-focused programs should continue aiming to convert unengaged consumers and should also consider targeting cosmopolitan consumers with appropriate strategies aiming at increasing the "engagement" of engaged consumers. While claims of lack of engagement with sustainable apparel among consumers are not new, this study identifies significant differences between groups of engaged and unengaged consumers. Specially for marketing practitioners, cluster analysis findings provide insights on how to target consumers who have engaged in sustainable apparel purchasing in the past. For example, given their strong CCO, sustainable apparel that communicates their cosmopolitan identity through cultural openness and global responsibility would be appealing to them. Also, since they tend to rely more on their referents' opinions and behaviors, marketing campaigns addressing authentic affiliation and conformity with social norms stressing global benevolent, egalitarian and universal identity are likely to be more engaging. Additionally, according to cluster analysis results, marketers should target metropolitan and younger consumers when designing and creating campaigns. Finally, since they have a more favorable attitude toward sustainable apparel and consider themselves more knowledgeable about apparel sustainability, these consumers have already a predisposition toward sustainable apparel. Therefore, marketers need to make sure that the qualities that make apparel sustainable are communicated, so that the consumers can identify it as sustainable apparel and their sustainable characteristics do not get unnoticed. For example, marketers can use labels on their clothing to communicate important information. According to Rogers , cosmopolitan consumers are early adopters of innovations and critical for marketplace success. This study found that adopters of sustainable apparel tend to be more cosmopolitans than non-adopters. This finding supports Rogers' argument about cosmopolitan consumers. Considering that the segment of cosmopolitan consumers is growing , practically our findings imply a future much-needed mainstream consumer adhesion to more sustainable lifestyles embeding a positive future outlook for the diffusion of sustainable apparel in society. --- Limitations and avenues for future research Our work focused on US consumers as an appropriate starting point to study the phenomena, and the study revealed relevant relationships between the studied constructs. A natural avenue for future research is to replicate the study in other countries and geographies with different cultural orientations and/or different levels of economic development. Cross-national comparisons between countries should provide a rich analysis to validate results and explain differences when predicting sustainable apparel beliefs and purchase behavior in different geographies. It would be interesting to analyze if different levels of economic development and cultural orientations impact the direction of the relationships found in the US. A cross-cultural study would strengthen results to consider or discard the possibility of finding in cosmopolitan consumers a homogeneous global consumer segment with similar attitudes, knowledge, norms and predisposition toward sustainable apparel. These perspectives would definitely enrich and extend theoretical and practical implications; for example, they will suggest a new global paradigm and the possibility of global strategies applicable at larger scales by apparel firms. A very current topic in academic literature and of great interest for managers is the influence of social media on consumers. Based on our study results, CCO influences sustainable apparel knowledge and perceived norm positively. Since cosmopolitans tend to be innovative and prone to the use of technology, plus they have a global orientation and they tend to be young, it seems natural that they would be interested in social media. A future research study may investigate how cosmopolitans obtain sustainable apparel knowledge, and if their social media act as relevant referent to guide their sustainable apparel consumption. Such a study would contribute to theory and practice, since it would reveal channels of communication between cosmopolitan consumers and media and also provide insights to managers about effective/ineffective channels to reach cosmopolitans. Finally, this study partially addressed the attitude-intention gap through the integration of perceived norm in the conceptual model and utilized cluster analysis to investigate actual consumer ownership of sustainable apparel addressing the intention-behavior gap. However, based on the unexpected unsignificant relationship between sustainability apparel knowledge and attitude , we suggest that an avenue for future research lies in the examination of additional factors such as perceived aesthetic risks, greenwashing concerns, lack of appropriate/effective labeling or perceived behavioral control. Shifting the focus from knowledge to other factors counteracting the positive formation of attitudes might explain more of the variance in purchase behaviors among consumers. This in turn will portray a clearer picture of the phenomenon and enable managers to act proactively on design, reputation, marketing, labeling or even pricing distractors. In regards to the intention-actual behavior gap, literature discusses observation studies and constructs addressing actual behavior. A study using actual behavior as a result variable would help address the interest in examining actual behavior, thus, allowing for an alternative way to analyze the phenomenon and diminishing the intention-actual behavior gap. 62% of group 84.3% of group Notes: KNOW -apparel sustainability knowledge; ATT -attitude toward purchasing sustainable apparel; PI -purchase intention toward sustainable apparel; NORM -perceived norm. Cases with missing information were not considered in the cluster analysis. Categorical denominations of low and high were assigned to group predictors based on sample mean scores. Cluster mean scores higher than sample means were categorized as "High," and cluster mean scores lower than the sample means were categorized as "Low." Past purchase behavior is the most important predictor of clusters, followed by NORM, PI, KNOW, CCO, MAR, ATT and age *The age category 31 -40 was the most frequently appearing category in the group **The age category 25 -30 was the most frequently appearing category in the group . Kruskal -Wallis H tests showed that clusters differ statistically in CCO = 40.71, p < 0.001), KNOW = 90.14, p < 0.001), ATT = 12.90, p < 0.001), PI = 103.45, p < 0.001), NORM = 117.98, p < 0.001), Age = 16.17, p < 0.001) and MAR = 19.15, p < 0.001) Source: Table created by Author

Purpose The purpose of this study is to empirically examine the impact of cosmopolitan consumer orientation (CCO) on sustainable apparel consumer behavior. Design/methodology/approach A total of 469 US responses collected using MTurk were retained for the analysis after screening for unengaged responses. Structural equation modeling was used to confirm the factor structure of the measurement model and to analyze the structural model. A two-step cluster analysis using log-likelihood distance measure and Akaike's Information Criterion was conducted to explore consumer profiles and past behavior. Findings Based on the model results, CCO positively impacts apparel sustainability knowledge, attitude toward purchasing sustainable apparel, perceived norm and sustainable apparel purchase intention. Attitude and perceived norm also impact sustainable purchase intention. The two-step cluster analysis, based mainly on sustainable past behavior, reveals that the group of sustainability engaged consumers knows more about apparel sustainability, has a stronger intention to purchase sustainable apparel, is more cosmopolitan and shows a higher tendency to follow social norms. Consumers in this group also tend to live in metropolitan areas and are slightly younger than unengaged consumers. Originality/value This study expands CCO research linking two major trends in society and industry: cosmopolitanism and sustainable apparel consumer behavior. The study reveals that CCO uplifts consumers' sustainable behavior and provides evidence in support of CCO as a driver of sustainable consumer behavior. Moreover, results imply a positive future outlook for the diffusion of sustainable apparel, as well as a much-needed mainstream consumer adhesion to more sustainable lifestyles. Given the repercussions of the findings, this research has numerous theoretical as well practical implications.

Background Access to palliative care is a universal human right to health, however of the 40 million people globally in need of palliative care, only 14% of people receive it [1] Palliative care is an integrated care approach that supports patients with chronic or life-threatening illnesses and their families by creating awareness and education about the illness and symptom management, providing resources for counseling and bereavement services, and overall promoting quality of life [1]. Individuals with advanced-stage chronic or life-limiting illnesses have complex physical, psychological, social, and existential needs [2,3]. Ideally, Palliative care should be initiated alongside a diagnosis of a serious chronic or life-threatening illness to address holistic care needs of patients and families, allow adequate care planning, faciltate treatment choices, advance goals of care, and reduce distress among patients and families [4]. Evidence indicate that adequate access and provision of palliative care improves pain and symptom management [5], enables informed decision making [6], reduces hospital readmissions and healthcare costs [7,8]. Despite multiple benefits, palliative care access remains a challenge in the United States especially in smaller towns and rural communities. This paper aims to highlight various barriers and facilitators in access to palliative care services in smaller towns and rural communities in Indiana, USA. --- Palliative Care from a Rural Context Recent health indicators in US, suggest that there is an increased prevalence of chronic illnesses such as heart diseases, cancer, unintentional injury, chronic lower respiratory disease, and stroke [9] among rural residents in comparison to urban population as evidenced by the higher death rates in rural or non-metropolitan areas as compared to urban areas [10]. According to National Review of State Palliative Care Policies and Programs report there is a limited availability of palliative care professionals, resources, and services in smaller towns and rural communities [11]. These geographic inequities are expected to increase with the growing aging population and increased rates of life-limiting illnesses and deaths [12][13][14]. In the US state of Indiana, there are approximately 6.3 million people are living with at least one chronic disease, and 1.6 million people with 2 or more chronic diseases. Most common are heart conditons followed by cancer chronic obstructive pulmonary disease , and diabetes [15]. Majority of these people reside in rural areas and represent the agricultural workforce. Most reported reasons for the high prevalence of chronic illnesses and higher death rates among these people are poorer health, limited socioeconomic resources, restricted access to high-quality emergency and specialty care services including palliative care [10]. The Center for Rural Engagement, Indiana University reported lack of access to health care services as one of the greatest needs of people in rural Indiana. According to them, most people live in counties that are designated as 'Health Professional Shortage Areas' , where there are only 230.8 active physicians per 100,000 people [16]. Patients in rural communities lack multiple healthcare services including but not limited to poststroke care, mental health services, complex therapies, hospice services, specialized providers, pain management, however, those services are not found locally and therefore both patients and families experience major healthcare disparities [17]. Medicare beneficiaries also encounter multiple barriers to palliative care access [17]. Limited or no access to palliative care can give rise to further complexities of care, inappropriate treatment choices with increased healthcare risks, caregiving burden, frequent hospitalization, and higher healthcare costs [18]. Integrating and improving access and provision of palliative care should be considered as a major priority area to enhance the quality of healthcare delivered to all Americans especially in rural communities. A strong commitment and engagement are essential in order to improve the access and quality of palliative care services in rural settings, The following paper highlights major barriers and facilitators experienced by the rural communities from the perspectives of healthcare providers and community stakeholders in Indiana, USA. --- Theoretical framework A theoretical framework of 'access' by Penchansky and Thomas [19] was used to guide the study. This model of access has been used widely by several researchers to identify and improve the access and provision of quality healthcare services particularly in both urban and rural contexts of the USA, Canada, and Europe. Penchansky and Thomas define 'access' as 'the degree of fit between the consumer and the service ' . The theory of access incorporates and addresses five specific dimensions that include: accessibility, availability, affordability, acceptability, and accommodation of services [19,20]. These five dimensions identify a set of specific areas that link the patient and their family and the health care service, essentially a supply-demand relationship, which recognizes both service and user requisites [20]. It is important to note that all these five dimensions are interdependent, addressing one will not significantly improve the access to care unless the other four have not also been addressed [21]. The particular framework served as a comprehensive guide for developing the study tool and provided an analytical framework in identifying perceived barriers and facilitators to access palliative care services in rural Indiana. --- Method --- Study design Given the exploratory nature of the study, a descriptive qualitative study using a community-based participatory research approach was designed. A CBPR approach offers a unique collaboration between researchers and community members and allows co-learning and reciprocal transferring of expertise and outcomes [22]. The approach also encourages innovative adaptations of existing resources, explores local knowledge and perceptions, empowers people to act as investigative agents in their own situations, and acknowledges the community as full and equal partners in all phases of the research process, making it an appealing, credible, and useful model for research with vulnerable populations [22]. The qualitative approach in the study recognizes the importance of context, or situation in which an experience occurs influences the way people perceive, experience, and interpret the world around them [23], this contextual perspective contributes greatly to perceptions held by rural residents, in this case among the rural residents of Indiana, USA. --- Research setting The study was done in Indiana, US. Indiana is located in the Midwestern US, with a total population of 6.8 million in 2020. Indiana has 730 towns and cities where 491 towns are considered as small towns and 72 as rural. The study was done in collaboration with the Indiana Rural Health Association , a non-profit agency working to improve the health of residents in Indiana. Other partners include Purdue Nurse Managed Clinics and Purdue Health and Human Sciences Extension Educators. There are four nurse-managed and nurse practitionerled clinics in northern rural counties of Indiana affiliated with Purdue University School of Nursing. Purdue Extension Teams are in 92 rural counties and support social and economic welfare of the rural communities. The study was funded by the AgSEED short for Agricultural Science and Extension for Economic Development, Purdue in April 2020, however, the study got delayed for a year due to the COVID-19 pandemic. Data collection began in Jan-April 2021. Ethical approval for the study was obtained from the Institutional Review Board, Purdue University, West Lafayette, IN, USA. The study followed the revised ethical guidelines as provided by the University during the pandemic. --- Participant recruitment and sampling Participant recruitment was done with the assistance of community partners as listed above. Online flyers and posters containing the study information sent through emails, and social media using the Facebook page of collaborative partners in the study. The snowball approach was also used for recruiting eligible participants. The participants approached the research team using emails and phone calls. The study information and consent form were sent online to the eligible participants along with the preferred timing for scheduling the interviews. A purposive sample of rural healthcare providers and community stakeholders was obtained. Healthcare providers included physicians, nurse practitioners/ clinical nurse specialist, registered nurses, case managers, social workers, and chaplains, whereas stakeholders were defined as managers and administrators of different homecare/hospice agencies, clinics, and hospitals. All the participants were 18 years and older and experienced providing generic or specialized palliative care services to the patients and families residing in the rural communities of Indiana, USA. Participants were recruited in assistance with the IRHA, Purdue Ag Extension, and Purdue Nurse Managed Clinics in Indiana. --- Data collection and analysis Data was collected using semi structured interview guide. The interview guide was developed using Penchansky & Thomas [19] framework and relevant palliative care literature and was finalized after having an expert opinion from IRHA and palliative care experts from the university. The five specific domains of access such as accessibility, availability, affordability, acceptability, and accommodation of services described in the framework is added in the interview tool. Addiitonal questions were added including awareness, cultural and psychosocial concerns, and transitions in care. Semi structured interview guide allows an open dialogue between the researcher and the participants, often supplemented by follow-up questions, probes and comments. Using this flexible interview tool, the researcher is able to collect rich and in-depth meaningful data while exploring participants' thoughts, feelings, expreinces, and beliefs about a particular topic under study [24]. Interview guide is provided in Table 1. Interviews were conducted by the PI along with a doctoral student. PI is primarily a qualitative researcher and has done multiple qualitative interviews before in her previous studies. During the initial five interviews, PI sat with the doctoral student during the inteviews, facilitated the interviews and provided the training and advice to the student to conduct the future interviews. Once the student felt comfortable, rest of the interviews were conducted by the student herself. All the fifteen interviews were done online using Zoom or Microsoft Teams as preferred by the participants. Before each online interview, participants were explained about the study, asked to provide verbal consent along with permission to record the interviews. Each interview was about 30-60 min on average. Interviews included asking participants about the demographic information as well as their experiences and available palliative care services in the area. All the interviews were recorded, transcribed, and stored in the institutional online data repositoryr after removing all the personal identifiers. A thematic analysis approach was undertaken to analyse the data using a constructvist interpretive paradigm [25]. The particular paradigm assist the researchers to understand and interpret the underlying meanings in the data in order to better understand the phenomena under study. The data analysis process involved inductive reasoning, constant engagement, testing, challenging preliminary interpretations and, finally, conceptualizing to understand the meaning, concepts, and patterns generating in the data. All the recorded data was transcribed, read, coded, and put into separate categories by a team of 3 researchers including PI along with two doctoral students. Each category formulated was defined, discussed among the team members and were put under separate themes. Regular meetings were held among the team members to discuss the codes and categories emerging in the data as well as to process continuous reflection, ongoing critical questioning and coherent reasoning to avoid any premature closure or overdetermination of the patterns and meanings in the data [25]. An iterative process of analysis was undertaken, emerging categories, linkages, patterns, and themes in the data were shared among the research team members and a final thematic list was generated with consensus among the team members [25]. To ensure trustworthiness and and rigor, a separate audit trail was kept about all the research activities. Separate logbooks, field notes, and reflexive notes were also maintained throughout the study. --- Findings Demographic profiles of the 15 participants include age, gender, job title, facility type, years of experience in palliative care and palliative care training experience. The demographic findings are given in Table 2. Findings obtained from the interviews are categorized under various themes. The themes reflect barriers and facilitators to palliative care access among patients and families in the rural communities. Seven themes describing barriers to palliative care access emerged as: misconceptions about palliative care as an underrecognized specialty; lack of trained palliative care providers; late involvement of inpatient palliative care and community hospice services; inadequate palliative care training for clinicians; financial barriers to palliative care development; cultural barriers and geographical barriers. Three themes describing facilitators to palliative care emerged as: hospital administration supporting palliative care growth because of financial benefits; hospital nurses helping identify patients with palliative care needs; word-of-mouth recommendations of hospice services in the community. --- Barriers to Palliative Care Access Misconceptions about palliative care as an underrecognized specialty The lack of a clear and well-disseminated definition of palliative care as a specialty impedes its recognition and understanding among healthcare professionals. Participants had multiple concerns regarding the clear definition and regulation of palliative care services: --- "So typically, the timeframe that people really look at palliative care is two years. People with a twoyear survival from their illness … two years or less … I don't think it's official [definition], but that's sort of a commonly accepted [definition]. " . "…defining palliative care [is important] …just making it distinct from hospice. " . "I have been concerned in the past about hospice agencies and other health care systems kind of terming a program palliative care when it's really not at all. " . Participants also verbalized not having clear guidelines for practicing palliative care in outpatient and community practice settings as compared to inpatient areas leading to confusion, misunderstanding, and delay in providing adequate palliative care services. don't think there's much guideline at all when we're talking about palliative care in an outpatient or in the community. " . Often palliative care is understood as 'giving up on you' or 'it's the end of life' which makes it further difficult for the clinicians to communicate about or make referrals for palliative care services. Such an understanding is common among the clinicians as well as the general public. --- "Oncologists still have this idea that If I tell people I'm gonna send you to palliative care, I'm giving up on you and I'm not going to tell my patient. " . "We [inpatient palliative care team] get barriers in that we tend to deal with other physicians sending mixed messages [about palliative care] to patients. " . Similarly, patients and families often equal palliative care with hospice care. --- "And people think it [palliative care]'s the same thing as hospice, you know, so those words don't mean different things to people" . Other than lack of understanding and clarity about the palliative care, another major barrier is the lack of trained palliative care providers in the area, another theme in the study. --- Lack of trained palliative care providers Most participants shared that there are not enough trained or specialized palliative care providers. In inpatient settings, there are only a small number of specialists, including both physicians and advanced practice nurses. First of all, palliative care is not a well-paid specialty: "It's not a field that pays that well, you know there's a lot more ways to make a lot more money. " . --- Also, some palliative care providers may only practice part-time: "There was one provider at X [a local hospital] who was an oncologist. He really only did inpatient palliative care 25% of the time, which was ridiculous, like you got a bigger hospital than we do…" . Despite rising demands and multiple benefits of palliative care, it was sad to note that some hospitals had downsized their palliative care team and were planning to close these services in the near future. "Hospital decided to downsize the palliative care team, so now they're no longer in house. They are more with the oncology group and mostly outpa- --- tient. " . Participants conveyed that there lacks outpatient palliative care in the community and therefore, the quality of care suffers. They feel limited in providing comfort care, pain management services, adequate consultation, and guidance to improve the quality of life of their patients and families. "We don't have a palliative care team [in primary care clinic], and I mean actually palliative care. It's something that's come up quite a bit in our discussions in our meetings because quite a few of us are now getting palliative care patients all of a sudden, and how do we provide the best quality care? How do we assess where they are?" . --- "We don't really have anything at all in our eight county area for outpatient services. " . Following up on patients residing in rural and remote areas after their discharge was an additional concern among the service providers. --- "It sometimes is difficult to try to make sure you can get follow up with those people [after hospital discharge] and sometimes one of the reasons we [inpatient palliative care team] end up following them is because I can't really get anybody to write the opiates [in the community]. ". Participants also added that most resources are available and limited to hospice agencies only. --- "In some areas and some counties in Indiana, less hospice agencies even go to those areas [remote or rural areas]. What a lot of people don't understand is that out in the community there's really not palliative care resources outside of Hospice. " . Lack of resources and trained providers in the community often resulted in delayed care and lengthy care transition process, another major barrier in the adequate access to palliative care in rural communities. --- Late involvement of inpatient palliative care and community hospice services Participants shared that both inpatient palliative care consultations and community hospice services are introduced too late as the disease progresses which causes additional sufferings for the patients and families managing illness, decision making, and coping with additional care responsibilities. --- "I feel like [inpatient] palliative care is brought in a little too late. We find that the consultancies are happening when the patient is already to a point where it looks like they're going to pass sooner than later. " . --- "They [patients]only get to us [Hospice] when it is the only option. " . "Still, because people [clinicians and patients] don't get the idea that palliative care should begin way upstream, you know it should begin literally at the time of or shortly after diagnosis of a life limiting illness. ". Participants also reported that clinicians' training background, self-attitudes, and practice philosophy may impede the early involvement of palliative care for patients with life-limiting illnesses and therefore, needs to be considered a high priority for improving the palliative care services in the area. --- "I think for a lot of doctors. It's a little tiny bit of a pride thing. Hospice means they lose their patient. " . "Especially for physicians and surgeons who can take really aggressive measures to treat a patient. They know all of the advanced treatments and surgeries and technology and that they want to exhaust all of their actions before considering getting palliative care involved. " . Training is essential in the provision of adequate palliative care in the community, discussed as another theme in the study. --- Inadequate palliative care education and training Many of the participants acknowledged that palliative care education/training for clinicians in all care settings is inadequate and poses major communication and illness management barriers for the rural communities. --- "Palliative care wasn't a thing when they [physicians] were training, so they never knew they never had it. They never used it; they never knew how to incorporate it into their practice. " . "I only had like two nurse practitioners when we first started. But during that time, you would think that people would get to know what palliative care does, and yet for the vast majority of my colleagues, It doesn't seem to resonate with them and some of that is sort of cultural within their specialties. ". "Palliative care, I'm not too sure because that to me is actually been a new term, especially this last year I've been learning about it, so I feel like our services [in the nursing home] for palliative care probably not as great as Hospice care, since I feel like we've had known about Hospice care for a long time. " "Even our staff they probably really don't know the difference between palliative and Hospice care. " . Participants indicated that there is no requirement for continuing education specifically in palliative care for clinicians other than palliative care specialties: --- "I don't think anyone's required to have continuing education hours on palliative care. " "I don't find my hospital pushing for it [palliative care continuing education]" . Consequently, most clinicians lack confidence and are often not comfortable talking about uncertainties, psychosocial and spiritual concerns, advanced care planning, pain management and comfort care treatment options available in the community with the patient and families. --- "We have had our nurses talk to the families about it over the phone, but I think depending on the nurse and how much experience they have communicating with the family on hospice care, a lot of the families come at the nurses and are like just getting so angry at them. " . "Why we [inpatient palliative care team] get involved is that the nurse has been talking to the patient and they'll start bringing things up, and they'll be like 'I can't talk about that [palliative care] with them' . " . "I have had some primary care providers call me [palliative care specialist] and say, hey, here's this patient. Sometimes they just want to ask me questions. You know, I think they need more support with this. " . Participants reported that often it is difficult for them to communicate or respond to unique palliative care concerns and cultural needs, especially among specific minority rural communities. --- "Every time I see the palliative conversations take place with the Amish specifically, I feel like it's like a brand-new concept. I'm not sure in their culture what palliative care means to them. " . It was also interesting to note that spiritual or existential care i.e. an important aspect of palliative care philosophy is often relegated to chaplains in the care settings. For most nurses and doctors, attending to spiritual and existential concerns is the responsibility of chaplains or faith healers. For some, due to time constraints and workload, spirituality is often given less priority among the other domains of palliative care. "It's a very valuable thing to have dedicated spiritual care and social work. Because they both can have much more fruitful conversations with families, more so than I [palliative care physician] and my NPs can. We're still…though we may be a little more enlightened. My skill set is still not there. With either chaplaincy or the social work part of it, so I'm mostly medicine to begin with. " . --- Financial barriers Among other barriers, financial barriers were also reported impeding palliative care access and provision among communities. Participants highlighted that finding financial support for palliative care development can be challenging: --- " 'We [hospital administration]'ll keep you [palliative care team]around now' but it doesn't mean that they're willing to fund anything out of the ordinary that you still have to go to him and say, OK, I have to beg and beg and beg and beg and beg, then we have another nurse practitioner. " . "Really our [palliative care team] mission is the same as everybody else. They [hospital administration] want to make money so everything that you present to the administration has to have the financial component front. " . "I know that X [a local hospital] had an outpatient clinic for a period of time but could not continue to support it from a staffing perspective. " . "We [nursing clinics] don't have the resources to provide it [palliative care] … I haven't looked a lot to see what HRSA has, but I don't think it's one of their priorities right now for funding. …the funding base needs to be there… we [nursing clinics] would have to be able to apply for funding or something. " . Participants conveyed that financial constraints can be a barrier to palliative care access among patients and families, especially those who are not eligible for Medicare: --- "I think definitely the financial piece of it is a huge concern also for families. You know whoever has a life ending illness ends up on disability and then the partner can't work, and the family circles financially. That's a major concern. " . "Especially like some of our clinics where we have that high Hispanic population where they are our self-pay and don't really have any insurance or anything like that to be able to afford it [palliative care]. " . Funding and resources were found essential to continue and support palliative care training and services among providers and therefore, need special consideration at both systems and policy level. Besides financial and other barriers, participants also reported several contextual and cultural barriers in accessing PC services which are reported as the following theme in the study. --- Attitudes and beliefs around palliative care Participants perceived that often patients' and families' own attitudes, values and beliefs around palliative care serve as a barrier to its access and provision. --- "People who are struggling with their diagnosis [would not want palliative care]. You know they don't really like the idea that they have a life limiting illness and that's what we [palliative care team] talk about. " . "It [palliative care] is a discussion that is still very uncomfortable for a lot of my nurses. I think a lot of it too is it's uncomfortable at a level of even the patients just refused to hear about it, so there is a lot of them that are in denial" . Patients and families deny or lack a clear understanding of palliative care or are unable to accept such care for their family member suffering from advanced illness due to fear, familial and cultural roles and expectations, values and beliefs about life and death. --- "When a physician initiates it, it's just that fear from the parent. it is pushed back in there. They shut it out, you know like 'no, I don't want to hear that' , and I think sometimes it can make the medical team a little uncomfortable. " . "People think that a miracle is going to happen, and which is not going to, so they kind of disregard the discussion regarding how serious it is with their loved ones. " . It is significant to note that most patients and families residing in the rural communities in Indiana belong to farmers' communities. Farmers' healthcare beliefs including palliative care delays were highlighted by participants. "I find that a lot of the patients that are farmers. They usually don't seek medical treatment until it's like absolutely necessary, like last case resort. I feel like palliative care is more proactive and you should be getting involved super early. It's more of a preventative thing in my eyes and I don't know if that's something they prioritize in their health care. " . --- "A lot of people from rural populations on farms or you know they maybe work in their town and so they might not take the time to be seen until it's truly an emergency and there have been patients who got chest pain for months and are just getting around to getting it looked at. " . Farmer families may have a better acceptance of death and thus don't perceive the need for palliative care. --- "I think farm people in general are a lot more accepting of death and a lot less afraid of death. I think they just get it. It's inherently part of life on the farm that you have life on the farm but there's also death on the farm. " . --- Geographical barriers The geographic distance can be a barrier to access. Long travel to appointments is especially challenging for patients in poor health conditions. --- "One of the last guys I [palliative care physician] saw as a home visit was a gentleman who was getting transfused all the time. He had leukemia and a different kind of cancer and he had heart failure, whole bunch of stuff and he was in a wheelchair and he had chronic swelling in his legs and it was just a big ordeal every time his wife had to come and bring him in to do anything. " . "Location is another huge barrier because if they have to go into you know into Lafayette or into Indy or whatever a lot of them don't have that transportation. " . For home-based services, hospice agencies often have distance limits that professionals may not be able to offer home visits to remote areas. This further gets complicated during harsh weather situations. --- "[Long drives can be a barrier to access], of course, if they [patients] are very remote. Our Hospice tries to make sure we can always get to a patient's home within an hour, so obviously we're not a rapid response team. " . Other than multiple barriers, the study also found out certain facilitators to improve the access and provision of palliative care services. These themes are as follows. --- Facilitators to palliative care access Financial gains supporting palliative care growth Few participants believed that the provision of inpatient palliative care has resulted in reduced length of stay, repeated hospitalization, and cost-effectiveness, all of which are considered as key reasons why palliative care has received a lot of administration support in recent days. This has served as a major facilitator in upbringing the support for palliative care. "We [inpatient palliative care team]'re following up after we've seen them in the hospital is because we're more worried about them getting re-admitted and all of those kinds of things. " . --- "We have huge savings in length of stay, so you identify early on what people goals of care are and what's a you know the best thing for their care. You can keep them out of the ICU's, and you can keep him from having expensive procedures, so the costs are down, the length of stay is down. " . Other than financial gains, nurses' role recognition in palliative care was another major facilitator found in the study as follows. --- Advanced nursing practice roles in identifying patients with palliative care needs Participants felt that bedside nurses as well as nurse practitioners in rural clinics can play a key role in facilitating and supporting palliative care. Nurse practitioners in rural outpatients clinics should be informed about adequate resources and training tools. These tools may help nurses to identify and communicate timely about patients who can benefit from palliative care. Advanced nurse practice roles could promote early goals of care planning, decision making, and proper illness management especially in low resourced rural areas. "We had a little like pocket card. That are the things that would make this patient appropriate for palliative care, and we hand those out to the nurses to try to get them used to. Hey, if you're seeing somebody that has multiple comorbidities, they've been re-admitted many times or within the last 30 days or whatever, they have complex family dynamics, they have psychosocial issues, they have uncontrolled symptoms. All these things were on the card. Please call palliative care. " "Now we have a pretty, I'd say, well informed nursing staff that works very closely with us, and often they're the facilitator for getting us consults. " . "My role is mainly trying to get palliative care on board and suggest to our primary physicians. Can we consult the palliative team to just talk to family about goals of care?" . Nurses as primary care providers can play an integral role in early palliative consultations resulting in better coordinated and effective holistic care. --- Creating awareness and informing the community about palliative care Participants felt that often people in rural communities are not aware of the services available in the area or sometimes services are located far from their own reach. As a result, they experience difficulties in managing care, go through additional burdens and stress. Healthcare providers conveyed that the hospice agents they work for have well-established multidisciplinary hospice teams and great service coverage, especially for Medicare beneficiaries. In the communities, good experiences with hospice spread within social networks that facilitate the usage of hospice services: "We [hospice] have a team. Every patient has a chaplain, social worker, nurse, home, health aide. We have a music therapist also, so the direct questions and discussions about their spiritual concerns. " "We've had families that need food. Need help with food and we can do some. You know some small grocery gift cards or whatever but as far as what the patients need directly. It's pretty much all supplied, I mean under pads for the bed, diapers, bathing stuff. " . "It [hospice]'s word of mouth and if people have had a good hospice experience, or a bad one, unfortunately, but it travels, 10 to 20. Maybe at their funeral they talk about how great it was. Or you know, whatever, and I think the rural population is actually pretty aware of hospice now. " . --- Discussion Study findings provide insights into several barriers and facilitators to palliative care access from the perspectives of providers who serve rural Indiana. To the authors' knowledge, this is the first study looking into palliative care access in rural Indiana. Within the context of inpatient palliative care, findings are evident that barriers to palliative care access among rural populations were critical. Palliative care, as a newer specialty that comprises end-of-life care or hospice, has not gained adequate familiarity among both clinicians and the public in rural Indiana. However, community hospice care has gained more familiarity as it has been long-standing and well covered under Medicare. The findings in this study highlighted that misconceptions about inpatient palliative care such as equaling it to end-of-life care or hospice were common among clinicians and patients and families. Similar results were found in our recent scoping review where multiple studies pointed out to the fact that healthcare practitioners including primary care providers, nurses and practitioners either lack the general understanding and clarity around palliative care concepts [26]. Despite the recent development in palliative care awareness, education and training, providers remain unable to recognize the needs and significance of it which further results in delayed care for the patients and families. There could be several reasons for this emerging argument. Often the reason for the inadequate provision of palliative care is not limited to only the knowledge and awareness of providers, it is how the service is perceived or practiced in the facility. The organizational climate and culture are not unable to support the ability of providers for adequate palliative care referral and consultations in the facility. Providers' own values and belief system, lack of continuing education, and not having the necessary tools or education to recognize the needs and requirements of chronically ill patients and their families are often additional barriers in the provision of palliative care. Multiple definitions and regulatory guidelines around palliative care cause further confusion and lack of clarity towards its better understanding in care practices among providers [27]. The lack of community-based palliative care approach found in our study aligned with the findings of previous researches [28][29][30][31]. Inpatient services with minimal community outreach can be a critical barrier to care access as this approach excludes the vast majority of people living with life limiting illnesses in the community. Within nurse managed primary care clinics in rural communities, providing palliative care was found to be an overstretch of practice scope due to the lack of funding support. Even the existing inpatient services were involved too late, mostly towards the end of life. Providers' training background and practice philosophy and the culture of specialty areas can vary a lot that root their exposure to and beliefs in palliative care. Inadequate palliative care training for all clinicians was fundamental to provider misconceptions and late or no referrals to palliative care for patients with life-limiting illnesses. Our findings also indicated that the lack of acceptance of palliative is associated with the fear and denial among patients and families. Palliative care is often viewed as 'giving up' or 'end of life' as evident in the findings. Such cultural meanings and values associated with palliative care were also major barriers towards appropriate communication, referrals, and consultation. Literature also supports that in rural communities, there are multiple contextual barriers impeding PC services including occupations and regional and generational cultures [30,32,33]. From the rural Indiana context, for example, a lot of people are farmers who commonly play down preventive health services. Death and dying may be well accepted spiritually among these farming communities but the benefits of palliative care beyond the end of lives need to be further disseminated. Moreover, the geographic location, lack of transportation, and financial constraints were found to be other barriers in accessing palliative care, especially in rural communities. The nature of work and employment, healthcare facilities located far from home, limited availability of transport esp. during winters or harsh weather conditions often create additional challenges for the people in the rural areas to access services [30,31,33,34]. In addition to the limited availability of palliative care providers and facilities in rural areas shown in other similar studies [35][36][37], our study highlighted the challenging atmosphere for palliative care to grow in hospitals as a not well-paid specialty and fewer providers choose to practice palliative care. Furthermore, some facilities were downscaling the available palliative services and resources in rural areas due to a lack of funding. It is also important to note that this study was done during the second phase of the Covid pandemic, this was a significant time when palliative care service was highly needed among the patients and families. Lack of palliative care availability during these times could have further added to healthcare disparities and sufferings among the underserved rural communities in the area. Apart from previous studies, our study highlighted some novel perspectives around facilitators of palliative care including strengthening advanced practice nurses' roles to improve palliative care access and utilization in the rural communities. Findings are evident that rural clinics run by nurse practitioners were providing referrals for the palliative care consults in the areas. Literature is evident that nurse practitioners are cost-effective, resourceful, and found to be effective in responding to the primary care needs esp. in rural and underserved communities [38]. Advanced practice nurses can play a key role in reducing disparities and improving the community outreach. Findings also supported that palliative care trigger tools were found to be useful among nurses to recognize palliative care needs and consultations for patients and families, especially in inpatient settings. More robust palliative care training and education and additional incentives to participate in these educational training sessions were also highlighted as major facilitators of PC in healthcare and community settings. Few providers in the study did perceived financial gains or economic benefits as a facilitator of palliative care in the study however given the limited scope of the study, further studies are suggested to examine in detail about the economic or financial benefits of PC provision in rual settings. --- Strengths and limitations of the study The study is first of its kind among rural communities of Indiana and provides an enriched understanding of palliative care access and delivery from a diverse perspectives of healthcare providers including nurses, nurse practitioners, social corkers, chaplains and nurse managers from both inpatient and outpatient settings. The rich descriptions generated may help us to critically understand and identify ways to improve the palliative care delivery in the rural communities. Limitations include non generalizability of the findings as the findings are limited to rural Indiana context. Moreover, there is a possibility of social desirable responses of the participants in the study. At some instances, we faced internet connectivity issues while doing the interviews and had to reschedule some interviews at different times. It is possible that some participants may not be comfortable using the online medium for interviews that may have possibly influenced findings in the study. --- Implications Palliative care is a human right, recent studies have shown its multiple benefits from specific health, social, and economic perspectives [39][40][41][42]. Especially in rural communities, access to palliative care requires a lot of attention to reduce the healthcare burden and prevent disparities. Our study strongly suggests the need to increase the awareness of palliative care among both users and providers at education, policy, and practice levels. Healthcare policies need to put more emphasis on the utility and provision of palliative care. Efforts need to be in place for creating a clear definition of palliative care. Advanced practice nurses' placement in rural regions can further enhance the availability and access of palliative care in rural communities. Proper policy guidelines and regulatory mechanisms tailored to the cultural and contextual needs of the rural population needs to be in place and emphasized among the healthcare community. Future funding support is needed to enhance community outreach services and resources to improve palliative care access and availability in rural areas. --- Conclusions Lack of palliative care access can be a major factor affecting the healthcare conditions in rural and underserved communities, overall influencing rural growth and wellbeing. Further emphasis and efforts are needed at both practice and policy levels to enhance the awareness, importance, and availability of adequate palliative care services in rural communities. Strengthening advanced practice nurses'roles and appropriate allocation of funding and reosurces can create new pathways for ensuring accessible palliative care services in the rural communities of Indiana. --- --- Authors' contributions N.L. designed the study. N.L. and Y.C. performed the online interviews and analyized the data. N.L contributed to the introduction, literature review, methods, and discussion sections of the manurscript. Y.C. contributed to the literature review, results and discussion sections of the manuscript. The author read and approved the final manuscript. --- --- --- Competing interests We confirm that there are no known conflicts of interest associated with this publication. Received: 29 July 2021 Accepted: 7 February 2022 • fast, convenient online submission • thorough peer review by experienced researchers in your field --- • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year --- • At BMC, research is always in progress. --- Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: ---

With the growing aging population and high prevalence of chronic illnesses, there is an increasing demand for palliative care. In the US state of Indiana, an estimated 6.3 million people are living with one or more chronic illnesses, a large proportion of them reside in rural areas where there is limited access to palliative care leading to major healthcare inequities and disparities. This study aims to identify common barriers and facilitators to access palliative care services in rural areas of Indiana from the perspectives of healthcare providers including clinicians, educators, and community stakeholders. Using a community-based participatory approach, a purposive sample of palliative care providers (n = 15) in rural areas of Indiana was obtained. Penchansky and Thomas (1981) theoretical framework of access was used to guide the study. A semi-structured individual in-depth interview guide was used to collect the data. All the interviews were conducted online, audio-recorded, and transcribed. Barriers to palliative care include: misconceptions about palliative care as an underrecognized specialty; lack of trained palliative care providers; late involvement of inpatient palliative care and community hospice services; inadequate palliative care education and training; financial barriers, attitudes and beliefs around PC; and geographical barriers. Facilitators to palliative care include financial gains supporting palliative care growth, enhanced nurses' role in identifying patients with palliative care needs and creating awareness and informing the community about palliative care. Robust education and awareness, enhancing advanced practice nurses' roles, increasing funding and resources are essential to improve the access of palliative care services in the rural communities of Indiana.

Introduction There is growing acknowledgement that, with respect to both crude and standardised measures, England's coastal communities have significantly higher needs for NHS, social care and public health services than their inland counterparts. The Chief Medical Officer's 2021 annual report highlighted the substantially higher burden of physical and mental health conditions in coastal communities, often with lower life expectancy. This is partly explained by the fact that that coastal populations tend to be both older and more deprived than non-coastal populations. However even after adjusting for these factors , there still appears to be a "coastal excess" in the prevalence of disease and risk factors . There is also evidence of a significant health service deficit in terms of recorded service standards, cancer indicators and emergency admissions in coastal communities. Health Education England's analysis for the 2021 CMO report found that, despite coastal communities having older and more deprived populations, they have 14.6% fewer postgraduate medical trainees, 15% fewer consultants and 7.4% fewer nurses per patient . Reasons for the mismatch between workforce and disease prevalence in coastal areas are not understood. However, it is worth noting longstanding concerns that, due to the inherent circulatory of activity-based resource allocation models, systematic biases introduced in the 2002 AREA formula may still be disadvantaging areas serving older demographic populations. There are also concerns that NHS funding formulae do not adequately capture the difficulties providers face in achieving economies of scale, as their coastal location gives them a 180-degree catchment area and the populations they serve tend to be geographically dispersed. With no fundamental changes to the system of NHS resource allocation in sight, challenged health and care systems in coastal areas need to start working differently. In this paper, we propose that, by shifting the balance of care from reactive to preventive, supporting early diagnosis and care at home and using advanced data analytics and Artificial Intelligence to improve clinical workflows, digital transformation has the potential to promote the "quadruple aim" of better population health and reducing inequalities, improved experience of care, healthcare team well-being and lower system costs. However, there are a range of barriers to realising this potential, including variations in digital maturity which tend to disadvantage coastal areas. We discuss these barriers and identify strategies that policy leaders and health service managers can deploy to enable digital transformation. --- The coastal excess The "coastal excess" in disease burden in England largely reflects the fact that coastal populations tend to be both older and more deprived than non-coastal populations. 16.6% of coastal residents live in one of the 10% most deprived Lower Super Output Areas in the country, compared with 8.4% of non-coastal residents. At the other end of the scale, 5.1% of coastal residents live in one of the least deprived LSOAs, compared with 10.6% of people living in non-coastal areas. Moreover, although the proportion of people aged 65 and above varies with deprivation, it is always far higher in coastal areas. Overall, 21.0% of people living in coastal LSOAs are aged 65+, compared with 17.8% in non-coastal LSOAs. With both age and deprivation associated with an increased risk of disease, it is hardly surprising that analysis carried out for the CMO 2021 report found higher prevalence rates of a range of diseases in coastal communities , the coastal excess of cardiovascular diseases and chronic obstructed pulmonary disease being particularly stark. However, age and deprivation did not fully account for the difference. Indeed, once age, socioeconomic status and ethnicity were accounted for, COPD and mental health prevalence rates were found to be nearly 11% higher in coastal than non-coastal communities. There was also evidence of very poor outcomes for children and young people, with high rates of hospitalisation for self-harm, alcohol and substance use. Reflecting a shift in the distribution of children living in poverty since the 1990s, this may be an early indicator of a future public health crisis in these communities . The reasons for this coastal excess are not fully understood. However, research points to the emergence of new and worrying patterns of deprivation and associated complex inter-linked challenges amongst "lagging" coastal communities, including high levels of unemployment, low incomes, seasonal jobs, low skills and poor educational outcomes, detrimental patterns of selective in-and out-migration, unaffordable housing, hidden homelessness and high rates of anti-depressant and opioid prescribing . Economic decline and socio-economic deprivation in coastal areas exacerbate the risk of developing non-communicable diseases , particularly at a younger age . Low pay and low job security reduce access to material resources such as housing and healthy food and increase exposure to occupational hazards . Low job status with less autonomy and income insecurity are also key risk factors for chronic psychological distress , a risk factor for chronic inflammation and in turn the development of NCDs . As noted above, a significant excess has been found in coastal LSOAs with respect to poor mental health. Insofar as they are known risk factors for poor parenting and associated impacts on children's development, financial insecurity and poor parental mental health may be one of many factors behind the increasing gap in health inequalities among children and young people between the core and periphery. Education can provide a protective effect in the longer run. Predicting employment, income and access to material resources as well as psychosocial well-being and health behaviours, education is arguably the single most important modifiable social determinant of health . Various mechanisms have been proposed . First, education leads to better-paid and more stable jobs with greater autonomy and less exposure to psychosocial stress. Second, there is growing evidence that education plays a direct role in developing psychological resilience . Resilience has been identified as a predictor of health in children and adults , possibly through protection from stress-induced immune changes . A third causal pathway linking education, resilience and health is self-efficacy, a concept that refers to an individual's belief in their ability to exert control over their health risks behaviours. Against this background, the fact that educational performance is significantly worse in the coast then in inland areas should be a cause for public health concern. --- The role of digital transformation in shifting the balance of care A vast array of digital health technologies exist that can prevent ill-health, promote well-being and support the delivery of health and social care. This includes smartphone apps, wearable devices , smart home devices and environmental sensors, virtual reality, surgical and care robotics and platforms that provide remote healthcare . As the uptake of such technology increases, the potential of artificial intelligence (and associated machine learning and national language to monitor individual patient risk, design intelligent triggers and support health care decision making will also escalate. It is not within the scope of this paper to provide a comprehensive review of the potential of DHTs to support the "quadruple aim". Instead, we highlight areas where there are likely to be technological solutions to key coastal challenges. --- Digital mental health interventions Among children and young people who generally enjoy good physical health, mental health problems are a particular concern, in part because of their relatively high prevalence, in part because of their impact on future developmental trajectories, including risk of premature NCDs. As noted above, CYP in coastal areas are showing higher prevalence rates of mental distress. Could DHTs offer a solution to managing this higher burden? As we discuss below, one of the challenges to assessing the role of DHTs in health prevention and promotion is that relatively few are "evidence based" . One of the few systematic reviews in this area found some support for the clinical benefit of DHTs, particularly computerised cognitive behavioural therapy for depression and anxiety in adolescents and young adults but uncertain evidence about the benefits for managing attention deficit/hyperactivity disorder, autism, anxiety, depression, psychosis, eating disorders and post-traumatic stress disorder . Indirectly, DHTs may promote young people's health chances by e.g., mediating parenting behaviours . Plausibly, more effective innovations for coastal youth would address the complex interlinkages between poor employment prospects, low aspirations, poor mental health and the potential of e.g., outreach from metropolitan-based corporations to both mentor and offer online job opportunity experience. We are not aware of either initiatives focused on coastal outreach or evaluation evidence to support investment in this approach. Uncertain evidence about the benefits of DHTs needs to be balanced against known concerns about the current state of Child and Adolescent Mental Health Services in the UK, which has been overly medicalised , operates a high threshold for referral-prior to the backlogs exacerbated by Covid19, only 25% of children or young people referred to specialist CAMHS were seen and which are now subject to waiting times that commonly exceed one year. Higher levels of digital engagement by children and young people also suggest that DHTs may offer a unique opportunity to address unmet needs. Against this background, this may a fruitful area for practitioners and policy makers in coastal areas to explore. Robust evidence of the effectiveness of DHTs in improving the mental health of adults is similarly lacking. Again, in the context of a shortage of mental health provision, there is growing interest in the potential of AI-based chatbots, avatar therapy and companion bots in managing psychiatric symptoms and augmenting therapeutic treatments . A failure to meet evidential requirements together with clinician concerns about safety, accountability and legal responsibilities have limited the implementation of such technologies into current care pathways. A balance may need to be struck in highly challenged health care systems between risk aversion and an openness to exploring the potential of innovation in ways that are methodologically robust but, in the context of rapidly evolving DHTs, may not be conducive to classic randomised control trial methodologies, a point to which we return below. Again, considering the role of DHTs within the vision of addressing the wider determinants of health, more research needs to be conducted on the potential of digital technologies to promote self-efficacy, capabilities and well-being in a more holistic understanding of people's lives as opposed to narrow patient activation measures. --- Digital support for primary and secondary prevention The evidence base for DHTs is more established with respect to secondary prevention than for primary prevention. For example, the use of wearables such as Apple watches to monitor a range of measures from physical activity, heart rate, blood pressure, sleep patterns to fertility patterns is defined as "consumer" health care and often strictly separated from DHTs that are deemed medically safe. Insofar as such devices offer the opportunity to exploit continuous as opposed to intermittent data in diagnosis and management, there may be a case for being more flexible about incorporating data from consumer devices into health care records. In contexts that allow the real-time transfer of physiological data from wearables to electronic patient records, there is evidence of benefit to e.g., elderly care generally , as well as supporting the early diagnosis of deterioration in e.g., cardiovascular, neurological and pulmonary diseases and the identification . Smart devices and environmental sensors can also facilitate the remote monitoring of the home environment as well as important physiological signs and activities of occupants, supporting functions from the simple to the more complex identification of anomalies in usual behaviours , including diet, physical activity and social interaction through, e.g., acoustic sensing, Radio Frequency sensing and passive environmental sensing. Novel data science and machine learning techniques offer enormous potential for analysing such data and improving safe care in the home including, if necessary, from a distant facility. Virtual reality is increasingly used in healthcare as an educational tool. Hardware limitations of headsets have raised questions about the practicality of prolonged use by vulnerable or frail patients, though recent advances in mobile-based light-weight headsets have been shown to be feasible to deploy in challenging healthcare settings. There are now plausible studies demonstrating the potential value of VR in supporting a range of physical, mental, or psychosocial health outcomes , including pain management , cognition and depression . Care robotics are an emerging technology in the UK and can support independence for both adults and children as well as addressing staff shortages. When robots are equipped with AI and depth cameras, they can monitor patients as they go through prescribed exercises, tracking progress more precisely than the human eye. Social robots can interact adaptively with patients to provide coaching, encouragement and monitoring, including in long-term care environments for older adults and those with dementia . They may encourage patients to comply with treatment regimens or provide cognitive engagement, though again, more research has been done on the internal validity of social robots than the factors that affect their implementation in real-world practice. There are also numerous applications of AI/ML, including risk stratification in population health management, AI-based platforms leveraging real world data to provide insights as to which patients can benefit from home-based technologies to manage the risk upfront to lower downstream costs. Voice assistants can be used to set simple reminders to e.g., take medicines. As the uptake of wearables, personal devices and sensors increases, the potential of AI based technologies to monitor individual patient risk and design intelligent triggers will also escalate. --- The smart hospital As with supporting prevention, early diagnosis and management at home, there are numerous types of technologies that can transform the quality and efficiency of care within the hospital environment including: location recognition and tracking technology, coordination and communication services, Internet of Things-based technology, mobile devices and wearables, telehealth, AI, robotics and extended reality and 3D printing . Further, the rise of intelligent buildings in health and social care such as smart hospitals, will accelerate digital technologies deployed specifically to support the entire lifecycle of estates as part of Building Management Systems and Building Information Modelling processes. Though lacking in formal definition, notionally, smart hospitals are born when a matrix of digital solutions converge in the real world through their considered integration into the fabric, footprint and flow of healthcare facilities-this includes in-flow, within flow and outflow. Ultimately such a convergence of technology and associated data will bring with it an embodied intelligence capable of improving patient outcomes and experiences, staff satisfaction and retention, operational efficiencies and environmental sustainability. While intelligent buildings are commonplace in the commercial sector, there remains a paucity of guidelines specific to health and care facilities for which managers and decision makers can follow. In large part this is due to the considerable variation in contextual nuance such as needs, resources, policies and practices. Importantly, smart hospitals are not a status to be attained per se, but a concept heavily influenced by contextual facture such as the baseline digital maturity of a facility and region as well as the innovation and transformation appetite and opportunity. Major healthcare estates projects, particularly greenfield constructions or redevelopments, present significant opportunities to accelerate digital adoption. Not least because they allow for technology to be incorporated into the fabric of the building in ways previously unattainable thanks to modern methods of construction, but also because they allow for core digital infrastructure to be architected from inception to allow facilities to avail themselves of contemporary technologies-including those yet to be conceived. Practical examples of technology within hospitals include robots which are increasingly used to assist in surgical procedures, manual handling, and to supplement and enhance cleaning [e.g., using Ultraviolet light ], perform routine tasks and support patients with anxiety and rehabilitation. However, within the published literature, much of the focus has been on the use of advanced data analytics and AI in conjunction with Electronic Patient Records to improve in-patient outcomes and, by supporting an optimal allocation of resources , cost-effectiveness. The American EPR, EPIC, is the most studied system internationally, in part because it holds the current leading market share in the US, where studies have demonstrated improvements in both clinical outcomes and efficiency . The example of just one of its 30 "Cognitive Computing" models, Clinical Deterioration, demonstrates the potential of improving prediction in ways that support patient outcomes as well as overall costs. Approximately 5%-10% of hospitalised patients encounter a severe adverse event during their hospital admission, including cardiac arrests and admission to an intensive care unit. Delays in the recognition of deterioration lead to delayed diagnostic and therapeutic interventions resulting in increased morbidity and mortality and associated costs. Since 2012, acute hospitals in the NHS have used the National Early Warning Score to detect acute illness severity and clinical deterioration . This is based on a simple aggregate scoring system in which a score is allocated to physiological measurements already recorded in routine practice when patients are present or being monitored in hospital. The EPIC Clinical Deterioration AI model, currently in use in 12 countries comprises a far wider range of variables and uses trained models to predict risk of critical illness and clinical deterioration in hospitalised patients. US studies using the EPIC clinical deterioration model have shown a dramatic improvement when the model was used in comparison to the US equivalent of NEWS. This is but one example of the possibility of replacing a simplistic and largely reactionary system with a proactive, predictive approach. However, the reasons why EPIC AI models are not being deployed in any of the UK hospitals that have procured the EPR shed light on the complex barriers to unlocking the potential of DHTs in the NHS and most particularly in coastal areas. --- Barriers to digital transformation Variations in infrastructural readiness To make the most of digital innovations, we need platforms for the real-time transfer of data from wearables, sensors etc. to EPRs. Despite a strong policy push and the catalyst of the COVID-19 pandemic, the NHS has had variable success with embedding a digital infrastructure and has consequently struggled to adopt eHealth innovations. Recent research for the Health Services Journal reveals that, in 2022, 27 acute trusts in England did not even have comprehensive EPRs . While some of these use smaller-scale electronic systems in individual departments, several continue to rely on largely paper-based patient records. It is important to note that hospital trusts in coastal and rural areas on England's peripheral are overrepresented among this group, reflecting previous investment, e.g., in Local Health and Care Record Exemplars and Digital Innovation Hubs, that benefitted only areas that were already judged to be "digitally mature". Areas such as Greater Manchester and London are now galloping ahead with respect to their digital health ecosystems and, as such, attracting further investment from both Government and industry. As a result, digital maturity would appear to be geographically concentrating rather than trickling down . While other parts of the health and care system have EPRs , the absence of a system within large acute hospitals tends to be associated with a lack of data integration in the wider local ecosystem. Innovative general practitioners may be monitoring their patients with e.g., wearables. However, in most cases, they will have had to purchase Application Programming Interfaces to link programme software to patient records and data will be manually monitored. This is manageable for small groups of patients. However, to scale up, data need to be integrated into an orchestration layer and, for large datasets, processed using advanced and automated data analytics. The goal would be to have a centralised "command and control centre" that can monitor system demand and have processes to proactively respond to it, e.g., by alerting responders to suitably calibrated triggers and to regulated responders. At present, large swathes of England have no orchestration platforms and there is systematic variation in infrastructural readiness. Indeed, in many areas work is still in progress linking primary, secondary and other data into "shared care records", the integration of which tends to be more advanced in LHCRE areas than in those that lacked LHCRE status and have pursued a "minimum viable solution". There is growing interest in the potential of shared care records to go beyond the better coordination of an individual's care by embedding linked information into population health management schemes. This should support the shift in the balance of care by providing intelligence on where to target more preventive and proactive care. Again, however, this is huge variation in the maturity of PHM across the country which reflects previous rounds of investment into digital integration. Restricted access to integrated data also limits its potential for shaping strategic intelligence. While this reflects legitimate concerns about privacy and security, it is important that current approaches to information governance are aware of the increasingly sophisticated methods that are available for data anonymisation and protection, lest such systems fall behind the technologies they oversee. --- The regulatory landscape By finding undiagnosed disease or indicators of clinical deterioration and thereby identifying people who could benefit from early intervention or who require urgent care, an effective "command and control" system that is using AI technology to process millions of data items clearly offers potential to promote population health, improve quality and outcomes, reduce demands on staff time and reduce overall costs . Given the significant financial pressures facing the NHS, particularly coastal systems serving older demographical areas, slow progress in data integration and orchestration is a lost opportunity cost. Yet, local health and care systems must negotiate wider institutional challenges to unlocking the potential of DHTs. For example, since 2021, AI that influences clinical decision-making has been classified as "AI as a medical device" , which thus requires regulation by the Medicines and Healthcare products Regulatory Agency . While the intention is to create a regulatory framework that is proportionate, precise guidelines about how to demonstrate that AlaMD meets regulatory approval have yet to be published. This creates a somewhat fuzzy area for procurement. Some, usually US-based clinical data platforms such as Orion and CioX are already being procured by some Integrated Care Boards and individual trust providers. These will comply with US federal and state regulations with respect to the development of their policies, procedures, and technology but not necessarily UK regulations. Slow adopter ICBs and providers may be more cautious due to other institutional drivers, such as requirements concerning cyber security, information governance and legal responsibilities. A tendency to prioritise compliance over ambition may be associated with the digital maturity of a health and care ecosystem as well as operational and resource pressures, which give rise to more siloed thinking . NHS procurement is also expected to be informed by robust evidence of e.g., quality, safety, effectiveness and costeffectiveness. Despite a proliferation of DHTs, few meet these evidential requirements. This lack of evidence probably says more about methodological expectations than of the effectiveness of DHTs themselves. The evidence standards framework for digital health technologies published by the National Institute for Health and Care Excellence provides a useful guide to standards relating to design, value, performance and deployment and the kind of information that companies developing DHTs needs to collect to meet each standard . However, the standards set out for even the lowest tier of technologies, such as smartphone apps and standalone software are quite exacting, particularly for the small and medium sized enterprises that dominate this sector, and which are unlikely to have the capacity to engage in the comprehensive evidence gathering set out in the guidance. This has given rise to an evidence gap, NHS procurers expecting information that many developers are unable to provide. It is common to hear complaints about cold calling by "digital snake oil salesmen" operating in the digital equivalent of an unregulated "Wild West". --- Cultural resistance Cultural resistance to digitalisation may stem from a lack of confidence-the digital health competencies expected of health and care professionals are now very wide ranging . Staff need to have confidence in the role of technology in delivering high quality, financially viable, resource efficient and sustainable services for their patients. There are also some concerns that, far from reducing demand, digitalisation can cause extra workload. There is not robust evidence for this, in part because of the methodological weakness of studies and it is generally assumed that health and care staff who feel burdened by their digital systems are working with outdated systems that have been poorly designed and implemented. Against this background, an eloquent account by Atal Gawande on how using the relatively new EPIC EPR is potentially contributing to physician burnout is a sobering read . To make the most of home-based DHTs designed to help people stay healthy and independent, citizens require reassurance too. Unfortunately, the very groups who are at increased risk of NCDs are also more likely to be digitally excluded. Using data from the 2020 Use of Communication Services survey, an estimated 10% of adults in the UK did not use the internet or have access to it at home. Old age, living alone, having a limiting or impacting condition and being financially vulnerable all significantly contributed to the likelihood of being digitally excluded in terms of use of or access to the internet . Broadband coverage levels are also lower in rural areas . For example, around one in six rural premises cannot access superfast broadband and over half cannot get an indoor 4G mobile connection on the four main networks. The reliability of digital networks in rural areas is also an issue . --- Concerns about environmental impacts Insofar as digital technologies are expected to reduce travel and smooth flows of activity within health care facilities, they are generally associated with a reduction in carbon emissions. In fact, the production and disposal of DHTs are not carbon neutral, the raw materials required for wearable technologies, robotics and devices requiring large mining operations, while a very small proportion of electronic technologies are recycled, raising concerns about e-waste . The energy required to store and process very large health datasets is also a concern though again there are many technological solutions to this, from green cloud computing to the use of intelligent building design to e.g., improve cooling and air flow management in large data centres. --- Enablers of digital transformation Central guidance and support In recognition of variations in digital maturity, NHS England committed over £12 m to its Digital Productivity Programme. However, the flagship target to have EPRs in 90 per cent of trusts by the end of 2023 is unlikely to be met. There are concerns that organisational turbulence at the centre, together with cost cutting and the departure of key digital champions have reduced the NHSE's commitment to driving through digital transformation, particularly in areas that need to level up. At the same time, the strong central mandate on productivity savings in the acute sector to balance the books is drawing attention from transformational efforts to shift the balance of care by strengthening prevention . The mandate from the centre needs to be more ambitious, retain its focus on digital levelling up and be properly resourced. Many learnings relating to the development and deployment of technologies are not shared in traditional academic journals with knowledge held by industry or service providers who embark on internally reported quality improvement programs. If we are to reduce duplication of effort and accelerate digital transformation far greater transparency in digital innovation methodologies and the implementation science driving change are needed, both successes and failures. --- Transparency and shared learning --- Enhancing digital readiness and architecting a learning health system To support improved healthcare outcomes and ensure the sustainability of public health systems, data-driven decisionmaking and continuous learning are essential. The coordination of large scale linked data including patient specific health and wellbeing information, patterns of health service utilisation and the ancillary operational processes and societal costs will accelerate value-based health care. Notwithstanding the fact that many systems in digitally immature coastal and rural areas are at the start of their transformation journey, there may be investments that could yield benefits in terms of both patient outcomes and cost savings. Focusing on prevention as opposed to the more common approach of supporting hospital discharge through "virtual wards" may be more cost effective in the long run. Given some variation between Integrated Care Systems in the deployment of digital innovation to support discharge vs. avoidable admission, it would be helpful to accrue some evidence of the relative cost-effectiveness of both approaches, a task for NHSE, but not one we suspect would be amenable to available data. The digital immaturity of coastal communities may prove opportunistic by allowing ICSs to architect the core infrastructure and data needed to effectively support value-based health care e.g., the routine deployment of preference-based patient reported outcome measures to support cost-effectiveness analysis and help to better demonstrate return on investment over a longer time horizon. Additionally, non-direct data is needed to contribute to evidence-based innovation and decision-making such as carbon emission. As part of the NHS net zero carbon agenda, digital transformation with the UK is inextricably linked with all construction and supply chain partners who will need to demonstrate their carbon impact and strategies to proactively reduce emissions by 2030, reaching net zero by 2045 . Greater transparency in the environmental impact of the health services ecosystem through disaggregated indices is therefore essential to facilitate accuracy in accurate carbon accounting and adequately demonstrate emissions reductions as a result of digital transformation. If ICSs are to invest in the use of DHTs for prevention, early diagnosis and care within the home, they need to set up orchestration layers that draw data from various platforms into a single "command and control" centre in which clinical teams identify triggers for intervention, an approach that should be increasingly supported through AI. Some coastal ICSs are doing this through pilots. However, integrating data, negotiating information governance barriers etc., is proving to be highly labour intensive. Perhaps more attention needs to be paid to understanding why such an obviously useful approach is so difficult to enact in practice and which institutional barriers need to be overcome to make better use of shared data and e.g., AI . Finally, active steps need to be taken to ensure that socioeconomic variations in digital inclusion do not exacerbate socioeconomic inequalities in health. This is a particular issue in coastal areas due to older demographic populations , variable broadband on the periphery and fears that, in responding to the current cost of living crisis, financially challenged groups are more likely to prioritise basic needs than internet subscriptions. In many such areas, the Voluntary, Community and Social Enterprise Sector is playing a critical role in addressing disparities in access to technology, skills, confidence and systems, in some cases such as Devon, through financial support from ICBs. Imaginative solutions are also being developed with respect to promoting access through e.g., community buildings. In short, the technological solutions are there, the vision and infrastructure are not. ICSs/ICBs are increasingly looking to USbased companies to provide such technological solutions. Whether such solutions fit into existing NHS systems or support the UK Department for Science, Innovation and Technology's remit to promote digital innovation as an economic driver is debatable. --- Creating digital living labs It is generally agreed that, to address the gap between the development of DHTs and their implementation, it is helpful to support partnerships between industry, academia and end users . There is international interest in potential of "living labs" to this end. Defined as open innovation systems where end users and other stakeholders are involved in the exploration, co-creation and evaluation of solutions in realistic circumstances , living labs offer test-beds for co-creation , exploration , experimentation and evaluation . Living labs could address several barriers identified above, opportunities for co-design ensuring better organisational fit and potentially reducing cultural resistance, while embedding evaluation supports improvements in evidence-based digital health. Indeed, as there are international networks of living labs, such as the Australian Living Lab Innovation Network and the European Network of Living Labs, there is the potential to carry out international evaluations of DHTs, opening up the market for innovators as well as reducing replication of effort. To date, however, crosscultural differences in research design remain a challenge , raising questions about methodological consistency. There are some digital health living labs in the UK, some are in situ facilities within homes and health services. More commonly however, they are off-site and established by academic institutions or community innovation hubs. These are randomly distributed and their creation has tended to be dependent on the efforts of local champions. Mobile digital health living-labs are a nascent concept. Much like mobile medical vans, these facilities are intended to overcome barriers to digital innovation and adoption by meeting people where they are-a particular model that would benefit coastal communities for all the geographical and sociocultural factors previously discussed. Living labs are a key lever for innovation and adoption in a digital health ecosystem, and there is a strong case for UK Research and Innovation bodies, NHS England and individual ICBs to explore opportunities to rolling out labs more systematically. To this end, it would be helpful to think outside the box with respect to key system partners. For example, social housing organisations provide a unique opportunity to develop, test and evaluate DHTs within the home and village hubs to provide dispersed coastal and rural communities with an opportunity to explore technology. --- Understanding evidence The limitations of the current evidence base in digital health are one factor behind their relatively slow adoption in the UK. Small and Medium sized Enterprises are particularly challenged with respect to meeting the evidential requirements of commissioners . Commissioners themselves may be overly reliant on randomised control trial evidence, although RCTs are not often appropriate for rapidly developing DHTs. Thus, there is a need to provide generic training for SMEs on how to develop their value proposition and produce evidence of how their product fits into the system and whether it delivers an effective and cost-effective solution as well as key business and accreditation skills . At the same time, clinicians require reassurance about clinical safety and accountability, while managers could benefit from training in the more agile assessment of DHTs. There are existing providers of such training in the UK. However, there are some concerns that their costs are too high for smaller SMEs. Again, local health and care systems could benefit from tapping into international networks such as the European Connected Health Alliance Group , many members of which are training providers. Importantly, part of the NHS net zero carbon agenda, digital transformation with the UK is inextricably linked with all construction and supply chain partners who will need to demonstrate their carbon impact and strategies to proactively reduce emissions by 2030, reaching net zero by 2045 . Greater transparency in the environmental impact of the health services ecosystem through disaggregated indices is therefore essential to facilitate accuracy in accurate carbon accounting and adequately demonstrate emissions reductions and return on investment. --- Digital skills training If health and care practice is to be increasingly supported by new technologies, much more needs to be done to upskill the current workforce and prepare future professionals. Health Education England provides several NHS Digital Academy learning programmes, including a Chief Nursing Information Officer masterclass series and there has been a drive to appoint a chief nursing information officer in every organisation to help support the digital agenda. With respect to health informatics, nursing appears to be further ahead than other professions , perhaps because nurses appear to be taking on the lion's share of responsibility for entering and monitoring digital data . However, digital innovation influences all areas of medicine, nursing and allied health professions and the strong focus to date on health informatics is not preparing professionals for working in a digitised health and care system . Again, as much training is experiential, staff working in digitally immature ecosystems are less likely to become confident and skilled than those located in digital exemplars. Within undergraduate courses, there is significant variation between UK universities in the extent to which digital training is embedded into medical, nursing and allied health profession education. There is currently no central guidance on how to embed digital competence into the medical curriculum, let alone how to train doctors to manage the rapid pace of technological innovation. Published in 2019, the Topol Review presented a compelling vision for preparing the healthcare workforce to deliver the digital future. Going beyond health informatics to consider the role of genomics, artificial intelligence, digital medicine and robotics, the report's recommendations are currently being explored in curriculum development plans and are likely to provide a much-needed driver for the improvement of both undergraduate and continuing professional education. Importantly, given the sustained proliferation of health technology, multiple methods must be offered to professionals to adequately support skills development. Embedding opportunities for digital awareness and skills development within traditional training programmes as a cross cutting theme as opposed to a bolt on module is needed to adequately support the digital readiness of professionals whereby technologies are seen as an enabler of routine care. Real-world case studies and opportunities for practical exploration of technology such as through virtual and physical technology libraries will enable continuous exploration and learning while more bespoke postgraduate programmes are needed to support specialisation within the digital roles such as digital hospital in the home programs. Given the rapid evolution of technology and the considerable variation in personal skills, competencies, career trajectories and variation in what technology is available in any given setting training is more likely to occur outside of traditional programmes such as through short courses. The development of micro competences and the scaffolding of learning must ideally be CPD accredited wherever relevant to demonstrate professional growth and pave the way for new career opportunities. Irrespective of delivery type, it is widely acknowledged that digital technologies must foremost be designed to be intuitive in their use thus eliminating skills-related barriers to their adoption and optimisation. --- Conclusion In this paper, we have noted the challenges facing coastal health and care systems in England, which are characterised by a "coastal excess" in the prevalence of disease and risk factors and evidence of a service deficit. The latter may stem from a system of resource allocation which may disadvantage areas serving older demographic populations and that does not adjust for the difficulties systems serving peripheral and dispersed populations face in achieving economies of scale. We have made the case that digital transformation offers the potential to shift the balance of care from reactive, expensive hospital management to prevention, early diagnosis and management within the home. However, coastal areas are also more likely to be digitally immature, raising concerns that variations in digital maturity may be exacerbating coastal health inequalities. Some of the barriers to digital transformation require a response from central bodies to e.g., level up digital maturity, a role for NHS England; streamline routes to standards compliance, regulatory approval ; and provide clear guidance as to how undergraduate and continuing education curricula can better prepare the workforce for maximising the potential of DHTs, a role that is being assumed by Health Education England but that also requires involvement from the professional bodies. In other cases, health service managers within ICBs, ICSs and local trusts can play an important role in accelerating digital transformation. We have proposed that priority areas should include investment into orchestration layers in particular to allow remotely monitored data to be integrated into single platforms and analysed so as to allow early diagnosis and intervention to take place at a large scale. Rolling out a more ambitious programme of digital living labs is also within the gift of local organisations. Focusing testbed opportunities around people's homes and local communities would again facilitate a shift in emphasis from the hospital to the role that should be played by wider partners in an integrated care system. In the scope of this paper, we have only been able to touch upon the enormous potential of DHTs to promote better population health, reduce health inequalities, improve the experience and outcomes of care, reduce demands on the health and care workforce, and ultimately reduce costs. Yet, the NHS, particularly in coastal and rural areas, remains years behind countries such as the USA in its deployment of DHTs, particularly those that allow the remote monitoring of patients. We have outlined some of the complex reasons why the NHS has struggled to embed digital health innovations and offered some solutions. What is essentially required, however, is the vision and the willingness to embrace a digital future. We hope that this paper can make a contribution towards winning the hearts and minds of the policy leaders and health service managers who can effect such change. --- --- --- Publisher's note All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Healthcare systems worldwide are striving for the "quadruple aim" of better population health and well-being, improved experience of care, healthcare team well-being (including that of carers) and lower system costs. By shifting the balance of care from reactive to preventive by facilitating the integration of data between patients and clinicians to support prevention, early diagnosis and care at home, many technological solutions exist to support this ambition. Yet few have been mainstreamed in the NHS. This is particularly the case in English coastal areas which, despite having a substantially higher burden of physical and mental health conditions and poorer health outcomes, also experience inequalities with respect to digital maturity. In this paper, we suggest ways in which digital health technologies (DHTs) can support a greater shift towards prevention; discuss barriers to digital transformation in coastal communities; and highlight ways in which central, regional and local bodes can enable transformation. Given a real risk that variations in digital maturity may be exacerbating coastal health inequalities, we call on health and care policy leaders and service managers to understands the potential benefits of a digital future and the risks of failing to address the digital divide.digital health technologies (DHTs), variations in digital maturity, NHS (National health service), coastal health inequalities, the quadruple aim of health systems, shifting the balance from cure to prevention

Introduction Social Enterprise or SE is an entity that generates revenue from the production or distribution of goods or services with clear business goals to solve problems and develop communities, mainly social or environmental. It is not the only goal of generating maximum profit for shareholders and owners . The operation of social enterprises focuses on social goals. The Social Enterprise Fund, Inc set three goals for potential performance, also known as the "triple bottom line," consisting of 1) financial goals, 2) social goals, and 3) environmental goals. The goal of solving social and environmental problems while generating profits so that businesses can operate sustainably and concretely change social problems. The operation of social enterprises is similar to the Company Social Responsibility , but social enterprises are different from CSR, where social enterprises have main social or environmental goals. At the same time, Company Social Responsibility is usually a social activity or project of a for-profit business. Consequently, we can sum up that Social Enterprises are businesses that create a positive impact by reinvesting profits into their social or environmental aims. They trade for the everyday suitable addressing social needs, strengthening communities, and helping to improve lives. They may also achieve their aims through training and employment opportunities . Figure 1 -Social enterprises: A hybrid spectrum Nowadays, more people are paying attention to society's problems and the environment. Social enterprises have come into play and are worth watching, but more importantly, social enterprises benefit people more broadly, including returning income to community society. With the economic conditions' recession caused by COVID-19, running a business to survive and be sustainable is challenging. The concept of social enterprise in Thailand has been discussed and valued more in the last ten years, partly due to the government's policy of supporting and encouraging social enterprises. Many countries recognize the importance of social enterprises, believing that social enterprises will be the sector driving equality . For Thailand, the nature of social enterprises may be different from the emergence of social enterprises abroad. Most government policies, such as co-operatives and community enterprises, developed from the sector's CSR concept. The implementation of the past period, amid the problems of Thai society, has been complicated, and inefficient solutions have created the need for new, more effective solutions. However, social enterprises in Thailand are often young. Over half of the social enterprises were established from 2008 to 2017. The average start-up rate per year in 2018-2020 was49% higher than the start-up rate between 2008 and 2017 . The Thai government has had a policy of promoting social enterprises to improve citizens' quality of life and enable the private sector to work with the government to help communities and society . If we look at why some social enterprises go out of business, the study found that the top three barriers for social enterprises are relevant in Thailand -cash flow problems , access to both loans and investments , and lack of understanding from the public and consumers about social enterprises . --- Research Questions 1. What factors of workers' contributions to social enterprises significantly affectthe contributions of social enterprises? 2. What is the suitable model for managing workers' contributions to social enterprises? --- Literature Review --- Social enterprise in Thailand According to the Regulations of the Office of the Prime Minister on the Promotion of National Social Enterprises BE 2554 , "social enterprise" means private individuals, groups of individuals, or communities that operate with clear goals to solve problems and develop communities. Their production process or management is not aimed at generating maximum profits for shareholders or owners . Social enterprise has a long history in Thailand. Co-operatives that most fit the social enterprise criteria have operated in Thailand for over a century. Early entrepreneur-led social enterprises in Thailand date back to the 1970s. The most recent social enterprise wave was established from 2010 onwards when the government introduced a social enterprise promotion policy . Although Thailand has a legal system of social enterprise registration, such registration does not provide juristic person status. For this reason, both enterprises registered as social enterprises and unregistered have a wide range of legal and organizational forms. Most of them are registered as juristic persons in the form of limited companies , with the rest registered in the form of co-operatives and foundations or associations , partnerships , and community enterprises . --- The organizational context The literature defines the organizational context as "a set of circumstances, or facts related to an event" . The context refers to the organization's characteristics and individual and his role inside the organization. In a study regarding the influence of the context on research in the organizational behavior field, the context was defined by Gary Johns as "situational opportunities and constraints that affect the occurrence and meaning of organizational behavior as well as functional relationships between variables". According to the author, the organizational context could be integrated into two categories: considering the context in extension and focusing on the contextual variables shaping the behavior and attitudes . In addition, the organizational context is also directly influenced by organizational culture, climate, goal setting, equity, teamwork, and affiliation. These variables are established in the organizational settings and are less susceptible to change than motivators . --- Intrinsic motivationfactors Intrinsic motivation refers to the condition of a person who wants to learn or seek something on his own without being involved . For example, an employee works hard in his sense of ambition. To sum up, intrinsic motivation factors are generated internally as brain-directed or cognitive factors. Therefore, they are not tangible yet have a practical impact on motivation. They include feelings of responsibility, achievement, and accomplishment. Intrinsic motivations lead to performing meaningful work . --- Extrinsic motivation factors External motivation refers to the condition of a person who is motivated by the outside. External motivation leads to the goals or expectations of the person. For example, the probate employee aims to be included in the workforce and therefore tries to work hard. There are also other behavioral stimulations such as compliments, titillation, contests, competitions, and even tests that can be classified as a tool that can cause any behavior . Consequently, extrinsic motivation factors are tangible and visible to others, and they are being distributed by an external party, such as pay, benefits, promotions, and action to save the culprit from punishment. Such factors are helpful in drawing people in, recruiting, and keeping them within the organization. Extrinsic factors are contingent and may change with the level of performance. They are often used to inspire employees to achieve higher levels of goals. --- Worker contributions In this research, worker contributions refer to any type of commitment that workers make to their advantage programs. They may not accompany matching commitments from managers. Much of the time, representative commitments emerge from laborers' checks before charges, permitting them to lessen charge risk for the time being and put resources into an advantage that will payout over the long haul. Representative commitments are willful yet fundamental, assuming workers wish to partake in manager-supported advantages programs. --- Research Method This research focused on analyzing the business operating of the organization in Thailand. G*Power 3 refers to the statistical power analysis program for the social sciences, selecting the proper sample size based on statistical analysis. Sample size calculated from G power 3.1.9, Linear Multiple regression, the sufficient sample size must be greater than 312. The questionnaire was framed on a five-point Likert scale ranging from rating 1 to rating 5 , and a ten-point rating scale, rating 1 to rating 10 was used to collect data. Data were analyzed using descriptive, Linear Multiple regression, and Pearson correlation. --- Research Results Most of the respondent is female, aged between 41 and 45, married, have no children, have College / University background education, have income between 10,001-20,000 Baht, experienced 1-2 times job changes, and obtained 1 -3 years the current job experience . Regarding intrinsic factors, social service and pride in working or providing services to others are the most significant factors . According to extrinsic factors, compensation is significant but, on average, less significant than the intrinsic factors .Considering the social enterprise model, employees may tend to be less focused on compensation than intrinsic factors. --- Discussion and recommendations Previous studies have shown that workers understand their intention to work from a number of environmental factors to indicate their commitment to the organization . Workers create awareness about their dedication to their work by adjusting their attitudes and understanding of the organization in various areas. Some researchers suggest that both intrinsic and extrinsic motivation conceptual factors must be added to enhance the positive work experience. It also increases the commitment of the organization and reduces resignations and workers' turnover rates, with such experiences being the result of effective personnel management policies and practices . The results present the factors of workers' contributions to social enterprises that significantly affect the contributions of social enterprises. The results of this study reveal that the workers' contributions are significantly correlated with some intrinsic and extrinsic motivation consistent with previous research. It still has other intrinsic and extrinsic factors that may influence the workers' commitment to any employees. Individuals may value their relationships with colleagues and their sense of security at work over other intrinsic factors, but surprisingly, according to the results of the survey in this research, workers who worked in the social enterprise have less emphasis on social service factors than another intrinsic factor as moral values and working activities. This may be because most social enterprise recruitment is done in a hurry as the need for the manpower. In addition, social enterprise is not the first choice for the typical job seeker. In the same way, the extrinsic factors that workers who had worked in social enterprise highlighted were the matter of achievement of outcome more than any other external factor. Because working with a focus on results clearly reflects success and can make those workers proud of themselves and encourages them to see even better results . On the other hand, remuneration and compensation, which many corporate executives think is probably the most important factor for workers, is the external factor that has the least influence on workers working in social enterprises instead. It may come from reasons of success that affect society and the collective rather than the compensation that affects only themselves. The study also examined the influence of organizational context in the context of strategic objectives, working procedures, and corporate culture. The organizations and executives have to prepare the support for the worker's performance in anticipation of better results because a good organizational context can reflect the worker's good performance in the organization . Besides, all three impacts of social enterprise have an influence on workers' contributions because mostly Social enterprise in Thailand is necessary to operate in accordance with the sustainable development goals that the Social Enterprise Management Regulations in Thailand have already established making all those impacts determine the guidelines for the operation of social enterprises. In conclusion, this study has theoretical and practical implications. Most of the workers of social enterprises is committed to working hard to achieve positive results both for society and for the organization. Nevertheless, only for those social enterprises must there be clear goals in terms of living and social development, profit-making, and environmental management . Moreover, social enterprises also need to be a balance between intrinsic motivation and extrinsic motivation, especially in terms of improving the relationships of all workers to work in harmony, create results and develop collaboration together. Social enterprises themselves have to adjust the organizations' contexts to be balanced and consistent with the outcomes. In addition, the welfare of those workers must be improved to make them feel stable and secure at work so that they will also contribute themselves to their work to make the best impact on social enterprises.

This research examines the trends and elements of workers' contributions to social enterprises. The research applies the quantitative research method, the sample size calculated from G power 3.1.9, Pearson's correlation the sufficient sample size must be greater than 312. The reviewed questionnaire was used to collect data and analyzed using the descriptive and Pearson's correlation Research result: most of the respondent is female, aged between 41-45, married, has no children, with college/university background education, income between 10,001-20,000 Baht, experienced job changes 1-2 times, and obtained 1 -3 years the current job experience.Factors are considered the most significant factors. The respondent's perception of Extrinsic Factors and Organizational Context is likely the same. Social enterprise worker contributions seem to be a less important issue than others.

Introduction Coronavirus Disease 2019 has caused a healthcare crisis all over the world. Social distancing, travel restrictions and self-isolation strategies have led to heavy socio-economic burdens. On a national basis, schools and factories closed, while demand for medical supplies and food increased tremendously, prompting healthcare providers to approach the pandemic with revolutionary solutions [1]. Since the declaration of the pandemic, countries are racing to slow down the spread of the virus by employing test regulations, contact tracing, travel bans, and lockdowns [2]. Today, vaccination tops the list of strategies to counter COVID-19 spread, with many approved marketed vaccines available for use [3]. Vaccine hesitancy is defined as delay in accepting or refusal of getting vaccination despite its availability [4]. Although the relative safety of marketed vaccines has been established [5][6][7], there are many other factors that affect community response towards vaccination, these include personal background , social and political projections along with conspiracy theories, and safety concerns due to wide-spread myths and false messages spread on social media [8][9][10]. In Jordan, presently one of the countries most affected by the pandemic [11], the government has established an online portal to facilitate registration for vaccination. Despite many governmental campaigns on vaccination importance, the turnout on the portal was weak, uncovering alarming levels of hesitancy. Previous studies showed that vaccine hesitancy is a rising global phenomenon with multifactorial background [12,13]. Most common related factors reported were safety issues, religious beliefs and lack of scientific knowledge [14]. In this study, we aim to explore the factors affecting attitudes towards vaccines at a tertiary hospital in Jordan, a middle eastern country. The primary objective is to study the effect of educational background and work field on responses towards vaccination. The secondary objectives are studying the most encouraging and discouraging factors to take the vaccine and how significant is the role of social media in shaping the attitudes of people in the Jordanian community. --- Materials and Methods --- Study Design This correlational cross-sectional study was conducted between 22 January 2021 and 28 February 2021 using an online structured self-administered questionnaire. We included medical field workers from Jordan University Hospital , which is a tertiary teaching hospital in Amman, the capital of the Hashemite Kingdom of Jordan. The included medical personnel were those in direct contact with patients at inpatient or outpatient departments, including those who has history of COVID-19 and those who does not. Out of 2823 current employees at JUH, 2055 were considered in direct contact with patients, including physicians, nursing, pharmacists, technicians, and transporters [15]. Considering an acceptance rate of 28.4% for vaccination in Jordan [13], 5% precision, 95% confidence interval [23.4-33.4], and a standard error of 0.0255, a sample size of 282 was required from current JUH employees [16]. We reached 350 medical personnel, as well as 500 nonmedical adult subjects as a comparator group. The response rate was 82% among medical field workers, and 71.8% among comparator group. The included participants in both groups were subjects who have not taken the vaccine yet, and do not have any family member who has been administered the vaccine yet. It is noteworthy to mention that Jordanian COVID-19 vaccination program was arranged by registering through an online platform [17]. Since the mid of January 2021, vaccination was scheduled using this automated registry, taking into consideration the age of the citizens, their medical illnesses, and their field of work [18]. --- Questionnaire The questionnaire was designed using Google Forms, which is an online survey creator tool developed by Google. The questionnaire was divided into four sections. The first section included an introduction page, in which we introduced the topic and explained the study objectives and the participants consented to fill out the questionnaire by proceeding to the next section. The second section was the demographic section, in which we inquired about general demographic data and previous history of COVID-19 among the participants and their families. The third section questioned their knowledge about COVID-19 vaccines, their concerns regarding these vaccines, and the rating of their trust in those vaccines' efficacy and safety on a scale of 10, where 10 signifies full trust, while 1 means minimal trust. In addition, this section also investigated their willingness to take the vaccine and to encourage others to take the vaccine. Moreover, it investigated the rate of influenza vaccine administration in the last year, as well as whether the participants consider themselves antivaxxers, which is a term that reflects a movement calling against vaccination in general [19]. In the fourth section, we investigated the most common rumors received via social media regarding COVID-19 vaccines, along with the most common circulated information regarding potential side effects. After that, we inquired about the most encouraging factors for vaccination, then the extent of influence of social media, medical personnel, teachers, professors and employers on the decision to get a vaccine was rated on a scale of 10, where 10 implies very high influence, while 1 means minimal influence. For the questionnaire validation, it was reviewed by 10 consultant physicians, after which it was modified based on their review. The calculated Cronbach's alpha value was 0.76, indicating adequate internal consistency [20]. --- Ethical Approval The study was approved by the Institutional Review Board of JUH . An informed consent was obtained at the end of the introduction page of the online questionnaire for the approval of participants to proceed to the questionnaire. We did not include any personal information, and the anonymously collected data was used solely for statistical analysis. --- Statistical Analysis We used the Statistical Package for the Social Sciences version 25.0 for statistical analysis. Descriptive statistics were applied, and data was presented as mean ± standard deviation for numeric variables, and number for categorical variables. We used Mann-Whitney U test for the comparison between medical field workers and non-medical individuals in age, income, the aforementioned scales that measured the trust in the safety and efficacy of these vaccines, and scales measuring the extent of influence of social media, medical personnel, teachers, professors and employers on the decision to get a vaccine. Chi-squared test was used to compare categorical variables between the aforementioned two groups. Univariable binary logistic regression analysis was applied to predict factors associated with willingness to administer COVID-19 vaccine, and the odds ratio as well as the 95% confidence interval of OR were demonstrated. Subsequently, the variables that were significant in the univariable regression model were then included in a multivariable binary logistic regression analysis. A two-sided p-value <0.05 was used as the significance threshold in all aforementioned statistical tests. --- Results Overall, 646 individuals were enrolled in this study, of which 169 were males, and 477 were females. The mean age of the studied population was 28.2 ± 10.8 years. Of the studied subjects, 133 have previous history of COVID-19. The demographic data and previous history of COVID-19 are illustrated in Table 1. The comparison between medical and non-medical work fields showed no difference between the two groups in being concerned regarding getting COVID-19 in the future . 137 of medical field workers have previously read a scientific article about COVID-19 vaccines, while only 112 of non-medical personnel have read such a scientific article . The belief that COVID-19 vaccines are safe was higher among medical field workers compared to non-medical persons . The belief that those vaccines are able to protect against COVID-19 followed the same trend , with 149 of medical personnel believing in the vaccine protection against this disease, compared to only 123 among non-medical persons. Upon inquiring about the most trusted vaccine according to the latest available information, 153 of non-medical persons did not have enough information to choose between the available vaccines . Remarkably, medical staff were more concerned regarding the availability of the vaccines . Overall, 226 of the studied population were planning to take the vaccine once available. Of those who has previous history of COVID-19, 38 are planning to take the vaccine once available, while 95 were not. In comparison, 188 of those who did not have COVID-19 before were willing to take the vaccines . Reading a scientific article about the available vaccines showed a significant increase in the rate of willingness to take the vaccines , with 130 willingness rate among those who read a scientific article, compared to only 96 willingness to vaccinate among those who did not. Even among medical personnel alone, 85 of those who read a scientific article were willing to be vaccinated, compared to 47 willingness rate among those who did not read such an article . Of the 287 medical field workers, 131 were willing to take the vaccine, compared to 94 of non-medical workers . Moreover, 131 of the medical personnel and 94 of non-medical workers were willing to encourage others to take the vaccine . Interestingly, 47 of non-medical staff considered themselves antivaxxers , and only 37 of them took influenza vaccine . The circulated information and rumors about COVID-19 vaccines on social media platforms were illustrated in Table 3. The rumor that was believed the most was that these vaccines are not safe, with a prevalence of 283 . This rumor was mostly believed by non-medical persons . On the other hand, 163 of medical persons did not believe in any of these circulated rumors, compared to 126 of non-medical persons . Upon investigating the most encouraging factors for vaccination, we found that short scientific videos were the most attractive method, being chosen by 390 persons. Social media awareness campaigns , doctors' advices , television and radio-based advertisements , national medical studies to prove vaccines efficacy , and short scientific videos were all chosen significantly more by medical field workers as an encouraging factor for vaccination, as illustrated in Table 4. The rating of trust the safety of these vaccines was higher among medical staff compared to non-medical field workers . In addition, the rating of the trust in efficacy of these vaccines and the rating of trust in efficacy against mutated SARS-CoV-2 were also higher in medical group, compared with 4.1 ± 2.8 and 3.5 ± 2.5 respectively among non-medical personnel. The rating of the effect of teachers, professors and employers' advices on the decision to get a vaccine was also higher among medical staff, with a rating of 6.3 ± 2.3, compared to 5.9 ± 2.7 among non-medical personnel . We applied multivariable regression analysis , in which we found that following factors are independently associated with the willingness to take COVID-19 vaccine once available: being concerned regarding getting COVID-19 in the future , the rating trust the safety of these vaccines , the rating of the trust in efficacy of these vaccines , the rating of the effect of medical personnel advice on the decision to get a vaccine , and the rating of the extent of social media effect on the decision to get a vaccine . OR: odds ratio; 95% CI: 95% confidence interval. **: We used the marital status "single" as reference standard for all comparisons. --- Discussion Vaccines, when available, will likely be our best tool to counter the COVID-19 pandemic. Herd-immunity is one of the earliest ideas to come across at the beginning of the pandemic, but many studies showed that immunity acquisition from previous infection with COVID-19 is transient, and relatively inferior to immunity acquired through vaccination [21][22][23]. In our study the results showed no significant difference in willingness to take the vaccine in patients who were previously infected with COVID-19 when compared to those who were not. Previous studies assessing attitudes towards vaccines revealed huge regional variability in perceiving safety and effectiveness of marketed vaccines [24]. Interestingly, higher-income countries were the least certain regarding vaccine safety. Significant variability was noted on agreeing to vaccine safety between Northern America , western Europe and eastern Europe . However, majority of people in lower-income countries agreed on vaccination safety, with the highest proportions in South Asia and in Eastern Africa [25]. The assessment of these patterns can be invaluable in countering vaccine hesitancy. Upon comparing people in medical and non-medical working fields, no significant difference was found regarding concern of getting COVID-19. Ideally, all healthcare workers should be worried about getting the infection as COVID-19 has been associated with increased mortality in these people [26]. Although there are other studies that showed no difference in terms of worrying [27], a previous study held in Jordan on frontline doctors dealing with COVID-19 related their concerns to lack of proper protective equipment [28]. Information resources about COVID-19 vaccines are many, including scientific journals, internet pages , friends, traditional media , lessons on COVID-19 , medical staff, and family members [29]. Scientific articles are labelled as the best source of information for medical and non-medical workers. In this study, there was a significant difference in reaching scientific articles between medical and nonmedical personnel . Since COVID-19 is a relatively new pandemic and evidences with high level of confidence are being published only recently, one can expect the reluctance in reaching scientific articles as definite source of truth at the early stages of the pandemic. Nevertheless, the low percentage of reaching scientific articles among healthcare workers is worrying, as depression, anxiety, and distress are more likely to occur among healthcare workers if they were not exposed to the rightful information on COVID-19 [30]. Regarding safety of the vaccine, the belief that the vaccine is safe was higher among medical personnel . This can be partially attributed to higher reach of scientific articles among medical personnel compared to non-medical ones. As a percentage, 56.8% is a relatively low percentage compared to other studies [31], but it points to the huge role other sources of information may play in the Jordanian community. 45.6% of medical field workers are planning to take the vaccine when available, compared to 26.2% of nonmedical workers. The reported percentages among medical personnel worldwide ranged from 27.7% to 63% [32,33]. For non-medical group, only 50% of people in USA reported willingness to take the vaccine once available, while in France the percentage reached 74% [34]. Both medical and non-medical people showed readiness to encourage other people to take the vaccine with similar percentages. On the other hand, 13.1% considered themselves antivaxxers. Objections to vaccination can turn to strong attitudes if they are based on a religious background, associated with infringement of personal liberty, or being backed by suspicion of scientific articles [35]. The biggest challenge facing the employment of a vaccination strategy in any community is rumors [36]. In this study, the rumor that was believed the most was that the vaccines are not safe. There are many elements standing behind this rumor, most notably is the fake news spread over social media, described by the WHO as 'infodemic'-indicating that the fake news spreads faster than the virus itself! [37]. Rumors on social media intermingles disinformation and misinformation about the COVID-19 vaccine, and even though many social media companies are employing strategies to fight these rumors, the spread is inevitable [38,39]. 46% of United Kingdom population and 48% in the united states reported exposure to rumors about COVID-19 and vaccines [40,41]. One of the most reported rumors among antivaxxers is that the virus is either manmade or produced by powerful organizations, and many people reported to believe in that in the United Kingdom and United States [40,42,43]. Regarding vaccination campaigns, the study found that the most effective method to encourage people to take the vaccine is short scientific videos spread on social media. Although one can argue with the definition of 'scientific videos', many experts around the world provide scientific knowledge through these platforms. The problem from vaccination perspective is that many of the social media published materials are claimed to be scientific, and till date, no efficient governmental or service provider strategies were able to validate this behavior [44,45]. Additionally, among the factors independently associated with the willingness to take COVID-19 vaccine are the rating of medical personnel advice on the decision to get a vaccine and the rating of the extent of social media effect on the decision to get a vaccine . On a study targeting US adults' attitudes towards COVID-19 vaccine, factors that were independently associated with hesitancy included younger age, lower educational level, and not having received the influenza vaccine in the prior year [46]. Another study implicated that perceived severity and fear of COVID-19 was associated with vaccine acceptance, while negative attitude towards general vaccination was associated with low vaccine acceptance [47]. During the data collection period of this study, the number of daily new cases in Jordan rose from 730 cases on 22 January 2021 to 4594 cases on 28 February 2021, after which the number of new cases continued to rise till it reached a peak of 9535 new cases on March 17th [11]. These dynamic changes in the numbers of new cases and their burden on the healthcare system and the society may influence the concern of the included persons of contracting COVID-19 and may eventually impact the vaccination acceptance rate. Based on our results, we recommend that legal action should be taken by the government and the public security directorate in order to stop the spread of social media rumors and fake news about COVID-19 vaccines. Moreover, we encourage adequate funding for national medical studies to prove the safety and efficacy of these vaccines. Social media awareness campaigns are also encouraged, as social media was shown to have significant influence on the willingness to vaccinate against COVID-19. The main limitation of this study is that it did not investigate the awareness among individuals who does not use social media. Even though that recent estimates showed that most Jordanian adults own their own mobile phones [48], Those who do not use social media and depend on other sources for information should be considered in future studies. Moreover, the sample included in this study is not representative of the whole Jordanian population, for which future studies in which the distribution of occupation, age and, gender of the included sample should be similar to that of the country's population. Furthermore, this study was conducted on individuals who have not taken the vaccine yet, and does not have any family member who has been administered the vaccine yet. Although this was considered a strength of this study by studying the influence of social media alone, the increasing numbers of vaccinated citizens in Jordan necessitates the need for investigating the influence of the feedback of family members, peers, and colleagues on the willingness to vaccinate for COVID-19. --- Conclusions In conclusion, COVID-19 vaccines are considered the most promising intervention to control the spread of this pandemic. Vaccine hesitancy is a global challenge, and one of the main factors associated with it is the spread of misinformation on social media. In this study, we demonstrated that 35% of the studied population were planning to take the vaccine once available. The trust of the population in the safety and efficacy of these vaccines were positively associated with the willingness to vaccinate. Moreover, medical personnel and social media play a crucial role in increasing the inclination of the society towards vaccination. This imposes a huge responsibility on those sectors in providing the community with updated and evidence-based information about COVID-19 vaccines from trusted sources, in order to correct the previously spread misinformation and raise the population awareness regarding the importance of these vaccines as an efficient medical countermeasure. --- Data Availability Statement: The data presented in this study are available on request from the corresponding author. --- Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. ---

Vaccination is the most promising strategy to counter the spread of Coronavirus Disease 2019 . Vaccine hesitancy is a serious global phenomenon, and therefore the aim of this cross-sectional study was to explore the effect of educational background, work field, and social media on attitudes towards vaccination in Jordan. We compared between medical personnel who were in direct contact with patients and non-medical individuals at Jordan University Hospital in terms of demographics, knowledge about COVID-19 vaccines, rumors received via social media, their trust in these vaccines, and the encouraging factors for vaccination. 646 individuals were enrolled in this study, of which 287 (44.4%) were from medical field, and 359 (55.6%) from non-medical field. 226 (35%) were planning to take the vaccine once available, with a positive response from 131 (45.6%) medical field workers, compared to 94 (26.2%) non-medical individuals (p < 0.001). The social media rumor that was believed the most was the unsafety of these vaccines (n = 283; 43.8%). Only 163 (56.8%) of medical persons did not believe any of the circulated rumors, compared to 126 (35.1%) of nonmedical persons (p < 0.001). The effect of medical personnel advice (OR = 0.83; 95% CI = 0.70 to 0.98; p = 0.026) and social media (OR = 1.21; 95% CI = 1.04 to 1.41; p = 0.012) were significantly associated with the willingness to take COVID-19 vaccine once available. In conclusion, medical personnel and social media play a crucial role in increasing the society's inclination towards vaccination by providing the community with updated evidence-based information about COVID-19 vaccines as an efficient medical countermeasure and by correcting the previously spread misinformation.

Introduction Men who have sex with men are among the fastest growing risk group in China's current HIV epidemic [1,2]. HIV prevalence among MSM had increased from 0.9% in 2003 to 6.3% in 2011 [3]. In 2009, only 8.6% of existing HIV infections was transmitted through male to male sex, but in 2013, this proportion dramatically increased to 21.4% [4]. The Chinese government has taken great steps to address the growing HIV epidemic in this population, providing various HIV/AIDS prevention services [5][6][7]. These services include provision of free condoms and lubricant, peer education, free voluntary HIV counseling and testing , as well as diagnosis and treatment of sexually transmitted infections [8]. These measures are effective in some extent in decreasing risk behaviors and HIV transmission [5,9,10]. However, despite high levels of self reported risk behaviors [2,11], the HIV testing rates among MSM remains low, with only about half reporting having ever been tested [1,2,11,12]. It seems to get more complicated in this understudied population due to their unidentified and yet unaddressed risky sex behaviours [13]. Despite its low visibility in contemporary society, homosexuality is nothing new in China. References of same sex practices can be traced back to the Shang Dynasty and throughout Chinese literature and history [14]. Homosexuality was not only tolerated but intrinsically interwoven with political and cultural life in feudal China [15]. After the formation of the People's Republic of China in 1949, homosexuality came to be regarded as a corrupt lifestyle, believed to originate from perceived evils of capitalism. Homosexuality was not decriminalized until 1997, and later in 2001 this term was removed from the list of psychiatric disorders by the Chinese Psychiatric Association [16][17][18]. Although same sex practices are not illegal in contemporary China, gay men still experience significant negative social and cultural ramifications [19][20][21][22]. Since sexual behavior is a product of one's social and cultural environment, any intervention aimed to change behavior must be rooted in the target population's socio-cultural landscape [23]. Literature on homosexual practices has emphasized the importance of understanding the complexities of sexuality, embedded in cultural meanings [24]. Globally, researchers have begun exploring the importance of gay men's cultural identification and its impact on health [25][26][27][28]. In1992, the sociologist Yinhe Li, initiated the first Chinese study exploring life experiences, sexual practices, and social networks of gay men in Beijing [29]. Further scholarship was undertaken from using a socio-cultural perspectives, to look into how gay men's behaviors were influenced by the traditional Chinese culture, the impact of social-norm and stigma had on their social network, and the contradiction they faced when balance between family and self-identity [22,[30][31][32][33][34][35]. There were also studies explored the implication of socio-cultural factors on HIV transmission among gay community in China [1,14,24,33,36,37]. For example, Liu's study [37] explored the association between the response to stigma and HIV prevention in Shanghai, and Fung's study [33] focused on the impact of socio-cultural factors on HIV transmission among rural gay men in China. These pilot studies provided valuable insights in understanding the relationship between socio-cultural factors and HIV transmission among MSM. Nevertheless, China is now under more complicated situation than ever before: on one hand, Confucianism, the traditional Chinese philosophy [33], has strong emphasis on duty, filial piety and moral values that brought homosexual people enormous family and society stresses to get married to continue the family line. On the other hand, with the economic revolution in the late 1970s, impacted by western culture with the strong characteristics of individualism, social atmosphere dramatically changed, with a reflection on people's sexual beliefs and behaviors. It resulted in more people, especially the new generation, starting to look for a balance between collectivist family interests and individualistic desires and conducts. This rapid developments in societal norms with a changing cultural landscape, are giving rise to new challenges on health issues. In this given condition, the aim of this study is to give a holistic social-cultural view to explore influence factors of gay men's HIV-relevant beliefs and behaviors, and the implication for HIV epidemic. --- Methods We performed in-depth interviews with 61 gay men from eight selected sites in China between July 2015 and May 2016. The eight sites were: Nanchang and Xinyu in Jiangxi Province, Zhengzhou in Henan Province, Harbin in Heilongjiang Province, Guangzhou in Guangdong Province, Nanjing and Suzhou in Jiangsu province and Chongqing Municipality. --- Study sites The eight sites were selected for their quickly increasing HIV epidemics among MSM and varied HIV prevalence among MSM in the country-level. Located in southwest China, Chongqing has an increased HIV prevalence among MSM, precipitously from 11.6% in 2009 to 15.4% in 2010 [38]; Henan Province is in the middle China, with a HIV infection rate among MSM increased from 4.69% in 2008 to 8.33% in 2013 [39]; Zhengzhou, as the provincial capital of Henan province, has a HIV incidence rate of 7.4/100 Person Years among MSM [4]. Located in northeast China, Harbin has a moderate HIV epidemic compared to other areas [40], its prevalence among MSM increased from 1.0% in 2006 to 7.5% in 2010 [41]; Both Jiangxi and Jiangsu province are located in southeast China. Jiangxi Province had a HIV prevalence of 8.44% among MSM population, with low rate of consistent condom use rate and HIV testing [42]; for the two study sites of Jiangxi, Nanchang had a dramatically increased HIV prevalence among MSM population, from 5.91 in 2013 to 12.56% in 2014, whereas Xinyu had a relatively low prevalence: from 0 in 2013 to 0.5% in 2014 [43]; HIV incidence in Jiangsu Province increased from 5.10% in 2011 to 6.62% in 2015 [13]. Nanjing, which is the provincial capital of Jiangsu province, has a HIV incidence rate of 5.4/100 Person Years [4]; Suzhou, together with Nanjing, were reported had the top two most HIV positive cases among MSM in Jiangsu province in 2011 [44]. Guangzhou, the capital of Guangdong province located in south China, is remarkable for its rapid economic growth and attracts gay men from all over the country for its relatively tolerant social atmosphere. Its HIV prevalence of MSM has also increased significantly, from 5.0% in 2008 to 11.4% in 2013 [41]; --- Recruitment and participants In-depth semi-structured interviews were conducted to explore themes related to participants' self-identity, perceptions of homosexuality and sexual behaviors, understanding of gay community culture, as well as utilization of local HIV related health services. To be eligible to participate, interviewees had to be male, have had penetrative sex with a man in the preceding six months, and be aged 18 or above. The local centers for disease control and prevention and MSM serving non-governmental organizations approached prospective participants through their networks and referred to us those who volunteered to participate. We sought a diverse sample in terms of age, education level, and marital status as prior research revealed differing perceptions as well as behaviors among gay men of varied demographics [45]. --- Interview procedures Sixty-one interviews lasting approximately one hour each covered a variety of topics related to the objective of this study. Informants were asked to discuss life experiences focusing on: 1) perceptions and attitudes about homosexuality among family, friends, and other social relationships; 2) personal feelings about their sexuality and decisions about disclosure to families and friends; 3) perceptions on casual sex and partner relationship; 4) personal understanding of local gay community and its contemporary culture and 5) personal practices and thoughts on health care services seeking, especially on AIDS related services such as HIV counseling/ testing or health promotion activities. Respondents were encouraged to talk openly about more general topics, including experiences with emotion and societal pressures. The investigator guided specific main topics to be covered and provide suggestions or follow-up inquires according to the reply of the participants. Data collection ended after 61 interviews as the investigators determined that the data saturation had been reached. Each interviewee was given 100 Yuan to reimburse them for time and transportation costs. --- Ethical statement Interested and eligible participants met with a study investigator individually in the privacy of an NGO counseling room, local CDC office or hotel room. After a briefing on study objectives and participants' rights, written informed consent was obtained from all participants who were assured of confidentiality, the use of pseudonyms, and safe storage of data. They were also assured that they could end the interview at any time or refuse to answer any question without consequence. The study protocol and informed consent form received ethics approval from the Institutional Review Board of the National Center for AIDS/STD Control and Prevention, China CDC. --- Data collection and analysis Before commencing the interview, all participants' basic demographic information was collected. The interview schedule was developed from a comprehensive literature review, piloted and adjusted according to feedback from expert panel as well as interviewees. The content of 61 interviews was de-identified before transcribed in Mandarin Chinese. The transcripts were translated into English by a native English speaker, and the co-authors cross-checked the content validity of the translation. Two coders reviewed the transcripts when listening to recordings to check accuracy and anonymity. In line with guidelines by Braun and Clarke [46], coding was conducted by using NVivo 11 software . Both coders have epidemiology and biostatistics experience and interview coding training. The initial coding was conducted by two coders independently, with data broken down into basic elements. Five transcripts were analyzed to establish and ensure the reliability of coding styles and structure. The two coders coded the whole transcripts independently after the coding framework was established. Data points were written into categories which formed the initial themes. In the next step, over-arching themes which provided more intricate understanding of the subject, were proposed and tested. The validity and usefulness of overarching themes and subthemes were regularly evaluated by coders to ensure coherency and consistency. Final themes and their subthemes were discussed and developed through a consensus within the research team. --- Results Sixty-one key informant interviews were conducted with participants varying in demographic characteristics. The majority of respondents were between the age of 19-30, had completed college, and were unmarried . --- Conflicts in self identity and negative coping toward sex Respondents made mention of the constant challenge to uphold expectations from their surrounding community to a degree that they perceive themselves as living a double life, going between their families and the MSM scene, as quoted by one participant: "Whenever I go home, I feel like my wife and I are strangers. Yet we still keep up the charade in front of friends. It's depressing. When I'm with my gay friends [gay sexual partners] we have to hide our activities, it's extremely stressful.". It is not uncommon that gay men in China present themselves as bisexual, or marry women to manage stresses put on them by family and society as a whole:"Self proclaimed bisexual men just want people to acknowledge their 'normalcy', to show they have sex with women, that having sex with men is a mere game to them." "A number of my gay friends are considering marriages to women. This might be one of their biggest issues." The stress of fulfilling the family obligation may vary geographically. Firstly, people in rural areas were more conservative than those in urban. They were more likely to obey the social norm to continue the family line: "It's less stressful for those in urban cities, but in countries the stress [to marry] is high, because parents are so eager to have grandchildren." . Secondly, some areas of China have more strict restrains on people's social behavior, for example, the Chaoshan area located in Guangdong Province. One participant who came from Chaoshan described: "In my hometown , people should get married at a very young age, basically around 22 or 23, and are expected to have grandsons as much as they can! So how could I let my parents know I am a gay? They would definitely be mad and kill me!" A poor sense of identity and, the stress of living a double life led negative attitudes toward sex and relationships. One participant reported that: "I told one of my sex partners about ZT [a local MSM NGO providing health services and support] but he was skeptical, thinking that it was just another place to find sex partners. I've tried explaining that things on the apps aren't a real --- Generational differences among gay men One recurrent theme throughout the 61 interviews concerned differences between age groups. Our sample contained participants ranging from19 to 65. Commonly noted was the way in which the gay community culture varied. Compared with seniors, young gay men in their 20s, exhibited a generally pride in their sexual preference, and did not appear to conform to negative images of homosexuality set forth previously. "I don't think it's wrong for me to love a man, I should not be blamed. At school, I am still a good student, and in my family, I am still my father's good son. . Sexual beliefs and practices had also changed: "I was 19 when I realized I was gay, but was still a virgin at 23. That was when I finally entered the gay community, and learned where the venues were. But now, things are totally different, people changed. Guys born after 1980 are still slightly conservative, but those born after 1990 have extremely liberal, impulsive views on sex". . For the youth, insufficient and poor health communication skills accompanied by a lack of social experience and safe sex knowledge set the stage for naivety, and HIV-risky behavior. "If I see a hot guy, of course my first thought is sex rather than HIV" . Another obvious change among generations can be attributed to the fast development of social media. GPS enabled mobile APPs are now very popular among gay men, which grant users increased flexibility to find nearby partners in real time. Gay men, especially the youth, now could approach potential sex partners much easier than ever before. However, for the senior or illiteracy, they still prefer traditional gay venues such as public restrooms, parks and gay saunas. One participant commented:"when I was young, people usually look for friends at parks or saunas, while, it is so different now. Far less people go to such places. I guess now most of them look for friends just by using APPs, you know, such as BlueD [a Chinese social software designed for gay men] or some other similar ones" The changed pattern for approaching partners impacted HIV intervention and control. APPs providing gay people more privacy but less visibility limited the effect of traditional HIV intervention, such as face-to-face health education and condom-promotion. "Many gay men now hunting by APPs, which makes our routine intervention doesn't work now. The traditional venues, like gay bars, are losing more and more customers because more and more people prefer to use dating APPs, they are much easier! We are losing our intervention subjects now.". Moreover, health-relevant information provided by the social media was neither thought to be enough nor effective:"people who use this APP are directly for dating, for one night stand, so how can you expect them to really care about health information?" --- Cultural barriers for "safe sex": Look into several features of sub-culture of gay community Open minds toward casual sex and partnership. Conservative sexual behavior and single sex partner could potentially reduce the risk of HIV transmission [47]. However, based on many participants' comments, their perspectives on sex relationship were quite tolerant. On one hand, open mind toward casual sex were commonly described among participants. --- Such as the quotations: "Personally, I don't really like 'one night stand' relationship. But I respect people's choices, no judge. I guess it is just a way of life, nothing to do with morality, and the only thing you should keep in mind is safety." . "If I hook up with some one, first of all, I will make it clear that this is just a one-night stand, don't look for further relationship. I don't know if others think the same way, but this is my opinion." On the other hand, a new style of relationship called "open relationship" was observed among some of the young gays. "Some of my friends are in 'open relationships' which means more than two guys are included in one relationship. I think it is OK if they appreciate this way and can live a healthy and happy life, it is not a big deal, noting to do with morality." Region-crossing sexual behaviors. Despite the geographical difference, participants commonly mentioned that gays like travel, to meet new friends and have sex adventures. As one quotation:"One of my friends is called '419 Queen' because he travels a lot, looking for his 'prey' all the time." "As typical of gays, they love to meet new guys. When they visit a new place, the first thing to do is definitely finding out where the local gay venues are, and go there look for some "adventures" . Another region-crossing sexual behavior often occurred in gay saunas, especially in North China. As commented by one manager of a gay sauna: "for some gays who are on their travel for business or some other things, they would prefer go to saunas than hotel because sauna is much cheaper and it is much easier to have sex". . As a venue where casual sex easily takes place, it facilitates HIV transmission among gay population, as commended by one pioneer working in a NGO: "you would be shocked when seeing what happened in the sauna. I still remember that one day when I go to a sauna for condom promotion, I saw several guys standing around one boy, having sex with him in turn without any protection! This is so crazy, so horrible!" Health was considered a low priority. Prioritization of perceived needs in the Chinese gay community is of nominal importance. Numerous respondents reflected that primary concerns as gay men in China have little to do with health issues. "If someone lives under huge stress, stripped of their most basic levels of respect, dignity, and rights, how could he possibly be concerned with his own health? HIV is among the furthest things from his mind." "The people who come to us [MSM serving NGO] do get testing here, but they more often approach us about issues relating widely to navigating the landscape of being gay: marriage issues, how to communicate with their families, self-acceptance, and dating. These are what they actually worry about and care about. They come to us because we help to address these issues." Apart from these, a diagnosis of HIV may further exacerbated gay men's suffering and sense of inequality: "Many people won't get HIV tests, as people will talk. I definitely wouldn't get an HIV test." Poor treatment and ostracizing stems from not only society at large, but also within the community itself: "Oftentimes stigma is far worse within the gay circles than in general society. Gay guys are afraid that if their positive status is revealed, they'll be marked as a pariah in the community." The stigma inside the gay community further obstructed individuals from seeking health care services. --- Discussion This study explored socio-cultural influences on gay men's attitudes, the community sub-culture and their implications for HIV prevention in mainland China. Revealed by our study, participants, especially those from rural or conservative areas faced much more stress on fulfilling the family obligation, struggling to live a double life. Although historically, homosexuality in China has not been subject to the same degree of persecution in other countries, it was still stigmatized in contemporary China due to traditional values emphasizing on procreation and social order [17,18]. Producing offspring and subsequent child-rearing still remain culturally important in Chinese culture, and these conventions often keep Chinese men from disclosing their same sex orientation. These pressures often lead MSM elect to marry women in order to avoid shame for their families [8,24,48]. Married gay men were more reluctant to disclose their sexual orientation to heterosexual partners for fear of divorce and subsequently broken family ties [49]. Chinese gay men experience immense pressure, discrimination, which may contribute to greater vulnerability for health issues including HIV [50][51][52]. Despite similar HIV risk trends among gay men worldwide, sociopolitical circumstances in the conservative Chinese cultures often relegates individuals into a space of secrecy and further risk. Similar with previous study [25], a generational difference was observed in our study. This generational difference was commented caused partially since the inauguration of the Open Door Policy in China in the late 1970s and the economic reforms which commenced in the mid 1980s [15]. In the following decades, with more international communication and culture integration, chase for freedom, individualism and pleasure seeking became a new fashion among youth. China seems in the middle of a great generational transformation: Older people at least pretend to follow the traditional narrative, to have respect for traditional sex culture for their elders, but many younger people, they seems don't care very much about any of that. Meanwhile, online dating has now become more popular among gay community because of its ease of access, anonymity and high acceptability [53][54][55], especially in young groups [56]. Pilot studies suggested that men who had sex with an online partner were at higher risk for contracting HIV and STIs compared to those who did not approaching partners online [53,57,58]. Some previous studies revealed that, the majority of gay men prefer receiving AIDS related intervention via internet rather than via CDC or hospitals due to the anonymous nature of the social media [8,59]. It implicated that promotion of non-face-to-face internet based intervention strategies could be developed to better match the target population's needs. Based on our study finding, "open relationship" was common among gay population. According to the national behavioral surveillance report, the proportion of MSM who had multiple sex partners in the past six months increased from 68.0% in 2008 to 85.4% in 2011 [60]. Some previous studies also revealed the common nonmonogamous relationship in MSM community [3,61,62]. Regardless of this situation, being a long time, "loyalty" was emphasized and underlined in health promotion activities. This strategy did not reach expected effect due to its confliction to the current gay community sex culture. Therefore, for the future intervention, promotion on safe sex without moral judgment on nonmonogamous relationship should be considered. Apart from this, although there were dramatic changes in sexual beliefs and behaviors, communication on the subject is largely nonexistent, posing considerable issue in grappling with the HIV/AIDS epidemic and its prevention [63]. Effective sex education can reduce risky sexual behaviors among adolescents, yet it is still regarded as a sensitive topic and thereby ignored largely in school and at home, resulting in tremendous lack of knowledge among youth [63][64][65]. As a feature of gay's subculture, region-crossing sex activities brought obstructs for HIV prevention in many aspects. Mi's study [66] revealed that around 30% registers on the social media dating APP were migrants with the most preferred inflow areas of metropolis such as Guangzhou and Beijing. It facilitated HIV genetic communication among infected individuals, as previous studies showed that there was highly dynamic of HIV virus infected among MSM in China [67,68]. Gay saunas and bathhouses, which had higher risk for HIV acquisition than from other venues due to its high rates of UAI and group sex [56,[69][70][71], were more popular among floating population. This situation made HIV control more complicated. Venue based intervention strategies, as well as regional integrated intervention services were greatly warranted in the future. Apart from these, another feature of gay sexual subculture is the ignorance of health issues. Societal stigma and severe discrimination towards homosexuality has driven much of the target population to a largely hard to reach status. Due to rapid loss of CD4+T cells leading to fast progression to AIDS [67], concerns on better linkage to care should be underlined. However, engagement of the demographic has been challenging, for health education and intervention, even with the changing tides of acceptance. Aggressively climbing HIV infection rates in the population cannot be ignored, nor can the distinct role that MSM serving NGOs play in public health. These NGOs are well positioned to reach gay people and offer a multitude of support, including that which pertains to general mental health and navigating the landscape of the gay community [23]. China's Twelfth Five-year Action Plan for HIV/AIDS Prevention andControl , mandated that the country "Give full play to the role of social organizations to better deliver integrated HIV/AIDS prevention work". Although for years the majority of publicly funded HIV programs in China have failed to engage NGOs and experience limited success in HIV prevention [65], China has recently founded "China AIDS fund for non-governmental organizations" since the year of 2015, endorsement and support in the form of official documentation for NGOs from the governmental settings has been initiated to promote work on HIV control and health services delivery. Findings of this study have some valuable implications for addressing the increasing HIV epidemic among China's gay population. First and foremost, HIV prevention intervention should go beyond merely condom promotion and HIV test promotion to sensitively address issues of stigma towards the gay community. Intolerant social environments pose enormous perceived barriers for testing and further utilization of health services. Secondly, prevention strategies should be better designed for gay men's needs rather than merely for HIV prevention. Other STIs are areas of concern for gay men, but services for their prevention and treatment are rather limited, and not currently linked with HIV efforts. Apart from these, in China, basic HIV prevention and control measures saw implementation through CDCs, where staff is often insufficiently trained to work with the gay population. Inclusion of culturally competent materials within MSM serving NGOs with experience in providing care and psychological support for clientele is imperative. Lastly, effective education is imperative to better equip individuals with knowledge such that they may engage in safer sexual and other risk behaviors. Yet despite the tremendous need for reliable, culturally sensitive information, our respondents were vocal about the pressing issue of addressing the psychological pressure, and stigma they experienced, from their surrounding environment. More than techniques for prevention of HIV/AIDS or other sexually transmitted infections, their answers reflected a desire for ways to exist with dignity and support, ways to negotiate their gay self identity in a society and world that currently discriminates against them cruelly. To date, silence on such issues has compounded stress, and led to mental health issues not limited to depression and suicide [18,72]. Taken together, these findings complement the body of knowledge from qualitative studies regarding socio-cultural factors impacting gay men's sexual and health seeking behaviors. However, this research has several limitations. First, because we recruited only in eight sites, it is possible that results are not easily generalizable to other parts of the country. Second, recruiters were the link which enabled us to reach those who might otherwise have refused participating due to stigma and prejudice surrounding research topics. However, this kind of recruitment has limitations as it mainly relies on the recruiters' personal networks and is potentially not representative of the target population. --- Conclusions To our best knowledge, this study is among the first to utilize a socio-cultural angle to evaluate present issues and priorities within the gay community culture in mainland China. It uses a lens which allows for a more holistic view of the community's perceived stresses and barriers to care. Future directions are numerous. Transformations within the gay community call for effective sex education and positive engagement of the demographic to build a healthier community culture. Targeted HIV prevention measures which take into consideration of sociocultural influencing factors are imperative to future successful efforts in wellness and health of the community. --- All relevant data are within the paper. --- --- Data curation: HH FL. Formal analysis: HH NZ.

Current Chinese studies continue to view male homosexuality through a disease focused lens which pays limited attention to socio-cultural aspects of sexual behavior and HIV transmission. This qualitative study aimed to investigate how socio-cultural factors influence gay men's sexual beliefs and behaviors in contemporary China, and their implications for HIV epidemic.Qualitative methodology was used in this study. During 2015-2016, in-depth interviews were conducted with 61 self identified gay men in Jiangxi, Henan, Heilongjiang, Guangdong, Jiangsu provinces and Chongqing municipality of China. Our study revealed that: 1) influenced by Chinese traditional culture, gay men have conflicts on self-identity, which led to low self-acceptance and negative attitude on sex, and huge socio-psychological stress; 2) a generational differences within gay community was observed, reflected in varied sexual attitudes and practices as well as way for approaching new friends, both of which have implications and challenges on HIV control and prevention; 3) socio-cultural barriers, including open minds towards casual sex and nonmonogamous relationship, and low priority of health demands were widely observed and led to negative coping with AIDS among gay community.It is essential to take a holistic view into gay men's HIV epidemic in China. Socio-cultural barriers for HIV control and prevention found in this study call for serious and imperative

Introduction Reducing child mortality by two thirds between 1990 and 2015 is the target of the fourth Millennium Development Goal . Recent analyses indicate that there has been major progress towards this goal. Current estimates show that the global under five mortality rate dropped from 93 per 1000 live births in 1990 to 72 per 1,000 in 2006 [1]. The targets to measure the progress towards the achievement of the Millennium Development Goals assess the progress of countries at the national level. However, national averages could be misleading, particularly in countries with great inequalities, as progress could be attained by improving the health of the well off while neglecting the health off the worst off. A recent report to monitor progress towards the attainment of the MDGs in 68 priority countries, which account for 97% of all maternal and child deaths, found that the poorest 20% of the population are less likely to be covered by effective interventions capable of preventing most maternal and child deaths than their wealthier counterparts [2]. It is therefore crucial to produce within country estimates to properly monitor progress of the MDGs and share countries' experience on policies and interventions that have been successful in reducing health inequalities and improving the health of the poor. In the Latin America and Caribbean Region, the large inequalities in living conditions within countries together with the increase in the percentage of U5MR attributed to neonatal mortality jeopardize the MDG-4 target [3]. In Latin America, neonatal mortality represents more than half of overall infant mortality and 42% of under five deaths, and most of the neonatal deaths happen during the early neonatal period [4]. Brazil is the largest economy in Latin America [5] and it is considered to be one of the five most important emerging economies of the world along with Mexico, China, India and South Africa [6]. However, despite Brazil's economic achievements, it is still at the top of the list of countries with the highest income inequalities [7]. The poorest 20 percent earn 2 percent of the income while the richest 20 percent earn 63 percent [5]. Brazil's national territory is divided administratively into 5 regions, 27 states and 5,507 municipalities . There are wide variations in population, wealth, climate and size across geographical areas. For example, 25% of the municipalities have population below 5,000 inhabitants, while a few have populations above 1,000,000 inhabitants . In recent decades Brazil has achieved very important health gains [8] and is on track to achieve most of the MDG targets at the national level [9]. National level estimates show major progress in the reduction of U5MR. Between 1990 and 2005 child mortality dropped from 53.7 to 28.7, a declined of 46.6% and infant mortality dropped from 33.7 to 21.1, a declined of 37.4%, a drop of more than one third during the period, suggesting that Brazil is on track to reach the MDG-4 target by 2015 [9]. The health sector reform of the 1990's and the public health and inter-sectoral interventions implemented by the Brazilian government have contributed to the reduction of U5MR. During the last two decades Brazil has undergone a series of health reforms aimed at: 1) providing universal access to health services free of charge to the entire population -Unified Health System), and 2) decentralizing the decision making to the municipality governments . This resulted in the expansion of services, the most significant being the Family Health Program , focused on providing and improving primary health care, particularly maternal and child care, among the poor [10][11][12]. Other public health and inter-sectoral programs have also been introduced to reduce infant and child deaths, such as access to clean water and sanitation, education of the mothers and immunization coverage. In addition, the federal government launched other programs to fight against hunger and poverty [13], such as the Zero Hunger program [14] and the cash transfer program [15]. In 2006, 99% of 1 year olds were covered by immunization of measles vaccine , diphtheria and tetanus toxoid and pertussis vaccine , hepatitis B vaccine and haemophilus influenzae type b and 92% of neonates were protected at birth against neonatal tetanus [16]. However, despite the efforts made to improve access to health care services for the poor and the policies implemented to reduce child mortality, in practice there remain major health inequalities. The poorest region had the highest under five mortality rate of 39 per 1000 live births in 2005, whereas richer regions had the lowest child mortality rates of around 18.5 per 1000 live births [9]. Moreover, in 2004, 6% of the children less than five years old in the poorest region died from diarrhea, compared to 2% of the children in richer regions [17]. An evaluation of three maternal and child health programs found less access to health care among the poor [18]. A recent study also found that poorer municipalities had a lower proportion of deliveries attended in health facilities and a lower proportion of pregnant women covered by antenatal care than richer municipalities [19]. This is particularly important as most of the deaths of newborns are directly related to inadequate care during and after pregnancy and child birth [4,20]. Although the challenges involved in reducing child mortality are widely recognized, there is not enough evidence currently available to monitor whether newborn and child health interventions have been successful in reaching the poor at municipality level . Most of the existing evidence in neonatal and child mortality shows differences across regions or states [9,[21][22][23][24] but very few analyses have been undertaken to monitor inequality trends across municipalities throughout the country [25,26]. Furthermore, evidence suggests that particularly in the poorest regions , national civil registration data are not an accurate source of information on infant deaths [27][28][29][30]. In this paper, we develop accurate estimates of neonatal mortality at the municipality level. We then use these estimates to measure inequality trends in neonatal and under five mortality across municipalities characterized by their socio-economic status, in a period where major health reforms and several pro poor policies were implemented in Brazil focused on decreasing child and neonatal mortality. We then infer whether these policies and interventions in newborn and child health have been successful in reaching the poor as well as the better off in Brazil. --- Data and methods To analyze inequality trends in under five and neonatal mortality at the municipality level we constructed a dataset that compiles information on mortality rates and socioeconomic indicators for the 5,507 Brazilian municipalities for 1991 and 2000. The most recent year of analysis is 2000, as it is the year with the most complete and accurate data to monitor differentials in child mortality at municipality level. Brazil has two systems to monitor vital events: the Mortality Information System and the Information System on Live Births . Several evaluations suggested however that neither system is complete enough to monitor differentials in infant and child mortality [27][28][29][30]. It is estimated that the Mortality Information System [31] underreports 25% of deaths in the Northern Region and 29% in the Northeastern Region [28]. Similarly, the coverage of the Information System on Live Births [32] is only 73% in these two regions compared to the national coverage of 93% [33]. These inaccuracies are likely to be higher among children under one year of age and in municipalities with less than 50,000 inhabitants [29]. For these reasons we used Census data in this analysis. Data on the under five mortality rate per 1000 live births, estimated from the 1991 and 2000 population Census of Brazil, were obtained from the Institute of Applied Economic Research , which compiles and produces publicly available data of socio-economic indicators of the municipalities [34]. The neonatal mortality rates per 1000 live births are a predictive estimate of the under five mortality data. First, we produced sub-national rates of neonatal mortality and child mortality by applying direct life table methods to birth histories from the various Demographic and Health Surveys . DHS is a survey designed to collect detailed information on social and demographic characteristics and maternal and child health over a sample of women aged 15 to 49 years old in developing countries. The DHSs of Brazil used in this study were conducted in 1986, 1991 and 1996 and are sub-nationally representative: 1) 1986 is representative at regional level; 2) 1991 was conducted in the North-eastern region and is representative at the state level; and 3) 1996 is representative at the regional level and in four states in the North-eastern region, Rio Grande do Norte, Bahia, Ceará and Pernambuco [35][36][37]. Second, we used the rates produced from DHS to investigate the relationship between neonatal and under five mortality per 1000 live births at sub-national level using a log-log regression model. In addition, we included binary variables in the models to control for the different years of DHS. Finally, we extrapolated these relationships to predict neonatal mortality rates for the municipalities of the Southern and Northern Regions using the estimates of under five mortality rates at municipality level for 1991 and 2000. Data on the proportion of population below the poverty line , were obtained from IPEA [14]. These data were then used to group municipalities by poverty quintiles, where the fifth quintile represents the poorest 20% of municipalities and the first quintile the richest 20%. We also categorize municipalities as poor or non-poor . The first two quintiles were considered as poor as they comprise the municipalities with more than 50% of their population below the poverty line. To monitor whether improvements reached the worst off as well as the better off populations, there are two measures of the extent of mortality inequality. The first one, the relative measure, defined as the ratio of child mortality in the poorest quintile to the richest quintile; and the second one, absolute measure, defined as the difference in mortality between the poorest and richest quintile [38]. We also constructed an index to classify municipalities in four categories depending on the under five mortality reduction between 1991 and 2000 and the level of under five mortality in 2000 to identify the municipalities that require urgent policy interventions. --- Results The results are presented in two subsections . First, we show how we produced the data of neonatal mortality at municipality level for 1991 and 2000. Then, using these data and the existing data on under five mortality, we present the results of inequality trends in neonatal and child mortality across poor and non-poor municipalities in Brazil. --- a. Prediction of neonatal mortality The log-log relationships between neonatal and under five mortality across sub-national units in Brazil are presented in Table 1. We found two different relationships; one for the three southern regions and a second one for the North and Northeast Regions. For the northern regions, a one percent increase in U5MR is associated with a 0.79 percent increase in NNMR, while for the southern regions, a one percent increase in U5MR is associated with 0.94 percent increase in NNMR. For the Southern Regions the model explains 94% of the total variance while for the Northern Regions the model explains 61%. It is likely that there are other socioeconomic characteristics that were not taken into account in the model for the Northern Regions that are likely to explain the remaining variance. None of the dummy variables included in the models to control for differences in the survey years were significant. We used these relationships to predict neonatal mortality from existing estimates of under five mortality at municipality level for 1991 and 2000. We compared the averages of our predicted estimates of neonatal mortality by region with the numbers produced for the same year by two official sources of information; the Brazilian monitoring report on the MDGs, 2004 [25] and the 2008 Report on Basic Health Indicators [17]. We found that our predicted estimates are very similar to the numbers presented in these two reports, which are corrected from the underreporting of deaths in the Northern Regions and used different methods and sources of information to produce their estimates . This consistency implies that our predicted estimates of neonatal mortality are reasonable and can be used in the analysis presented in the following section. The advantage of our estimates of neonatal mortality is that they can be disaggregated at the municipal level which is the level where the decisions on the administration and provision of services are made. --- b. Inequalities in neonatal and under five mortality at sub-national level The distributions of the predicted neonatal mortality and under five mortality across poor and non-poor municipalities are shown in Figure 1 a, b, c and1d. In general, we found that between 1991 and 2000 there has been a decline in neonatal and under five mortality across poor and non-poor municipalities. Despite these declines, in 2000, poorer municipalities still have much higher rates of neonatal and under five mortality than richer municipalities. We looked in more detail at these differences by disaggregating the data into the five poverty quintiles. The average of the predicted neonatal mortality and under-5 mortality across municipalities grouped by poverty quintiles are shown in Figure 2 and in Table 3. Between 1991 and 2000, the averages of neonatal and under five mortality have decreased steadily across all socioeconomic groups. However, we found great inequalities across economic groups: the poorest municipalities had higher neonatal and under five mortality than any other economic quintile and this problem has not changed over time. For example, the neonatal mortality among the poorest municipalities in 2000 is similar to the average national rate of this indicator in 1991 and to the national rates of low income countries like Eritrea and Kenya , while the average rate among the richest municipalities is similar to the national rates of uppermiddle-income countries like Mexico and Romania [20]. From the relative ratio between the poorest 20 percent and the richest 20 percent of municipalities, we found that the inequity gaps have also increased in the period analyzed. The poorest municipalities had double the neonatal mortality of the richest municipalities in 1991 and this gap has increased to 3 times higher in 2000. These gaps are even larger in under five mortality: in 1991 the poorest municipalities had U5MRs 3.7 times greater than the richest municipalities and this gap increased to almost five times more in 2000. This implies that the policies and interventions introduced in the late 1980 s and 1990 s are failing to improve the relative position of the poor municipalities. The absolute differences show that the neonatal mortality between the poorest and richest municipalities differed by 17.5 * Estimates are produced using census data for some states [27] and for the remaining states estimates are produced using data for the Mortality Information System and from the Live births Information System . ** Estimates are produced combining census data and survey data of the Pesquisa Nacional por Amostra de Domicílio -PNAD. deaths in 1991 and 17.7 in 2000, while the difference in under five mortality between these two groups fell from 83 deaths in 1991 to 65 in 2000. In terms of the decline experienced between 1991 and 2000 at national level, there was a decline of 33% in under five mortality suggesting that Brazil is on track to meet the MDGs' child health target by 2015. However, this decline is not homogenous across poverty quintiles. We found an increasing gradient in the percentage decline by poverty quintiles, such that richer municipalities experienced a faster decline in under five and neonatal mortality than poorer municipalities. For example, the poorest municipalities experienced a reduction of 24% in neonatal mortality and of 28% in under five mortality, while the richest municipalities had a decline of 49% in neonatal mortality and of 45% in under five mortality. Figure 3 shows that inequalities within regions and states are also very wide we found that inequalities in neonatal mortality between the poor and the rich have increased in a large majority of the states during the period analyzed. However, the highest inequalities were found in states belonging to richer regions -Southern Regions-. For example, we found that the states of Sao Paulo and Rio Grande do Sol had the highest inequalities in neonatal mortality in the country, with a poor non-poor gap of 2.4 and 1.7 in 2000 respectively. These are also the states with the highest inequalities across different socioeconomic indicators, which may explain our findings [41,42]. Sao Paulo for example, combines the poorest and the richest populations, thus it is not surprising that we also found the highest inequalities in neonatal mortality. We related the under five mortality reduction between 1991 and 2000 with the level of under five mortality in 2000 across poor and non poor municipalities . We added as cut-points the national values of both variables. The cut-point of 33% represents the national decline in under five mortality between 1991 and 2000 and the cut-point of 33 per 1000 lb represents the level of under five mortality at national level. From this plot we can observed four groups of municipalities: 1) group one, municipalities with high under five mortality -defined as rates above the national cut point of 33 per 1000 lb-and with low decline -defined as reduction below 33% national decline-; 2) second group, municipalities with low under five mortality -defined as rates below the 33 cut point-and low decline; 3) third group, municipalities with high under five mortality and high decline -defined as reductions above 33% national decline -; and 4) fourth group, municipalities with low under five mortality and high decline. A large majority of municipalities in group one -low decline and high under five mortality-are poor municipalities, while almost all municipalities in group four -high reduction and low under five mortality-are rich municipalities. For policy purposes, we constructed an index using the groups of municipalities of Figure 4 and mapped its distribution across municipalities to identify the critical geographical areas . Not surprisingly, we found that the majority of municipalities in group one -low decline and high under five mortality-belong to the poorer regions where the majority of poor municipalities are concentrated. Specifically, a large majority of municipalities in this group are concentrated in four states of the Northeast Region: 24.4% in Bahia , 13% in Maranhão , 10.4% in Piauí , and 8.8% in Paraíba , which are also the states were we found the highest levels of neonatal mortality among the poor in Figure 3. However, we also found other states from wealthier regions with a significant number of municipalities in group one. For example, 47% of the municipalities in Mato Grosso and 33% of the municipalities in Espirito Santo were classified in group one despite the fact that these states belong to richer regions like the Center West and Southeast Regions. Despite the socioeconomic conditions, there are some poor municipalities in the poorer regions that have been successful in reducing under five mortality to levels below 33 per 1000 lb : these municipalities are mainly concentrated in the states of Pernambuco , Rio Grande do Norte , Alagoas , and Ceará from the Northeast Region and Pará from the North Region. --- Conclusions This study highlights the importance of conducting sub national level analysis to monitor progress on MDGs. Sub-national level analysis is also important in countries that have undergone decentralization reforms, specifically at the level where the decisions on the administration and provision of services are made in order to determine whether policies and interventions have been successful in improving the health of the population. Several policy implications may arise from this study. As shown by the case of Brazil, the policies and interventions focused on improving newborn and child health have been more effective in reaching the better off than the worst off. However, the magnitudes of these inequalities are not shown by the national numbers. We found that the achievement in the reduction of under five mortality at national level has mainly been reached by reducing the level of under five mortality among the rich. Poor municipalities suffer from worse newborn and child health than richer municipalities and the poor/rich gaps have further increased, thus jeopardizing the possibility of reaching the MDG-4 target by 2015. This findings are consistent with the report to track progress on the attainment of the MDGs on child and maternal mortality by 2015 [2]. Most of the deaths between one year and five years old are related to infectious disease and could be averted by very cost effective interventions [43]. However despite the implementation of specific interventions to decrease child mortality in poor areas, still 6% of children under five in these areas die from diarrhea [17]. In addition, despite the implementation of actions to increase access to clean water and sanitation among the poor, still 83% of the households in rural areas do not have access to improved drinking water as compare to 9% of the households in urban areas [44]. It is therefore crucial to pursue intersectoral interventions to improve the socioeconomic conditions of poor municipalities. In the other hand, most of the deaths of newborns could be prevented with access to adequate care during and after pregnancy and child birth [4,20]. However, a recent study found that poorer municipalities had lower proportion of deliveries attended in health facilities and proportions of pregnant women covered by antenatal care than richer municipalities, the differences being attributed to lack of human resources, qualified personnel, and local health facilities [19]. Thus, further efforts are required from municipal, state and federal authorities to make health systems more equitable and to identify the interventions that have the ability to reach the poor and reduce socioeconomic inequalities in maternal, newborn and child health. These results contribute to a major understanding of inequalities in newborn and child health within Brazil. Brazil's MDG report 2007, has pointed out that policies and programs should be targeted to improve the health of the population in the poorest Regions to address health inequalities [9]. In this study, we provide further evidence and found that not all the municipalities in these regions require additional policy interventions. In fact, despite being economically disadvantaged, some poor municipalities in the North and Northeast are performing very well and their policies and interventions can be used as an example for further actions by municipalities with similar socioeconomic conditions. Furthermore, the majority of the municipalities with low decline and high levels of under five mortality have also very high levels of neonatal mortality. These municipalities are concentrated in four states in the Northeast Region, but some are found in richer states and richer regions. The conclusions drawn from this study should take into consideration the limitations of the data. The data used in this study are from 2000, which is the most recent year with valid estimates of child mortality at municipal level . It is therefore likely that the level of neonatal and child mortality in this study overestimate the current numbers. Although, several national and sub-national interventions have been implemented to reduce inequalities in neonatal and under five deaths across socioeconomic groups since 2000, the most recent information shows that in general, between 2000 and 2005, there have been relatively few improvements in neonatal and child health inequalities between the poorest and richest regions [9,17]. Data quality may have affected the pattern of our estimates of neonatal mortality in the Northern Regions. We have therefore performed a sensitivity analysis producing the estimates of neonatal mortality for the municipalities in the Northern Regions using the relationship of the Southern Regions . We found that the estimated neonatal mortality in the Northern municipalities is higher if the Southern relationship is used than the estimates when the Northern relationship is used. We also found that the patterns found across municipalities do not change and even more the inequalities are more accentuated when the Southern relationship is used. This implies that the results and conclusions found in this study do not change when using different extrapolation models. We found that the model of the Northern Regions explains 61% of the total variance while for the Southern Regions it explains 94% of the variance. Although the lack of explanatory power of the model in the Northern relationship may have affected the quality of our estimates the sensitivity analysis demonstrates that the patterns found remain the same even when using different models. Despite the limitations associated with the data, this study has highlighted some critical issues in terms of the persistent inequalities in neonatal and child mortality within Brazil. This is particularly important for most developing countries and countries with great inequalities, thus further efforts are required to improve the measurement and monitoring of trends in neonatal and under five mortality at sub-national level. --- Additional material Additional file 1: Appendices 1-6. --- Contribution of authors AS conceived the study, conducted the analysis and interpretation of the data and drafted the paper. KH advised on the methodology, contributed to the interpretation of results and writing of the paper. MDP contributed to the interpretation of results. All authors read and approved the final manuscript. --- Competing interests The authors declare that they have no competing interests.

Objective: Brazil's large socioeconomic inequalities together with the increase in neonatal mortality jeopardize the MDG-4 child mortality target by 2015. We measured inequality trends in neonatal and under five mortality across municipalities characterized by their socio-economic status in a period where major pro poor policies were implemented in Brazil to infer whether policies and interventions in newborn and child health have been successful in reaching the poor as well as the better off. Methods: Using data from the 5,507 municipalities in 1991 and 2000, we developed accurate estimates of neonatal mortality at municipality level and used these data to investigate inequality trends in neonatal and under five mortality across municipalities characterized by socio-economic status. Results: Child health policies and interventions have been more effective in reaching the better off than the worst off. Reduction of under five mortality at national level has been achieved by reducing the level of under five mortality among the better off. Poor municipalities suffer from worse newborn and child health than richer municipalities and the poor/rich gaps have increased. Conclusion: Our analysis highlights the importance of monitoring progress on MDGs at sub-national level and measuring inequality gaps to accurately target health and inter-sectoral policies. Further efforts are required to improve the measurement and monitoring of trends in neonatal and under five mortality at sub-national level, particularly in developing countries and countries with large socioeconomic inequalities.

INTRODUCTION The progress of time is growing rapidly. One of them is the internet, where the internet provides everything we need, ranging from information, entertainment, and various media facilities to communicate. Often the internet is abused by teenagers with a form of cybersex behavior. Cybersex behavior is the activity of viewing pornography, engaging in sex chat, using devices such as web-cams to engage in sexual activity online and seeking sexual partners online or engaging in 3D sex role play. . Cybersex behavior is an activity that increases sexual arousal and is carried out online such as looking for a sex partner, communicating between two people discussing sex and seeking sexual content in the form of photos or audio. . In the United States there are many internet users, especially teenagers who engage in cybersex behavior. About 60% of internet users visit sites or websites that are sexual in nature and 87% of students have virtual sex . According to survey data from the Ministry of Information and Communication , ranking of porn site users in Indonesia in 2013 Indonesia was ranked sixth for porn site access, then in 2014 it increased to third place, and in 2015 to second place. Sexual cases involving school-age children continue to increase from year to year Ali & Asrori, . The number of cases increased by 50% from 2014 to 2015. Of these, 10% of cases occurred in schools. In Indonesia, the result was that of the 238 respondents who accessed pornographic or sexual content found, there were around 175 respondents who were in the adolescent age category. . Research in East Java also shows the results of 76.8% of cybersex connoisseurs aged 15-17 years by watching porn on gadgets . Based on a preliminary study on February 2022 at SMA Antartika 2 Sidoarjo, researchers conducted interviews with 10 class XI students and found that 100% of students said they had accessed pornographic films out of curiosity and then they searched for pornographic films on the internet. The cause of cybersex behavior is due to the low level of knowledge about reproductive health and the high behavior of premarital sex . Factors that cause cybersex behavior are because parents are not open enough to talk about sex with adolescents, there is parental prejudice against adolescents and internal and external problems do not really affect adolescent monitoring . Excessive Cybersex behavior can lead to problematic Cybersex behavior such as premarital sex. Problematic cybersex is generally defined as excessive and uncontrolled involvement in online sexual activity which is characterized by several symptoms such as a persistent desire to behave Cybersex . Cybersex behavior is the lack of parental supervision in supervising adolescent activities so that they run well and are not misdirected. The importance of the role of parents is needed for adolescents by limiting youth activities to positive activities, parents must also know with whom teenagers associate and establish good and positive communication between parents and adolescents. The role of the teacher is also important in providing guidance and counseling and providing additional activities in positive learning. --- RESEARCH METHODS The design used in this study is a correlation analytic research design with a crosssectional approach. This research was carried out on 02 June - --- DISCUSSION The results showed that the distribution of respondents based on the type of sexual communication in the respondents showed that the majority of parents' sexual communication with adolescents was in the low category of 56.2%, the medium category was 33.1%, the high category was 27.2%. There are several factors that cause this, including the perception of parents that talking about sex is taboo and parents of teenagers showing negative attitudes towards access to cybersex behavior, as well as a lack of parental knowledge . The majority of mothers do not provide sexual and reproductive communication to their children, and one of the contributing factors is the low knowledge of parents about cybersex behavior . There are several factors that trigger low cybersex behavior, namely self-control, where a negative relationship between self-control and adolescent cybersex behavior is acceptable, the higher self-control that adolescents have, the lower cybersex behavior occurs in adolescents who use the internet . Selfcontrol is needed when accessing the internet, when a person does not control himself and uses the internet inappropriately it can cause addiction to cybersex behavior . The results showed that the majority of family monitoring of adolescents was in the low category as much as 39.7%, the medium category was 33.1%, the high category was 27.2%. Family monitoring or parental supervision is how closely parents monitor their teenagers. Low family monitoring, where parents do not monitor and control their children will tend to experience violations, in contrast to parents who control or supervise their children properly . When teenagers don't get enough self-monitoring from their parents, this can be an opportunity for teenagers to engage in risky behavior. Supervision from parents can form an environment that requires adolescents to obey and minimize risky behavior . From the statement above, low family monitoring relationships affect the incidence of cybersex behavior, the role of mothers in parenting such as directing adolescent behavior by advising adolescents, involvement of family members, trust and expectations of mothers , communication and the quality of the relationship between mothers and adolescents . Meanwhile, internal and external problems did not really affect the monitoring of adolescents causing cybersex behavior to be influenced by family monitoring. --- CONCLUSIONS There is a relationship between sexual communication and family monitoring to adolescent cybersex behavior. It is hoped that families will pay more attention to adolescents in monitoring their sexual behavior. In addition, it also implements open communication with adolescents regarding their sexual behavior, so that adolescents avoid cybersex behavior.

Cybersex is the activity of accessing pornography on the internet, engaging in real-time, like conversations about online sexuality with others. The purpose of this study was to analyze the influence of family monitoring and sexual communication adolescent cybersex behavior in Sidoarjo. The study used cross-sectional analytics. Probability sampling technique with simple random sampling. Samples were 121 adolescent students of SMK Antartika 2 Sidoarjo. The research instrument used sexual communication questionnaires, family monitoring, and ISST questionnaires to assess cybersex behavior. Statistic test using rho sperman test. Results showed adolescent sexual communication with parents was 56.2% in the low, 29.8% moderate and 14% high categories. Family monitoring related to sexual behavior was 39.7% in the low category, 33.1% in the medium category, and 27.2% in the high category. Teen cybersex behavior was 61.9% riskless, 17.4% were low-risk, and 20.7% had high-risk behavior. The results of the spearman rho test between sexual communication and cybersex behavior showed p value = 0.000 which means that there is a relationship between sexual communication and cybersex behavior. In the family monitoring variable with cybersex behavior, it is obtained p value = 0.000, meaning that there is a relationship between family monitoring and adolescent cybersex behavior. It is hoped that families will pay more attention to adolescents in monitoring their sexual behavior. In addition, it also implements open communication with adolescents regarding their sexual behavior, so that adolescents avoid cybersex behavior.

To understand how citizens assess representation, we draw on classical and constructivist representation theory. In her landmark work, Pitkin theorizes representation as a factual outcome of legislative politics. She distinguishes substantive , descriptive , and symbolic representation . Constructivist approaches see representation as a claim. Representatives-inside or outside of parliament-claim to speak and act for the represented, and in doing so, constitute political groups and their interests. Despite their differences, political actors are central to both theoretical approaches. In this study, we put the spotlight on citizens . Drawing on twelve focus group interviews , we compare citizens' assessments across three state categories: Dutch citizens with Turkish, Moroccan, and Surinamese "migration backgrounds." The Dutch proportional electoral system with preferential voting offers favorable conditions for the election of ethnically/racially minoritized candidates in comparison to other European democracies . We find that citizens' self-identified political groups do not always align with state categories that are often used in studies of representation. Citizens associated political concerns important to them with self-identified groups outside the scope of state categories . For most participants, racism matters more to their political self-identifications and concerns than migration background-the commonly used state category in scholarship on Europe. In participants' assessments of political actors, representation emerges as a "me ´lange" of constitutive, substantive, symbolic, and descriptive representation . Participants care most about whether parties or politicians act on their self-identified group interests: substantive representation. But they feel that established parties sometimes fail to do this. In this context, many research participants see political parties led by ethnically/racially minoritized leaders, as well as activist groups, as claim makers who represent issues that mainstream parties neglect. Even if substantive representation is decisive for most participants, the symbolic presence of role models appears to be significant too specifically for those who are descriptively most underrepresented . Finally, participants experience "moments of ambivalence" that illustrate dilemmas in assessments of representation. Next, we develop a conceptual framework of political representation from the perspective of ethnically/racially minoritized citizens. Subsequently, we detail case selection and methods. The empirical body consists of three parts. First, we investigate how ethnically/racially minoritized citizens constitute and politicize self-identifications and interests. Second, we explore how citizens assess parliamentary representation, and finally, we highlight the role of extra-parliamentary politics, taking BLM as a case. --- Political Representation: A Citizen Perspective Feminist theorists argue that historically disadvantaged citizens benefit from group-level representation to counter shared political obstacles . Hence, recognizable social groups and their interests are central to conceptualization and measurement of political representation. To study representation empirically, researchers mainly identify a group based on proxy "state categories" used in social surveys, the census, or policy, like "women/men" or "African American" . Subsequently, researchers often predefine what the political interests of this group are . Three critiques on predefining groups and interests inspire our approach to intersectionally study how citizens themselves constitute and politicize their self-identifications and interests, in relation to how researchers and political actors do so. First, using predefined categories as proxy for sociopolitical groups overlooks the diverse ways in which citizens politically self-identify . While some citizens self-identify with state categories, others reject or redefine them . Thus, citizens who fall in the same state category do not necessarily politically self-identify in the same way or share group-based interests . Many European countries avoid ethnic/racial markers and categorize citizens based on "migration background." By focusing on shared migration history, the migration background as a state category disguises group identifications and interests based on shared racialization. Given this limitation of predefined state categories, we ask ethnically/racially minoritized citizens: how do you politically self-identify, and what are your political interests? Second, predefined state categories allow for measurement along one social axis , assuming that minoritized citizens' political selfidentifications and interests are uniform. This risks overlooking how minoritized citizens experience representation differently, depending on their gender, sexuality, age/generation, social class, or ability . We apply an intersectional framework to study how the perspective of individuals and groups in society is influenced by a range of interconnected structural positions . This permits a more nuanced account of variation both "within" and "between" predefined state categories . We adopt state categories provisionally to unpack how ethnically/racially minoritized citizens' experiences vary within and between self-identified political groups. Third, sociopolitical groups and interests are considered stable and transparent in classical representation theory . Constructivist scholars stress how political actors construct groups and/or their interests through claiming to represent them: constitutive representation. Constructivist researchers do not predefine groups and interests and instead investigate how political actors do so. But we cannot be sure whether minoritized citizens agree with how political actors constitute their self-identified political group and interests. Representative claims that constitute groups can speak on behalf of, but also against, a group ). Disadvantaged citizens may lack the resources or networks to object to claims. Citizens could also experience that politicians address their interests without mentioning their self-identified political group . To empirically study how citizens constitute their political selfidentifications and interests, we draw on Young's argument that minoritized citizens share social perspectives derived from their structural intersectional social positions. Social perspectives become visible when a diverse group of minoritized citizens discuss their experiences . From this, individuals constitute shared, but also diverse, political self-identifications and interests . We explore how minoritized citizens constitute political self-identifications and interests in conversations. To understand how citizens assess representation, we depart from Pitkin's three dimensions of representation. Substantive representation occurs when politicians act in the interests of the electorate, in a manner responsive to them. Descriptive representation refers to shared social backgrounds between citizens and politicians. Symbolic representation captures symbols that evoke people's relevant attitudes or emotions towards what they stand for, or what is being represented . While Pitkin foregrounds substantive representation, feminist theorists argue that descriptive representation by politicians who share citizens' experiences is crucial for minoritized groups . Constructivists conceptualize representation as a claim by political actors consisting of constitutive, substantive, and/or other aspects . This fits Celis and Childs' proposition that citizens assess representation as an interrelated "me ´lange" of different aspects. Classical and constructivist theories inspire us to investigate how citizens assess representation, which aspects matter most, and to whom. Citizens may accept or reject claims, or they may be undecided or hesitant. Emejulu talks about ambivalence, when "a range of opposite emotions about a person or a situation comingle." Moments of ambivalence allow us to study citizens' emotional or cognitive orientations towards political representation in relation to the prevailing social norms, and the raced, gendered, and other social structures underlying these tensions . To reveal who citizens believe represent them, the extant literature provides some pointers. Descriptive representation can advance minoritized groups' substantive and symbolic representation . Minoritized politicians' presence in powerful positions can flip negative stereotypes by showing that historically excluded groups are "able to rule" . A meta-analysis demonstrates that voters assess candidates with whom they share the same ethnic/racial background significantly more positively than candidates with whom they do not share this background . We expect ideological and intersectional variation in how citizens value descriptive politicians. Citizens may prefer descriptive representatives with stronger connections to their self-identified political group . Preferences for descriptive representatives intersectionally vary within and between self-identified political groups , depending on the political context and issue . How citizens value role models can also vary by age and gender, with role model discourses tending to be male centered . At the same time, minoritized politicians can downplay their identity to conform to dominant party or societal norms , triggering disappointment among citizens . Next to descriptive MPs, ethnically/racially minoritized citizens might favor certain political parties. Historically, left-wing parties have been relatively open to minority rights and socioeconomic mobility, although they have become less vocal on these issues . In this respect, the electoral system matters. Countries with open systems, like the Netherlands, allow greater space for new parties that take up minority interests to emerge . Even if they are small and do not enter government, they can be influential through contagion effects . However, a study of parliamentary politics alone likely cannot distil who ethnically/racially minoritized citizens feel represented by. Ethnically/racially minoritized citizens may have interests that remain invisible in electoral politics because politicians and parties fail to address them . Furthermore, politicians may favor prototypical over intersectionally disadvantaged groups and interests . From a representative claim perspective, political actors include a broad range of players such as nongovernmental organizations, interest groups, and activists . Social movements may provide substantive representation and create spaces where minorities politicize self-identified groups and define shared political priorities . To capture both parliamentary and extra-parliamentary avenues of representation, we ask ethnically/racially minoritized citizens who they believe represents them. --- Case Selection and Methods The Netherlands has a proportional electoral system. Political parties compose a party list and citizens can vote for a politician of their preference on Political Representation and Intersectionality the list. It has no pre-set electoral threshold and a single, nationwide electoral district . This has allowed the so-called "ethnic/racial minority" parties, led by ethnically/racially minoritized politicians with an explicit anti-racist agenda, to gain parliamentary presence. In 2015, the first political party led by ethnically/racially minoritized politicians, DENK-meaning "think" in Dutch and "equal" in Turkish-gained seats in national parliament. The party is left-wing on economic issues with an anti-discrimination focus. 3 In the 2017 and 2021 parliamentary elections DENK gained three seats . In the 2021 national elections, the first party founded and led by a Black woman entered parliament: BIJ1. The BIJ1 program is intersectional, anti-racist, and decolonial. 4 DENK and BIJ1 put marginalized anti-racist critiques on the political agenda . This breaks with the past, in which the dominant establishment often upholds an innocent self-image, denying racism and the continuing impact of colonialism , while small, ethnic/racial minority parties have significant impact by expressing political ideas, requesting parliamentary debates and by building cross-party coalitions to get legislative proposals accepted. To explore citizens' assessments of political representation we conducted focus groups. This design facilitates studying how citizens collectively negotiate, debate, and potentially agree over who represents them. By composing groups based on state categories, we looked at how participants politicize selfidentifications and define group-level political issues in relation to how they are categorized . To explore the relationship between state categories and political self-identifications, we selected participants based on state categories: "non-western" migration backgrounds in the Dutch context. We draw on McCall's "intercategorical" and "intracategorical" operationalization of intersectionality, whereby ascribed categories are taken as starting points before systematically unpacking the inequalities that are imbricated "within" and "between" categories. The intracategorical approach involves studying a single social category "to analytically unravel one by one the influences of gender, race, class, and so on" ). We study how ethnicity/race, gender, and social class intersect in individual's experiences within the same migration background category. The intercategorical approach investigates relationships of inequality among predefined categories "as imperfect and ever changing as they are, and . . . explicate those relationships" ). Similarly, we compare assessments of political representation between different "migration background" state categories. Among each migration background state category, two focus groups were of mixed gender, and one comprised women only-to reduce social heterogeneity and power inequalities among participants . We selected Dutch citizens with Turkish, Moroccan, and Surinamese migration backgrounds: the largest electorates with "non-western backgrounds" and the most racialized and problematized in political debates . Turkish-and Moroccan-Dutch citizens arrived as semi-skilled labor migrants in the late 1960, followed by family migration. Most Surinamese-Dutch citizens migrated after the independence of Surinam, followed by labor, political, and family migration. They are all diverse in ethnic/racial and religious terms. Dutch citizens with a Turkish background are largely Muslim, but also include secular and ethnic/racial and religious minorities including Kurds and Alevis. Surinamese-Dutch are mainly Afro-Surinamese and East Indian-Surinamese, descendants of African enslaved people and contract laborers from India, respectively. Moroccan-Dutch are predominantly Muslim, rendering them a religious minority in an increasingly anti-Islamic political context . Next to Moroccan-Arabic speakers, most Moroccan-Dutch descend from the Berber minority . The focus group topic list aimed to generate discussions, while ensuring comparability between focus groups . We instructed the moderators to pick up on political self-identifications and representatives that participants themselves introduced and started by asking what feeling represented means to them. Subsequent questions zoomed in on substantive, descriptive and symbolic representation, and representative claims, for example, "Does it matter to have a representative who is like you? If so, why?" Data collection was undertaken between October 2020 and February 2021-a few months after the BLM protests in June 2020, and before the Dutch national elections in March 2021. The topics that participants brought up reflect this political context. The first author and research assistants recruited participants through their own extended networks, social media, neighborhood and ethnic/racial and religious minority organizations, flyers, and snowballing. To reach less politically engaged citizens, we offered a monetary incentive . Interested participants filled out a survey and an informed consent form . We finetuned the focus group composition based on interested participants' profiles . The final sample consisted of sixty-five participants distributed over twelve focus groups: four comprised Dutch citizens with a Turkish, Moroccan, and Surinamese migration background; three per migration background state category were mixed gender, one was women only. Twenty participants were first generation and forty-five second generation, both were present in almost every focus group to study generational differences in the conversations. The sample was slightly skewed towards university-educated, politically engaged, and left-wing oriented participants . We conducted one focus group with Moroccan-Dutch women aged over fifty-a particularly hard-to-reach profile-who knew each other and met regularly; most participants, however, did not know each other in advance. Except for the pre-existing Moroccan-Dutch focus group, the conversations were organized online using Zoom due to the COVID-19 pandemic. Participants mainly joined from urban areas. Attendees ranged from three to six per group, a size consistent with recommendations for online focus groups . With participants' permission, we videorecorded the focus groups. Participants actively took part in the conversation. Conversations lasted between 90 and 180 minutes and were mainly in Dutch with occasional statements in Turkish or Arabic translated by the moderator. Two research assistants moderated the focus groups. One moderator was a veiled Muslim woman with an Egyptian migration background, the other a secular man with a Turkish migration background. The first author occasionally jumped in. She is a white secular woman without a migration background. All moderators took notes during the entire research process and discussed the focus groups together. All team members had the impression that participants, especially the Surinamese-Dutch groups, saw the first author as a white out-group member. Positionality of the team was dynamic and changed across and during conversations depending on group composition and topic . The research assistants transcribed and pseudo-anonymized each focus group. The first author coded the transcripts in Atlas.ti in three stages: indexing and data reduction, applying analytical codes, and refining concepts through constant comparison . First, we used the focus group autocoding tool to link demographics with participants' statements. We indexed the transcripts by broadly coding political issues discussed , representative actors , political self-identifications , and conceptual themes . We identified recurring or notable aspects of the conversations. Together with notes on the research process and team meetings, these formed the starting point of the next stage of analytical coding. We coded the indexed segments in which participants discussed political actors in relation to the political self-identifications and interests they mobilized, and the reasons they felt represented by these actors. We made use of Atlas.ti's analysis tools such as code/document co-occurrence tables to explore similarities and differences within and between focus groups. Participants' names were replaced by pseudonyms to ensure confidentiality. --- Identifications and Interests What self-identified political groups and interests do participants constitute, and how do they relate to their ascribed state categories? The political issue that concerns nearly all participants are experiences with racism. When narrating these stories, participants sometimes adopt state categories, but in a reflexive way, mocking how state categories of non-white Dutch citizens change, while still maintaining racialized distinctions. For instance, they jokingly ask what their "politically correct" label is nowadays . Participants redraw the rigid boundaries of state categories by eschewing the state logic that divides them into distinct Turkish, Moroccan, or Surinamese migration backgrounds, talking more generally of racialized Dutch citizens. Participants refer to citizens with "migration backgrounds," but also to "minorities," "non-white Dutch," or "foreigners." Rachelle notes: "[i]n the seventies, eighties, it was the Surinamese [who politics problematized], then it became the Antilleans, now it's the Moroccans and soon we'll have the Turks again." Rachelle analyses how different racialized groups are commonly categorized and politically problematized in turn. She rejects state category divisions and highlights common experiences between ethnically/racially minoritized groups. In a similar way, Emir refers to ethnic/racial profiling to illustrate shared injustice. My two best friends are Black . . . one is Surinamese, and one is half Surinamese, and with them I sometimes talk about ethnic profiling. . . . That's something that people who live in the Gooi region or Barendrecht or a very expensive neighborhood don't understand, while for other people in my environment it's a daily thing from the time you're a teenager . . . yes, then at moments like that, you really see that it's a sort of a minority problem. Emir distinguishes himself from people who live in the Gooi showing how social class and ethnic/racial disadvantage intersect. Like Rachelle, he highlights common minority experiences across racialized citizens, rather than experiences from his migration background state category alone. Socioeconomic inequality is another collective political issue that participants discuss, occasionally related to racism. They mention poverty and a lack of affordable public services such as housing and healthcare. Angela : "I didn't know where to go after my [student] room contract [ended]. . . . I didn't have parents who had a pot of money. . . . for me the right to exist is important, that everyone can develop themselves and use public resources." Young participants mention economic pressures, some as first-generation students without familial financial support. Most believe that combating racism and socioeconomic inequality are collective interests that cut across migration background state categories. In self-identified political groups, ethnic/racial and social class intersect. Participants mention political interests specific to their self-identified political ethnic/racial and/or religious group. Turkish-Dutch citizens mostly politically self-identified hybrid: Turkish-Dutch, occasionally intersecting with Muslim. In these focus groups, participants discuss how Turkish political selfidentification and concerns arise from racialization; being quizzed about Political Representation and Intersectionality loyalty to the Netherlands during job talks, for example, or if they support Turkish president Erdo gan. They also discuss Turkish political selfidentification in terms of belonging to a Turkish ethnic group, which more closely corresponds to state categories. Here, participants differ in how they interpret Turkish ethnic self-identifications and interests, sometimes leading to debates. For some, political interests are transnational, not bound by borders, whilst for others Turkish ethnicity is a private matter with Dutch and Turkish political spaces and interests separated. Deniz: I think it's very important that we leave Turkish issues in Turkey. . . . Let the Turks worry about their own politics. Erve: But when there are issues in France, the Netherlands interferes with that too. Then we don't leave it in France either, right? Deniz: Well, maybe we should. . . . there are more important issues, which should be urgent, which concern us primarily, because we live in this country [the Netherlands]. Turkish self-identification differs across focus groups. Most saw racialization as Turks relevant to political self-identifications and concerns. The division over whether Turkish ethnic self-identification is a private or a political matter, contrasted with the a priori assumption in state categories that "Turkish migration background" politically matters to citizens. In the Surinamese-Dutch focus groups, participants often talk about themselves as Black or person of color Dutch citizens. Some individuals self-identify as Muslim, Christian, Hindustani, or Afro-Surinamese, but these did not become collective political self-identifications in the conversations. Participants collectively discuss anti-Black racism as a deeply rooted institutional problem stemming from Dutch colonialism. This becomes apparent in the following statement by Jennifer : The history of slavery . . . what bothers me is . . . to approach it from the idea that it's about the other person. No, it's always about you. So, when it comes to street names [of former colonizers], it's [removing them] not because it hurts other people. . . . No, keep it to yourself and say, we [white people] have been responsible . . . and we still have some cleaning up to do. POC/Black is the salient self-identified political group that formed the basis for common interests in the Surinamese-Dutch focus groups, illustrating that experiences of racialization matter to understanding their constitutions and politicizations of self-identified groups and interests rather than migration background. For most participants in the Turkish-and Moroccan-Dutch focus groups, self-identifying as Muslim trumps other political selfidentifications. Politically, they discuss "Muslim" as a religious, but primarily as a racialized self-identification. They consistently talk about anti-Muslim racism and threats to the freedom of religious practice, such as countering attacks on mosques, negative stereotypes, and access to religious education. Intersectional disadvantage becomes apparent in participants' constitutions of intersectional self-identifications and political priorities. For Muslimidentifying women, especially those in all-women focus groups, religious/racial and gender minorization intersect negatively. In the Turkish-Dutch all-women focus group, Muslim-identifying women discuss a recent law that partially forbids face covering. This illustrates how intersectional racial/religious and gendered selfidentifications, rather than Moroccan migration background state categorizations, matter to citizens' constitution of political priorities and assessments of representation. Self-identifying as Muslim women, they disagree with the partial ban on face covering. Young women participants across focus groups consider the political underrepresentation of ethnically/racially minoritized women-women of color, Muslim women, and Black women-problematic. Naomi says: If you grow up in the Netherlands and you see only white people in politics, in positions of power, it gives a kind of feeling like you're a guest in the country while you are born there . . . children [should] see, you have it in you to be in politics and lead this country. That it's not reserved for a certain kind of person. Naomi describes underrepresentation in relation to intersections of ethnicity/ race, gender, and age. Age/generational political issues matter in other respects too. Young participants mention the climate crisis as an urgent issue. Age/ generational concerns intersected with migration history and social class when participants discuss the lack of government communication in non-Dutch languages. The relationship between state categories and how citizens themselves constitute and politicize their self-identifications and political interests is fuzzy. --- Political Representation and Intersectionality Generally, participants experience state categories as commonly divisive, politically self-identifying instead as 'disadvantaged minorities' that face economic inequalities and racism. To most, racialization as Turkish, Muslim, or POC/ Black is more politically relevant than migration or ethnicity. This was more complex in the Turkish-Dutch focus groups, where Turkish ethnicity was political to some, irrelevant for others, and intersected with racial/religious political self-identification for the rest. The emphasis on racialization differs from common European research practice of referring to ethnicity or migration. In all cases, the political self-identifications that participants themselves constituted were key to understanding their political concerns. --- Citizens' Assessments of Representation Focus group participants relatively agree over what representation means. The political interests that politicians and parties constitute should match their self-identified political interests. Then, politicians or parties should act on these self-identified political interests, concerns, or values, by making policy, or at least by speaking out on their behalf. We interpret this as substantive representation. Participants also discuss politicians with whom they share lived experiences stemming from intersecting race, class, gender, and other social positions in society: descriptive representation. Finally, participants discuss whether descriptive politicians make their self-identified political group "stand for" something positive. We interpret this as resembling symbolic representation. Participants thus discuss representation as a me ´lange , but as we will highlight, prioritize some aspects of representation over others. In the following, we adopt participants' understandings of constitutive, substantive, descriptive, and symbolic representation when referring to these concepts. --- Parliamentary Actors Most participants discuss three parliamentary actors as representing them somewhat: established progressive parties, ethnic/racial minority parties, and in both, descriptive representatives. We discuss how they evaluate these actors in turn and consider dilemma's participants experience between them. Most participants discuss established progressive parties such as GreenLeft , Labor , and Democrats 66 as substantively representing some important political concerns such as social class, education, and climate crisis. But most participants find established progressive parties inadequate in constituting, let alone substantively representing, their other concerns: anti-racism and religious rights. The criterion that participants apply to evaluate descriptive politicians is mainly substantive representation. In theory, many hold that descriptive politicians' experiential knowledge makes them more likely to substantively represent them. However, opinions vary. Turkish-Dutch-identifying participants are more doubtful that descriptive politicians share their views due to the ideological diversity of their self-identified political group. Participants across focus groups regard descriptive MPs in established progressive parties as unable to substantively represent them, because they must toe the white party line: "If you as a Black man, Black woman, join parties like D66, GreenLeft, CDA, then you are polished [mat geslepen]. . . . They [descriptive MPs] are not at all concerned with their origins. . . . They try to be as Dutch as possible" . Although most participants doubt whether descriptive politicians in established progressive parties substantively represent them, many still see them as symbolic representatives. In the following conversation in a Moroccan-Dutch focus group, participants discuss former Labor Party member and prominent house speaker Khadija Arib, a Muslim woman with a Moroccan background. Participants discuss Arib as an example of an assimilated descriptive politician who does not represent them substantively by speaking or acting on their behalf. Karima sees this as exemplary of the first generation. Participants still consider Arib a symbolic representative: a role model. Nonetheless, most focus group participants prioritize substantive over symbolic representation in their assessments. Overall, to most, established progressive parties thus score high on substantive representation of some concerns , but low on others . Many participants see these parties as constraining descriptive MPs, providing limited opportunities to substantively represent them. Although of secondary concern to most, participants nevertheless value them as symbolic representatives. Participants discuss ethnic/racial minority parties as claim makers on their behalf: "When I think of diversity and someone who really fights for the Political Representation and Intersectionality interests of minorities, I immediately think of BIJ1" . Similar to the discussion about Arib, participants note how the second generation claims a position as equal Dutch citizens with distinct roots. Some participants place the emergence of ethnic/ racial minority parties in this generational trend: "now you have another generation participating" . Compared with established progressive parties, some participants are less satisfied with how ethnic/racial minority parties substantively represent economic, student, and climate issues. However, many value that these parties address-constitutively represent-other important concerns such as antiracism, colonialism, and religious rights that established parties ignore or neglect. Despite their small size, most participants further argue that ethnic/racial minority parties substantively represent these issues, both directly and indirectly. Ethnic/racial minority parties at least speak out on their marginalized concerns. Layla indicates feeling represented by minority party NIDA for this reason. Political representation doesn't immediately have to lead to action . . . in society, sometimes a dissenting voice is enough. The moment you only hear voices that don't represent your opinion, then it's already very nice to have one who presents a counter message, your voice. Many participants see ethnic/racial minority parties as advocating for them on critical matters that established politics neglects, which directly influences the political debate. In addition to direct substantive representation by speaking out on their behalf, some perceive ethnic/racial minority parties' advocacy as having contagion effects on other parties, indirectly providing steppingstones to policy-making. Benny remarks: "I always say that if Sylvana [Simons, BIJ1 party leader] wasn't there, there wouldn't be so much change either." Brigitte comments: "That's quite complicated . . . to get done with one seat in parliament." Brigitte adds that BIJ1 needs to work together with "larger parties to promote inclusiveness . . . to contribute certain themes." Participants in other focus groups similarly express that ethnic/racial minority parties' advocacy may lead established parties to constitutively and substantively represent these issues. Benny applauds Simons for daring to stand up for marginalized issues. This reflects how most participants assess descriptive MPs within ethnic/racial minority parties more positively than those in established parties. Many consider them brave and outspoken, and as having more opportunities to substantively represent minoritized citizens. Participants thus assess ethnic/racial minority parties as constituting and substantively representing marginalized issues that established parties neglect. Most discuss descriptive MPs in these parties as more capable of speaking out on their behalf, next to secondarily seeing them as symbolic role models. Up until this point, we described participants' assessments as quite clear and straightforward. However, this does not capture moments of hesitation, doubt, and trade-offs in participants' assessments towards parliamentary actors in the focus groups. In the following, we highlight ambivalence stemming from two main dilemmas of representation participants discuss. Many participants positively assess substantive representation by established progressive parties on some issues , and by ethnic/ racial minority parties on other issues . Hence, a first dilemma participants across focus groups discuss is having to choose between partial substantive representation by established parties on some issues versus by ethnic/racial minority parties on other issues. Many participants describe their intersectional political self-identifications and concerns as "split" between parties. and Najar nod]. When it comes to religious freedom, then specific representation is very important. So, I'm thinking a lot about this, especially now that [ethnic/racial minority party] NIDA is emerging, I'm thinking, gosh, why should a party represent itself so specifically, you know, and is that a good thing? But such a broad people's party as once the PvdA [Labor Party] is not quite right either [laughs]. So that's kind of my thought-struggle right now. . . . One moment you are a Muslim woman and the next you are a nurse who cares for the disabled. Many participants state that they feel like they should support ethnic/racial minority parties because progressive parties fail to address racism or religious freedoms. At the same time, some state that ethnic/racial minority parties' anti-racist advocacy comes at the expense of substantive representation related to other political self-identifications and concerns. Emir states: "I think that some economic measures should be taken regarding inequality. But I can't vote SP [Socialist Party] because, I see them as a bit xenophobic." Emir concludes that the SP and Labor Party "just seem to have completely given up in people with my profile." Others are more positive about established parties or prioritize climate or social class issues over anti-racism. Participants' ambivalence stems from a dilemma between partial substantive representation by either progressive parties or by ethnic/racial minority parties, reflecting a context of anti-minority hostility. I think it [representation] is a bit [political] issue-dependent [Fatima Most participants prioritize substantive representation. But young Muslim, POC, and/or Black-identifying women, whose intersectional raced/gendered political self-identifications are most underrepresented in politics, are an exception. For them, symbolic representation by descriptive political role Political Representation and Intersectionality models weighs heavily, next to substantive representation. A second dilemma that these specific participants in all-women and mixed-gender focus groups consider is whether they should vote for a party that best represents their interests , or one that offers a role model . Sandra and Naomi, two young women, and Brigitte, a more senior woman, discuss: Sandra: I think so [that descriptive representatives matter] now because there are so few. Although I do think that representation is a bit more important than the [political] views but generally many people who look like me in politics have similar views, so I don't mind a few of my views being put aside for representation. Naomi: . . . Indeed, I think that presence alone does something, just to show it is possible that someone of color reaches a high-ranking position. But at a certain stage, if enough people of color get into that position, I wouldn't necessarily be happy that that person with those [different political] views is there like I am now. Brigitte: Yes, but that's also very important to keep looking at what the [politicians' political] position is. Participants state that having politicians who substantively represent them matter most. Simultaneously, Sandra and Naomi value symbolic representation by descriptive politicians. They reconcile ambivalence by stating that descriptive representatives often share their views, and that substantive representation can be reprioritized once there are enough descriptive politicians. Young Muslim women-identifying focus group participants similarly hesitate between voting for the first Dutch candidate wearing a headscarf ever-Greenleft candidate Kauthar Bouchallikht-despite disagreeing with the Greenleft's, in their view, at times assimilationist politics, constraining descriptive MPs. Sara : "Do I want to vote for her [Bouchallikht] because she resembles me and represents me even though I don't completely agree with the party? . . . I still struggle with that." Most ethnically/racially minoritized citizens prioritize substantive representation in assessing parliamentary politics. Although many see established left/ progressive parties as substantively representing some issues , ethnic/racial minority parties substantively represent other neglected issues . Participants mainly judge descriptive representatives by their capacity to substantively represent them. Many see descriptive MPs in established parties as constrained, and those in ethnic/racial minority parties as more likely to act. Participants' ambivalence in the discussions reveals complex trade-offs within a context of structural racism. An intersectional perspective exposes the specific importance of symbolic representation for young women of color, whose intersectional raced/gendered political self-identifications are nearly invisible in politics. --- Extra-Parliamentary Representation In addition to politicians and political parties, participants speak of several extra-parliamentary representatives, BLM being the most extensively discussed. We explore how participants assess representation by BLM and how they relate experiences of representation by the movement to parliamentary politics. Participants assess representation by BLM in similar ways as they do ethnic/racial minority parties, namely constituting and substantively representing concerns such as racism and colonialism that are neglected by the mainstream. Romana comments that: "There's a real awareness . . . Black Lives Matter, for example . . . that the Netherlands . . . colonized a lot of countries. And that simmers through in society." Like Romana, participants note that racism and colonialism were publicly perceived as belonging to the past, and therefore, made invisible. BLM constitutes and advocates for these concerns, and by doing so, shifts the political agenda and makes the white majority attentive to their continuing salience. Participants note how constitutive and substantive representation by BLM has a direct, positive effect on their everyday lives. For instance, by reducing racist imagery and instances of everyday racism: "I've talked to people who have changed their minds very much about Zwarte Piet [a blackface caricature that plays a prominent role in annual St Nicholas celebrations in the Netherlands] . . . they just didn't know" . Like ethnic/racial minority parties, participants discuss how BLM's constitutive and direct substantive representation furthermore indirectly triggers substantive representation by parliamentary actors through contagion effects. Ishaan: Look we're talking about Black Lives Matter . . . that whole spectrum of engaging in protest has broadened. And so, you can call attention to the things that you consider important. . . . Rachelle: Yes, I also think you shouldn't forget that politics is dragging behind. Ishaan: Yes. Rachelle: So, where politics used to be leading, now movements in society are leading. And on that basis, politics starts acting, speaking, shouting, jumping, you name it. I enjoy observing that the spectrum is changing. Because we see that the power isn't, like you learn, with Political Representation and Intersectionality politics [Samantha nods], no, the power lies with us. Participants discuss BLM's constituted interests and claims as close to their own. The movement can push parliamentary politics to substantively act for them. To paraphrase Weldon , to turn protest into policy. Some, however, emphasize that parliamentary politics still needs to respond to movements' demands: "So, these movements and then pushing people to the front to use this space in parliament'" . Most participants in the Surinamese-Dutch focus groups discuss BLM as an obvious and positive representative claim maker on their behalf. However, some argue that the movements' framing of injustice was polarizing. One Surinamese-Dutch participant did not feel represented by BLM for this reason. She introduced this view as "I am going to say something shocking," reflecting how BLM's claim is accepted by most. Most, however, agree that to succeed BLM's message needs to be clear and inclusive. Like ethnic/racial minority parties, Black/POC-identifying participants discuss BLM as engaging a younger generation that experiences insufficient representation by parliamentary politics. Rachelle says: "It often takes movements from the outside [of politics] to mobilize . . . young people from a different cultural background than the native Dutch background for what really matters." Experiencing representation by BLM applied across focus groups for most who politically self-identified as Black, and as a disadvantaged minority, thus highlighting the importance of studying political self-identification. However, there are differences within and between focus groups in how participants discuss BLM. Participants in the Turkish/Moroccan-Dutch focus groups who politically self-identify as a minority explain more to each other and to moderators why they experience representation by BLM. This suggests that they do not see themselves as self-evidently part of BLM's representative claim. Nevertheless, minority-identifying participants in the Turkish/Moroccan-Dutch focus groups note how racism and discrimination are shared concerns of Dutch minoritized citizens, who equally benefit from BLMs' constitutive and substantive representation. Ays ¸e narrates how the movement positively impacts her everyday life: I think that since this movement has come to the fore, that, also with my [white] friends . . . I can talk about it [racist microaggressions] more easily and they understand me better, because now they are aware of it. . . . They can empathize more, because I can say, the comments that used to be normal are racist and discriminatory towards a group. And it's only now that that's becoming clear to some people. They're really getting a wake-up call. In Ays ¸e's story, BLM represents her by constituting racist microaggressions as an issue, triggering positive everyday changes. Her comment further shows how politically self-identifying as a disadvantaged minority, and perception of shared interest on that basis, underlies her experience of representation by BLM. All-women focus groups discussed BLM extensively. In the Turkish/ Moroccan-Dutch all-women focus groups, participants relate BLM to the racism experienced by Muslim women wearing a headscarf. Nonetheless, participants sometimes hesitate whether BLM's representative claim should include them. This is reflected in the following conversation in a Moroccan-Dutch allwomen focus group. Participants initially see themselves as represented by BLM, because the movement constitutes and substantively represents issues such as discrimination that concern all those who politically self-identify as a minority. However, participants are ambivalent, as they indicate not being sure whether they should experience representation by BLM, which may come at the expense of addressing anti-Black racism and colonial continuities more specifically. Similarly, Ishaan, a Surinamese-Dutch participant who self-identified as Hindustani, comments: "Black Lives Matter for me very much concerns slavery history and Afro-Caribbeans . . . maybe I should not involve myself but support it and give others the space to talk." Ambivalence arises from how racism differently positions participants, creating hierarchies of privilege/disadvantage within the self-identified political group of disadvantaged minorities. How participants who politically self-identify as disadvantaged minority differently assess representation by BLM reflects their self-identified political group history and their perceptions of relative privilege and disadvantage. --- Political Representation and Intersectionality In other cases, participants do not feel represented by BLM either because they do not politically self-identify as disadvantaged minorities, or because they do not see their own interests as shared with minoritized citizens generally. In the Turkish-Dutch all-women focus group, Hatice notes that she does not feel represented by BLM, explaining that she notices racism in her social environment, but does not experience it on a daily basis. The others try to convince her that she should care about ethnic/racial inequalities. Sarah: "I understand Hatice's dilemma [tweestrijd]. . . . But I think you have to transcend that for the greater good [Hatice nods]." Arzu agrees: "I have to make myself heard because [BLM] is a minority and I am also a minority. . . . We have to support each other." A lack of common political self-identification or shared political concerns also arose in the Turkish-Dutch groups with regards to participation in BLM and Kick Out Zwarte Piet, an activist group against the racist blackface caricature. Senna: What's funny about Kick Out Zwarte Piet, I didn't have anyone to go to the demonstrations with, so I went with my father . . . and he literally spent half an hour preaching to me, how come there are no Turks here, and Moroccans, and this and that. . . . because Kick Out Zwarte Piet and the Black Lives Matter demonstrations now stand as a sign for all anti-racism, anti-discrimination. Why aren't we active in that? Some participants thought that Turkish-Dutch are more concerned with Turkish politics, while others point to racism or lack of shared political selfidentification between minoritized citizens as a barrier. Some Turkish-Dutchidentifying participants thus hold that they, or their self-identified political group, should experience representation by BLM. But for various structural reasons, including relative privilege, lack of shared political self-identification or interests, they do not experience being part of BLM, leading to ambivalent orientations. Most Black/POC-identifying participants argue that like ethnic/racial minority parties, activist groups address, and advocate for, concerns that parliamentary politics neglects: constitutive and substantive representation. Pushing electoral politics into neglected spaces also increases the potential for substantive representation by parliamentary politics. Participants' intersectional political self-identifications are important to understand assessments of representation by BLM. Participants who experience representation by BLM were not only those who politically self-identify as Black, but also as a disadvantaged minority. An intersectional analysis, focused on moments of ambivalence, contributes to understanding how participants assess representation differently by laying bare distinct self-identified group histories and relative privileges and disadvantages in racist structures among participants who politically self-identify as disadvantaged minorities. --- Conclusion Representation is about citizens. Yet, in empirical work on representation, the citizen's perspective is marginal. The perspectives of ethnically/racially minoritized citizens are often even absent, their political identities and interests a priori assumed, or only exist as socially constructed claims by political actors. Bringing citizens back in, we find that the relationship between state categories and ethnically/racially minoritized citizens' political selfidentifications is fuzzy. Only in the Turkish-Dutch focus groups, Turkish ethnicity was politically relevant to some participants, triggering discussions between participants. Other participants mostly highlight socioeconomic disadvantage and racialization as a disadvantaged minority, Turkish, Muslim, PoC and Black in the constitution of their self-identified political groups and interests. The prominence of racialization in participants' conversations contrasts with much European political science, where state categories based on migration background or ethnicity are default descriptors. An intersectional perspective exposes overlapping religious, gender, age, and social class selfidentified political groups and interests within and between 'migration background' state categories, as well as specific intersectional political selfidentifications and interests, like the freedom to wear a headscarf among Muslim-identifying women. In their evaluations, participants consider elements that roughly approximate constitutive, substantive, descriptive and symbolic representation. Citizens discuss representation as a "me ´lange" of different aspects. However, most prioritize substantive representation in this me ´lange. Participants experience that established progressive parties represent them partially: they substantively represent their socioeconomic or climate concerns. But these parties sometimes fail to represent their other concerns: anti-racism and religious rights. Most participants hold that ethnic/racial minority parties fill this gap by putting these issues on the political agenda. Outside parliament, BLM similarly constitutes and substantively represents anti-racism and the continuing impact of Dutch colonialism. Participants sometimes discuss challenger parties and activism as part of a generational trend, in which the second generation claims a position as distinctive, but equal, Dutch citizens. Some hope that ethnic/racial minority parties and BLM's advocacy will lead established parties to act on these issues through contagion effects. Ethnically/racially minoritized citizens' substantive representation should therefore be regarded as multisited. Participants primarily judge descriptive representatives by their likelihood to substantively represent their interests, again highlighting the importance of substantive representation in participants' assessments. Many consider descriptive MPs within established progressive parties as constrained, while those in ethnic/racial minority parties were felt to have more agency. Although secondarily to substantive representation for most, participants Political Representation and Intersectionality nonetheless value descriptive representatives in both established and minority parties as role model symbolic representatives. Some participants clearly favor representation by a particular party, and most prioritize substantive representation. Yet, many participants experience dilemmas of representation stemming from ambivalence: a comingling of opposite emotions or orientations . In relation to parliamentary actors, participants experience dilemmas between substantive representation on some issues by established progressive parties, and by ethnic/racial minority parties on other issues. Moreover, despite the strong focus on substantive representation among most participants, young women of color, whose intersectional raced/ gendered political self-identifications are nearly invisible in politics, specifically hesitate between substantive representation or symbolic representation by a descriptive role model. This echoes Mansbridge's argument that descriptive representatives create a social meaning of "ability to rule," which is crucial for historically subordinated groups. Finally, participants hesitate about experiencing representation by BLM due to differing self-identified groups, histories, interests, and privilege/disadvantage among those who politically self-identify as disadvantaged minorities. We show that how ethnically/racially minoritized citizens themselves constitute political self-identifications and interests is central to comprehending assessments of political representation. The state categories that researchers and policy-makers frequently use are analytically weak, not least because they deny the subjectivity of European racialized citizens to constitute their own political self-identifications. Predefined state categories are often misaligned with citizens' political self-identification and prioritize ethnicity/migration over race, othering, and citizens' multiple, intersecting political self-identifications. An intersectional perspective is indispensable in analyzing how citizens "between" and "within'" state categories assess representation differently. It reveals significantly greater nuance than single-axis approaches in variation in preferences of representation, and in evaluations of political actors. Our bottom-up approach provides insights into how citizens assess representation as a "me ´lange" . We contribute to this approach by delving into the relationships and relative importance of constitutive, descriptive, substantive, symbolic dimensions of representation as well as parliamentary claim makers, to self-identified political groups of citizens. We show that citizens' assessments are often not straightforward. Citizens carefully consider the options for representation available to them and discuss dilemmas of representation. Focusing on such moments of ambivalence towards representation and parliamentary actors reveals "the broader sociopolitical context which sets the terms of feeling in this moment" . Experiencing representation only partially, further highlights the need for citizens' assessments of representation outside of claim-making frameworks to be more rigorously explored . Only when minoritized citizens' views are taken seriously can we get to the bottom of the meaning of equal representation. Our study has several limitations that we hope future studies will pick up. First, our focus on political self-identification cannot simply replace state categories in measuring how citizens prefer to be represented in politics. Political self-identifications such as "foreigners" can be used by insiders or group members. But it seems unlikely that ethnically/racially minoritized citizens want to be constituted as such by others, such as researchers or majoritized politicians . Future research should consider asking citizens how they prefer their self-identified political group to be constituted by various political actors. Second, we focus on ethnically/racially minoritized citizens' prevailing assessments of representative claim makers. Yet, claims can change depending on context and audience , and so could citizens' assessments, which may be further explored. Third, citizens' views on descriptive representation could be further investigated by intersectionally studying when citizens perceive MPs as "descriptive" and when and how shared, lived experiences with MPs matter to citizens' assessments of representation. --- Supplementary Data Supplementary data can be found at https://[email protected]. --- Conflict of interest statement No potential conflict of interest was reported by the authors. --- Political Representation and Intersectionality Notes 1. Considering the fuzzy distinction between the ways ethnicity and race socially construct difference, we use the two concepts interchangeably . We use "minoritized" instead of "minority" to emphasize that citizens are actively constructed as a minority by the dominant group . We use "ethnicity" only in cases where participants refer to group belonging based on shared history or culture. 2. The Dutch Bureau of Statistics categorizes citizens with a migration background if they, or their parent, are born abroad. At the time of our study, the CBS distinguishes between "western" and "nonwestern" migration backgrounds, based on the socioeconomic position of the country of birth

How do ethnically/racially minoritized citizens feel represented by increasingly diverse parliaments? We approach this question intersectionally and study how ethnically/racially minoritized citizens (i) constitute and politicize self-identifications and interests, (ii) assess political representation, and (iii) discuss who represents them. We draw on twelve focus groups with Turkish, Moroccan, and Surinamese-Dutch citizens (N ¼ 65), and find that citizens' political self-identifications, rather than predefined group labels, are key to understanding assessments of representation. Citizens prefer politicians who act on their substantive concerns but feel that mainstream parties sometimes fail to do so. Parties led by ethnically/racially minoritized politicians and social movements fill this void by contesting the status quo. An intersectional perspective reveals that symbolic representation by descriptive representatives specifically matters for young women of color who lack role models. The Black Lives Matter (BLM) protests in the United States triggered demonstrations in cities across the Netherlands, with protesters expressing solidarity and declaring that racism is a Dutch problem too. Politicians, activists, and political parties increasingly address the lack of ethnic and racial diversity in politics. This heightened attention raises new questions about how ethnically/racially minoritized citizens 1 make sense of, and assess, political representation. Who, in their view, represents them and why? In unpacking this question, we provisionally identified minoritized citizens based on so-called "state categories" (Monk 2022)-predefined group labels often used in studies of representation and derived from government statistics and policies-of having a "migration background." 2 We then asked minoritized citizens which political self-identifications and concerns they believed that politics should address. We aim to reveal how political self-identifications and concerns intersect within and between "migration background" state categories depending on gender, social class, religion, and age (cf. McCall 2005).

Definitions and Trends What is Poverty? In many ways, the definition of poverty that is used to benchmark American families' economic struggles is unambiguous: A specific dollar value for yearly income is used as a threshold for families of a given size in a given year, and if family income falls below that bright line, the family is considered to be "in poverty." However, when social scientists step back to consider the root causes of poverty, its impacts on development, and preferred program and policy interventions to tackle poverty, it is clear that poverty can be defined in several other ways . For example, "relative" poverty illustrates how far a family's income is from the national median for families, while "subjective" poverty reflects individuals' perceptions and local economic conditions. The "family self sufficiency standard" serves as an additional measure, taking into account what it would take to meet family's basic needs without relying on external, government, or charitable support. Among these definitions and estimates, the official U.S. poverty line represents the lowest "bar" and therefore, the figures we noted previously represent lower bound estimates of poverty in this country. In this report, we primarily will use the U.S. official definition of poverty and child poverty , despite its limitations, because the vast amount of basic and applied research in the U.S. uses these official definitions. As mentioned earlier, children in families with incomes less than 100% of the poverty threshold are considered poor. In addition, children in families with incomes below 50% of the poverty threshold currently are defined as in deep poverty and those in families with incomes below 200% of the poverty threshold are considered "low income." --- Why a Focus on Children, Families and Poverty? Poverty serves as a major risk factor for non-optimal child development. Poverty in childhood, and especially deep poverty in early childhood, is associated with a very broad range of problems in physical-biological, cognitive-academic and social-emotional development and evidence grows that these problems persist into adulthood . After years of debate, it is now clear that low family income actually has a causal impact on non-optimal health and development. Important questions remain about the mediating processes through which poverty affects child health and development. But rapid progress in understanding these causal processes is being made, which contributes to improved scientific understanding of child development and offers new insights into important targets for program and policy interventions. Since poverty leads to substantially higher risk of non-optimal outcomes for both children and adults, citizens and policymakers have important reasons to be concerned. Poverty itself is a major contributor to many of the most serious social problems facing our nation, including widening health and academic achievement gaps between high and low income children , high rates of school failure and dropout , declining rates of college attendance and graduation , and a growing skills gap between what the U.S. economy requires from young workers to make a middle class income and what many young adults possess . Notably, while poverty disparities in academic achievement have grown over the last several decades, corresponding racial disparities have actually declined . Thus, while race and ethnicity are correlated with poverty in this country, we focus on poverty in this paper given the substantial differences in outcomes between poor and nonpoor children. In sum, according to the best scientific research and to policy analysts across a broad range of the political spectrum, reducing poverty and ameliorating its negative effects on human capital development are essential to returning the U.S. to its historical legacy of being the land of opportunity . --- And Why Now, in the Fall of 2012? Fifty years ago, Michael Harrington published The Other America , a wake-up call to the nation describing invisible pockets of poverty in a nation of plenty. Two years later, Lyndon Johnson declared his famous "War on Poverty." In this section, we briefly outline trends in child poverty over the last 50 years, describing the nation's historic policy responses to poverty . We then assess the impact of policy on poverty, describing why, despite considerable effort , we find ourselves back where we were 50 years ago, facing a daunting set of challenges. --- Trends in Child Poverty Figure 1 charts changes in the child poverty rate from 1960 to 2009. After a dramatic decline in child poverty over the 1960s , rates have risen and fallen with the business cycle ever since. Two periods of increasing poverty rates were followed by two periods of decreasing rates . Figure 1 also charts trends by race/ethnic group. While the trends follow generally the same pattern of rise and fall over time for White, Black and Latino children, their average rates vary dramatically. Over the last 35 years, rates for White children have varied between 10% and 15%, but rates for the Black and Hispanic children have varied from just above 25% to just below 50%. Different factors push poverty rates up or down. Factors that exert upward pressure on child poverty rates include low work rates, low wage rates, and adverse changes in family structure and trends in education . Women's work rates were relatively low in 1980 , increased steadily from 1980 to 2000, but have flattened out in the face of two recessions from 2000 to 2010. Single and never married mothers, whose children experience especially high poverty rates, increased their work effort quite dramatically from 1992 to 2000 in response to both welfare reform legislation and a booming economy. In contrast, men's work rates were relatively high in 1980 and did not change much until they began to decline from 2008 to 2010 in the face of the Great Recession. Especially hard hit in periods of economic downturn over the last three decades have been young Black men. Haskins describes these secular trends in U.S. work rates as "a mixed bag," with trends "toward less work by males, especially Black males, and more work by females, including low-income mothers" . Wage rates also have an important impact on poverty rates. According to calculations by the Economic Policy Institute using U.S. Census Bureau data from 1973 to 2009, wages for the 10TH percentile of the income distribution fell through the late 1980s and then returned to their 1973 levels. In contrast, wages for workers in the top half of the income distribution rose steadily from the early 1980s, increasing by 10% for the 50TH percentile group and 32% for the 90TH percentile group. Haskins summarizes the combined work and wage trends succinctly: "people with low skills and little schooling cannot escape poverty unless they work, but even if they do work and their only income is wages, they are likely to be poor" . Family structure also affects poverty rates. About 11% of children in married couple families are poor, but 44.3% of children in female-headed families are poor . Over the last 50 years, the proportion of families with children that are headed by females has nearly quadrupled, from 6.3% to 23.9% . The increased proportions of children being raised in female-headed households has exerted clear upward pressure on child poverty rates over the last 50 years. Parent education is another factor influencing child poverty rates. Recent census data indicates that in 2009, the median income of families headed by a college graduate was about $99,700 while that of the average high school dropout is about $31,100 . While college graduation rates have increased for all race/ethnic groups over the last 30 years, Hispanics and Blacks complete college at one-third to half the rate of Whites . Notably, on measures of college entry, persistence, and completion, the increases over time have been largely concentrated in the top end of the income distribution . In fact, the disparity is quite stark: with the top two quartiles increasing college entry rates by 22 percentage points and the bottom increasing by only 10 percentage points-resulting in a growing gap in college entry rates between the rich and poor. And, even among those who enter college, poverty continues to exert a toll on degree receipt . If declining work rates for men, stagnant wages for the low-skilled worker, increasing rates of children raised in female-headed families and growing gaps in educational attainment exert upward pressure on child poverty rates, a logical, if surprising question might be "Why have child poverty rates not increased more?" The preponderance of the evidence suggests that over and above temporary improvements caused by upturns in the economy, it is public policy that has exerted consistent downward pressure on child poverty rates. We turn next to summarizing the research that supports this judgment. The impact of anti-poverty programs and policies on U.S. poverty rates-A set of new studies have identified the policy steps that the U.S. has taken to reduce poverty and examined whether those steps have been effective . For example, Haskins identifies 82 specific means-tested federal programs that provide assistance to the poor, relying on longterm trend data for means-tested spending from the Congressional Research Service. He focuses on spending levels for 10 major means-tested programs, and estimates outlays of $621 Billion in 2010, with spending having increased six-fold from 1968 to 2004. In comparison, Ben-Shalom et al. include both means-tested programs such as Medicaid, the Supplemental Nutritional Assistance Program , the Earned Income Tax Credit and Temporary Assistance to Needy Families and also social insurance programs . Because Ben-Shalom et al. take the most comprehensive view on antipoverty programs and policies, we emphasize their findings here. For FY 2007, Ben-Shalom et al. estimate about $523 Billion in annual expenditure on nine major means-tested programs for low-income persons and families. In addition, while not targeted primarily on low-income persons or families, five social insurance programs accounted for another $1.105 Trillion in expenditure, much of which goes to poor and low-income individuals . While this demonstrates substantial support for the poor, it is notable that U.S. expenditure on social programs was only 75% of the OECD 30-country average in 2005 , leaving the U.S. behind most of our peer countries in the OECD . Regarding time trends, an initial look at Ben-Shalom et al.'s numbers would suggest that U.S. spending on the poor went up: They show that both means-tested and social insurance programs grew faster than did GDP before 1995 and at about the same rate of growth as GDP from 1995 to 2007 . But those data also highlight the relatively stable average growth in social spending in the U.S. since 1995 is a result of two very different trends: dramatic declines in TANF and very large increases in SNAP. And, their evidence suggests that there was significant redistribution of expenditures on the poor over the period from 1984 to 2004: relatively more for the elderly and disabled and relatively less for singleparent families and the non-employed. In summary, both in aggregate and per capita, the U.S. has very significantly increased expenditures on both means-tested programs and social insurance programs over the last four decades. And, this form of spending benefits lower-income families. As shown in Table 2, analyses of the Survey of Income and Program Participation data show that in 2004, without transfers, 29% of U.S. families would fall below the poverty line; but after transfers the poverty rate falls to 13.5%. In 2004, the U.S. system of means-tested and social insurance programs reduced poverty by 16 full percentage points . What can we conclude? First, these analyses highlight the ways that poverty is not a "natural state" dictated by the exigencies of labor policy and recession, but instead can be clearly remediated by anti-poverty policy. Second, we argue that, as successful as the U.S. system appears to be in cutting "pre-transfer" poverty rates by more than 50%, we have substantial room for further improving our nation's track record in tackling poverty. The current social-economic and policy-political context-Questions regarding the causes and consequences of poverty for children come at a critical juncture in American economic history. We are just beginning to emerge from the Great Recession of 2007, one of the worst economic periods of our time . Yet most of the research on child poverty trends and the effectiveness of antipoverty policies dates from before the Great Recession. What about now? As would be expected, in the wake of the recession, both unemployment rates and poverty rates increased. Low income families have substantial difficulty finding jobs in the current economic climate. And, since we have largely placed the social safety net on top of employment, tying benefits for poor families to their work effort , they are hit doubly hard by the recession. In short, not only have poor families lost ground on employment like their middle and upper income neighbors, but they have also lost their opportunity to take advantage of key benefits, like the Earned Income Tax Credit, that were so pivotal to move them out of poverty in recent times. The Great Recession could not have come at a worse time for America's poor children and families. Growing federal budget deficits and resulting increases in national debt have led to calls from both major political parties to cut the federal budget deficit . The need to cut the deficit in the long term clashed with the need for government to provide a significant short-term stimulus to the economy to get it going again and to bring down the unemployment rate. The Obama administration fought for and won passage of the American Recovery and Reinvestment Act . Many key provisions of the Act supported children, especially low-income children, by increasing public resources through approximately two dozen federal programs that benefit children . An estimated $153 Billion of the $787 Billion in ARRA, or approximately 20% of the total, was dedicated to children and children's programs . These temporary investments did reduce the unemployment rate and partially protected federal, state and local investments in low-income children as well. But the ARRA provisions are ending in 2012. How to get the economy going again and generate new jobs are the paramount issues in the Presidential and Congressional elections of 2012. Alarmingly, while there is hot debate about how to protect and revive America's middleclass families, there are many fewer debates or compelling new ideas about how to protect America's low-income children and families and help them move into the middle class1 . Finally, in addition to major national elections and huge decisions to make about the economy and federal budget deficits, a large number of pieces of federal legislation that directly impact low-income children and their families are coming up for re-authorization or face an uncertain future . In short, the U.S. is still in the throes of the biggest economic crisis since the Great Depression; poor families have been hit doubly hard-losing both employment and the social safety net tied to employment; the short-term policy actions that protected low-income children and families from the Great Recession have begun to expire; child poverty rates are up; the 2012 Presidential and Congressional election campaigns have not included much discussion about poverty; and major policies affecting poor children and families will be reconsidered in 2013. For all these reasons, it is a very compelling time to address issues of children, families and poverty. --- The Emerging Science of Poverty A long history of research has examined questions of whether, how, and for what outcomes income "matters" for children's well-being. The answer to this question is key to informing policy decision-making about whether and how to effectively target policy to improve the lives of low-income children. In this section, we discuss the state of the science on income effects. In doing so, we first highlight the way in which the last decade has "put to bed" some key questions about income effects, while also opening new questions about pathways that are key for further investigation. A wealth of research has been devoted to addressing the question about the magnitude of income effects and, most critically, whether or not the associations between income or poverty and outcomes for children are mere associations or represent causal relations . Fortunately, a number of studies have tried to address this question by leveraging planned experiments ; "natural" experiments ; policy changes outside of families' control , or, finally, by conducting rigorous analyses of nonexperimental data . The consensus from this work is that there are "modest" positive effects of income on multiple domains of children's development. These studies together show that income effects range from about .10-.20 of a standard deviation with $1,000 increase in family income for families at the low end of the income distribution. Using these estimates, a grant the size of the EITC could generate moderate to large increases in outcomes for children. What remains unclear, however, is the precise pathways by which income effects occur beyond the economic investment and family stress pathways that are typically invoked. New findings from related social-science disciplines, such as neuroscience, economics, and sociology have not been fully integrated into a conceptual frame for this area of research. In the next section, we begin to build this more nuanced conceptual frame in order to understand how to improve the trajectories of lowincome children. --- Common Theories of Income Effects The two most commonly considered complementary theories about how income affects outcomes for children are drawn primarily from the fields of economics and psychology, respectively. From economics , income is thought to affect children through the investments their parents can make in their developmental outcomes . Income enables parents to purchase both material resources for their children as well as nonmonetary goods . From developmental psychology , seminal studies conducted on families losing income in the Great Depression and during other periods of economic shocks like farm failures have highlighted the role of income in increasing parents' stress and impairing parents' child-rearing practices. This theoretical work on the stress effects of poverty has been extended to single-parent African-American families, showing how poverty may impinge on children's social-emotional development through disruptions in parents' psychosocial well-being and parenting . Importantly, parents as well as children can be affected physiologically by poverty, and these effects may explain some of the effects of poverty on parents' health. Both of these traditional theories have largely focused on the family unit and effects on behavior of parents and children, while simultaneously paying relatively less attention to both the internal processes and external factors beyond the family unit. Fortunately, advances in neuroscience and sociology/ecological science allow us to expand the conceptual frame to address these issues. The issues raised from these disciplines are largely not new, but they are only recently being integrated to an interdisciplinary view of understanding the effects of income and poverty on children. We focus on a few key findings below by way of example. --- Biological Processes in the Effects of Poverty: Advances in Neuroscience What is going on inside the bodies of low-income children that would make poverty disrupt their behavior and achievement? Recent work on the physiological impacts of stress suggests that the initial response may lie in the brain, particularly in the emotional systems of the brain, with far-reaching physiological consequences . Current thinking argues that the stress-system is characterized by "allostasis" rather than homeostasis , such that rather than return to baseline following the experience of a stressful event, the stress-sensitive system matches internal processes with external demands . In the case of poverty, this means that the child in a low-income family, faced with a stressor, will respond physiologically by adjusting parameters to a range of functioning that is appropriate to that stressor. The condition of poverty does not entail a single stressful event. And thus, the value of the allostatic model is that it can account for the physiological effects of repeated or chronic exposure to stressors on outcomes and address questions about how and for what outcomes poverty will have long-lasting consequences. Under repeated stressor exposure, which is more likely to occur in cash-strapped households and communities, the body "anticipates" the stressor by setting new set points in physiological systems . While preparing the body for the stressor in a number of ways , these new set points also have long-term physiological costs to physical and mental health outcomes. They do so through adverse changes to the cardiovascular system , the immune system , and/or the neuroendocrine and cortical systems , in ways that are "toxic" . Importantly, parents as well as children can be affected physiologically by poverty, and these effects may explain some of the effects of poverty on parents' health. Low-income adults have consistently been found to experience substantially deteriorating health, such as compromised immune functions and/or higher risk of cardiovascular disease, with increasing exposure to cumulative poverty-related stressors and at much earlier ages than their more advantaged counterparts . These health effects may be a consequence of higher levels of allostatic load, with concomitant disruptions in both sympathetic and parasympathetic nervous system response, as indicated by altered cortisol, norepinephrine, and DHEA-S release and uptake that would explain these health effects . Thus, adults who are caring for children under conditions of chronic and high economic stress are also likely to have health problems of their own. The implications of parent health for health care policy and the role of health care policy in ameliorating or reducing that burden are substantial. This new thinking helps to specify more precisely the outcomes on which poverty might impinge, while also highlighting potential avenues for intervention. Much of the child development research on outcomes for children in poverty has focused on effects on academic achievement and social-emotional outcomes, with less attention to health effects and underlying brain-related processes. This work would suggest that more attention be paid to both-the health consequences of poor economic conditions as well as implications of such conditions for processes such as executive function and regulation . With regard to intervention, these models also imply addressing the stress-response system , which is a critical addition to poverty-fighting policies. Considering the neurobiological consequences of poverty also highlights another key nuance in our thinking about income effects on children-that poverty may or may not be a static state. Interestingly, such income instability or volatility has been largely understudied in the poverty literature, especially with regard to its implications for children . Yet, income volatility may especially disrupt children's development by reducing the regularity of routines in their lives and impinging on their health. Policy features, such as income eligibility limits for benefits and short recertification periods may exacerbate the problem of income volatility among families, while the availability of short-term low-cost loans could facilitate the smoothing of family income across periods when resources are scarce. In this way, attention to the issue of income instability may help us better develop a system of policies that promotes the development of low-income children. --- Environmental Factors in Poverty Effects: Advances in Sociological/Ecological Science Children and families living in poverty have inferior and sometimes toxic housing conditions, and those environments likely also impinge on the outcomes of low-income children . For example, housing conditions of low-income families include higher exposure to substandard physical characteristics as well as higher density ; the implications of such conditions for family processes may be substantial . Household characteristics such as lack of safety, noise and crowding are associated with greater cognitive and neuroendocrine indicators of stress and lower levels of child adjustment . For example, a compelling study on the negative impact of noise shows how that the introduction of sound absorbent panels can improve preschoolers' letter-word-number recognition and language skills . Recent quasi-experimental and experimental studies of low-income families' housing mobility and their corresponding decreases in symptoms of anxiety and stress suggest that low housing quality may be an additional dimension of poverty-related risk to consider, even after taking income and psychological dimensions of poverty into account . The effects of poverty also extend to neighborhoods. Less-advantaged neighborhoods provide fewer enriching opportunities for children such as parks, libraries, and children's programs , while dangerous neighborhoods present physical and social hazards to children. A large body of research highlights how even living in close proximity to violence in the community can negatively affect children's social-emotional and cognitive development . Mechanisms for the negative impact of more violent and more socially disorganized neighborhoods on child outcomes may be both direct and indirect . The peer and parenting environments of low-income communities also likely differ from those in higher income communities . Peers exert strong influences on children's behavior for a number of reasons, resulting in children's higher rates of delinquency in higher risk neighborhoods. For example, children may imitate their peers' behaviors, emulating the behavior of those around them. But at the neighborhood level, other processes are at play, as stigma is reduced for delinquency when such behaviors occur more frequently, and the likelihood of getting caught is reduced because of congestion effects in law enforcement . And even beyond peers, adults in a neighborhood can influence young people who are not their children by acting as role models or, more importantly, by exerting social control on their own children's as well as their friends' behavior . These findings demonstrate that poverty can dictate the broader "contextual" state of the family rather than merely be a measure of the amount of resources within a given household. Thus, changing family income alone, without attending to the broader context of poverty, may not be sufficient to improve the outcomes of children. As we discuss in the next section, policies can be expanded to address issues in neighborhoods or communities, and doing so may be key to making a substantial difference in the lives of low-income children. --- Advances in Applied Intervention and Policy Research As previously proposed, the U.S. system of antipoverty programs and policies has served to reduce poverty. Nonetheless, the general and child poverty rates remain stubbornly high. In this section, we turn to research over the last decade that may provide insights about program and policy innovations and reforms that could directly reduce poverty, or protect and enhance the human capital development of poor children or parents. In this way, they map closely onto the investment and family stress theories of income effects. The goal of this is to selectively and strategically focus on a few initiatives that hold special promise and merit serious consideration. --- Poverty Reduction Models Particularly striking recent changes in U.S. antipoverty policy include a decline in cash assistance through the welfare system and an increase in cash assistance through the tax system. In this section, we focus on the latter in part due to the fact that the former is currently serving an increasingly smaller proportion of the low-income population . The largest form of cash assistance for low-income families is delivered via the Earned Income Tax Credit . Eligibility for EITC is contingent on work. A family earns 40¢ of a tax credit for each dollar earned from $1 to $10,000 dollars of earned income. None of the credit is lost until a family's earned income reaches beyond the poverty line. Then the credit is phased out at a slower rate until it reaches zero at about $40,000 of earned income per year. The tax credit is first used to offset tax liabilities. But if the value of the credit exceeds a family's tax liability, the remainder of the credit is refundable. Both the tax offset and the refundable features of the credit are critical to its antipoverty effect. Another form of cash assistance that is targeted on a broad range of families, not just poor families, with children under age 17 is the Child Tax Credit . Under current law, families can receive a CTC of $1,000 for each child. Like the EITC, the CTC is partially refundable and conditional on earnings, where some families may either earn too little , or earn too much to get the credit. Together, these refundable tax credits are estimated to lift 7.2 million Americans, including 4 million children out of poverty, which is more than any other program or category of program at any level of government . And, by tying support to work, they encourage rather than discourage work effort. Not only do refundable tax credits reduce child poverty; they also promote child development. Dahl and Lochner found that with each increase of $1,000 brought about by income tax credits, children's reading and math achievement increased by 0.06 SD. Importantly, these positive effects were somewhat larger for lower income families. In light of the higher level of political and public support for efforts to combat poverty via the tax system , tax policy is a particularly promising avenue to directly reduce poverty. A variety of changes could be made that would increase the antipoverty effects of the EITC and the CTC. As examples, Congress could raise the maximums and increase the phase-in rates of both credits; the CTC could become fully refundable. Of course, these changes would require additional federal expenditures. But decisions could be made to redistribute government expenditures to strengthen tax credits for poor families and their children, and to consider additional tax credits beyond those tied to employment alone. Even so, the question is whether such approaches are sufficient to substantially improve outcomes for poor children, whose outcomes fall so far behind those of middle-and high-income families. --- Human Capital Development Models A number of programs designed to advance the human capital of low-income children have been evaluated and show promise for such children, at least in the short term. These span the age range from early childhood to adolescence, targeting children primarily outside of the family-in the child care and education services where children spend a substantial component of their day. These interventions often target the quality of interactions between children and adults or peers-that is, those same interpersonal processes that are eroded by poverty-related stressors . Certainly, none of those programs can be said to completely make up for the deficits of growing up in poverty. With that caveat in mind, we discuss promising programs below, to illustrate progress made over the last several decades in identifying programmatic strategies that can improve the outcomes of low-income children. Preschool is a centerpiece of policy efforts intended to reduce the achievement gap of lowincome children, in part because of the early emergence of the achievement gap before the start of school and in part because of cost-benefit calculations that show that even very expensive preschool efforts nonetheless can be cost-effective . Comprehensive early childhood programs for children at risk have the potential to improve cognitive and academic outcomes and have extended benefits over time. In fact, a metaanalysis estimated that the average effect size on cognitive measures was half a standard deviation in size . The two most widely touted comprehensive programs -Perry Preschool and Abecedarian, have both been tested in small-scale randomized trials and show sizeable short and long-term benefits . Comprehensive early childhood programs for children at risk have the potential to improve cognitive and academic outcomes and have extended benefits over time. The more recent findings from the Head Start impact study, providing a rigorous test of the effects of Head Start across 84 nationally-representative delegate agencies, have shown more modest, but still positive, short-term effects on outcomes for children . But there also appears to be considerable fade out of effects as children advance to elementary school . Now, a "next generation" of preschool studies shifts from the question of whether preschool benefits poor children, to asking how best to provide sustained, high-quality opportunities for learning within multiple educational contexts. Recent cluster-randomized trials have demonstrated that effect sizes for the impact of preschool intervention on low-income children's cognitive and behavioral outcomes can be substantially larger when the quality of the classroom environment is targeted . From that perspective, child care, preschool, pre-kindergarten, and kindergarten programs are then best conceptualized not as a patchwork system of care, but as multiple early educational platforms that bridge home and formal schooling. It is to those two respective contexts of home and school that we now turn. For very young children, fewer promising interventions have emerged, in part because younger children spend a substantial amount of time with parents and it has proved to be more challenging to intervene with parents than with teachers. Several models capitalizing on early childhood settings and/or home visitation such as Early Head Start and Nurse Family Partnership have shown small positive effects on quality of parenting and school readiness for children in the infant and toddler years. Two newer approaches to changing parenting are also worth noting and show promise in changing child outcomes. First, some interventions rely on an interventionist who video-records the parent and child and uses review of the video with the parent to support positive parenting . Second, Family Check up is a homebased, family-centered intervention that utilizes an initial ecologically-focused assessment to promote motivation for parents to change child-rearing behaviors, with follow-up sessions on parenting and factors that compromise parenting quality. For school-age children, comprehensive school reform models have burgeoned over the last decade. These models target not only teachers' professional development and practice in individual classrooms, but strive to alter the entire school context by offering curriculum aligned across grade levels. Some of the most well-known of such programs, such as Success for All , have been examined in over 100 studies, with SFA demonstrating improvement in children's early reading skills across several RCT evaluations . In addition, other models for school-age children have also targeted children's social and emotional learning , recognizing that children's social experiences, emotion skills and regulation can support or impede their own and others' academic success in school . A recent meta-analysis finds that SEL programs successfully improve children's academic performance, with effects ranging from a fifth to a quarter of a standard deviation . Key to these initiatives has been the recognition that intervention-driven academic gains that are earned within a given grade or school year may not be sustained unless that intervention or school reform effort is extended across grades. That said, these models provide another promising strategy for intervention with school-age children. For the oldest children, there are a few programs with promise in reducing the intergeneration cycle of poverty. Evaluations of small schools of choice in the New York City system, for example, have shown benefits to children's graduation rates and prospects for college attendance . The feature that differentiates these schools is not only their size, but their emphasis on academics, personalized attention, and community relationships. Another example of a promising approach for this age group is a set of programs called "Career Academies" that connect older children to the world of work . These programs have substantial effects on the earnings of lowincome young men as they moved into young adulthood . While there are a number of approaches that can boast success, the human capital development models for all age groups of children could do better. The core challenge is that while they all attempt to "make up" for the challenges of growing up in poverty they do nothing to change the economic conditions of families or communities. The result is that children get some dose of nutritious interactions, but these effects are continually undermined by the stressors of growing up in poverty. As we discuss next, there are innovative models that attempt to do both in the form of "conditional cash transfers." --- A Model Designed to Reduce Poverty and to Promote Human Capital Development: Conditional Cash Transfer Programs Remarkably, while the research community has realized that the separation of the investment and family stress pathways is a false dichotomy, our policy efforts have largely proceeded on separate and parallel tracks. As the Haskins andBen-Shalom et al. analyses describe, the U.S. system of antipoverty programs and policies include major investments to reduce poverty in the short-term via cash and in-kind assistance. They also include major investments aimed at stimulating the human capital formation of poor children, youth and parents. But in the U.S., there are few policy efforts designed to address both of these goals simultaneously and synergistically: that is, the reduction of income poverty in the short-term and increases in investment in children's human capital development in the intermediate-and longer-term. However, in the southern hemisphere, one such approach known as holistic or comprehensive conditional cash transfers has been tried. Comprehensive CCTs have been developed and implemented over the last 15 years, first in Latin America and increasingly in Africa and Asia. CCTs aim to 1) reduce poverty in the near term by providing cash support to families and to 2) increase parent investment in their own self sufficiency and their children's human capital by making cash support contingent on certain behaviors that are consistent with such investments. In their review of the first wave of well-evaluated CCTs, Fiszbein and Schady concluded that CCTs can reduce poverty, increase the use of health, nutrition and education services and improve health outcomes, especially in early childhood. But there is less evidence that CCTs have an impact on children's learning and academic achievement despite their positive impacts on school enrollment and attendance. Comprehensive CCTs are one variant of the broader use of financial incentives to change behavior. Other types of financial incentives designed to change low-income people's outcomes also have been evaluated over the last two decades with the range of targeted domains, value, and resulting impacts of different packages of incentives varying widely. Conceptually, one would not expect domain-specific incentives of less value to have the same impact on behavior and outcomes as multi-domain, higher value incentive systems. Thus, comprehensive CCTs are expected to have more synergistic effects across several domains of behaviors and outcomes. Moreover, they resolve some of the challenges presented earlier in terms of tying all incentives to employment highlighted in section I-in times of high unemployment, CCTs that target education and health as well as employment can still support low-income parents' investments in children and help reduce poverty. The first effort to adapt the holistic CCT approach to combat poverty in a developed country was undertaken by New York City government and named "Opportunity NYC/Family Rewards" . Families were paid cash rewards of about $3K a year on condition that parents and children undertake a variety of activities to advance children's education, families' preventive health care and parents' employment . A rigorous random assignment evaluation of Family Rewards, undertaken by MDRC, is following 4,800 families and 11,000 children over the 3 years of participation of the program and 2 additional years after the CCTs are ended. Interim results after 2 years of the initiative are promising in some respects . Family Rewards reduced current poverty and material hardship; increased savings and banking; increased school attendance, course credits, grade advancement and scores on standardized tests ; increased families' health insurance coverage and receipt of medical and dental care . This is an impressive array of short-term outcomes, very much in keeping with the dual goal of CCTs: to reduce poverty in the short-term and to stimulate investments in human capital development in the long-term. But Family Rewards did not impact achievement test scores for most children; and while it increased employment in jobs not covered by the unemployment insurance system, it reduced employment in UI covered jobs. We have argued elsewhere that Family Rewards represents CCTs 1.0 as the first attempt to implement this kind of program in the United States . New York City government and the Obama administration concurred that there was sufficient promise in the strategy to revise the CCT policy based on interim results, and to conduct a new random assignment trial of CCTs under the Obama administration's Social Innovation Fund in New York City and in Memphis, TN. Why do we think that CCTs can do better than either poverty reduction or human capital efforts alone when the history of such programs is positive but not yet remarkable? What CCTs do really well at is reducing poverty, but they also encourage parents to take advantage of human capital development services. Yet, they have not yet generally been paired with "supply side" services to boost the quality of services, focusing instead on "demand side" adjustments . If we can do all three-help encourage parents to take up services, while reducing family poverty and ensuring the services children are receiving are high quality-we may finally make a substantial dent in the effects of poverty for the next generation. The recent work in the U.S. on CCTs is but one example of the many ways that the U.S. can adopt new, innovative approaches to combating poverty. Policy makers and prevention scientists may consider these interlocking components of poverty reduction , allostatic load and stress reduction , human capital promotion and behavioral change , and, finally, community-improvement, as multiple "working parts" of comprehensive intervention. This example could serve as a catalyst for other innovative models that strategically take a "bothand" rather than "either-or" approach to reducing material hardship and increasing the likelihood of children's positive health, educational, and behavioral outcomes over time. --- On a Way Forward In this final section, we reflect on implications for moving forward on the program and policy front and on the research front. --- Program and Policy Reflections First, there is no single magic bullet to address the child and family poverty problem in the U.S. Rather, a combination of cost-effective and publicly supportable strategies is needed to reduce child poverty and to promote human capital development for poor children. But our both/and approaches in the past have either not been enough to reduce child poverty or have been poorly coordinated, failing to protect poor children from the worst effects. Second, to more effectively reduce child poverty and to enhance children's health and development will require a greater proportion of U.S. public expenditures being devoted to these efforts and creative redesign of poverty reduction and human capital development initiatives to achieve greater synergies across our investments. Alarmingly, antipoverty programs and policies remain in silos that are insufficiently coordinated to obtain optimum impact for our investments. Third, the costs of doing nothing more and/or differently in poverty reduction and human capital investment are high. Poverty's impact on children's educational achievements and health result in lower productivity of the nation's economy and higher health care costs. Indeed, cost-benefit analyses of many antipoverty programs suggest they more than pay for themselves . Finally, programs that offer greatest promise may be those that consider ways to remedy both the material and the psychosocial conditions faced by families in poverty. Our broad review indicates that the rate of poverty among our nation's children and their families is not immutable. A productive policy strategy would be for policy makers, leaders, and community members at the state and federal level to set a target for how much of a reduction in the rate of U.S. poverty among children that we hope to accomplish in the coming 5 to 10 years . Without a transparent commitment to an ambitious, yet achievable target for anti-poverty policy, it will be difficult to build a roadmap of steps that we might take, as a nation. --- Scientific/Research Implications The emerging science suggests several issues that deserve high priority attention from the broadly interdisciplinary child development research community. First, perhaps the most important and exciting area of research is in the search for the processes that mediate the impact of poverty on children's health and development. As the history of scientific research on income effects shows, there were great scientific challenges to establishing the causal influences of poverty on children's health and development. There are even greater challenges to demonstrating that a complex process like allostatic load is a causal link to low and unpredictable family incomes and poor health and development . But these challenges can be met by joining the power of basic scientific research with rigorous experimental evaluations of antipoverty programs and policies. Merging prevention science and developmental science can help program developers and policymakers identify new ways to support optimal outcomes among families facing an increasing level of economic pressure, in increasingly tough times with expanding levels of inequality. As but one example, the recent research on neuroendocrine processes of povertyrelated "wear and tear" and increased allostatic load in adults and children exposed to chronic deprivation suggests that much of what we will learn in the next 5 to 10 years will have high levels of significance for their health as well as educational outcomes. This means that increased attention to prevention programs to support positive outcomes among poor families may offer major benefits to not only increased human capital but lowered health care costs for our nation over time. Finally, programs that offer greatest promise may be those that consider ways to remedy both the material and the psychosocial conditions faced by families in poverty. Building integrated platforms of service delivery that target poverty reduction and health and human capital promotion is not a small task. That said, such integrated models of family behavioral change and material support may yield benefits in unanticipated ways. In sum, the prospects and problems for children in poverty are daunting. There is now an opportunity to deploy the tremendous empirical and policy tools at our disposal to make major improvements in the lives of poor children in the next decade. We can use these tools to accurately identify points of inflection in the life course and economic trajectories of families and children. It is our task as scientists and citizens to maximize the ways that children's and families' economic and developmental trajectories can be set on a positive course, in the years ahead. ■ ---

Now, more than ever, it is crucial to address the topic of children and poverty in the U.S., given current scientific knowledge about poverty's influence on children and effective strategies to mitigate its negative impact. In this report, we summarize the best available information on definitions and trends in child poverty, policy responses to child poverty and the impact of poverty on children's health and development. Research suggests that various factors exert upward and downward pressure on child poverty rates. Upward pressure is exerted by declining work rates for men, stagnant wages for low-wage workers, increasing rates of children raised in female-headed households, and growing gaps in educational attainment. Downward pressure is exerted by the U.S. system of antipoverty policies and programs, which appears to be cutting "pre-transfer" poverty rates by more than 50%. Nonetheless, child poverty rates in the United States are high by both historical and international comparison. We then review the emerging science on biological and ecological processes by which poverty affects child development and key findings regarding the efficacy of comprehensive strategies to reduce poverty and to promote the human capital development of poor children. In the final section, we reflect on implications for moving forward in science and policy. In 2010, in the shadows of the Great Recession, 46.2 million Americans lived in "poverty" using the federal government's official measure of poverty (National Poverty Center, 2012). Fully 15.5 million of poor Americans are children under the age of 18 (Addy & Wight, 2012). Thus 15.1% of the country's population and 21% of all its children are poor. Over the last 40 years, the national poverty rate has fluctuated from a low of 11.1% in 1973 to a high of 15.2% in 1983 (National Poverty Center, 2012). Such numbers, percents and time trends are the basic facts of poverty in America that the media, politicians, policymakers and the concerned public typically track. In this Social Policy Report, we examine definitions of poverty, the numbers and time trends that derive from the definitions, and what they mean for the nation's children and families. First, we ask "what is poverty?" and "why is it important at this time to address the topic of children, families and poverty in the U.S.?" Second, we review some of the emerging basic science on the influence of poverty on children and families to better understand its grave and complex consequences. Third, we turn to applied (intervention and policy) science for insights about effective strategies to address the multiple challenges that children and families in poverty face. Finally, on the basis of these reviews of concepts, definitions, and basic and applied research, we conclude with critical reflections on a way forward.

Introduction The use of social media has grown exponentially over the last two decades, particularly among young adults [1,2]. These platforms have become an integral part of the lives of people all around the world, with more than half of the world now using these platforms [3]. Usage rates are high in many of the advanced economies including twothirds or more of all adults in the United States, Australia, South Korea, Canada, Israel, and Sweden using social media platforms [4]. Developing countries are quickly catching up, with Middle Eastern and North African countries reporting a median of 68% adult population using social media platforms [5]. Communication across these platforms range from private messages to group messages to public posts, from sending texts to voice calls to video calls, with almost all social media sites providing multiple modes for communication. Most social media platforms state that their aim is to connect people with their families and friends [6]. Arguably, the increase in social media-driven communication has been influencing interpersonal relationships such as parent-child relations [7]. There is emerging evidence on how parentchild relationships are evolving in the ever-changing social media landscape [7]. Findings from various studies highlight how unique social opportunities afforded by social media , cultural norms , and increasing diversity of communication options impact various ways in which young adults choose to communicate with their parents using social media platforms. Most studies have focused on parent-adolescent child relationships, and only a few have examined how these relationships differ across gender in context of social media use [8,9]. Sampasa-Kanyinga et al. [10], in their survey of 9,732 adolescents in Ontario, Canada, found that heavy social media use was associated with greater odds of negative relationships between mothers and daughters, fathers and daughters, and fathers and sons, but not between mothers and sons. A handful of studies have also examined the impact of a parent connecting with adolescent children on social media on the quality of their relationships. Abar et al. [11] and Yang [12] in their studies found that adolescents frequently included parents in their social networks and granted them similar levels of access to their personal information as their friends. However, when it comes to parents sharing content about their adolescent children, Verswijvel et al. [13] found that adolescents, particularly boys, viewed this negatively. Research on parent-adult child connectedness using social media has not kept up to speed with increase in social media use. A recent systematic review on the effect of social media use on family connectedness found that social media use between parents and adult children has generally been examined from a narrow viewpoint, mostly focusing on monitoring by parents or coviewing and "friending" [14]. A key limitation of existing evidence on the dynamics of social media use between parents and young children is its emergence from largely individualistic societies as in the United States, United Kingdom, Canada, and Australia. In these societies, parents emphasise children's independence, and there is a general expectation for adult children to leave home within a few years of finishing high school or college and be financially independent [15]. Less attention has been paid to patterns of social media use among parents and adult children from diverse racial and ethnic groups with varying family structures. For example, Asian and Latino families are considered classically as large, patriarchal, collectivistic, and joint families. Such traditional families function as a dominant influence in the lives of their individual members [16]. People who identify as collectivistic are individuals whose focus is the family or the group, rather than the individual [17]. It is possible that the influence of online and offline connections on parent-child relationships varies by culture, particularly in those from collectivistic backgrounds. However, this supposition has yet to be fully tested in the published literature. Pakistan, one of the most populated countries in the world, traditionally had a joint family system. However, like many other Asian countries, over time, balance is shifting towards the nuclear family system [18]. Despite the shift, adult children are likely to live with their parents well into their adulthood, until they are married. This young adult group is the early and significant adopter of social media platforms. Pakistan had 71.70 million social media users in January 2022 [19]. YouTube and Facebook remain the most popular social media platforms followed by WhatsApp and Twitter [19]. Most of the country's internet and social media users are those aged 18-35, and most Facebook users and Twitter users are male. Other studies on young adults in Pakistan indicate that social media usage was higher among this cohort with almost 90% using social media sites [20,21]. Facebook and YouTube were reported to be the most accessed sites among this cohort [20]. Pakistan, for its demographic, cultural and strong religious ties, has been a region of interest to examine the adoption of digital technologies and their social impact [22]. Like other South Asian countries, Pakistan also has the highest reported gender digital divide, with the widest gaps in mobile and internet usage [23]. This also has implications on social media usage, and as reported by Schoemaker in his survey of 900 mobile data users , 85% of the male respondents reported that they mostly use Facebook, compared to only 47% of female respondents. By contrast, 45% of women said that they used WhatsApp, compared to only 13% of men. While preference for WhatsApp among women can be seen for its financial affordance while enabling privacy and reinforcement of religious social norms, there is limited evidence on how parents, in collectivist societies like Pakistan, can influence social media use [24]. In the evolving social media landscape and increased intergenerational usage of these applications, Pakistan offers a unique opportunity to investigate how social media use across gender can be influenced by parents, who play an instrumental role in shaping the lives of their adult children [25]. There is growing interest in exploring how social media is impacting family relationships in Pakistan, with a primary focus on exploring implications associated with its usage. Several preliminary studies have reported a common negative perception about social media and family communication among young adults. A study across 175 young adults reported that 85% of participants agreed that social media sites made family relationships nonexistent [26]. A recent study on college students in Pakistan reported that around 74% of the participants perceived that people prefer spending more time on social media than with family. However, some also report positive perceptions around family social media use. A recent study on 200 college students from Faisalabad, Pakistan, found that 75% of the participants believed that social media family groups play a vital role in family connectedness [27]. While there are prevailing negative perceptions, research is limited on whether and how young adults in Pakistan engage with their parents on social media. This includes exploring any potential links between social media interactions and the sense of connection between parents and children, as well as young adults' views on parental monitoring 2 Human Behavior and Emerging Technologies within the realm of social media usage. Therefore, in this exploratory study, we aim to investigate patterns of social media use among young adults and their parents and its perceived association with family connectedness and parental monitoring. We also examine any differences in these perceptions across different gender groups. --- Materials and Methods An online cross-sectional survey was used to collect data from young adults in Pakistan. The survey was disseminated using convenience sampling strategies. We used a convenience sampling approach due to its affordability, efficiency, and availability of participants. We acknowledge that a random sampling approach would have produced more representative findings. The link to the questionnaire was sent to eligible participants in a university where one of the authors is a full-time academic. The survey was also advertised via relevant group pages on social media platforms . A total of 421 participants aged 17-30 years took part in the survey. The participants reported the following: demographic information: age, gender, marital status, parental employment status, participant employment status, family income, family size, and living arrangement; access to electronic devices: shared computer at home, shared computer at work/university, personal computer/laptop, smart phone, and tablet; frequency of use of social media platforms: which platforms are used and frequency of their usage: Facebook, YouTube, WhatsApp, Instagram, Snapchat, Twitter, LinkedIn, Tumblr, Tinder, Tango, imo, Viber, Skype, Hangouts, Pinterest, TikTok, Duo. Responses ranged from never to several times daily [28]; use of social media with parents: participants were asked if they connected with their parents on social media, and if yes, how often they used each social media platform with their parents . Those who did not use social media platforms to connect with parents were given a list of reasons for not connecting and ask to choose. The options included "My parents don't have smart phones," "My parents don't use social media," "I don't want to," "I like connecting with only my friends," and "I like my own privacy"; family connectedness and parental monitoring: We adapted family connectedness [29] and parental monitoring scales [30] from existing studies where these have been previously validated. As there have been no previous studies around this topic in the unique cultural context of Pakistan, these scales were adapted for specific age and cultural context ; therefore, the psychometric properties of the original scale are not applicable. We therefore measured the reliability of our adapted scales as reported in the next section. Participants' family connectedness was measured by 7 items on a 5-point Likert scale ranging from "strongly agree" to "strongly disagree." This multi-item scale was adapted for young adults from the family connectedness scale validated by Waters and Cross [29]. Parental monitoring was measured directly and indirectly. Firstly, participants were asked if their parents monitored their internet/mobile phone usage or computer-related activity. Secondly, parental monitoring was measured using four items adapted from The Good Starts study [31]. Items in both scales were modified to suit the age and cultural values of the participants. Ethics approval for the study was obtained from the university ethics committee in Pakistan and ratified by human research ethics at QUT, Brisbane. 2.1. Data Analysis. A data matrix was produced from the completed questionnaires using SAS 9.4 for Windows . Descriptive statistics were used to present demographic data and to evaluate the frequency of use of social media platforms. We summarized data as frequencies for categorical variables. Appropriate statistical tests were used to conduct analysis. Results were converted to mean values, analysis of frequencies, and associations using chi-square tests, where required; p ≤ 05 was considered significant. Total scores of family connectedness were summed to range from 7 to 35, with higher scores representing higher self-perceived family connectedness. Total scores of parental monitoring were summed to range from 4 to 20, and higher scores represented higher self-perceived parental monitoring. To assess the measurement properties of the family connectedness and parental monitoring scales in a sample of Pakistani young adults, factor analysis with Varimax rotation was performed. This measured the unidimensionality of the scales. Cronbach's alpha was also computed for the factors. The index for the family connectedness scale showed a moderate internal reliability with a Cronbach's alpha of 0.65. The parental monitoring scale also showed an acceptable internal reliability with a Cronbach's alpha of 0.8. The Box-Cox technique was used to find the optimal normalising transformation for both family connectedness and parental monitoring [32]. Multiple linear regression analyses were used to examine the relationship between social media use to connect with parents and family connectedness and parental monitoring separately. Both crude and adjusted beta coefficients are reported. Covariates include age and gender. --- Results The mean age of survey participants in this study was about 22.6 years . About 61% of the participants in this study were male, and 86% were current undergraduate or postgraduate students. Almost 96% of the participants were unmarried, and about 72% lived at home with family. All participants had access to at least one electronic device like smart phone or laptop. More than 90% of participants had a personal smart phone, and 82% had a personal laptop or computer. 3.1. Patterns of Social Media Use. All participants in the study were using at least two social media platforms with 88.4% engaging with more than five platforms, irrespective of the types of devices participants accessed. --- Human Behavior and Emerging Technologies WhatsApp was the most used platform irrespective of gender or age, with about 91% of the participants reporting daily usage. All participants who had a personal smart phone were users of WhatsApp. This was followed by YouTube , Instagram , and Facebook as the most popular social media sites used daily. Only 43% reported 4 Human Behavior and Emerging Technologies that they used Snapchat, and 7% used TikTok daily. The use of Facebook and YouTube was significantly higher in males , while female participants were more likely to use Instagram and Snapchat . There was no gender difference in the use of WhatsApp, which was the most widely used social media app among this cohort of Pakistani youth. 3.2. Patterns of Social Media Use with Parents. Overall, 63% of participants connected with their parents using social media. This varied significantly among male and female participants with 69% of female participants connecting with their parents on social media compared to 59% of males . WhatsApp was the most reported social media app used by participants to connect with parents with 71% reporting using it all the time to communicate with parents. More than half of the participants reported using Facebook to communicate with parents, with 24% reporting using it very often. YouTube , Instagram , and Snapchat were the other commonly used social media sites that participants reported to connect with parents. The platforms used to connect with parents did not vary by gender with males and females connecting on Facebook, YouTube, WhatsApp, and Instagram. However, females reportedly used Snapchat to connect with parents in significantly higher numbers than males . Living with or away from family was found to have no association with connecting with parents on social media. For those who do not connect with their parents on social media , when asked about the reasons for not doing so, the most common reason cited by participants was that their parents did not use social media . The second most cited reason was participants concerned about their privacy and preserving it by opting not to connect with their parents. --- Social Media Use, Family Connectedness, and Parental Monitoring. Family connectedness was measured using seven questions. The median family connectedness score was 31, very high for this cohort of participants; however, it is reflective of the patterns in this region, with family being a very strong unit. The score did not vary with gender, age, family size, or if coliving with family. Results from multiple linear regression analyses examining the association between the use of social media to connect with parents and family connectedness are presented in Table 3. Both adjusted and unadjusted models show no association between any of the social media platforms used to connect with parents and participants' perception of family connectedness. Overall, 19% of the participants reported that their parents monitored their internet/ device usage when asked directly . Parental monitoring was also measured indirectly using 4 items. The median parental monitoring score was 18.0, and it varied with gender with more than 55% of females reporting experiencing higher monitoring compared to 45% of males . Parental monitoring did not vary significantly with age, family size, or if coliving with family. Table 3 presents results from multiple linear regression analyses examining the association between the use of social media to connect with parents and parental monitoring. Connecting with parents on both Facebook and Instagram was highly associated with participants' perception of parental monitoring before and after adjusting for age and gender. This was not the case for the most used platform WhatsApp. --- Discussion Our study is aimed at exploring patterns of social media use between Pakistani adult children and their parents. Pakistan's cultural context is notable for its prominent display of heritage and ancestral traditions. Like other densely populated South Asian countries in the region, it emphasises traditional, collectivist values, focusing on interdependence and community ties [33]. Young adults well into their adulthood live with their parents and obeying them is central to the family values [34]. To the best of our knowledge, this is one of the very few studies that explore the impact of social media platforms on parent-child relationships in the collectivist family cultural context of Pakistan [35]. Our study advances the much-needed evidence on how adult children's increased social media consumption possibly relates to traditionally collectivist, interdependent values and culturally relevant parental goals of South Asian parents. All participants were active social media users using diverse available platforms, despite some variation in device access with 90% having a personal smart phone. Our findings confirm that although the percentage of men and women using social media is comparable, there are gender differences in terms of preference and frequency on specific social media platforms. WhatsApp is the most popular social media platform irrespective of gender and age. The nature of WhatsApp, which is mostly to communicate with your own contacts and not an unknown audience, highlights that the principal role social media has is communication with others rather than entertainment. The next most popular platforms were Facebook and YouTube, which are mostly used for entertainment activities such as media consumption and sharing. Importantly, Facebook also includes a texting function that is used for communication. When looking at gender differences, Facebook and YouTube were more popular among males whereas female participants were more likely to use Snapchat and Instagram. These patterns corroborate with those reported in other studies and reported social media statistics [36,37]. In terms of the use of social media platforms with parents, 63% of participants connected with their parents on at least one social media platform. Our study provides insight into patterns of connecting with parents across specific social media platforms. WhatsApp was the most preferred platform used by adult children to connect with their parents and used frequently. This pattern aligns with the ongoing research on how users choose different social platforms based on the presence of specific contacts [38]. As highlighted by Taipale and Farinosi [39] in their examination of WhatsApp for family communication, the WhatsApp platform offers many modalities which allow Human Behavior and Emerging Technologies choosing of the most desired and most suitable mode of communication for every family member individually in a safe environment [39]. Popularity of WhatsApp between parents and children therefore may be indicative of their preference to keep the communication private as much as possible. Privacy is not usually afforded on that level by Facebook and Instagram, and the other popular social media platforms are used to connect with parents but used less frequently. Privacy as a preference was also indicated as one of the most common reasons by those who do not connect with their parents on social media. Our findings also highlight gender differences in choice to connect with parents-female young adults are more likely to connect with their parents using social media. This perhaps affirms the idea that females tend to use social media plat-forms to maintain existing social relationships as compared to males who use these platforms more for expressing their opinion [40,41]. In terms of family connectedness, 72% of participants in the study lived with their parents and had very high perceived family connectedness. This is indicative of how families are the strongest unit in Pakistan driven by collectivist family values [42]. This pattern of strong family connectedness is similar to that reported in studies examining family connectedness in other collectivist societies in the Arab countries. Dwairy et al. [43] reported high family connectedness in Arab societies, and when compared to the connectedness of American adolescents, their study reported Arab societies scoring higher than the American adolescents [43]. The high family connectedness score in our study also 6 Human Behavior and Emerging Technologies adds to the evidence on a significant difference in the connectedness between the individualistic western and collectivist eastern societies. As demonstrated by Dwairy and Achoui [44] in their study, adolescents in France, Poland, and Argentina were less connected to their parents than adolescents in Kuwait, Algeria, Saudi Arabia, Bedouins in Israel, Jordan, and India [44]. Our findings also extend the evidence to support that despite the increase in urbanisation, modernisation, and industrialisation in collective cultures, the closely knit family patterns continue to exist [45]. Our study found no association between adult children's use of social media platforms with parents and their perceived family connectedness. In line with the collectivist family culture, young adults tend to live with their families, in parental homes for a much longer time. Therefore, bonding and relationship development is less reliant on social media technologies. However, the frequent use of social media platforms directed to communication such as What-sApp does show the importance for social media for broader communication process without being associated with family connectedness. An implication of this finding is that technology-based solutions or interventions oriented to families might not be as widely accepted. While participants in our study indicated a high degree of family connectedness, they were conscious of maintaining their privacy and their parents monitoring their social media use. This was indicative of the complexity and asymmetric nature of the parent-adult child relationship, meaning that parents and children have different roles and responsibilities in it [46]. During adulthood, the relationship is constantly changing due to the desires of independence and self-sufficiency of children, and it presents challenges that are unique to this period: parents go through a stage of acknowledging the child as an adult, and there is an interplay of autonomy and dependency between both parties [47,48]. Surveys conducted in the United States indicate a high prevalence of parental social media monitoring in the context of parent and their teenage children [49]. It is likely that in Asian cultures including Pakistan, where adult children tend to colive and be financially dependent for much longer, parents monitor their adult children's social media usage. Several potential limitations of this study should be noted, as they also suggest opportunities for future research. First, because only cross-sectional data were used, the results of this study indicate associations between measures that should not be interpreted as causal relationships. Also, the use of a convenience sample of participants limits the generalisability of the study results. Further, as the study focused on young adults in Pakistan, a collectivist society at large that emphasise interdependence and family connectedness, it is unclear whether the patterns of results can be generalised to other cultures and individualistic societies. Our findings also highlight the prevalence of parental monitoring of social media use. Previous studies have demonstrated that such practices may increase conflict in individualistic families [50,51]. The impact of such parental monitoring on family conflicts in collectivist cultures, especially those with strong religious ties, needs further investigation. Further, it is important to consider that the sample in this study had the means to access the social platforms and internet frequently. Many families, especially those living in rural regions in Pakistan, have limited access to the internet with some estimates suggesting 46% of all Pakistanis accessing the internet daily [52]. Future research therefore should explore how the lack of access to the internet and social media platforms might affect the degree of family connectedness and their possible association with family conflicts. Another possible limitation of this study is that social media use with parents was assessed by a simple dichotomous measure and frequencies. Some variables of interest were measured using single items, which might not fully capture the underlying complexity of the constructs. However, the use of single items improved the efficiency of data collection by preventing participant fatigue and minimizing incomplete responses and contributed to expanding the diversity and comprehensiveness of the sample. An essential avenue for future research is to integrate qualitative in-depth investigations and employ finer-grained culturally relevant measures to 7 Human Behavior and Emerging Technologies comprehend young adults' specific interactions with parents on social media platforms. This will enhance our understanding of how social media usage influences parent-child relationships. --- Data Availability The data is restricted as per the ethics policy of the study administering institutions . --- Conflicts of Interest No potential conflict of interest was reported by the authors.

An increasing number of young adults are using social media platforms in collectivist family cultures like Pakistan, but little is known about how social media use is associated with family connectedness in this population. This study is aimed (a) at examining the social media usage patterns in Pakistani young adults and how they use it to connect with their parents and (b) at exploring the possible association between social media use and perceived family connectedness and parental monitoring. Data came from an online survey conducted among young adults in Pakistan. Patterns of use across various popular social media platforms were recorded. Additionally, family connectedness and parental monitoring were measured. Multiple linear regression analyses were used to examine the relationship between social media used to connect with parents and family connectedness and parental monitoring. A total of 421 participants responded to the survey. All participants regularly used at least two popular social media platforms. WhatsApp was the most used platform irrespective of gender or age, with about 91% of the participants reporting daily usage. Overall, 63% of participants connected with their parents using social media. This varied significantly among male and female participants with 69% of female participants connecting with their parents on social media compared to 59% of males. Connecting with parents on both Facebook and Instagram was highly associated with participants' perception of parental monitoring before and after adjusting for age and gender. The study highlights that while family connectedness in collectivist societies like Pakistan is not associated with social media usage, adult children perceive parents monitoring their social media activities.

Background In 2019, the Australian Government established a Royal Commission into the violence, abuse, neglect and exploitation of people living with a disability. The Royal Commission was charged with investigating, among other questions, the prevention, protection and best practise responses to violence and abuse of people living with a disability. One of the key questions raised in the Commission's issues paper was "What are the experiences of First Nations people with disability regarding violence, abuse, neglect and exploitation"? [1]. In Australia, the Indigenous population consists of the Aboriginal and Torres Strait Islander peoples. Although estimates vary, Aboriginal peoples are thought to have arrived on the Australian continent at least 50,000 years ago, while the Torres Strait Islanders first settled the islands of the Torres Strait approximately 3000 years ago [2,3] For many Indigenous peoples worldwide, racism intersecting with colonisation has led to a much higher risk of experiencing violence from both Indigenous and non-Indigenous people across their lifetime [4][5][6]. In Australia, Aboriginal and Torres Strait Islander people have been subjected to multiple policies of forced removal from land and family, systematic discrimination from education, employment, and services, and currently experience vast inequalities in poverty, health, and overall quality of life outcomes [6][7][8][9]. For Indigenous people living with disability, the connection between disability and violence is complex and affected by multiple factors of identity. International studies from the general population show disability and violence are interconnected, as people with disability face increased risk of exposure to violence and experiencing violence can often induce or cause disability [10][11][12][13]. This relationship is influenced by the disproportionate poverty and high rates of violence experienced across the life course compared to people without disability, as well as dependence on carers common among people living with disability [14][15][16][17]. Exogenously, cultures of silence, encouragement to ignore violence as part of everyday experiences, and imposed norms of helplessness or worthlessness perpetuate cycles of violence [15,18,19]. For Indigenous people with disability, racism may intersect with ableism to ultimately create overlapping forms of marginalisation and vulnerability to violence [20]. In Avery's study of disability among Aboriginal and Torres Strait Islander people, among 9 of 41 interviews conducted for the research , participants spoke about exposure to violence or a traumatic death of a family member. These included references to close family members who had been murdered or 'lost,' gender-based violence, violent attacks, suicide, the death of a young child through medical neglect, and exploitation by a human trafficking ring [20]. This underscores the need for studies that specifically examine the intersection of indigeneity, violence, and disability [21][22][23][24]. However, there continues to be a dearth of quantitative research at this intersection. The current literature indicates that the mechanisms that link indigeneity or disability with violence, intersect and amplify for Indigenous people living with disability [25]. In this paper, using nationally representative data, we seek to examine several questions about experiences of physical violence and violent threats reported by Aboriginal and Torres Strait Islander people living with disability. Firstly, are Aboriginal and Torres Strait Islander people living with disability more likely to cite incidents of physical violence or violent threats relative to those without disability? Second, does exposure to violence differ by type, severity or numbers of disability conditions? Finally, does the relationship between the perpetrator and victim of violence differ between Aboriginal and Torres Strait Islander people with and without disability? --- Methods --- Data To answer these research questions, we draw upon data from the National Aboriginal and Torres Strait Islander Social Survey with in-field operations managed by the Australian Bureau of Statistics between September 2014 and June 2015 [26]. The NATS ISS includes Aboriginal and Torres Strait Islander people residing in private dwellings and was sampled using a multi-stage design [27]. Following screening, a response rate of approximately 80% was achieved, with this figure marginally higher in remote Australia and marginally lower elsewhere. Critical to maximizing the survey response rate was the involvement of Aboriginal and Torres Strait Islander people in the design and conduct of the survey. For example, Aboriginal and Torres Strait Islander peak bodies were engaged at the design stage of the survey. The survey instrument itself varied geographically to ensure questions were culturally appropriate. Of relevance to our present study, the ABS adapted the wording of the concepts of disability and long-term health conditions to account for cultural differences [27]. When a selected respondent was unable to answer questions due to illness or injury, a proxy interviewer was sought when appropriate [27]. Our study is restricted to a sample of n = 6417 persons aged between 15 --- Measures Disability The ABS defines disability as "any limitation, restriction or impairment which restricts everyday activities and has lasted, or is likely to last, for at least six months" [27]. The 2014-15 NATSISS operationalises this definition utilizing the short disability module . The SDM includes a number of measures to identify a health condition that is ongoing and restricts day-to-day activities. For example, in Module 10.02 of the NATSISS, respondents are queried with a series of prompt cards as to the presence of a series of conditions. For those responding yes, they are further prompted "Are you restricted in everyday activities because of this health condition". Measurement of co-morbidities is included, and respondents are further requested to indicate specific conditions that "are likely to last or have lasted, for six months or more". The ABS notes that "the full SDM was used in both nonremote and remote areas, with some wording amendments to aid comprehension" [27]. ABS datasets utilizing the SDM show a higher prevalence of disability relative to the more detailed Survey of Disability and Carers. See ABS and Temple et al. for a discussion on disability measurement across ABS surveys [28,29]. Following other studies, we utilize several measures of disability [29]. First, a single measure indicating the presence of any disability. Second, a measure of the severity of disability, defined by the ABS . Third, a measure of the type of disability including sight, hearing or speech, physical, intellectual, psychological, head injury, brain damage or stroke, other restrictions. Finally, a measure of the number of disability types was included. --- Violence and violent threats In the NATSISS, physical violence refers to "any incident that involves physical assault, which is the use of physical force by a person with the intent to harm or frighten another person. It includes being pushed, shoved, hit or attacked with a weapon. Other forms of abuse are not included" [27]. The question used to measure physical violence was: "In the last 12 months, did anyone, including people you know, use physical force or violence against you?". Respondents reporting yes were asked the frequency of violence and how they knew the perpetrator. Interviewers used a prompt card to elicit the relationship as follows: current partner, previous partner, ex-boyfriend or ex-girlfriend, parent, child, sibling, other family member, friend, work colleague/fellow school student, neighbor, known by sight only or other known person. For physical threats, respondents were asked "In the last 12 months, did anyone, including people you know, try to use or threaten to use physical force or violence against you?". There is the possibility that due to the presence of the perpetrator , that the respondent may not feel comfortable in disclosing instances of violence. As part of the field operations, the ABS notes: "In order to conduct a personal interview with the selected person , interviewers made appointments to call-back to the household, as necessary. All interviews were conducted face-to-face. Due to the sensitive nature of the survey questions, it was suggested that interviews be conducted in private. However, interviews may have been conducted in private or in the presence of other household members, according to the wishes of the respondent" [27]. --- Statistical tests All statistical tests for this study were conducted using Stata via the ABS Remote Access Data Laboratory [26]. Firstly, simple weighted percentages of violence were disaggregated by the measures of disability outlined above. Following, a series of multivariable logistic and ordinal logistic regression models were fitted to measure the association between disability measures and exposure to violence. Following previous studies, separate regression models were estimated for each disability type. E.g., for the physical disability model, a categorical variable measuring no disability, physical disability, disability other than physical were included [29]. The NATSISS microdata file included 250 replicate weights to account for non-response and sample design features. As violence is a relatively rare event, previous studies have adopted relatively parsimonious models with few controls included. For example, in two recent Australian studies of violence and disability in the non-Indigenous population, controls are included for age and sex [30,31]. In a UK study, adjustments are made for age and gender as well as socio-economic disadvantage and neighbourhood quality [32]. Dammeyer and Chapman's 2018 study of disability violence in Denmark adjusts their analysis for gender, age and presence of either a physical or mental disability [33]. For our study of Aboriginal and Torres Strait Islander people, it is also important to include additional controls. Berry et al. find hospitalisation rates because of interpersonal violence increase with geographic remoteness [6]. Other studies note the importance of removal from natural family as an important factor heightening exposure to violence for Indigenous peoples worldwide [7,[34][35][36]. A recent Australian study also highlights a strong association between removal from natural family and racism [29] and another notes the increased prevalence of family removal for Aboriginal and Torres Strait Islander people living with a disability [37]. Following these studies, we present both adjusted and unadjusted odds ratios to examine the association between measures of disability and exposure to violence. The unadjusted odds ratios are calculated using only age and sex in each model. The fully adjusted odds ratios included a full list of control variables including: Age Sex Remoteness . Remote Australia is defined by the Australian Statistical Geography Standard [27]. Household Income . Specifically, this is gross household income, adjusted using an equivalence scale and coded into the income distribution bands defined above. Marital Status Removal from natural family . This was measured by asking "Have you been removed from your family by welfare or the government or taken away to a mission?". The need to estimate parsimonious models is pronounced due to 1) the rarity of violent events, and 2) the need to employ a jack-knife procedure on 250 separate replicate samples to calculate correct variance estimates for the regression coefficients and odds ratios. A further consequence of these design constraints is that there is insufficient sample size to split the regression analyses by other characteristics such as gender, due to matrix conformability issues. Nonetheless, we present selected split sample characteristics by gender in the descriptive statistics. Moreover, following the multivariable results, we present weighted descriptive statistics stratified by measures of disability and other characteristics on the relationship of the victim of physical violence to the perpetrator in the last act of violence. --- Results Before turning to the key research questions, it is useful to point to the characteristics of this sample of Aboriginal and Torres Strait Islander people aged 15-64 . Reflective of the younger age structure of the Indigenous population , approximately 45% of respondents were aged under 30 and about 10% aged between 55 and 64. The vast majority of respondents resided in non-remote Australia and the gender split is relatively equal. Just over 40% of respondents were married or living with a disability . --- Violence by disability and selected characteristics Table 1 disaggregates the weighted prevalence , odds ratios adjusted by age and sex and odds ratios adjusted by all covariates for physical violence, frequency of physical violence, violence with harm and violent threats. Across all measures of violence, people with a disability were at a heightened risk compared to people without a disability. Overall, about 17% of those with a disability reported experiencing an act of physical violence in the last 12 months, compared with just under 12% of those without a disability. In line with these results, people living with a disability were also more likely to report higher frequency of physical violence as well as violence with harm. Just under one quarter of people with a disability reported being subject to violent threats in the past year, compared with 14% of those without a disability. These descriptive statistics point to an association between disability and exposure to violence. However, it is important to control for confounding factors and to adjust for the complex sampling design of the NATSISS. As can be seen in Table 1, a number of factors independent of disability appear to be associated with violence. For example, prevalence tends to decrease with age and marriage, but increase with decreasing income, family removal and marginally, residence in a remote part of Australia. Controlling for these factors we find that disability is associated with a 1.55 to 1.75 fold increase in the odds of reporting physical violence , more frequent physical violence and violence with harm . Disability was associated with a doubling of odds of violent threats . Although not the key purpose of this study, it is noteworthy that family removal was associated with an approximately doubling of odds for experiencing physical violence, more frequent physical violence, violence with harm, and threats of violence . Table 2 re-estimated the above models by measures of the severity of disability, type of disability and number of disability types reported. Importantly, these results show that severity of disability is associated with exposure to violence. Aboriginal and Torres Strait Islander people with profound or severe disability were in excess of double the odds or risk of physical violence , more frequent violence , violence with harm and violent threats , relative to those with no disability. The odds ratios for those with other restrictions As a further proximate measure of the severity of disability, each additional disability type increased the odds of violence and violent threats between 1.4 and 1.5 times. Results in Table 2 also provide insights into the types of disabilities associated with violence and violent threats. Although the levels of exposure to violence were elevated for people with all disability types, the risk was particularly acute for people with intellectual disabilities; psychological disabilities; and those with head injury, stroke or brain damage related disabilities. People with psychological disabilities and those with head injury, stroke or brain damage related disabilities showed a 2.5 to 4-fold increase in the odds of reporting any form of violence or violent threats. --- Victim-perpetrator relationship The NATSISS includes additional detail about contextual aspects of physical violent events and selected detail on physical threats. Table 3 displays the relationship of the victim to perpetrator, by selected disability measures and gender. Tests of proportions between gender samples are also included in the table. An important finding of this research is the differences in the relationship of the victim to the perpetrator by gender. Whereas overall physical violence did not differ by gender in these data, women regardless of disability status were significantly more likely to report partner Similarly, levels of other familial violence was also higher for women. Men, on the other hand again, regardless of disability status were more likely to report other known persons , or unknown persons as the perpetrator of violence. Within gender groups, males with a disability were slightly less likely to cite partner violence and more likely to report other familial violence than women with a disability. Nonetheless, the differences in the relationship with the perpetrator are more pronounced between gender groups, rather than on the basis of disability presence alone. --- Discussion Motivated by the ongoing Royal Commission into violence, abuse, neglect and exploitation of people living with a disability, we sought to examine the exposure of Aboriginal people living with disabilities to violence. Our study found that within the Aboriginal and Torres Strait Islander population, presence of a disability was associated with a 1.5 fold increase in the odds of exposure to physical violence, a higher frequency of violence, and doubling of the odds of experiencing threats of violence. These findings are consistent with quantitative evidence showing people with disability and Indigenous people are at much higher risk of experiencing violence than those without disability or non-Indigenous people [5,6,11,16,[38][39][40]. Qualitative studies have theorised that when the two identities intersect, the systems of disadvantage that Indigenous people and people with disability navigate overlap to greater effect [25]. In Australia, narrative research on the experiences of Aboriginal and Torres Strait Islander people with disability found that more than a fifth of interviews referenced exposure to violence or a traumatic death of a close family member [20]. This is despite participants not being asked directly about exposure to violence. The high prevalence of Indigenous family and racially-motivated violence, incarceration, and poverty, in conjunction with the overrepresentation of people with disability in carceral settings and institutions, all work to increase risk of exposure to violence [21]. This is in addition to other external factors which increase likelihood of remaining in violent situations: disability often being underdiagnosed due to race, Indigenous spaces not having sufficient funding for accessibility measures, and the siloing of organisations into distinct Indigenous and disability-specific services [8]. The overall high prevalence of violence against people with disability is thought to be the result of lack of economic opportunity which increases the likelihood of living in areas with elevated crime rates and leads to the increased dependence on carers that people with disability experience, in combination with the colonial racism that has led to systemic poverty and trauma [41]. Cripps and Adams further conceptualise the factors contributing to family violence in Aboriginal communities into two key factors [42]. The first are factors attributable to colonisation, including dispossession and cultural dislocation, family removal and policies and practices that continue to impact Aboriginal people, increasing the risk of violence. The second group of factors increasing the risk of family violence in Aboriginal communities relate to those that can occur in any vulnerable population. For example, unemployment and welfare dependency, destructive coping behaviours, mental health issues and past instances of violence or abuse. Evidence also suggests that these latter factors indicative of poor socio-economic outcomes are also experienced by people living with disabilities in the broader Australian populationfor example, lower levels of education, employment, housing vulnerability and deleterious financial wellbeing [43,44]. The findings from this study that show an elevated risk of experiencing violence based on severity and type of disability are in line with the wider literature regarding levels of violence experienced by people with disability. Severity of disability is strongly associated with all measures of violence. Aboriginal and Torres Strait Islander people with profound or severe disability were at a minimum 2-fold increase in odds of reporting each measure of violence and almost 3 times more likely to be exposed to violent threats compared to Aboriginal and Torres Strait Islander people without a disability. As a further proxy measure of the severity of disability, multiple disability types increased the odds of exposure to violence by approximately 1.4-1.5 times with each successive disability type. This association is thought to be the result of the contribution of severity of disability to inability to leave violent situations and how that vulnerability contributes to the power dynamics of violence. Those with more severe disabilities are inherently more reliant on carers, thus are more likely to stay in violent carer relationships and be physically unable to escape violent situations [45,46]. They also face greater barriers to communicating to authority figures that they have experienced violence, among them being limited verbal communication or not being believed as a result of their disability [39]. These vulnerabilities make it more likely that a person with violent or controlling intentions seek out people with disability [47]. Violence was also more pronounced for those living with intellectual disability, psychological disability and/ or head injury/stroke/brain damage compared to people with other types of disabilities. In a national-level Danish study, it was also found that people with learning or mental disability were significantly more likely than people with physical disability to experience violence, harassment, or abuse [33]. A systematic review and meta-analysis of risk of violence against adults with disability found an increased risk for people with intellectual impairments compared to those with non-specific impairments; those with mental illnesses were found to have a higher risk than both other categories [11]. Similarly, in the UK, the risk of exposure to violent crime was elevated among adult with mental health problems in comparison to both adults with no disability as well as those with other forms of disability [16]. Interpreting this finding, it is also important to note that disabilities such as those related to head injury may in themselves be caused by a violent event [48]. Evidence from hospital morbidity data suggests that Aboriginal people, in particular Aboriginal women, are at a significantly increased risk of head injury due to assault [49]. Family removal, the practise of intentionally taking Indigenous children from their natural family and placing them in other settings, was found in the analysis to be a significant predictor of experiencing violence [34]. This practise has historically been documented in Australia, Canada, and the USA under the names of the "Stolen Generations", "Residential Schools", "the Sixties Scoop", and "residential boarding schools", in which the goal was to deny access to Indigenous heritage in order to assimilate Indigenous children into dominant cultures [34,36,50]. It has been theorised both as inherently violent by breaking family ties and connections with community, and as a pathway to experiencing violence in care institutions or foster homes [51,52]. Removed children, regardless of disability status, experienced high rates of physical, sexual, and emotional violence in their new care settings [36]. This was in addition to the constant degradation of their Indigenous identity and attempts to break ties to their culture [53]. Survivors also report poor housing, food, and educational quality within their care institutions [51,52]. Family removal currently manifests in foster or government care processes. It can include efforts to maintain connections to culture and heritage, such as placing children with other Indigenous families, but remains a traumatic experience for children and their families and is experienced at significantly elevated rates by Aboriginal and Torres Strait Islander people [54]. The impacts of experiencing family removal are widereaching and drastic. Survivors of family removal have an increased likelihood of being arrested or being charged with an offence, having alcohol or illicit drug use disorders, cancer, diabetes, heart disease, stroke, back and eye conditions, being unemployed or dependent on government payments, attempted suicide or having suicidal thoughts, worse educational outcomes, cognitive, behavioural or emotional disability, being a victim of violence, loss of parenting skills, living with an infectious disease, or having a mental illness [7-9, 35, 50, 55-57]. All of these factors increase barriers to socioeconomic mobility, which in turn, increase the likelihood of living in an area with higher crime [32]. Lower socio-economic mobility additionally limits the ability to escape violent situations by decreasing financial stability [32]. We further found that Aboriginal and Torres Strait Islander women, regardless of disability status, were more likely to report family violence, whereas men were more likely to report violence from other known individuals . The gendered sources of violence have been found in a national analysis of American crime victimisation, in which men were more likely to experience violence from a relative, whereas women were more likely to experience violence from an intimate partner [17]. The same was true in the Australian setting in a non-Indigenous-specific study [31]. Women with disability are considered to have a higher risk of experiencing intimate partner or family violence, as compared to men with disability. This is in line with other Australian research on the experiences of Aboriginal and Torres Strait Islander people with disability, where 7 of 9 interviews referencing experiences with violence were with women [20]. A potential explanatory factor is the socialisation of women with disability to believe that they are uniquely dependent on their partner for long-term care and economic stability, to be compliant, and that they are unable to access caring, loving relationships [39]. Alternatively, it has been argued that disability further adds to the power disparities and lack of access resources associated with traditional gender roles [58]. While there is a dearth of methodologically robust studies, it is clear that both gender and type of disability affect the prevalence of violence [11,58,59]. It is critical that strategies to prevent violence are developed taking into account these intersectionalities [11,58,59]. Overall, the findings of this study highlight the intersectionality between Indigeneity and ableism, which in turn has important implications for the design of services for Aboriginal and Torres Strait Islander people. It is an important complement to narrative research aimed at providing a voice to Aboriginal and Torres Strait Islander people with disability to share their unique experiences and their resilience [20]. National policy for the provision of disability support does include an Indigenous Engagement Strategy; however, the strategy has not been consistently implemented and needs to be considerably strengthened to address the complex needs for Aboriginal and Torres Strait Islander people with disability. This is particularly true in relation to ensuring that disability supports are responsive to the mental and emotional effects of violence, such as trauma or mental illness. Unlike many other sectors there is currently no overarching framework for selfdetermination, working with Aboriginal and Torres Strait Islander people or cultural safety in the disability sector. This study provides further support to research calling for Aboriginal and Torres Strait Islander people with disability to lead the development of frameworks and approaches for the disability sector [20]. --- Study limitations Our study contains a number of limitations. First, the NATSISS is a cross-sectional survey and we do not make causal inferences about the relationship between the various measures of disability and exposure to violence. Second, the NATSISS operationalises violence as either physical violence or violent threats, omitting specific questions related to sexual assault and sexual threats. Third, the violence measures were collected by self-report over a 1 year period, raising the issue of recall bias. This may lead to an under-estimate of levels of violence which may also exist due to the sensitive nature of the questions. Fourth, as noted earlier, the sampling frame for the NATSISS consisted primarily of people living in private dwellings. Respondents in institutions, for example, were not enumerated in the survey. Finally, Aboriginal people with very severe disabilities may have been under enumerated. --- Conclusions This study shows that among Aboriginal and Torres Strait Islander people, 1) presence of a disability was associated with a 1.5 fold increase in the odds of exposure to violence and doubling of odds of reporting violent threats, 2) that people with profound or severe disability experienced heightened exposure, in excess of double odds, 3) as were those with specific types of disabilities such as psychological or head injury, stroke and brain damage conditions. We further found that women, regardless of disability status were more likely to report family violence, whereas men were more likely to report violence from other known individuals . Independent of disability status, we further found that family removal was strongly associated with physical violence and violent threats. These findings underscore the unique position of Aboriginal and Torres Strait Islander people with disability, who are at heightened vulnerability of experiencing violence and threats due to overlapping forms of marginalisation and the invisibility of the Indigenous experience of disability. Accessing care is further complicated by the separate Indigenous, violence, and disability services which are difficult to navigate. These results underscore the need to develop policy approaches to ensure that service provision to Aboriginal and Torres Strait Islander people with disability is culturally safe and to ensure that disability support considers the vulnerability of people with disability to violence. --- --- Abbreviations NATSSIS: National Aboriginal and Torres Strait Islander Social Survey; ABS: Australian Bureau of Statistics; SDM: Short disability module; CURF: Confidentialised unit record file Authors' contributions JT performed the statistical analyses, participated in study design and contributed to the draft and final manuscript. HW, SA and MK participated in study design, as well as contributing to both the draft and final manuscript. YP and AF participated in the interpretation of the data, as well as contributing to the draft and final manuscript. The authors have read and approved the final manuscript. --- Funding JT was funded by the Australian Research Council through the Centre for Excellence in Population Ageing Research -CE1101029. This supported the data analysis and manuscript writing. --- --- --- Competing interests The authors declare that they have no competing interests. ---

Background: A recent Royal Commission into the treatment of Australians living with disabilities has underscored the considerable exposure to violence and harm in this population. Yet, little is known about exposure to violence among Aboriginal and Torres Strait Islander people living with disabilities. The objective of this paper was to examine the prevalence, disability correlates and aspects of violence and threats reported by Aboriginal and Torres Strait Islander people living with disabilities. Methods: Data from the 2014-15 National Aboriginal and Torres Strait Islander Social Survey were used to measure physical violence, violent threats and disability. Multivariable logistic and ordinal logistic regression models adjusted for complex survey design were used to examine the association between measures of disability and exposure to violence and violent threats. Results: In 2014-15, 17% of Aboriginal and Torres Strait Islander people aged 15-64 with disability experienced an instance of physical violence compared with 13% of those with no disability. Approximately 22% of those with a profound or severe disability reported experiencing the threat of physical violence. After adjusting for a comprehensive set of confounding factors and accounting for complex survey design, presence of a disability was associated with a 1.5 odds increase in exposure to physical violence (OR = 1.54 p < 0.001), violence with harm (OR = 1.55 p < 0.001), more frequent experience of violence (OR = 1.55 p < 0.001) and a 2.1 odds increase (OR = 2.13 p < 0.001) in exposure to violent threats. Severity of disability, higher numbers of disabling conditions as well as specific disability types (e.g., psychological or intellectual) were associated with increased odds of both physical violence and threats beyond this level. Independent of these effects, removal from one's natural family was strongly associated with experiences of physical violence and violent threats. Aboriginal and Torres Strait Islander women, regardless of disability status, were more likely to report partner or family violence, whereas men were more likely to report violence from other known individuals.

INTRODUCTION The area of Indonesia is 1.916.906, 77 km 2 , and it is famous as a number one archipelagic country in Southeast Asia by the number of Islands about 16.056. This condition causes Indonesia as heterogeneous nation of ethnicity, race, religion, culture, language, and customs. As information from BPS in 2010 that Indonesia has more than 300 ethnic groups which details about 1.340 ethnics1 . Indonesia is one of the world's largest multicultural countries with diversity of ethnic and culture. Each ethnic group has its own unique characteristics from language, customs, food, race, and culture. In addition, the Indonesian nation is also diverse in terms of religion. There are 6 legal religions in Indonesia namely Islam, Hinduism, Buddhism, Christianity, Catholicism and Confucianism. The above data shows that Indonesia is plural state. A plural country is a diverse country ethnicity, race, culture, religion, language and customs. This heterogeneity can be a great strength for the Indonesian nation to progress and develop. However, this diversity can also be a disaster or a threat towards the Indonesia integrity. Whenever it can happen, depends on how we respond to this diversity. If we are awareness respond it with mutual tolerance, respect for all kinds of the diversity, then this can create peace, unity and integrity as well as social justice which leads to the national progress. On the other hand, if our reaction is selfish, ethnocentrism, intolerance towards that diversity, then this can trigger various problems that will impact on divisions. The above matters can be avoided when Indonesian people are able to grow their awareness to respect each other. On the one hand, Indonesia's heterogeneity in terms of ethnicity, race, culture, religion, language and customs needs to be maintained and preserved so our that future generations will enjoy and perceive it. All of this can be done by education. According to Amin Abdullah, education is a powerful way to protect, preserve, defend, and conserve Indonesia's diversity from one to the next generation, from year to year, so that our descendants can enjoy, PAI Teacher Design to Develop Multicuturalism… Ilmuna: Jurnal Studi Pendidikan Agama Islam I Vol. 5, No.2 September ISSN: 2715-9981. EISSN: 2715-9434. DOI: https://doi.org/10.54437/ilmuna. see and perceive it. In addition, education can foster awareness within Indonesian society to respect one another among human beings. 2 One of the education that is able to stimulate and raise public or students' awareness to always maintain, preserve, and conserve diversity of ethnicity, race, culture, religion, language and traditions of the Indonesian nation that causes the forming of a sense of belonging, respect among human beings is multiculturalism. The implementation of multiculturalism education in madrasas is expected to be able to produce students who are polite, tolerant, humanist, respect and appreciate each other. However, the expectations have not matched the facts. It can be seen that recently education has been shaken by many acts of violence experienced both students and teachers which have snatched fatalities. Education should be a place for growing and seeding the peace values, but instead it becomes a vessel for violent behavior or actions that are inhumane and contrary to the humanism and justice principles. Recently, religious issues are often a source of social conflict 3 . The several cases appearance concerning about intolerance and SARA conflicts in society is not solely due to the religious education failure in madrasas, religious education nevertheless must be able to examine and find alternative solutions through the development of non-conventional-traditional religious education learning models. The Islamic Religious Education course is an important component in PAI learning which is expected to be able to provide a concrete description and comprehensive dealing with multicultural life as reflected from historical events of Islamic civilization. 4 The multicultural education theme has been widely researched, such as Irham's research which stated that PAI has a role as a motivator of moral pluralism for students. The implementation of PAI in the learning process is intended to shape students' attitudes and thoughts with 2 Programmable Logic Control, Togar Timoteus Gultom, and Stmik Itmi Medan, "Jurnal Ilmiah Maksitek Issn. 2655-4399" 5, no. 4 : 189-203. --- METHOD This study is a qualitative approach with a phenomenological model. Qualitative is a method that provides information in the form of sentences, or verbally derived from the observed object behavior. When going into the field, the researcher saw the actions of people in order to produce a valid source of information or in accordance with the research theme, and then the researcher selected informants to describe the problem under study. The type of this research is field research that is studying and understanding directly about the background conditions of the place, and social relations, both individually and as a group. This research relates to the design of PAI teachers in developing multicultural education at MAN 1 Karanganyar using the method of collecting data from observations, documentation and interviews. Observation is a type of activity that uses the five senses, such as sight, hearing, and smell, to gather information to be used in a study7 . Interview is meeting and conversation between two persons to get information directly about phenomena has been researched, to know what meaning of the topic which is studied. Based on this description, interviews can be conducted in an organized manner. Documentation is a record of things will be studied. It can be in the form of writing, pictures, or documents such as journals and other documents Interview can be done organized8 . The data then analyzed using reduction, display, and verification data. --- FINDING AND DISCUSSION --- A. The Nature of Multiculturalism Education Indonesia is a heterogeneous country in ethnicity, race, culture, religion, language and customs. So is called a pluralistic country. This plurality of Indonesia can lead to unity or division. This needs to be responded so that the pluralism and diversity does not trigger divisions by increasing Indonesian society awareness to respect one another among human beings. This awareness can be grown through multiculturalism education. Multicultural education is an educational process that provides equal opportunities to all generations without discrimination due to ethnic, cultural and religious differences. Furthermore, giving respect to diversity, and providing equal rights for ethnic minorities in order to strengthen national unity and nation's image to the world views 9 Multiculturalism education comes from two terms they are education and multiculturalism. Education is interpreted by some educational experts as a conscious effort to help students become more mature individuals or in other words maturing students by directing, guiding, training, understanding and explaining. Meanwhile, multiculturalism comes from 3 terms, namely multi, culture, and ism. The word multi means many, then culture is defined as custom, while ism is defined as an understanding or view. So multiculturalism means understanding cultural diversity in terms of equality 10 . Parsudi Suparlan argues that multiculturalism is an understanding that recognizes and respects heterogeneity in equality both individually and culturally. 11 Thus the conclusion is, multiculturalism education is an effort to guide, train, explain, direct, foster feelings to acknowledge, appreciate and respect the heterogeneity that exists both from ethnicity, culture, race, religion, and customs. Muhaemin el Ma'hady defined multiculturalism education as an education about cultural heterogeneity in responding to the population and culture of a particular community environment changes Heru Suparman stated that multicultural education at least has the characteristics of providing education to students to learn the following; --- Living in heterogeneous society Heterogeneity causes differences between one another, so the solution to avoid divisions is cultivating mutual respect for these differences, so that a peaceful, comfortable, united, harmonious and compact life will be realized. Heterogeneity has been explained in Q.S. Al-Hujurat verse 13 that humans were created by Allah SWT for various genders, ethnicities, nations, different cultures. --- Constructing three Mutual perspectives These three mutual perspectives are fostering trust between people, caring for each other, and upholding tolerance between people. These three can be provisions for living in the midst of pluralism or diversity. Because this is able to prepare students who have a tolerant attitude of humanism, this democracy can be actualized with an attitude of mutual respect by prioritizing unity and ukhuwah. --- Open Minded Through education, students are equipped with the ability to be openminded towards new cultural differences so that students are able to adopt, adapt, --- Overcoming problems without violence Looking for solutions of the problems with a cool head, or resolved by discussion. So the teachers are expected to be able to invite and grow themselves and students' souls to love peace and not be easily provoked. Islam is a religion that invites and teaches its people always to be peaceful and if people discuss a problem, it is advisable to solve by discussion without violence.14 --- B. PAI Teacher Design in Developing Multiculturalism Education in Madrasas The students of Madrasas learn both Islamic religious and general science however; madrasas more emphasize Islamic religious education. All madrasah students are Muslim. This similarity can be strength to live in harmony. Religious homogeneity nevertheless, can trigger divisions among students, because there are differences Madzab followed by students which be able to trigger students divisions, because they feel that their Madzab is the most correct of all. Besides Madzab, their heterogeneous backgrounds from ethnicity, race, culture, family socio-economic status, ability to environment. So, the role of PAI teacher is needed to provide their students' understanding related to these differences. PAI teachers of madrasah are expected to be able to produce generations that has tolerance or tasamuh, respects one another by changing the idea of exclusivism to be universalism. This is done to anticipate extremism, that is considering the most righteous, intolerant, and dehumanize with heterogeneity15 . In this case, PAI teachers have a big responsibility to design learning models that support the development multiculturalism understanding among madrasah students. In terms of terminology, the teacher is a person whose main duties and functions to guide and provide direction through learning that is carried out inside or outside classroom. In this case, the teacher is not only a person who provides understanding and explanation a material in front of the class, but they must be actively involved, creative and innovative and have humanist spirit in stimulating the development of their students to become more mature people, in other words the teacher have responsibility to guide the process of maturing students 17 Teachers play a crucial role in growing and actualizing the values of pluralism that are comprehensive and conservative and principled on human values in madrasas, thus, the role of the teacher will be realized. Besides, Teachers also play an important role in the multiculturalism education development. Whether or not the success of understanding multiculturalism education for students, depend on the teachers. Those who do not understand their own cultural background and other cultures are unlikely to be successful in implementing multiculturalism education. Therefore, the teachers must have an understanding of their own cultural background and across cultures. There are four teachers' roles to develop multiculturalism education, such as educator, facilitator, accommodator, and assimilator. All teachers understand their role as educator and facilitator. The teacher role as an accommodator and assimilator are: wise, understands what students need and is able to accommodate their diverse opinions, desire and abilities. Students in the classroom come from various ethnicities, religions, backgrounds, so the teacher's role as an assimilator can unite this difference which causes divisions. Teachers must also provide the students' special needs. Teacher roles in developing multiculturalism education, according to Muslimin as follows: 1. Teachers must be fair, democratic, and nondiscriminatory. 2. Teachers must be sensitive and responsive towards cases related to religion. For example, when there are fights, anarchic demonstrations, riots, and chaos, the teacher must be able to provide an explanation regarding his concern for the incident. 3. The teacher must explain that every religion prohibits violence, it only teaches love of peace, humanity, tranquility and prosperity. 4. The teacher must be able to stimulate and explain their students that all problems whatever the causes can be solved peacefully 18 . The learning plan must be well prepared in order to create a comfortable and calm atmosphere. The learning design must be based on social-multicultural values that integrate three student learning goals . To make the diverse students work together, teachers must prioritize the affective and psychomotor domains. Multiculturalism education does not only build students' knowledge, but also fosters and guides them become inclusive, respect and respect differences, democratic, and respect human rights. Therefore, diverse backgrounds educators and education staff are needed; the teachers are also implementing the "integration model" theory" 19 . --- B.PAI Teacher Design To Develop Multicuturalism Education At Madrasah Aliyah Islamic education with a multicultural perspective is an education that opens broader visions and horizons; they are able to cross ethnic group boundaries or religious traditions so that they are able to see "humanity" as a family that has different or similar ideas. Multicultural religious education carries social education to insert awareness of living together in diversity and cultural differences. This education is built on the spirit of equality relations, mutual trust, mutual understanding, and respect for similarities, differences, uniqueness and interdependence. This is an integral and comprehensive, innovation and reform in the content of religious education which provides new forms of information about religions that are free from prejudice and racism. Multicultural religious education provides recognition of pluralism, learning tools for cross-border encounters and transforming indoctrination towards a positive dialogue 20 . Teachers master a qualified understanding of the concepts and patrons of multiculturalism education. Teachers need to understand that multiculturalism education does not only explain other cultures to students, but they have to be able to create multicultural oriented conditions that prioritize social justice for students. 18 Muslimin, "Muslimin-Pendidikan-Multikultural-Sebagai-Perekat-Budaya-Nusantara-Menuju-Indonesia-Yang-Lebih-Baik.Pdf" . 19 This is in accordance with the theory that Islamic religious teachers in implementing their education and teaching are expected to implement multicultural values. According to him teachers must design innovative strategies. Such as: Islamic religious teachers must be aware of the students' ethnic diversity; the teachers must reflect on ethnic diversity with an attitude of respect for the differences; the values that are considered sacred by a community are not to be abused and not mentioned in educational practice21 . Teachers as educators must work together with school management and all education staff to develop the learning quality in order to improve student achievement, including: 1. The teachers must create classroom atmosphere to be comfortable, calm, fun, conducive and not stressful. 2. The teachers explain to their students, that they are free to access information sources and teaching materials that support the learning process from anywhere. 3. The teachers can apply various cooperative learning strategies. This learning strategy can establish cooperation between students so that it can eliminate the exclusivity of students who come from well-established families. In addition, cooperative learning can foster tolerance attitude that is manifested by mutual understanding and respect for all heterogeneity . 4. The teachers can encourage students to do their assignments based on in-depth studies. 5. The teachers must have a notebook contains all of students' progress during the teaching and learning activities, both individually and in groups. To minimize classroom division, the teacher can ask students to tell in front of the class about their background where they come from. It is hoped that students to know, understand and respect ethnic, racial, religious and cultural backgrounds among them. The teachers may use Imre Lakatos's theory about the truth in subject matter. According to Lakatos "truth is divided into two, namely truth that is absolutely true This shows that every human being or student has differences and characteristics. This phenomenon is in accordance with the theory about multicultural education is an educational strategy that is applied to all subjects. By using the students cultural differences the learning process becomes effective and easy 22 . In this case the teacher must explain the students that every human being has differences, therefore we must respect towards these differences in order to create a life that is harmonious, humanism, and comfortable. PAI teachers are able to do it by developing multiculturalism education for students. There are 4 clusters of PAI subjects in madrasas, namely akidah akhlak, fiqh, qur'an hadith and SKI. Through akidah akhlak, PAI Teacher is expected to be able to foster and emphasize the importance of living in harmony and establishing brotherhood among religious communities. For Qur'an hadith subject, PAI teachers explain the verses of the Qur'an and hadith regarding diversity, tolerance, brotherhood and living in harmony, so PAI lessons, is not only require students to be able to memorize the material but the students must be applied it in everyday life. And the Akidah Akhlak subject, PAI teachers must be able to create good moral generation either their relationship with Allah or human beings. Among the Islamic religious ideals written in the Qur'an is to know each other and respect various cultures, races and religions as one humanity. However, at the same time many conflicts in this world are caused by SARA. This gap between ideality and reality must be bridged by providing a multicultural understanding in the process of The next step is the teachers determine the learning strategies, models, methods, facilities, and learning media, then the teacher combines them. In Blended Learning online, the teacher provides a platform contains learning materials and sources which can be accessed by students at any time. Then offline learning services, the teaching and learning process is carried out face-to-face. When the face-to-face is conducted, the teachers make effort to be able to foster self-awareness in heterogeneous students through the strategies, and learning methods used. By using the cooperative learning model, PAI teachers at MAN 1 Karanganyar hope the students can work together, respect, help each other. --- CONCLUSION Multiculturalism education in Indonesia is a discourse to realize national unity over diversity and problems that often occur. To realize these learning objectives, the in accordance with the students' characteristics. Based on the explanation above, it can be concluded: . Madrasas must be able to teach and realize multicultural education earlier. The schools must develop all aspects based on socio-multicultural values from the vision, mission and objectives, curriculum, madrasah facilities, educators and students. Teachers must work together with school management to realize multiculturalism education. The diversity of ethnic, racial, religious, cultural and ethnic backgrounds in Indonesia can be represented by teachers and students, madrasas in this case must view them as individuals who are diverse and have their own characteristics. . The teachers use various designs, approaches, strategies and learning methods and appropriate with the conditions and students characteristics to foster awareness of multiculturalism education. . At least the teacher as an educator has four roles in realizing multiculturalism education, such as an educator, facilitator, accommodator, and assimilator. And In delivering learning material the teacher can use Imre Lakatos' theory so that the teacher is able to classify between absolute and relative truth. The steps to be a multicultural minded teacher is: First, the teachers do not discriminate against gender, for example selecting class administrators from male and female students. Second, Teachers must look over their students' different backgrounds and characteristics. Third the teacher must be able to foster an awareness of mutual tolerance, respect for each other. This can be done by choosing a learning model which creates cooperation and harmony, such as cooperative learning. Fourth, the teacher can roll student seats to achieve fairness for all students. Fifth, the teacher provides equal opportunities for all students, both male and female, to ask questions and present their arguments. And the six teachers can integrate learning materials with the diversity of ethnicities, races, religions, cultures, ethnicities and customs.

this research is aimed describe how PAI teacher design in developing educational multiculturalism at MAN 1 Karanganyar. This research used descriptive qualitative approach and PAI teacher of the school as informants. In collecting data, the researcher used observation, documentation, and interview. The research results are the teachers have an understanding of diversity, multiculturalism, and hold on to human principles and values to develop educational multiculturalism. When the teaching and learning process the teachers mostly has used various approaches, strategies, and methods in accordance students' characteristic that is implementing Negotiated Blended Learning. The teachers also avoided discriminatory effect by asking class administrator such as chairman and vice, also male and female students as gender representative. PAI teachers considered that the students as individual, who are from different backgrounds and characteristics, so that the teachers are able to foster awareness of mutual tolerance, respect each other. It was done by selecting learning model which cause to emerge cooperation and harmony that is cooperative learning. In creating equity for students, the teachers rolled students' seats, and all students male and female were given equal opportunity to ask, and express their opinion. The next, the teachers has integrated learning materials with tribes, races, religions, cultures, ethnicities, and customs diversity in Indonesia.

Introduction Depression is one of the most prevalent mental disorders in the elderly population and is associated with disability, mortality, and a lower quality of life . The prevalence rates vary by age and peak in older adulthood . Elderly individuals who have died of suicide or have a history of suicidal behavior are more likely to have experienced depression . Biological risk factors for geriatric depression include cardiovascular diseases , stroke , brain injury , inflammation , and other chronic medical diseases . Studies have shown that socially related factors contribute significantly to depression among the elderly population . Social health is an important dimension of health, alongside physical and mental health, according to the definition proposed by the World Health Organization . Social health contains both individual and social aspects . The individual level of social health is usually explained in terms of "well-being" or "adjustment" and can be measured through social support and social adjustment. Social support places an emphasis on the level of support the subject receives from others, while social adjustment focuses on the adaptive capacity of subjects to interact actively with their community. The social level of social health refers to a society in which there is equal opportunity for all and access by all to the goods and services essential to full functioning as a citizen , such as the perceived built environment and community management. Although there are numerous studies on the socially related factors among the elderly population in China , few studies have combined the two levels of social health in one study. There have been studies conducted to explore the relationships between social factors and depressive symptoms among community-dwelling elderly adults . A study conducted in Macau showed that poor family support and social network were related to depression among 31 communitydwelling old people . A crosssectional study in Shanxi found that social relationships were a significant influencing factor for depression among empty-nest elderly . Results from the National Social Life, Health, and Aging Project showed that relationships with family and friends were associated with depressive symptoms . However, previous studies failed to combine individual and community levels of social health into one study. Furthermore, evidence from the Chinese older population is still scarce. A comprehensive structured scale called the Social Health Scale for the Elderly has been developed to fill the gap in social health status measurement . In this study, we aimed to test the hypothesis that better social health status was significantly associated with a reduced risk of depressive symptoms among a large sample of communitydwelling elderly adults in Zhejiang Province, China. --- Materials and methods --- Study participants We conducted a cross-sectional study among the community-dwelling elderly adults in Zhejiang Province from Jan. 1, 2018 to Sept. 30, 2018. A four-stage convenient sampling method was implemented to recruit participants, as shown in Fig. 1. The inclusion criteria were age of 60 years or older and living in the local community for more than five years; these were conveniently sampled in the current study. Elderly people who were bed-ridden, had serious physiological or psychological illness, and/or had hearing disorders were excluded. A face-to-face interview was conducted with each participant to complete a structured questionnaire. The interview was conducted at the community health service center by well-trained medical staff or medical students. Proxy-based interviews or completion of the questionnaire by the participants themselves was not allowed in the current study. This study was approved by the Ethics Committee of School of Medicine, Zhejiang University, Hangzhou, China. Written informed consent was obtained from all participants before the face-to-face interview. --- Assessment of social health We used short version of the SHSE, the SHSE-S, to evaluate social health status. There are 14 items on this scale. The SHSE has been established and validated for the Hangzhou elderly population. The testretest variability was 0.77, internal consistency reliability was 0.79, concurrent validity was 0.64, and goodness of fit was 0.97 in construct validity . The raw score for the SHSE was calculated, standardized, and categorized into three levels for each participant according to the established Hangzhou norms . Participants with a standardized score of less than 40 were categorized as having "poor social health," those with a standardized score of 40-59 were categorized as having "moderate social health," and those with a standardized score of 60 or above were categorized as having "good social health." --- Assessment of depressive symptoms The short form of the Geriatric Depression Scale was used to assess depressive symptoms in this study. The GDS-S has 15 items coded as 0 or 1 . It has been widely used in numerous studies and appears to be a useful screening instrument for depression in geriatric populations . For an Asian population, the overall Cronbach's α is 0.80, the intra-class coefficient of test-retest reliability is 0.83, and inter-rater reliability is 0.94 . In this study, participants with a total GDS-15 score of 8 or above were classified as having depressive symptoms . --- Assessment of covariates Demographic characteristics-age, gender, region, living arrangements, occupation, ethnic status, marital status, education level, and income-were requested of the participants. Body mass index was calculated as weight in kilograms divided by height in meters squared. Comorbidities, the prevalence of 27 common chronic diseases, were determined using the question "Do you suffer, or have you been told by a doctor within the past year that you were suffering, from any of the following problems?" Chronic diseases included hypertension, ischemic heart disease, cerebrovascular disease, diabetes, chronic obstructive pulmonary disease, Parkinson's disease, arthritis, and osteoarthritis. The number of chronic diseases for each participant was also calculated. Multimorbidity was categorized into four levels . --- Statistical analysis Continuous variables with normal distribution were presented as mean±standard deviation , while non-normally distributed variables were presented as median±interquartile range . Categorical variables were presented as frequency and proportions. Differences between depressed and non-depressed participants were tested using a t-test or non-parametric Wilcoxon test for continuous . Covariates that differed significantly between depressed and non-depressed participants were further adjusted in logistic regression models. Three logistic regression models were used to analyze the association between social health and depressive symptoms. Covariates were selected for the logistic regression model on the basis of a significantly different distribution between depressed and nondepressed groups and literature showing that this factor was associated with depressive symptoms . Model I included social health , gender , age group , and region . Model II was adjusted for the same covariates in Model I plus living arrangements , marital status , and income per month . Model III was adjusted for the same covariates in Model II plus BMI , frequency of physical exercise , hypertension , diabetes , and coronary disease . The Bonferroni test was used for multiple comparisons in the logistic regression model. The enter method was used to perform the multivariate logistic regression. Tests for linear trends across different levels of social health status were examined by entering the social health levels into the logistic regression models. Subgroup analyses stratified by gender , age , region , living arrangement , marital status , and multimorbidity were conducted and tested to determine if any effect modification existed. All P values were two-sided and P<0.05 was considered statistically significant. All analyses were performed in R software 3.3.2 . --- Results --- General information on participants A total of 4106 participants completed the faceto-face interviews. We excluded 181 participants with incomplete information on the SHSE and 168 participants with incomplete information on the GDS, meaning that 3757 participants were enrolled in the final analysis. Of these, 1887 were female and the mean±SD age was years. In addition, 1444 participants lived in an urban area while 2313 participants lived in a rural area. With regard to living arrangements, 318 participants lived alone while 3439 participants lived with their family. A total of 1070 participants were illiterate or did not finish primary school while 706 had an education level of high school or above. --- Demographic characteristics of depressed and non-depressed participants Depressive symptoms were present in 25.92% of the study sample. The differences of demographic characteristics between depressed and non-depressed participants are presented in Table 1. Depressed participants had a higher proportion of females and were older than non-depressed participants . Those elderly participants with depressive symptoms were more likely to live alone and to be widowed . The difference in education level between depressed and nondepressed groups was not significant. --- Health-related characteristics of depressed and non-depressed participants Differences between depressed and non-depressed participants in terms of health-related characteristics are presented in Table 2. A higher proportion of nondepressed participants engaged in physical exercise than did depressed participants . Depressed elderly adults had a higher rate of hypertension , diabetes , and coronary heart disease . --- Social health differences between depressed and non-depressed participants Depressed participants had a lower social health status than did non-depressed participants . The raw score was standardized and categorized into three levels according to the previously established Hangzhou norms. A higher proportion of non-depressed participants had good social health than did depressed participants . The logistic regression showed an inverse association between social health status and depressive symptoms. Odds ratios ) for social health level and risk of having depressive symptoms are presented in Table 3. Participants with a moderate social health status had lower OR ) of having depressive symptoms than did participants with poor social health status, while the OR was 0.44 for participants with good social health status. The linear trend of association of depressive symptoms across different levels of social health was also significant . The results were consistent when other covariates were adjusted in Model II and Model III . Subgroup analyses stratified by gender, age, region, living arrangements, and multimorbidity showed that the protective effect of social health against depressive symptoms was significant in all subgroups . The results also showed that the effects were stronger in certain subgroups, such as participants living alone and participants aged 75 years or older . --- Discussion In this study, we found a significant inverse relationship between higher social health level and the risk of having depressive symptoms among 3757 Chinese community-dwelling elderly adults. Participants with a good social health level had a significantly lower OR for depressive symptoms than did those participants with poor social health status, after adjusting for age, gender, region, living arrangements, marital status, income, BMI, physical exercise, prevalence of hypertension, diabetes, and coronary heart disease. The results indicate that better social health status is associated with good mental health among community-dwelling elderly adults. --- Fig. 2 Subgroup analysis of association between social health status and depressive symptoms using the "poor social health" participants as the reference group Our study found that the rate of depressive symptoms in a sample of Zhejiang community-based elderly adults was 25.92%. A previous populationbased study using the Center for Epidemiologic Studies Depression Scale showed the prevalence of self-reported depressive symptoms to be 39.86% . Another study conducted in Taiwan by Chiu et al. showed the prevalence of depression to be 20.1%. This heterogeneity in the prevalence of depression may be due to the different populations and different methods of measuring depression. Further population-based studies in national representative samples are still needed. Previous studies have been conducted to explore the relationships between social factors and depressive symptoms among community-dwelling elderly adults . Chen et al. found that support from the neighborhood significantly affected depressive symptoms among low-income elders. Cao et al. found that social support mediated the relationship between social capital and geriatric depression. The study by Mechakra-Tahiri et al. indicated that social support and a lack of conflict in intimate relationships were associated with lower prevalence of depression in the elderly population of Quebec . Our study combined social support, social adjustment, and perceived neighborhood environment together and confirmed that social health was closely associated with geriatric depressive symptoms, which indicates that improvements to social health status may be an effective way to prevent depression in the older population. Our findings have some implications for clinical practice for the elderly adults. Given the close association between the social health status and geriatric depressive symptoms, it is essential to identify those elderly adults with social health deficits. The 14-item SHSE scale could serve that purpose alongside faceto-face interviews. Those elderly adults with vulnerable social health levels can be identified by comparing their scores to the established norm. Although this study helps us characterize the nature of social health status and its relationship to depression, there are still several things to do before the SHSE can be used to guide targeted clinical interventions. First, the normative data for the subscales of the SHSE need to be developed so that the social health deficits of specific individuals can be identified more precisely. Second, further detailed systematic assessments other than the GDS and their association with the SHSE need to be conducted. Our study also confirmed several health-related factors associated with depressive symptoms, including gender, physical exercise, and number of comorbidities . We observed that women had a significantly higher rate of depressive symptoms than did men, consistent with other studies . Physical exercise or activities may enhance social health among the elderly. A recent study found that horticultural therapy, which involves physical activity and social interaction in a park environment, improves the mental health of the elderly . Our study also confirms that elderly adults with more chronic diseases have a higher risk of being depressed . The protective effect of social health remained significant after adjusting for these covariates, which strengthened the robustness of our results. Mobile health technology has been found to be beneficial in managing various psychiatric conditions in the elderly . Further research is required to evaluate the impact of mobile health technology on social health in the elderly. The strengths of our study include its large sample size, detailed covariates, and reliable evaluation of social health status. However, there are several limitations that should be noted when interpreting our results. First, only association can be inferred from the current study, not causation, due to the inherent nature of the cross-sectional survey design. It should be noted that depressive symptoms can also have an effect on social health status, in terms of social participation, for example. Second, the representativeness of the study sample cannot be estimated due to the convenience sampling strategy and lack of data for those elderly people who refused to participate, limiting the generalizability of our results. The study sample was a little healthier than the elderly population as a whole. Caution should be taken when extrapolating our findings. Third, we used the GDS-15 scale to measure depressive symptoms among the elderly population, which may have caused an overestimation of the rate of depression when compared to structured interviews. We chose the GDS-15 to measure depression mainly because of the low time costs of this scale, which can overcome fatigue and distraction during the interviewing of elderly adults. The GDS scale also has acceptable reliability and validity and has been widely used for the elderly Chinese population. Fourth, selection bias was inevitable, although no cash was given to participants who completed the interview, only gifts worth 10 CNY. Finally, further studies are needed to explore how the SHSE scale performs in evaluating social health level in comparison to other scales. --- Conclusions In this study, we observed an inverse association between social health status and depressive symptoms among a large community-based sample of elderly adults. The association between social health and depressive symptoms was consistent after adjusting for confounding factors. The SHSE may serve as an efficient screener to identify those with social health deficits, but systematic assessment to guide intervention merits further investigation. --- Contributors Kun CHEN is the guarantor of this work and had full access to all of the data in the study. Hong-lei DAI and Zhe-bin YU conceptualized the study and designed the analysis plan. Zhe-bin YU performed all the statistical analyses and Hong-lei DAI drafted the manuscript. Liu-qing YOU revised the manuscript. Min-hua FAN, Hong-wei ZHU, Dan-jie JIANG, Mengyin WU, Shu-juan LIN, and Xiao-cong ZHANG contributed to the acquisition and interpretation of the data. All authors read and approved the final manuscript. --- Compliance with ethics guidelines Hong-lei DAI, Zhe-bin YU, Liu-qing YOU, Min-hua FAN, Hong-wei ZHU, Dan-jie JIANG, Meng-yin WU, Shu-juan LIN, Xiao-cong ZHANG, and Kun CHEN declare that they have no conflict of interest. All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation and with the Helsinki Declaration of 1975Declaration of , as revised in 2008 . Informed consent was obtained from all participants for being included in the study.

Objective: Mental disorders of the elderly population in China deserve attention. Social health is significantly associated with depression. This study aimed to evaluate the rate of depressive symptoms and to test the relationships between social health and depressive symptoms among a large sample of community-dwelling elderly adults. Methods: We conducted a cross-sectional study among community-dwelling adults aged 60 years or above in Zhejiang Province, China. Face-to-face interviews were used to complete a structured questionnaire for all participants. We used the Social Health Scale for the Elderly (SHSE) to evaluate social health status and used the short form of the Geriatric Depression Scale to evaluate depressive symptoms. Multivariate logistic regression was used to evaluate the association between social health status and depressive symptoms. Results: Of the total of 3757 participants included, 1887 (50.23%) were female, and the mean±standard deviation (SD) age was (70.0±8.3) years. The rate of depressive symptoms was 25.92%. The social health score was higher in non-depressed participants than in depressed participants (raw score 50.7 vs. 48.3, P<0.001). Participants with "moderate" or "good" social health had a significantly lower risk of depressive symptoms than those with "poor" social health (odds ratio (OR)=0.55, 95% confidence interval (CI): 0.46-0.66 for moderate social health; OR=0.45, 95% CI: 0.35-0.60 for good social health). The association between social health and depressive symptoms was consistent across several subgroups. Conclusions: Social health is significantly inversely associated with depressive symptoms. The SHSE may serve as an efficient screener to identify those elderly adults with social health deficits, but systematic assessment to guide intervention merits further investigation.

Our species is genetically quite variable. Taking advantage of the availability of public genome databases, a map was built of not less than 1.42 million single nucleotide polymorphisms . This nucleotide diversity varies greatly across the genome, but it is estimated that one SNP occurs at every 1.9 kilobases of DNA . Most recent estimates indicate that the number of SNPs in our genome well exceeds two million . These are large figures, but our closest biological relatives, chimpanzees and bonobos , are even more variable. Estimates have been made that for mitochondrial DNA these two species show 2-to 4-fold higher levels of diversity, and for the Y-chromosome 6-7 times more variation. For the X-chromosome and autosomal loci the situation is not so clear, but in general higher levels of nucleotide variation were found in Pan . The molecular methods which can be used to evaluate our diversity are diverse. Examples are those that determine restriction fragment length polymorphisms and short tandem repeat polymorphisms , besides the SNPs cited above, which can be assessed through sequencing. The changes involve not only single nucleotide changes, but also deletions, insertions, as well as more complex rearrangements. The picture that is emerging is that this variability is structured, as was verified by Rosenberg et al. , after studying 377 autosomal microsatellite loci in 1056 individuals from 52 world populations. Despite the fact that within-population differences among individuals account for 93% to 95% of the variation , without using prior information about the origins of individuals clusters have been identified corresponding to the major world geographic regions. Similarly, Barbujani and Romualdi verified that using more than 20 loci, by discriminant analyses, 80% of the genotypes could be attributed to the right continent. Bamshad et al. assayed 100 Alu insertion polymorphisms in a heterogeneous collection of 565 individuals, 200 of whom were also typed for 60 microsatellites. Correct assignment to the continent of origin reached 99% to 100% when more than 100 loci were used. Due to different evolutionary histories, the pattern of recombination found in the DNA of peoples from different continents is diverse. Gabriel et al. characterized haplotype patterns across 51 autosomal regions in samples from Africa, Europe, and Asia. They found that half of the human genome exists in blocks of 22 kilobases or larger in African or African-derived populations, while these blocks are much larger in European and Asian groups. To evaluate the significance of these findings in social terms, it is necessary to go back to the past, examining how these intercontinental differences had been considered and interpreted, returning afterwards to our time. --- THE RACE CONCEPT Inhabitants of different continents can be usually distinguished morphologically, and George Louis Leclerc de Buffon was the first scholar who applied the term race to different varieties of Homo sapiens . Initially the static idea of "racial types" was established. This concept was changed in the first half of the 20 th century, with the incorporation of population genetics principles in biological thought. It was then emphasized that race is a population unit . Templeton questioned the occurrence of human races. According to him these entities cannot be considered as distinct evolutionary lineages, since genetic differentiation alone is insufficient to define a subspecies , an additional requirement being historical continuity. Recent admixture would provide sufficient genetic contact to make all of humanity a single lineage sharing a common evolutionary fate. Another point to be emphasized is that race is also a social concept, and biologically different persons may also have different cultures. Cultural units can be characterized as ethnic groups . An ethnic group can be a nation, or groups defined by language, social customs or religion. Since the term is intentionally vague Ashley Montagu was one of the first scholars to suggest that the word race should be substituted for ethnic group. This suggestion has been followed in our most recent works. Independently of names the question remains, is that unit a biological entity? The studies mentioned in the introduction clearly indicate that continental groups can be clearly distinguished molecularly. --- CLASSIFICATION PROBLEMS Morphological features, such as skin color, type and color of the hair, nose and lip characteristics are often used by people for self-classification or the categorization of others. In certain communities, however, the descent rule prevails. If a person has an ancestor, no matter how remote, from a given continent, she/he would be classified as derived from that region independently of physical features. In communities where interethnic marriages are rigidly controlled by customs or law, and segregation policies are enforced, such classifications would be relatively easy to apply. This situation, however, is not generally true in Latin America, where interethnic crosses are widespread. Therefore, complex classifications such as the 16 categories developed in 18 th century Spanish America were never socially or legally enforced . In Brazil, traditionally, the census officers used four categories of skin color ; but when the National Research on Sampling by Domicile offered an open question for ethnic self-classification not less than 190 different denominations appeared . Matching problems can occur even with the first, rough classification. Table 1 shows the proportion of agreement between the interviewer classification and self-classification in a São Paulo, Brazil, survey conducted in 1986. As can be seen, there was only 79% of agreement in the classification of whites, and the concordance in the brown and black categories was even lower . --- ASSORTATIVE MATING Mating does not occur at random in relation to ethnicity, since persons from different ethnic groups may differ not only in morphology, but also in social class, culture, and other variables. Table 2 presents information about such unions in a large sample of the Brazilian population. There are about 11x more black x black unions and 42x more yellow x yellow unions than those expected by chance. Ethnic endogamy is clearly a prevalent event. The direction of the interethnic gene flow also differs according to the sex of the spouses; there are 1.35x more unions in the situation when the wife has a lighter skin than vice versa. Men, if they can choose, marry women with lighter skin. This may be a reflection of the higher social status generally associated with "whiteness". But the situation was different in the past, Colonial times, as will be shown by the genetic data to be presented in later sections. --- HISTORICAL GENETICS -ADMIXTURE QUANTIFICATION If a given genetic variant is present in one ethnic group only and appears in a mixed population, then it can provide a direct assessment of the historical, accumulated contribution of that ethnic group to the admixed one. Szathmáry and Reed provided details as how to deal with this type of data. The next simplest case is the one in which just two parental and one descendant population are considered. Allele frequency data can then be used in a model of simple dilution when just one system is available . An extension of this method was devised by Ottensooser that could deal with three parental stocks. More complex methodologies, that require electronic computation, were afterwards developed. Three much used methods are those of maximum likelihood ; weighted least squares and gene identity . The method of weighted averages, developed by Cavalli-Sforza and Bodmer has the limitation that it cannot be used for multiallelic loci . Bertorelle and Excoffier and Dupanloup and Bertorelle developed a method that is based on a coalescent approach, which explicitly takes into account molecular information as well as gene frequencies. Chikhi et al. devised a likelihood-based approach using a Markov chain Monte Carlo methodology that also takes into account drift since the admixture event, variation caused by sampling, and uncertainty in the estimates of the ancestral allele frequencies. Their method, however, can only be used in cases in which just two parental populations are involved. Early in these studies it was realized that the best ethnicity estimators are those with large allele frequency differences between the parental groups, and Shriver et al. provided a list of such alleles; they and McKeighe et al. also devised a Bayesian log-likelihood analysis for individual and population ethnic-affiliation estimation that was afterwards applied by Parra et al. to Brazilian data. Parra et al. and Collins-Schramm et al. also used alleles with wide allele differences between parental groups in their analyses. Relethford remarked that since the apportionment of genetic diversity in skin color is atypical it could not be used for purposes of classification. In agreement with that Parra et al. could find no clear association between individual ethnic evaluations based on 10 population-specific alleles and skin pigmentation coupled with other morphological features, subjectively evaluated, in 173 admixed Brazilians. But conversely, Shriver et al. found significant correlations between estimates of individual ethnic ancestry based on a panel of 34 ancestry informative markers and skin pigmentation, as measured by reflectometry, in African-American and African-Caribbean samples. --- HISTORICAL GENETICS -ETHNIC PARENTAL CONTRIBUTIONS IN LATIN AMERICA Sans made a review of the admixture studies performed in Latin America during the 20 th century. She classified the populations studied in three categories and was able to list estimates in Central and South America as follows: urban, CA:8; SA:17; Amerindian, CA:7; SA:17; and Africanderived, CA:3; SA:11. Most of these studies included blood group plus protein markers, and the number of systems considered varied widely. It was clear, however, that the process of admixture is universal in this region of the world, the relative contributions of African, European, and Amerindian gene pools varying in accordance with historical circumstances. As a part of their global evaluation of the Latin American microcosm, Salzano and Bortolini also investigated the processes and consequences of interethnic gene flow in the area. Their population classification, however, was different from that of Sans . The study included both previous and the calculation of new estimates, whose number can be categorized as follows: European-derived populations, previous estimates, CA:1; SA:17; new: CA:4; SA:5; African-derived, previous estimates, CA:12; SA:18; new: CA:12; SA:4; and Mestizo or unclassified in relation to continental groups, CA:19; SA:50; new: CA:15; SA:10. Therefore, no less than 63 Central American and 104 South American evaluations were considered, indicating a good number of studies of this subject. The quality of the data, however, varied markedly among surveys; the reader are referred to the indicated publication for details. As an example of whole-country evaluations, focus will be given to a Central American and a South American populations. Data related to the first is given in Table 3. While in former times it was believed that the people of Costa Rica was essentially of European derivation, the results presented in Table 3 clearly contradict this notion. Only about half of the genes in the Chorotega, Atlantic, and South regions are estimated to have this derivation, this fraction raising to 65%-66% in the two other regions. The African influence is most marked in the northern Pacific and Atlantic coasts, while the Amerindian contribution reaches more than one-third of the total in the above first mentioned regions. The Brazilian results are listed in Table 4. Results based on two types of autosomal , as well as mtDNA and Y-chromosome markers, are presented there. Comparing first the two sets of autosome figures, it is seen that the estimated European-derived fraction is markedly larger in the North and Southeast regions when the STRP system is employed . This is most probably due to the fact a Source: Morera et al. ). The maximum likelihood method of Krieger et al. that the sample tested for STRPs was derived from people seeking paternity determinations in a private institution, and most of them had to pay for the tests. Since there are marked differences in socioeconomic status between European-derived and African-or Amerindian-derived persons , it is possible that this sample presents a higher fraction of persons of European descent. On the other hand, the two sets of results give values which are consistent with the Brazilian history, pointing to a higher Amerindian influence in the North and Center-West regions. Also according to history, the blood group + protein results suggest higher African contributions in the Northeast and Southeast regions. The mtDNA and Y-chromosome data presented in Table 4 emphasize a now well-established fact, that interethnic crosses were asymmetrical in relation to sex in the Brazilian past. European males and African or Amerindian females contributed disproportionally more to the process. The Y-chromosomes are overwhelmingly of European origin, while African and Amerindian mtDNA types are especially frequent respectively in the Northeast and North regions. . The method used for the evaluations based on blood groups + proteins and STRPs was the same . The mtDNA and Y-chromosome estimates were obtained by the direct identification of ethnic-specific haplogroups. b Median number c Modal number d 302 base pairs of the HVS-I, 265 bp of the HVS-II regions, plus 16 restriction enzymes for the studies in the southeastern region; those involving the other regions included the HVS-I and restriction enzyme tests only. e Twelve unique-event polymorphisms plus PN2 ARMS determination using two specific primers, and SSCP tests for SRY-1532. Ribeiro ) asserted that the history of the conquest of the Americas by Europeans determined the establishment of three types of societies, the transplanted, witness, and new peoples. The first would be formed by descendants of the conquerors who adopted practically without changes the culture and ways of living of the dominant persons. Witness peoples are those who already had a highly developed and complex society, but were forced to witness the new events which occurred in the Conquest without being much affected by them. Finally, the new peoples would have been formed by the complex interaction of the three main groups of founders, Europeans, Africans, and Amerindians. An analogy can be made of this typology for the Latin American genomes. The Y-chromosomes of Latin Americans have been mainly transplanted from outside; their mtDNA, however, received much less outside influences, and can be classified as an witness of the process. This terminology is also adequate, because it indicates the unique possibility of rescuing part of the lost history of extinct aboriginal groups through their permanence in the women of the admixed populations. The nuclear genome, however, due to considerable reshuffling, would be entirely new . Cerda-Flores et al. recently reviewed the studies on interethnic admixture in México, providing new estimates of the accumulated proportion of parental groups using D1S80, HLA-DQA1, and 13 STRPs for the northeastern region of that country. The estimates they obtained were of 55%-60% European, 37%-40% Amerindian, and 3%-5% African influence in the present gene pool of northeastern Mexico. Martínez-Cruzado et al. were able to rescue mtDNAs from members of the Puerto Rican population that they interpreted to be derived from the Taino Indians, extinct as an independent ethnic unit for centuries now. The Native American maternal contribution to Puerto Ricans was estimated to be 53%; and the prevalence of type 2A genotype of JC virus among them confirms these findings . Similar results have been obtained by Carvajal-Carmona et al. and Rodas et al. in the Mestizos of Colombia . As for five African-derived communities of that country Rodas et al. found 19% of Amerindian mtDNA, while Ribeiro-dos-Santos et al. observed an even higher prevalence in Curiaú, a semi-isolated Amazonian Afro-Brazilian community. --- RACISM AND DISCRIMINATION The concept of an interethnic system, developed by R.C. Oliveira can aptly describe what happens when groups of different ethnic affiliation meet. They are dialectically joined by diametrically opposed, though interdependent interests. Dominance/subordinate relationships generate opposite goals for the two groups, although economically they may depend on each other. Racism, the assumption of inherent biological superiority of a certain ethnic group in relation to others, and the consequent discrimination against them, generally appears in these situations. Brazil was for a long time considered a "racial paradise", and really the interethnic tensions there did not reach the extremes that were present, for instance, in the United States or South Africa. But the sheer existence of laws about this matter in the country indicate that racist practices exist there. Presently infractors of these laws are not protected by bail. Table 5 presents some selected socioeconomic parameters indicating the different conditions in which European-derived and African-derived persons live in Brazil. As is indicated there, the latter are always in a disadvantaged position, presumably derived from historical times and the perpetuation of dominance/subordinate relationships. --- HUMAN RIGHTS AND AFFIRMATIVE ACTIONS The Universal Declaration of Human Rights, adopted by the United Nations on December 10, 1948 is a milestone in terms of what should be followed in the relationships between persons or between persons and governments. This document was followed, years afterwards, by the Convention on the Elimination of all Forms of Racial Discrimination, also issued by UN on December 21, 1965. Equally important for the problems discussed here is the United Nations Educational Scientific and Cultural Organization Declaration on Race and Racial Prejudice of November 27, 1978. A list of the most important documents issued by several organizations related to these matters can be found in Salzano and Hurtado . They generally emphasize individual's rights. The collective rights recognized by international law are relatively few, and they imply the existence of representative collective institutions. It should be remembered that not all traditional custom is by definition good, women's oppression in Moslem societies being just one example. The concept of justice should be universal . The so-called "affirmative actions" are private or public policies against the structural inequalities of groups most vulnerable to discrimination. The term was introduced in the United States especially through two Executive Orders, that of number 10925, issued by president John F. Kennedy in 1961, and that of number 11246, issued by president Lyndon B. Johnson in 1965. The idea was to favor underprivileged persons by extending benefits, such as quotas for admission to employment or universities, to minority groups in general. In Brazil, the decision to establish affirmative actions occurred after the Third World Conference Against Racism, Xenophobia and Related Intolerance, held in Durban, South Africa, in 2001. Examples are as follows: The Ministry of Justice reserved 20% of executive posts and head positions to Africandescendants; Fellowships were created by the Ministry of Foreign Affairs especially for this ethnic category; A quota of 20% for African-derived persons was established in the federal service; The Federal Supreme Court established a 20% quota of Africandescendants to be observed in the companies which furnish services to it; and The State University of Rio de Janeiro reserved 40% of its vacancies to persons of this ethnic group . There are doubts about the correctness of these actions. Especially in the case of the 2003 examination entrance to the State University of Rio de Janeiro, no less than 200 judiciary processes were started by persons who considered that they had been unjustly a Source: Heringer , Noronha . denied access to this institution. In a way, such measures are against the Brazilian constitution of 1988, since they could be considered as racism in backwards. Even in the United States, some persons believe that these programs may actually perpetuate rather than challenge racism . Moreover, the difficulty in the delimitation of ethnic groups or individuals in mixed populations such as those present in Brazil or Latin America as a whole was already emphasized. What level of European, African, or Amerindian ancestry should be considered "significant"? Would self-classification be enough? --- CODA The right to equal opportunity, assured by the Universal Declaration of Human Rights, should be strictly respected. Biological inequality has nothing to do with the ethical principle that someone's position in a given society should be an accurate reflection of her/his individual ability. --- ACKNOWLEDGMENTS My research is financed by Programa de Apoio a Núcleos de Excelência , Conselho Nacional de Desenvolvimento Científico e Tecnológico , Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul , and Pró-Reitoria de Pesquisa da Universidade Federal do Rio Grande do Sul . --- RESUMEN Son examinados los intentos pasados y presentes de clasificar y caracterizar a la variabilidad biológica humana, considerando el concepto de raza, los problemas de identificación étnica, el matrimonio selectivo basado en la etnicidad, y la historia genética. En relación con la última, se hace una revisión de los métodos disponibles actualmente para la cuantificación de la mezcla y de los estudios previos enfocados en la caracterización de la contribuciones parentales de origen continental a las poblaciones Latinoamericanas, con énfasis en las evaluaciones globales de los acervos genéticos de de Brasil y Costa Rica. Finalmente, se considera el tema del racismo y la discriminación, incluyendo la relación entre los derechos humanos y las acciones afirmativas. El derecho a iguales oportunidades debe ser estrictamente respetado. La inequidad biológica no tiene nada que ver con el principio ético de que la posición de cualquiera en una sociedad dada debe ser un reflejo exacto de sus habilidades individuales.

Past and present attempts to classify and characterize the human biological variability are examined, considering the race concept, ethnic identification problems, assortative mating based on ethnicity, and historical genetics. In relation to the latter, a review is made of the methods presently available for admixture quantification and of previous studies aimed at the characterization of the parental continental contributions to Latin American populations, with emphasis in global evaluations of the Costa Rican and Brazilian gene pools. Finally, the question of racism and discrimination is considered, including the relation between human rights and affirmative actions. The right to equal opportunity should be strictly respected. Biological inequality has nothing to do with the ethical principle that someone's position in a given society should be an accurate reflection of her/his individual ability.

BACKGROUND AND OBJECTIVES: Transgender persons face many barriers to accessing health care, including identifying a knowledgeable physician. Medical schools have made curricular changes addressing cultural competence in transgender medicine, but changes are inadequate to graduate physicians competent in gender-affirming health care. The aim of this study was to assess the current state of education on the comprehensive health care of transgender patients, including gender-affirming health care strategies in US and Canadian family medicine clerkships in addition to the beliefs and actions of the directors making those curricular decisions. RESULTS: Seventy-two percent of FM clerkship directors agreed transgender health care should be a required part of the medical school curriculum. Sixty-six percent report active advocacy within their institutions for increased curricular time devoted to transgender health care. Fifty-six percent treat transgender patients in their own clinical practice, but just 26% agreed they were comfortable teaching transgender health care to medicals students. While the presence of transgender patients within the clinical practice did not have a significant impact on FM clerkship directors' comfort teaching this subject, having transgender friends or acquaintances did. CONCLUSIONS: FM clerkships are primed for inclusion of comprehensive transgender and GAH education in their curriculum. Increasing comfort of FM clerkship directors in teaching this subject area by providing accessible curriculum may encourage further uptake of this content into FM clerkships. :676-84.) doi: 10.22454/FamMed.2021.509974 --- ORIGINAL ARTICLES Eighty percent of the respondents expressed interest in doing more to address these issues. 4,7 A recent survey of Canadian pediatric residency programs confirms the need for further comprehensive transgender and GAH education with less than 50% of residents citing comfort in specific areas of transgender care and 90% stating further education on gender diversity is required. 8 Many barriers to increased LG-BTQ curricula have been cited. These barriers include an absence of trained faculty, faculty perception that LGBTQ health care is not relevant to their specific curricular content, and a lack of faculty and attending physician role models with whom students can discuss issues of sexual orientation, attraction, or gender identity. 7 In the Canadian survey of pediatric programs cited above, pediatric program directors most frequently cited finding time in an already busy curriculum as a barrier to inclusion of gender diversity training. 8 However, evidence suggests that even small increases in transgender education increase medical students' and other provider's knowledge, attitudes, and comfort in treating future LGBTQ patients. 9 In one study of secondyear medical students, a single didactic session about gender identity, classic treatment regimens and monitoring requirements with additional small group discussion and subsequent testing of the material, was added into the required second-year endocrinology course. Authors noted a 59% drop in students' anticipated discomfort when caring for transgender patients. 10 A second study of a 1-hour didactic session provided to faculty, medical students, and residents at a single institution showed improved attitudes, knowledge, and comfort in caring for transgender individuals. 11 While medical schools have begun educating medical students both in culturally competent and medically comprehensive GAH of transgender patients, barriers remain to universally adopting this curricular change. FM clerkship directors charged with implementing clinical curriculum play a key role in making future changes. The aim of this study was to assess the current state of education on the comprehensive health care of transgender patients, including GAH, in North American FM clerkships, in addition to the beliefs and actions of the directors making those curricular decisions. Development of transgender-specific survey questions was guided by extensive review of available literature regarding known barriers to transgender persons seeking health care, including physician bias toward and discomfort with this particular patient population. [2][3][4][5][6][12][13][14][15][16][17] Additional questions highlighted common medical education resources that could be useful for implementation of a new curriculum. The survey utilized both dichotomous and interval questions. We used a Likert scale of 1-5 to measure responses, with 1 being extremely likely and 5 being extremely unlikely. Lower responses indicated a higher value of transgender information and education. The 12-question survey was estimated to take less than 5 minutes to complete . --- Methods --- Statistical Analysis We used nonparametric statistical measures to compare positive interest in education and implementation FAMILY MEDICINE ORIGINAL ARTICLES of transgender conditions and health care, respectively, compared to negative/no interest. We determined descriptive statistics on all variables including frequencies and percentages, means and standard deviations, or medians and interquartile ranges, where appropriate. To examine the association of various clerkship director demographics or program characteristics with current transgender curriculum in the family medicine clerkship, we used χ 2 or two-sample t tests. To examine the association of various clerkship beliefs with likelihood of transgender education implementation or agreement with teaching and access to transgender education, we used Kruskal-Wallis nonparametric tests. For the Kruskal-Wallis tests, we used the Dwass-Steel-Critchlow-Fligner multiple comparison test to adjust the overall α level for post hoc pair-wise differences. We assessed all P values using an a level of 0.05. We performed all analyses using SAS version 9.4 . --- Results Of 144 FM clerkship directors, 99 responded to the CERA survey. Ninety percent of respondents were from schools in the United States. Respondents-62 females and 32 males-indicated serving as clerkship directors for 0-20 years, with a mean of 5 years . Ninety four of 144 directors responded to at least some of the transgender survey questions. Of those who did not answer, all represent medical schools in the United States, four public and one private. Additional demographic data is available on three of the directors, all of whom were non-Hispanic White. Four indicated the length of time they had served as clerkship directors, ranging from 3-12 years. Seventy-seven percent of respondents indicated that no transgender health care education was currently included in their clerkships . Twenty-percent indicated inclusion of transgender-specific communication , 12% health care maintenance and preventive care in the transgender population , and 5% GAH . Analysis of the demographic characteristics of responding clerkship directors and their representative schools found no significant differences between programs having no transgender education and those programs with education on communication with or the health maintenance and preventive care needs of transgender patients. Institutions that included education on GAH had FM clerkship directors with fewer years of experience than institutions that did not . Seventy-two percent of FM clerkship directors agreed that education regarding transgender healthcare should be a required part of the medical school curriculum, with 42% agreeing this curriculum would be best taught in the context of family medicine. Fifty-six percent of FM clerkship directors currently treat transgender patients in their own clinical practice, and just 7% believe their clinical colleagues would prefer not to treat transgender patients . Twenty-six percent of FM clerkship directors strongly agreed or agreed they were personally comfortable teaching transgender health care, while 43% disagreed or strongly disagreed. Sixty-six percent strongly agreed or agreed there was active advocacy within their institutions for an increase in transgender-specific health care within the curriculum. Twenty-seven percent strongly agreed or agreed to having access to a preexisting curriculum in transgender health care, while 42% disagreed or strongly disagreed . **Respondents were asked to select all that apply so individuals may be represented more than once or not at all. Further analysis of FM clerkship directors and their representative schools found a significant difference by location influencing access to preexisting transgender educational materials. Clerkship directors located in Canada indicated significantly easier access to a preexisting transgender curriculum than US directors . The 72 FM clerkship directors who indicated they did not currently include any transgender health care education in their clerkship were subsequently directed to and answered questions regarding their likelihood of use of available resources regarding transgender health care. Twenty-eight percent , 18% , and 24% of FM clerkship directors, respectively, stated they were likely or extremely likely to implement an existing transgender curriculum, use their personal CME budget for transgender health care education, or invite a transgender person from their own community to teach medical students while 31% , 54% , and 43% , respectively, reported being somewhat unlikely or unlikely to do so . In analyzing how FM clerkship directors' personal beliefs and practices may influence their professional beliefs and behaviors, several statistically significant relationships were identified. FM clerkship directors with transgender friends or acquaintances reported being more comfortable in teaching transgender education and more likely to have access to preexisting transgender education . FM clerkship directors who currently believe transgender education should be a required part of medical school education had a higher likelihood of implementing an existing transgender curriculum into their clerkship and were more likely to indicate their medical school's administration, faculty, and/or students were actively advocating for and/ or working to increase transgender health care education into the medical school's curriculum . Believe my clinical staff and/or colleagues would prefer not to treat transgender patients in our office 7 7 --- Those FM clerkship directors who * Ninety-four clerkship directors responded to the questions. ORIGINAL ARTICLES believed transgender education was best taught in the context of family medicine indicated a significantly higher likelihood of implementing an existing transgender curriculum and inviting a transgender person from the community to teach medical students in their clerkship . Finally, FM clerkship directors indicated a belief their clinical staff and/or colleagues would prefer not to treat transgender patients in their office had a significantly higher likelihood of using their CME budget to obtain training in transgender education . --- Discussion It is evident transgender patients in our communities deserve better access to, and experiences with, health care systems. Encouragingly, our data indicate very low levels of perceived transphobia and high levels of care of transgender individuals .012 .725 .371 3 .068 .008 Believe genderaffirming health care is best taught and/ or practiced in the context of family medicine Yes 3 .005 3 .251 3 .004 --- .534 4 .937 3 .684 No Interestingly, FM clerkship directors reporting these personal connections with transgender friends and/ or acquaintances not only reported a greater level of comfort teaching about transgender health care, but also easier access to existing transgender health curricula. These differences suggest personal connections with transgender persons change belief systems or behavior, leading directors to seek out transgender curricula for their FM clerkships. This is in spite of the fact this group of FM clerkship directors did not have higher levels of administrative, faculty or student advocacy within their institutions for this curricular change. Advocacy does matter, however, with more FM clerkship directors reporting a belief transgender and GAH should be a required part of the medical school curriculum also reporting more advocacy for such a curricular change within the institution. It is not surprising those holding this belief also report a higher likelihood of implementing an existing transgender curriculum into their clerkship. Continued efforts at designing accessible curricular materials for clerkship directors may help further efforts toward improving education in this domain. One critical medical education issue to resolve is where the curricula for the clinical care of LG-BTQ patients should be housed in order to ensure universal introduction on the care of this patient population to medical students. This d ilemma may account for the number of clerkship directors remaining neutral on the question addressing implementing a transgender curriculum. Knowledge that most FM clerkship directors actively care for transgender patients in their own clinical practice suggests many medical students have opportunities to interact with transgender patients during the course of their FM clerkship. Yet, interaction with transgender patients alone is not adequate for comprehensive GAH of this population. While it is important to be able to comfortably communicate with diverse populations, including transgender persons, their health care needs are sufficiently different from the general population as to require specific knowledge and skills. The knowledge and skills needed to provide basic GAH, such as hormone prescribing and management, are missing from nearly all family medicine clerkships. Making curricula on GAH readily available to FM clerkship directors may improve the number of medical students graduating with a sufficient foundation of the knowledge and skills needed to comfortably care for transgender and gender diverse patients in residency and beyond. However, our results indicate that less than 30% of clerkship directors would be likely to implement an available curriculum in their clerkship. Active advocacy for inclusion of comprehensive transgender and GAH is necessary to encourage greater uptake of available curricula by FM clerkship directors given the number of competing educational topics vying for limited curricular time. Knowledge of the prevalence of practices caring for transgender patients may also convince FM clerkship directors, most of whom already believe transgender education is an important piece of medical school curricula, to support FM clerkships as the location within the undergraduate curriculum to house comprehensive transgender and GAH education. Current advocacy for inclusion of comprehensive transgender and GAH in FM clerkships not only continues to emphasize the broad nature of the specialty, but also stresses family medicine's inherent prioritization of preventive and patient-centered care. 18 These emphases create a starting place for teaching about the needs of transgender patients. As with all survey research, this study has limitations. The crosssectional nature of this survey limits the results to the views of FM clerkship directors at a single point in time. A risk exists that FM clerkship directors not responding to the survey have different beliefs than respondents or that FM clerkship director turnover limits applicability to current FM clerkship directors. However, given the representation of AAMC schools in both Canada and the United States, and the high response rate, it is likely this sample is representative. The possibility responses reflect perceived expected answers exists, but the anonymous construct of the CERA surveys and the regularity of use should mitigate this effect. An additional limitation is the academic affiliation of FM clerkship directors completing this survey, which may alter the makeup of their clinical practice as well as the beliefs of the people with whom they work from that of family physician practices as a whole. This may limit applicability of these results to FM clerkships utilizing large numbers of community practices as training sites. Arguably, FM clerkships are primed for inclusion of comprehensive transgender and GAH education. Continued institutional advocacy for inclusion of comprehensive transgender and GAH in the medical school curriculum as well as more specific advocacy for inclusion in the clinical years will accelerate medical student exposure. The development of curricular materials easily integrated into already busy FM Clerkships may expedite this transition. Further research into facilitators and barriers of comprehensive transgender and GAH education inclusion in FM clerkships, especially mechanisms for improving comfort teaching this topic, is warranted.

onservative estimates report 0.3% of the total US population is transgender, representing 1 million individuals. [1][2][3] Transgender persons face many barriers to accessing care, but the most commonly cited is difficulty finding a knowledgeable health care provider. 1,4,5 A 2010 study of physicians found that 40% of respondents had no training on lesbian, gay, bisexual, transgender, and queer (LG-BTQ) care in medical school or residency. The majority of those with education in LGBTQ care reported it was "not very" or "not at all" useful. Most physicians agreed that further training and education in the area of comprehensive transgender and gender-affirming health care (GAH) is necessary. 6 While many medical schools have attempted to address the lack of LG-BTQ knowledge prevalent in the health care community through increasing teaching in tolerance and cultural competency, that training alone is inadequate to graduate physicians knowledgeable in the necessary preventive and appropriate GAH (hormone therapy, surgical interventions) of transgender and gender-diverse persons. 1 LGBTQ curricula reported a median of 5 hours of content within the 4-year curriculum, compared to an average of 3 hours 26 minutes in a 1992 study. 4,7 A 2011 survey of US medical schools reported one-third of schools taught zero hours of LG-BTQ-specific curriculum in the clinical years. In a 2015 survey of all Liaison Committee on Medical Education (LCME)-accredited US faculty practices, with a 50% response rate, only 16% of institutions reported having comprehensive LGBTQ training programs and 52% reported having no LGBTQ curriculum.

Introduction Social isolation and loneliness can have negative effects on our health. A meta-analytic review by Holt-Lunstad et al. concludes that social isolation, loneliness, and living alone are associated with 29%, 26%, and 32% increased likelihood of mortality, respectively. Even though the literature examining the health cost of social isolation and loneliness is wellestablished, there are a few studies which focus mostly on Western countries in the literature analyzing the economic cost of these conditions in terms of increased healthcare spending and utilization. This article focuses on Japan to investigate if the results found for other countries hold true for an Asian country. It is important to focus on a non-western country because different cultures approach the issues of social isolation and loneliness differently. Japan also has a different healthcare system compared to, for example, the US, which likely affects the healthcare cost of social isolation and loneliness. Lastly, both social isolation and loneliness are important issues for Japan, which has an increasing number of "hikikomori", people with severe social withdrawal, and "kodokushi", lonely deaths. a We would like to thank our two anonymous referees for their valuable comments and suggestions. b Ankara Yıldırım Beyazıt University, Department of Economics, Ankara Turkey. email: [email protected] 0000-0001-7310-2225 --- Isolation and Loneliness in Japan Before proceeding any further, the terms "social isolation" and "loneliness" should be defined because careless use of these terms caused some confusion in the earlier studies and made it difficult to compare these. Social isolation is objectively measurable, while loneliness is a subjective feeling. Social isolation refers to having few social relationships or infrequent social contacts, while loneliness is a perception of being isolated irrespective of the number, frequency, or quality of relationships that one has. It follows that one could be socially isolated but not feel lonely or be surrounded by family and friends but feel extremely lonely. Lastly, loneliness is sometimes called "perceived" or "subjective" social isolation. An important part of defining the terms "social isolation" and "loneliness" is determining how they interact with each other. Cacioppo et al. argue, based on the evolutionary theory of loneliness, that loneliness plays a similar role to hunger, where one signals a lack of nutrition and the other a lack of social interactions. They argue that it is beneficial in an evolutionary sense to be in a group and have meaningful relationships, and feelings of loneliness signal to us when we stray from these. Although elegant, this theory is not entirely supported by empirical evidence. There is not a decisive study that investigates how loneliness and social isolation affect each other. However, there seems to be a consensus that there is not a one-way relationship between these two measures, and, although related, social isolation and loneliness seem to be distinct measures . The literature seems to find conflicting evidence on how social isolation and loneliness affect healthcare spending and utilization. However, Mihalopoulos et al. points out the difficulty of comparing articles studying the healthcare cost of social isolation and loneliness due to different measures used for these terms. Therefore, the findings of the following articles do not necessarily mean conflicting evidence. Landeiro et al. finds for Portugal that social isolation is associated with increased healthcare spending per person by €532, which increases to €905 for highly isolated people. This is due to delayed discharge from hospitals. Shaw et al. find a higher effect for the US, $1,643 per person annually. They also find that loneliness is associated with reduced healthcare spending in the US . However, Fulton & Jupp and McDaid et al. , without considering social isolation, find the opposite for the UK; loneliness is associated with increased healthcare spending by £800 and by £600 per person annually, respectively. Lastly, Mitsutake et al. find that social isolation is associated with reduced healthcare spending in Japan, though they do not mention the exact amount. Social isolation and loneliness can affect healthcare spending in various ways. 1) Social isolation and loneliness might directly affect health negatively through stress, high blood pressure etc., which would, in turn, increase healthcare spending . 2) Social isolation and loneliness might indirectly affect health through unhealthy behavior, such as drinking, smoking, a sedentary lifestyle, not taking prescribed medication etc. This would again increase healthcare spending. 3) Social isolation might affect healthcare utilization in both ways. First, it might increase utilization as socially isolated people might feel more anxious about even minor health issues and feel the need to see a doctor . In contrast, for people who are not socially isolated, someone in their social circle would usually give advice or calm them down when they get agitated, eliminating the need to use healthcare. Second, it might reduce utilization for minor health issues due to the difficulty for some people to go to a healthcare facility by themselves . So, if they are socially isolated without anyone to accompany them, and if the health issue they are having is not a critical one , they might prefer not to go to a healthcare facility. For people who are not socially isolated, someone in their social circle would accompany them , thus enabling their access to healthcare. Therefore, socially isolated people might use less healthcare for minor health issues. The review study of Valtorta et al. concludes that weak social ties likely increase the probability of hospital readmissions and lengthen hospital stays. 4) Loneliness might increase healthcare utilization and thus spending simply because some people, especially the elderly, might go to healthcare facilities to have someone to chat with to alleviate their loneliness. A survey reports that around 10% of all consultations were due to loneliness .1 There is a potential reverse causality issue concerning the first channel. Cornwell & Waite find that social isolation and loneliness are greater among those who have worse health, but the direction of the causality is unclear. It is likely there is a two-way causation such that this is a self-reinforcing cycle. If this is the case, the effects of social isolation and loneliness on healthcare spending would be overestimated. The size of this overestimation would also depend on how important the first channel is. However, there is a lot of research that associates both social isolation and loneliness with worse health outcomes, although there is still a need for research on the exact mechanisms. Therefore, keeping in mind that the magnitude of the effects of social isolation and loneliness on healthcare spending might be overestimated, this article still provides valuable insights into how social isolation and loneliness are associated with healthcare spending. The structure of this paper is as follows. Section 2 introduces the data and models used in the study. Section 3 provides the results, whereas Section 4 discusses the limitations of data and methodology applied in this study. Section 5 concludes. --- Data and Models This article uses data from the Japanese Study of Aging and Retirement survey, which is the Japanese counterpart of similar surveys such as the Health and Retirement Survey for the US, the English Longitudinal Study of Aging for the UK, and the Survey of Health, Aging, and Retirement for Europe. To the best of my knowledge, this study is the first to utilize the JSTAR dataset in the current context. The sample analysed in this study is retrieved from the 2007 wave covering 4,300 respondents over the age of 50 from five Japanese cities. 2 Social isolation is defined as "living alone" in the study. 3 The measure for loneliness comes from the statement "I felt lonely". There is a concern that respondents might not answer truthfully due to the stigma attached to the word "loneliness". However, the JSTAR dataset does not allow the construction of Hughes's 3-item Loneliness Scale. 4 Lastly, a generalized linear model proposed by Nelder & Wedderburn given in equation is used to estimate the association between social isolation/loneliness and healthcare spending: log = β 0 + β 1 S i + β 2 L i + β 3 X i where µ i = E is the mean of healthcare spending, S i is social isolation, L i is loneliness, and X i is a vector of individual characteristics such as age and income. Since healthcare expenditures generally follow a skewed distribution, using a generalized linear model allowing for loosening the normal distribution assumption, as opposed to a regular linear model, is required. Therefore, a generalized linear model which assumes gamma distribution of errors with a log-link function is widely used in the literature in the context of healthcare spending. In order to see if social isolation and loneliness are associated with an increased likelihood of inpatient or outpatient visits, the following latent variable model is estimated: 5 Y ′ i = β 0 + β 1 S i + β 2 L i + β 3 X i + ϵ i which takes the following form: P = Φ where P shows the probability of having inpatient or outpatient visits , and Φ is the cumulative distribution function. We do not observe Y; instead, we observe Y ′ , which satisfies the following condition. Y = 1 if Y ′ > 0 0 otherwise 3 There are numerous measures used for social isolation in the literature. Cornwell & Waite proposed a single scale to measure social isolation by combining the following indicators: social network size, social network range , the proportion of social network members who live in the household, the average frequency of interaction with network members, average closeness with network members, household size, living alone, having a spouse or partner, number of friends, number of children, number of grandchildren, attending religious services, attending meetings of an organized group, socializing with friends and relatives, socializing with neighbors, and volunteering. Unfortunately, The JSTAR dataset does not have all the variables necessary to construct this social isolation scale. 4 Detailed explanation of all the measures used in the estimations and descriptive statistics are available in Appendices A and B, respectively. 5 The JSTAR dataset does not allow for differentiating respondents regarding their number of hospital visits. Differentiating respondents would change the weights of the control variables in terms of how much they affect the likelihood of having inpatient or outpatient visits. However, we simply consider two alternatives as inverse measures of health separately. Therefore, note that the models do not consider/focus on differentiating/comparing inpatient and outpatient visits. World Journal of Applied Economics 2023 Inpatient Outpatient Figure 1: Loneliness and Inpatient/Outpatient Visits Source: Author's calculations using the JSTAR dataset. Note: Visits refer to having had an inpatient and outpatient visit in the last year. Figure 1 suggests that loneliness is associated with an increased likelihood of both inpatient and outpatient visits. However, the probit estimations suggest that there is not a statistically significant effect. Figure 2 suggests that living alone does not have an effect on the likelihood of inpatient or outpatient visits, and the probit estimations support this. --- Results and Discussion The GLM estimation results presented in Table 1 suggest that social isolation is associated with reduced healthcare spending by ¥79,000 per person annually in Japan. Loneliness, however, does not have any statistically significant effect on healthcare spending. Other characteristics, such as having serious health problems and old age, are associated with increased healthcare spending, as expected. Notably, employment is associated with reduced healthcare spending by ¥53,800 per person annually, while household income does not have any statistically significant effect. Next, the effects that social isolation and loneliness might have on healthcare utilization measured by inpatient and outpatient visits are analyzed. The results suggest that neither social isolation nor loneliness are associated with an increased likelihood of inpatient or outpatient visits. Although there are studies that find an association between social isolation/loneliness and healthcare utilization, a review of the literature by Valtorta et al. shows that most studies do not find an association. The results indicate that, as expected, having serious health problems increases the likelihood of both inpatient and outpatient visits, and so does old age. Notably, being male increases the likelihood of inpatient visits but reduces the likelihood of outpatient visits. The results support the findings of Mitsutake et al. that social isolation is associated with reduced healthcare spending in Japan, despite the differences in the definition of social isolation and the sample characteristics . They argue that socially isolated people might not get the healthcare they need because they find that social isolation is negatively associated with outpatient care use. However, the results of this article suggest that social isolation is not associated with less outpatient care use. This could mean that, although socially isolated people make the same amount of outpatient and inpatient visits as their non-socially isolated counterparts, they might not accept treatment or use their medications, resulting in reduced healthcare spending. It could also mean that socially isolated people might reduce their healthcare utilization other than inpatient or outpatient Isolation and Loneliness in Japan visits, such as long-term care services, which could not be tested due to data limitations. However, the fact that both this article and Mitsutake et al. find the same effect of social isolation on healthcare spending despite using different measures points to something inherent in Japan, such as culture or the healthcare system. The results of this article also weakly support the findings of Shaw et al. that loneliness is associated with reduced healthcare spending . However, this article finds the opposite effect of social isolation. This could be due to differences in either measures we use for social isolation or healthcare systems. --- Isolation and Loneliness in Japan Lastly, at the expense of sounding morbid, it should be said that social isolation and loneliness likely decrease healthcare spending through early deaths. 6 We know that healthcare spending is disproportionately concentrated in the older ages; the older a person gets, the more things go wrong in the body, which creates a need for medical intervention and leads to higher healthcare spending. Therefore, interventions that reduce social isolation and loneliness which, in turn, reduce the number of early deaths, might actually increase healthcare spending because these people will now live through older ages where there is a higher possibility of needing healthcare. This is something return on investment studies on this topic fail to consider. However, this does not mean we should not try to reduce social isolation and loneliness. It is quite the opposite. We should try to reduce these conditions even if our interventions do not reach their healthcare spending reduction goals. We should still try to eliminate human suffering and extend life. --- Limitations This study, using a novel dataset, examines the association between social isolation/loneliness and healthcare spending/utilization in Japan, contributing to our understanding of the subject. Focusing on Japan is especially important as the results of this article show that social isolation and loneliness are not universally associated with increased healthcare spending and utilization. Therefore, we should be careful before launching programs to tackle these two conditions with the hope of reducing healthcare spending. That said, this article has the following limitations. First, the estimated models do not allow for determining causality. Therefore, care should be taken when evaluating the results so as not to put too much emphasis on them. Social isolation has numerous aspects, so focusing on a single aspect of it, such as living alone and using it by itself as a measure of social isolation, is problematic. However, as it is mentioned above, lack of data prevents the construction of the social isolation scales used in the literature. That said, whether a person lives alone or not is likely more important compared to things like the number of friends that person has or how frequently he/she meets them in the context of this study. This is because someone who lives in the same household as you likely has more impact on your healthcare use by enabling you access to healthcare services, giving counsel and advice, following up on your medicine use etc., compared to a friend who does not live with you. The second limitation of the study relates to healthcare utilization. Living alone likely increases long-term care use and nursery services; however, the JSTAR dataset only has data for inpatient and outpatient visits. Lastly, this article only focuses on people who are age 50 and older in accordance with the literature. However, social isolation and loneliness affect the younger population as well, especially during adolescence. Therefore, additional research is needed that focuses on the younger population in the context of social isolation and loneliness and their effects on healthcare spending and utilization. --- Conclusion Social isolation is associated with reduced healthcare spending in Japan by ¥79,000 per person annually, while loneliness does not have a statistically significant effect. Neither social isolation nor loneliness has a statistically significant effect on healthcare utilization; that is, these conditions are not associated with increased or decreased inpatient or outpatient visits. The reason why social isolation is associated with reduced healthcare spending in Japan compared to increased healthcare spending found in the literature for the US and Portugal might be due to different measures used for it, or it might reflect differences in healthcare systems. On the other hand, the literature finds mixed results on the association between loneliness and healthcare spending/utilization. Overall, this study provides some evidence that social isolation and loneliness might not be universally associated with increased healthcare spending and utilization. Therefore, care should be taken before launching social programs to tackle these two conditions with the hope of reducing healthcare spending and utilization. 2) 1-2 days 3) 3-4 days 4) 5 days or more As mentioned, the "felt lonely" statement is removed from the scale because it is used as a separate variable. The following scores are assigned to each option, and they are summed for the total CESD score (For positively worded statements, this was done in reverse: 1 >> 0, 2 >> 1, 3 >> 2, 4 >> 3 --- Loneliness "We would like to ask you about your physical and emotional condition in the last week. Please indicate if any of the following has occurred in the last week. If something occurred, but did not continue for at least one day, answer 'not at all'." There are 20 statements in the CESD-20 depression scale, and one of them is 'felt lonely', which is used to form the 'loneliness' measure. This article also uses the CESD-20 scale in the models. However, the 'felt lonely' statement is omitted from the CESD-20 scale. This is standard practice. 1) Not at all 2) 1-2 days 3) 3-4 days 4) 5 days or more This article uses just 'loneliness' in one model and the other three dummy variables as an alternative in other models, where not lonely is the reference group. Loneliness = 1 if 3 or 4 to the above question. lonely severe = 1 if 4, 0 otherwise. lonely medium = 1 if 3, 0 otherwise. lonely mild = 1 if 2, 0 otherwise. not lonely = 1 if 1, 0 otherwise. --- BMI Body-Mass Index. BMI 20 covers BMI smaller than 20, whereas BMI 35 covers BMI greater than 35. A BMI between 20 and 25 is the reference group. --- Chronic Diseases "Please tell me if you have been diagnosed by a doctor for:" 1) Heart Disease 2) High Blood Pressure 3) Cerebral Accident 4) Diabetes 5) Chronic Lung Disease 6) Joint Disorder 7) Cancer Each of these is a dummy variable equal to 1 if the answer is yes. --- City dummies --- Sendai --- Isolation and Loneliness in Japan Valtorta, N., Kanaan, M., Gilbody, S., Ronzi, S., & Hanratty, B. . Loneliness and social isolation as risk factors for coronary heart disease and stroke: systematic review and meta-analysis of longitudinal observational studies. Inpatient "In the past year, have you spent one or more nights in the hospital?" =1 if the respondent had an inpatient visit in the last year. Outpatient "In the past year, have you been an outpatient at a hospital or clinic or received acupuncture, moxibustion, or bonesetting treatment?" =1 if the respondent had an outpatient visit in the last year. --- Smoking "Do you regularly use tobacco, or did you use it in the past?" 1) Yes, I smoke now 2) I smoked in the past, but I have quit 3) No, I have never smoked regularly. =1 if 1 or 2, 0 otherwise. --- Drinking "In the past 6 months, how frequently did you drink alcohol?" 1) Daily 2) 5-6 days a week 3) 3-4 days 4) 1-2 days 5) Several times per month 6) Hardly ever or never drink =1 if 1,2,3 or 4, otherwise 0. --- ADL "Do you have difficulties in the following everyday activities?" 1) Putting on or removing socks and shoes 2) Moving around the room 3) Bathing on own 4) Eating by self 5) Getting into or out of bed 6) Using western style toilet Each yes gets 1 , and the sum gives the ADL score. The higher, the worse. --- Male =1 if the respondent is male, 0 otherwise. --- Married =1 if the respondent is currently married, 0 otherwise.

Social isolation and loneliness are associated with worse health outcomes, and there is a growing literature that studies the economic cost of these conditions in terms of increased healthcare spending and utilization. However, a handful of existing studies mostly focus on Western countries. This article analyses the issue in the case of Japan using the Japanese Study of Aging and Retirement (JSTAR) dataset with the help of generalized linear and probit models. The results show that social isolation is associated with reduced healthcare spending, while loneliness does not have any statistically significant effect. Neither social isolation (living alone) nor loneliness has a statistically significant effect on healthcare utilization; that is, these measures are not associated with increased or decreased inpatient or outpatient visits. Therefore, care should be taken when launching social programs to tackle social isolation and loneliness to reduce healthcare spending and utilization because these two conditions might not be associated with increased healthcare spending universally.

Introduction The concept of health promotion, defined as the "process of enabling people to increase control over their health and its determinants, and thereby improve their health" [1], emerged in the early 1970s in Canada, following the realization on the part of public health authorities that despite significant increases in healthcare costs and investments, low health quality persisted among a portion of the population. The health promotion movement, which has since grown rapidly internationally, highlighted the inadequacy of existing healthcare models [2], calling for more patient engagement and a wider holistic view of health, one that considers all spheres of life and takes a multisectoral approach [3]. These ideas form the foundation of the seminal WHO Ottawa Charter for Health Promotion , stating that "health promotion focuses on achieving equity in health." Health promotion continues to play a key role in global health policy. Although new individual and public health challenges have emerged since the adoption of the Ottawa Charter for Health Promotion in 1986, health promotion's core message has remained relevant [4]. That is, health should be a means to a better life, rather than the goal of life itself [5]. Assuming appropriate and fairly distributed conditions that make health-promoting choices possible, individuals play a key role in sustaining their own health. In this regard, it is interesting to note that the Ottawa Charter definition, by putting the spotlight on personal agency, quickly gave rise to the idea of self-care as an integral aspect of health promotion [6]. Indeed, all those activities directed at strengthening individual health literacy and awareness, as well as individuals' activities to manage and maintain their health or address and prevent the root causes of illness, may promote health. Health promotion acknowledges both the importance of conventional disease prevention strategies and the social, economic, and environmental determinants over which individuals have limited control [7]. While highlighting the role of social determinants of health, health promotion also emphasizes the value of individual choices and behaviors such as embracing an active lifestyle, avoiding drugs and tobacco, maintaining a healthy diet, managing stress, and coping with mental health issues. The literature reports how the success of health promotion programs is often the result of a "bottom-up" approach that places healthy individuals and communities at the center, as responsible for and actively in charge of their own health, as opposed to passive recipients of decisions by health authorities [8]. This shift towards actively empowering people to stay fit and in good health presents an opportunity to overcome potential limitations of health systems to provide adequate services, especially in resource-poor settings. The concept of self-care offers opportunities for better health even in contexts where primary care is difficult to access, medical personnel are scarce, or management of acute illness is cost prohibitive; hence, WHO's endorsement of the value of self-care practices for alleviating global health burdens and achieving the Sustainable Development Goals [9,10]. Building on this background, this narrative review aims to illustrate how health promotion practices have been influenced and shaped by the emergence of new technologies. The scope of this piece is first to provide a broad perspective on opportunities presented by digital health promotion , and then to focus on the relevance of DHP for youth in low-and middleincome settings. Finally, we highlight enabling factors necessary to realize the potential of DHP for youth in developing economies. --- The rise of digital health promotion-DHP Since the early 2000s, together with the digitization of the health sector and the explosion of the use of big data for health [11], the increasing availability of digital technologies and their integration into self-care programs have reshaped the concept of health promotion, giving rise to DHP [12]. Although a clear definition of DHP is still lacking, for the purposes of this review, we define DHP as digital technologies utilized to propagate health awareness, or digital strategies and campaigns which engage and enable users to monitor their own lifestyle and health habits, fostering healthy behavior change. These technologies include social media platforms, webpages, mobile and web apps, wearable devices, computer/mobile games, AI powered chatbots, and texting services. Recent WHO initiatives promoting the worldwide adoption of digital health underscore the unprecedented potential of such technologies for improving self-care and public benefit [13,14]. Indeed, the growing success of DHP is due to several factors. --- Expanding access to health promotion services The widespread availability of the internet and smartphones enables individuals to search for and engage with health information everywhere. Information is available 24/7, privately, and without involving a healthcare professional. In settings with a shortage of healthcare workers or public health services, DHP can be especially useful for non-urgent health matters. The flexible and discreet nature of these tools makes them compatible with a range of time and lifestyle demands [15]. Digital platforms allow individuals to discuss health-sensitive topics, share their experiences, and seek support in discreet or anonymous ways [16]. For example, during the COVID-19 pandemic, web-based platforms emerged to support mental health. These programs aimed to reinforce positive habits by offering personalized emotional coaching that preserved user privacy and feedback via chatbots, peer support platforms, and content gamification [17,18]. --- Offering personalized services DHP strategies are characterized by their user centricity [19]. Users generate data in 2 ways: through active input or passive input, via tracking devices [20,21]. These tools run on a constant basis, creating measurements of temperature, breathing, heart rate, physical activity, blood pressure, body composition, and BMI [22]. Health apps make this health information accessible to individuals, aiming to encouraging physical activity or other positive health habits [23,24]. Yet, a primary appeal of DHP technologies lies in their ability to provide users with tailored feedback addressing individual health goals [25]. By leveraging individual preferences and adjusting feedback accordingly in real time, DHP holds potential for greater impact than traditional "one size fits all" public health promotion strategies. --- Supporting healthy lifestyles DHP applications aim to increase awareness of health habits, and to support users in healthrelated decisions, aiming to prevent future decline in health status. Studies have described how digital health technologies engage individuals, resulting in active behavior change [26,27]. Through gamification and creation of avatars, for example, DHP applications seek to nudge consumers to embrace "healthier lifestyles," while increasing their information receptivity and retention rates [28,29]. Leveraging reward strategies and social elements, DHP not only engages users, but also keeps them motivated to achieve their health goals [30,31]. --- Containing costs DHP interventions are an attractive choice for addressing health risks in part because the scaling costs are low. Because DHP strategies can be implemented via consumer technologies that most people already own, it is increasingly feasible to introduce initiatives targeting entire communities [32]. Examples include DHP strategies to improve health literacy, contact tracing apps alerting users of locations with high COVID-19 infection rates, or apps promoting immunization. Apart from the initial cost of creating software or designing the technology, upscaling a digital intervention is relatively inexpensive for developers [33]. By avoiding the significant expense often associated with traditional health promotion strategies, DHP have potential to reach a broader audience, maximizing impact. DHP is increasingly affordable, even for vulnerable individuals who face financial, cultural, logistical, or information barriers to accessing traditional healthcare or health promotion services [12]. --- DHP for adolescents and young adults Adolescents and young adults -namely people 10 to 24 years old-account for more than 1.8 billion people globally, or a quarter of the world's population [34]. A recent review of the literature revealed that in high-income countries , AYAs use digital technologies extensively to monitor their health and promote physical and mental well-being [35]. This finding aligns with earlier research reporting the feasibility, efficacy, and cost-effectiveness of digital interventions to promote health among young people in advanced economies [36,37]. In light of such trends, over the last years important international initiatives have focused on bringing young people into the discussion about digital initiatives, with the aim of achieving the SDGs [38,39]. For the field of health promotion specifically, young people may represent a breaking point with the past. Many diseases and behaviors that develop during adolescence and young adulthood carry negative consequences throughout the lifespan [40]. For example, malnutrition and lack of adequate activity are connected to the exponential increase in youth obesity in Kenya and Ghana, with serious consequences for future health [41,42]. Similarly, early pregnancies and sexually transmitted diseases , unaddressed mental health issues , use of tobacco and substances , violence, and poor physical activity constitute the greatest health risks faced by AYAs worldwide [40,43]. Targeting DHP solutions to a still healthy population may improve young peoples' behaviors and thus their state of health, both immediately and in the long term [44]. The available academic literature reports several reasons why DHP finds fertile ground among young people. First, youth appreciate the breadth of information available via digital media and the fact that this information is available at any time and without involving intermediaries [45]. Internet connectivity enables access to health-related information and services even in contexts with geographical barriers or health professionals' shortage [46]. Secondly, content is often personalized and engaging . Thus, the target audience is motivated to engage with the information, encouraging receptiveness to the subject at hand [47]. In addition, young people appreciate the ability to connect with peers, creating support groups while maintaining anonymity, and thus preventing potential stigmatization [48]. Finally, the low cost of such technologies allows the younger generation to access health promotion services without incurring out-of-pocket expenses [12]. --- Shedding light on DHP in LMICs While the contribution of digital technologies to health promotion has been critically examined in the context of HICs, research on the potential and challenges for implementing DHP in low-and middle-income countries is limited [49,50]. One possible explanation is that digital health is often associated with sophisticated forms of healthcare and technologies , which may be less accessible in LMIC settings. Yet, health promotion can also occur via other more common forms of digital media, from text messages to social media platforms [20,51]. So far, however, only a few studies have reported on the potential of these DHPs in emerging economies. Among them, for example, research conducted in South Africa shows how social media can rapidly disseminate health behavior information throughout the population [52]. Similarly, researchers in Tanzania have tested the effectiveness of animated health videos and interactive platforms for public health promotion in relation to HIV and tuberculosis [53]. In the coming years, young people in LMICs will have increasing access to technologies, and DHP adoption will no longer be weighted towards wealthy countries [54]. Already, data show a general increase in the uptake of smartphones in LMICs. Internet coverage continues to grow in sub-Saharan Africa, with a substantial increase during the COVID-19 pandemic [55]. Internet access has increased for youth in particular: in 2020, 40% of youth in Africa accessed the internet, compared with 27% of the general population [56]. Moreover, as digital natives, young people will be more tech-savvy and comfortable in adopting DHPs than older generations [57]. Young people across borders seem eager to harness the benefits of technology and make their voices heard concerning their needs [58,59]. Today's youth demonstrate awareness not only of their role in maintaining their own health, but also in leveraging technology to address pressing public health challenges [60]. Examples of such initiatives include the Young Experts Tech for Health and the Lancet Youth Network. These youth networks are driving action toward meaningful youth engagement in technology design, affordable training in digital skills, and broader access to health education and services. The Ndola Youth Resource Centre, a youth-led organization based in Zambia, co-developed the initial content for TuneMe, an online health promotion platform tailored to the health needs of AYAs of 7 African countries [61]. Its features include a collection of resources on HIV, STIs, reproductive systems, and safe sexual behaviors, as well as a direct booking system for screening services, and a platform to anonymously discuss issues of sexuality. Despite the anticipated benefits, significant knowledge gaps exist about how digital technology can effectively foster young people's sexual and reproductive health, mental health, or healthy lifestyle behaviors in LMICs [62,63]. Particularly, rigorous assessment of the perspectives, insights, and concerns of youth about DHP are still largely missing from the current dialogue. Since major benefits for individual and global health are at stake, DHP strategies for youth deserve more careful exploration. --- Main challenges and risks DHP initiatives have gained momentum among public health agencies, NGOs, and international organizations due to their potential for realizing and expanding access to health promotion also in LMICs . For example, the "Youth-centred digital health interventions" document published by WHO in 2020 aimed to foster and guide the development and implementation of digital health solutions for young people worldwide, reporting on successful case studies from Africa to South America [64]. Alongside the promises, the literature on health promotion, together with that on digital health, offers insights into the dangers and ethical challenges of these technologies [65]. Such concerns must be investigated and addressed promptly to avoid introducing technologies that bring more harm than good. --- Access and allocation of technologies The digital divide-understood as the discrepancy in access to digital infrastructure and literacy across settings and population groups-poses limits to the opportunities offered by DHP. While this is a well-known problem in the field of digital health, it is even more pronounced in contexts where economic resources are limited, and inequalities are significant [66]. In LMICs, the digital divide problem is stratified. First, some areas within resource-poor countries may lack reliable and interoperable information infrastructure and appropriate technologies due to economic, geographic, political, and cultural barriers. Internet infrastructure is uneven across and within countries; for example, between rural and urban areas [67]. Such technical limitations can result in tools that cannot be utilized, such as apps that require constant network access for data synchronization. Second, when infrastructures and technologies do exist, they are often not equally accessible across the population. As socioeconomic and cultural barriers stand between the availability of health-promoting technologies and their effective use, the most vulnerable demographic groups face the greatest disadvantage [68]. Because health promotion content is increasingly delivered digitally, population segments that lack digital literacy and have little experience with digital technologies may be excluded [69]. Despite the growing availability of infrastructure and tech exposure, lack of digital literacy still represents the main barrier to youth benefitting from digital health innovation [70]. Additionally, although much of the content offered by DHP is in English, vulnerable populations-including AYAs in LMICs-predominantly speak local languages [71]. Hence, the potential inability to benefit from such services and understand health messages [72]. Even more so, solutions that require a higher level of reading skills but lack gamification and interactive design aspects, as well as relevant cultural cues, would not be adopted by those with limited literacy [73]. Inability to afford expensive data plans is a further reason why AYAs might not adopt DHP. Accessing DHP in public spaces with free Wi-Fi can be a solution, but it may also create potential discomfort due to the sensitive nature of the topic; lack of access from home or other safe spaces might result in young people not taking advantage of the technology at all [74,75]. Third, contrary to what is generally the case in HICs, digital devices such as mobile phones are not necessarily available to single users in LMICs [76]. Oftentimes, multiple members of the same family share the same device, to limit costs of the hardware and the data plan. In order to maximize health outcomes and to minimize privacy-related risks, it is thus important to design DHP applications in a way that takes into account how digital platforms are actually used in a specific context. Finally, the lack of equal access to digital technologies leads to data poverty, which results in incomplete and nonrepresentative health datasets [77]. These biased datasets, in turn, may be used in the assessment and evaluation of DHP technologies, potentially creating inefficient, or even unsafe, implementation. That DHP may have poor outcomes is even more worrisome from a public health perspective, with potential ramifications of harmful health consequences and an increase in inequality, to the disadvantage of already underprivileged groups. --- Fair distribution of benefits Amid the discussion over the use of digital health to improve global health and expand health promotion strategies, some authors have raised concerns about the risk of digital and data colonialism as an effect of structural injustice [78]. A growing number of private companies from HICs are establishing themselves in the Global South by offering products, technological infrastructure, and human expertise in digital health. As a result, technologies and computational methods created in developed economies are deployed to acquire data from populations in developing economies [79,80]. Just as historical colonial practices involved the appropriation of land, the new form of data colonialism undermines individuals' sovereignty over their data and communities' capacity to address context-specific health needs [81]. While digital tools and large technology companies can potentially complement the role of public health institutions, the risk of misusing the health data of local populations generated through these new technologies is increased. In the name of open data access and to address the data poverty issue, companies transfer data acquired via digital technologies in the Global South to the West for research. While this harvested data can generate healthrelated knowledge, it primarily serves to create profit for those in control of the data, as corporations may sell the extracted information to third parties for consumer and business services [82,83]. Such data and techno colonialist practices are not unique to the private sector, but occur in academic health research as well. For example, in the field of global health, northern research institutions have for decades collected and monitored malaria data from endemic areas of Africa, without involving indigenous people in data analysis or presentation of results [84]. Today, DHP tools enable collect vast amounts of data about their users, but at the same time such data may be interpreted in developed countries by researchers lacking adequate contextualization. Therefore, while institutions that draw on data knowledge can respond to people's health needs in a timely manner through specific health promotion solutions, they also contribute to structural inequalities by maintaining control over data and exacerbating exploitation mechanisms [79]. The asymmetry of power described in this section between those who collect the data and those who provide it raises a burning yet unanswered question: Whose interests are served or denied when applying DHP strategies in LMICs? --- Privacy and misinformation DHP technologies may require or enable gathering and analyzing significant amounts of data. Some health data governance frameworks encourage the use of personal data as a means of promoting and improving health [85,86]. However, DH technologies can also put users' privacy and safety at risk. Because in most countries, these direct-to-consumer technologies are not vetted by a health or data authority, there is a risk for irresponsible data use. For example, fitness and wellness apps may share sensitive data with third parties without the user's permission or knowledge, and collect data points beyond those strictly necessary for their function. In addition, the personal nature of health information raises additional concerns about data leakage, disclosures, or misuse. Data could be used by marketing companies for targeted advertisements, by cybercriminals for blackmail and profit, or by governments to identify and prosecute certain minorities. This is especially relevant in contexts where data governance is not established or does not extend to all data uses; data security mechanisms are shaky, and individual and collective stakes are high. As of 2022, only 48% of the least developed countries have data protection and privacy legislation in place [87]. Breaches of health data collected via DHP can expose young people in particular to the risks of cyberbullying and online harassment. Because AYAs are the most active online and potentially least aware about how to protect themselves online, they are exceptionally vulnerable to such threats. For example, in Finland in 2020, sensitive data from students who adopted an easy-to-use app to boost mental health were leaked by hackers whose ransom requests were denied. Students experienced emotional, social, and reputational damage as a result, to the extent that some users changed phone numbers or social security numbers, or removed their presence from the web [88]. Because most DHPs are not considered medical devices, they do not have to abide by traditional oversight mechanisms. Thus, no regulatory authority is responsible to ensure that DHP technologies offer accurate evidence-based information, or recommendations that are demonstrably safe and effective. Misinformation and false sources could in particular harm young people, who are not in the habit of fact-checking online content . Similarly, misleading messages embedded in DHP may lead especially those with low levels of health literacy to engage in behaviors that may turn out as useless or even harmful. For example, AYAs might be persuaded by catchy visuals to smoke e-cigarettes instead of traditional cigarettes, if the former is labeled on social media as a healthier and fitness-enhancing alternative. However, scientifically speaking, this information would be misleading, as both types of cigarettes contain nicotine and can harm health. Therefore, questions persist about who is responsible for the negative consequences of unsafe and unreliable technologies, in the absence of responsible innovation regulations. --- The empowerment paradox Some authors have challenged the value of digital health technologies, with an argument rooted in a sociological critique of the neoliberal approach to healthcare [20,89]. According to this view, a main limitation of DHP is its focus on individual responsibility, with health promotion framed as an individual effort, and a lack of attention to socioeconomic determinants of health. This critique asserts that health promotion should not focus on changing individual behaviors, but instead address the power dynamics that indirectly influence the health behaviors of individuals and communities. Accordingly, some scholars argue that DHP is more aptly described as the digital re-creation of traditional health communication strategies [90]. While technology has been incorporated into traditional public health campaigns for some time, the actual implementation of health behavior is left to individuals. Health authorities might promote the use of apps and mobile services to keep physical fitness and mental health, but these tools are not broadly integrated into healthcare systems, and associated costs are borne by individuals [91]. Hence, DHP might not be the vehicle for addressing systemic issues and prioritizing the needs of underserved population segments. On the contrary, according to this critique, DHP could exacerbate inequalities, discriminating vulnerable groups. This sociological critique also disputes the concept of empowerment at the core of DHP. Behind the promise of more independence and self-determination, so the argument goes, lies the risk for surveillance and control. Individuals are not only nudged, but also rather manipulated towards a specific health goal, significantly limiting individual autonomy. Some authors in particular have argued that these technological approaches to health promotion translate, in practice, to a patronizing attitude of companies and government institutions towards citizens. Institutions impose health goals and priorities as defined by a system of surveillance, data analysis, rules defined by developers, and market laws, as opposed to the needs and preferences of individuals and communities. DHP technologies, therefore, far from serving autonomy, offer users only the illusion of individual control over their health trajectory. --- Considerations to unlock the potential of DHP for AYAs in LMICs While underestimating the complexities of DHP implementation could worsen conditions for the most marginalized and worsen health disparities, failing to recognize the possibilities offered by new technologies will leave the health needs of many unmet. --- Striving for empowering DHP Health promotion's overarching aspiration is to address sociopolitical determinants of health; this aim can be facilitated by the adoption of new technologies that promote self-care and wellbeing. From an ethical standpoint, helping people to become more aware of their health, and to take appropriate action based on that knowledge, is justified as an attempt to promote universally valued aims such as being in good health and avoid preventable health risks. Thanks to DHP, people may improve their life and health, making changes step by step. Individuals can boost their knowledge of health and wellness, learn about scientifically verified activities for staying healthy, and be supported in their efforts to reduce health risks. Especially in resourcepoor settings, DHP can mean a lot for people who lack access to other reliable sources of health information or to healthcare services altogether. Still, we argue that individual responsibility for health is not a substitute for governments to provide the necessary resources and conditions that allow people to maintain healthy lives. Therefore, we resist a divisive interpretation of DHP that overstates either the importance of the top-down approach or the bottom-up approach . Instead, we respond with a cautious but proactive attitude to the adoption of digital technologies for health promotion. This perspective is aimed at synergistically holding two different visions of health promotion, which are not mutually exclusive, but complementary. Especially in contexts where health systems suffer from systemic weaknesses and governments have limited resources at their disposal, solving complex problems takes time. Therefore, while working to address the social determinants of health, targeted low-cost DHP interventions could provide a timely response to the needs of a specific population. Although these interventions may only be accessible to certain groups , they could still progressively bring equity and positive outcomes for the whole population. By supporting some individuals, DHP can lessen the burden on the healthcare system, while allowing for more equitable redistribution of public resources. Namely, public health institutions could allocate more resources towards meeting the needs of those who, either by choice or out of necessity, do not use DHP technologies. That said, the literature reports several conditions that health institutions and DHP developers should meet to facilitate the success of DHP in LMICs and ensure that the benefits of technology are equitably distributed among stakeholders . --- Increase digital access and literacy Focusing on the digital divide, international institutions have called for mobile health tools to become more available and accessible worldwide [13]. Due to globalization and ongoing digital transformation, LMICs are gradually gaining stable access to broadband internet, modern IT infrastructure, and technologically advanced devices [92]. Yet, the private sector, nongovernmental agencies, and national governments should increase funding to lower the costs associated with newer infrastructure and services, encourage technological development, and train digital experts in the Global South. Local technology professionals and developers understand the needs of end-users, and they are best equipped to tailor solutions to those needs. For example, in LMICs it could be strategic to create health content in local languages with references to the relevant culture, as it will have the benefit of reducing the language barrier for young people, some of whom have not attended school and cannot read or write in English or another colonial foreign languages [93,94]. This would provide greater autonomy and independence from the Global North, not only regarding collection and management of data, but also for development and utilization of digital solutions [95,96]. As digital inclusion affects many aspects of living, scholars have recently argued for the addition of digital technologies among the social determinants of health [97,98]. Despite expanding capacity building and tech access, challenges such as inadequate digital literacy may persist in low resource settings [96]. Thus, enhancing young people's digital literacy and scientific fact checking skills, across genders and socioeconomic conditions, is a priority for reaching the full potential of DHP in LMICs. --- Involve end-users Only by effectively participating in the co-design of DHP interventions and associated technologies, and helping to define its purpose and rules, will young people have the opportunity to express their needs and priorities [60]. It is therefore necessary to involve prospective young users at various stages of technology development. Most importantly, individuals should be able to define their needs and consider whether and which technologies are the most appropriate to promote their health. This process must factor in cultural and social circumstances as well as local values. Enabling DHP co-creation might increase users' trust in technology and willingness to use it. Developers should therefore place technologies in context and respond to users' needs. This approach to DHP development contrasts with paternalistic and simplistic top-down methods that ignore technological relevance and reliability in context. --- Deploy safe tools Despite the potential for DHP to exert a transformative impact on youth health, such technological development remains a double-edged sword. Therefore, to realize its benefits, first and foremost DHP technologies must be safe [58]. Safety may be articulated in 3 ways. First, ensuring data privacy. This entails not only minimizing data collection to the data needed to develop and deploy DHPs, but also protecting data from cyber and ransomware attacks, to safeguard individuals and groups with sensitive health characteristics . Second, preventing the presence of unwanted bias and unfair interventions. Stakeholders must rigorously and repeatedly monitor DHPs to ensure the adequate representation of diverse demographic groups including ethnicity, gender, and wealth. Third, ensuring the scientific reliability of an intervention. While DHPs simplify complex health information, making it accessible to young people, they must not forego the nuances of scientific discourse, or worse, allow the proliferation of misinformation. Thus, alongside monitoring the quality of information provided, health institutions and DHP developers should contain any factors potentially distorting information [99]. --- Ensure technology oversight The use of DHP, particularly among vulnerable youth in developing economies, requires data governance policies and accountability mechanisms [100]. Strengthening the role of oversight instruments could ensure a more ethically aligned development and use of technologies. While the lack of adequately transparent policies and public involvement may threaten trust in entities developing DHPs, the presence of reliable, transparent, and efficient oversight systems could increase willingness to use DHPs [101]. For this reason, it is crucial to clearly define the ethical standards and boundaries of DHPs at the international level. These shared minimum standards should then be translated into more practical strategies and applied locally, according to cultural specificities and local needs. To ensure that relevant ethical standards are incorporated in DHP interventions, involved parties should appoint ad hoc ethics boards including local experts and community representatives. Such boards would specify international ethical principles in public health and health promotion into specific technical and use-related requirements to minimize risks and maximize benefit of DHP initiatives across target populations-especially in the case of vulnerable and marginalized demographics. --- Conclusions Digital health technologies become more pervasive worldwide, creating opportunities for individuals to engage in self-care and health promotion practices, intrigued by their potential benefit. As more and more young people engage in self-care and mobilize to harness technological power for their health, the discussion about DHP urges timely research and attention. Our analysis shows the practical strength of DHPs to benefit young people and, more generally, the population of LMICs as a whole. To fully exploit its potential, however, DHP cannot be considered as a substitute for other more structural public health efforts to keep populations healthy. Rather, DHP technologies are best conceived as one of many means to achieve universal health and adequate care of individuals and populations. To this aim, institutions, experts, researchers, and global health organizations should promptly address the limitations and pitfalls of existing DHPs by engaging end-users and targeted communities in co-creative models for technology development. --- Yes we are compliant with the PLOS Data Policy. We reported on the literature used for our conceptual analysis. ---

As digital technologies such as smartphones and fitness bands become more ubiquitous, individuals can engage in self-monitoring and self-care, gaining greater control over their health trajectories along the life-course. These technologies appeal particularly to young people, who are more familiar with digital devices. How this digital transformation facilitates health promotion is therefore a topic of animated debate. However, most research to date focuses on the promise and peril of digital health promotion (DHP) in high-income settings, while DHP in low-and middle-income countries (LMICs) remain largely unexplored. This narrative review aims to fill this gap by critically examining key ethical challenges of implementing DHP in LMICs, with a focus on young people. In the existing literature, we identified potential impediments as well as enabling conditions. Aspects to consider in unlocking the potential of DHP include (1) addressing the digital divide and structural injustice in datarelated practices; (2) engaging the target population and responding to their specific needs given their economic, cultural, and social contexts; (3) monitoring the quality and impact of DHP over time; and (4) improving responsible technology governance and its implementation. Addressing these concerns could result in meaningful health benefits for populations lacking access to more conventional healthcare resources.

INTRODUCTION Neighborhood ecological factors have been linked to crime and delinquency in decades of theory and empirical research. More recently, empirical studies have shown significant relationships between neighborhood/ecological factors and other problem behaviors, particularly substance abuse. However, there have been relatively few studies examining the relationships between neighborhood factors and problem gambling. Shaw andMcKay provided the early groundwork for the concept of social disorganization by demonstrating that high delinquency rates were linked to characteristics of certain neighborhoods in Chicago as opposed to the personal characteristics of the people living in such neighborhoods. They argued that ecological factors, such as poor economic conditions within square-mile areas and population instability, resulted in loss of social control in a geographic area and thus, resulted in increased levels of delinquency. Since this early work, social disorganization theory has been extended and empirically tested in a large body of scholarly work applied to the explanation of crime and delinquency. Wilson ) developed the concept of concentration effects or concentrated poverty to characterize impoverished neighborhoods with high proportions of poor, female-headed and minority households. These disadvantaged neighborhoods were associated with lack of access to jobs, lack of quality schools, and lack of exposure to conventional role models, resulting in a variety of negative outcomes including joblessness, low educational achievement and involvement in crime . The theory further maintains that regardless of individual characteristics, the ecological concentration of poverty gives rise to structural barriers and cultural adaptations that undermine social organization and the control of crime . Thus, the concept of social disorganization may be viewed as the inability of particular neighborhood communities to maintain effective social controls . Sampson, Morenoff and Gannon-Rowley analyzed over 40 studies focused on neighborhood effects and demonstrated that there are geographic "hot spots" for crime and other problem behaviors and further that these areas are characterized by the concentration of social and economic disadvantage. Studies of the effects of neighborhood disadvantage have been extended to substance abuse. Boardman, Finch, Ellison, Williams and Jackson examined the relationship between neighborhood disadvantage using census tract data and individual drug use from interviews with a general population sample of adults in the Detroit area. The authors found a positive relationship between neighborhood disadvantage and drug use even after controlling for individual-level socioeconomic status. Using a large national survey on drug use and health among 12-17-year-old youths, Winstanley et al. showed that self-reported neighborhood disadvantage was associated with alcohol and drug use and dependence. Martinez, Rosenfeld and Mares examined the relationships between social disorganization based on neighborhood-level census tract data, drug activity measured by official records of drug-overdose deaths, and neighborhood violent crime in Miami. The authors concluded that social disorganization predicts drug activity which in turn leads to higher levels of criminal violence. Karriker-Jaffe et al. found interaction effects in the relationship between neighborhood disadvantage and adult alcohol outcomes, such that neighborhood disadvantage was negatively associated with heavy drinking among whites but positively associated with heavy drinking among African-Americans. Empirical tests of problem behavior theory have long found that alcohol, other drug use as well as crime and delinquency co-occur and constitute a problem behavior syn- Addictions 2, pp. 82-89 | 83 Neighborhood disadvantage on problem gambling drome . More recently, problem gambling has been linked with these other problem behaviors and thus, common predictors such as neighborhood disadvantage may provide insights into the occurrence of gambling behaviors as well as substance abuse and other problem behaviors. --- Journal of Behavioral There are relatively few studies examining the effects of neighborhood factors on gambling behaviors and problem gambling. Clotfelter and Cook , in their seminal work on lotteries in America, document that disadvantaged, poor neighborhoods had a high density of lottery outlets and that the financial burden associated with state lotteries fell disproportionately on the poor. Ecological studies of video lottery terminal gambling in Canada also have shown strong neighborhood effects. The spatial distribution of VLTs in Quebec was strongly correlated with poorer neighborhoods . Similarly, Wilson, Gilliland, Ross, Derevensky and Gupta examined VLT access among high school students in Montreal and found that VLT access was more prevalent near schools located in socioeconomically-deprived neighborhoods as compared with schools located in more affluent neighborhoods. These studies were spatial analyses of VLT locations and were not designed to link individual gambling behaviors with neighborhood disadvantage. Martins, Storr, Lee and Ialongo , in a study of 596 low income, urban young adults, found that higher neighborhood disadvantage concerning the inhabitants , was associated with gambling frequency and gambling problems. --- GOALS OF THE STUDY In the national Survey of Gambling in the U.S. , neighborhood disadvantage was significant in predicting adult respondents' frequency of gambling in the past year as well as past year problem gambling, even after taking into account respondents' socioeconomic status . The present investigation extends this investigation in several ways. This study examines the effects of neighborhood disadvantage across the lifespan from 14 years to 90+ years using a combined sample of two comparable national samples of youth and adults . Furthermore, gender as well as age, socioeconomic status, and race/ethnicity are included in the analyses to examine possible interaction effects of neighborhood disadvantage and socioeconomic status on gambling behaviors and problems after controlling for interactions of neighborhood disadvantage and SES with these other demographic factors. An additional consideration in the present paper is the inclusion of alcohol use and alcohol abuse/dependence variables, as well as gambling outcome variables, to permit a comparison of the neighborhood and other socioeconomic predictors of both gambling and alcohol behaviors. The hypotheses for this investigation are as follows: 1) Based on social disorganization theory, it is hypothesized that neighborhood disadvantage will predict gambling and problem gambling over and above the effects of gender, age, race/ethnicity and individual socioeconomic status. 2) Because tests of problem behavior theory have shown that problem gambling and alcohol abuse co-occur, it is hypothesized that the same neighborhood and sociodemographic factors which predict gambling and problem gambling will also predict alcohol use and abuse. The first survey of adults carried out in 2000 resulted in 2,631 completed interviews and the second survey of youth and young adults was initiated in 2005 and resulted in 2,274 completed interviews. Despite the time lag between the two surveys, analysis of gambling data from both surveys indicated there were no meaningful differences and thus, no evidence of a chronological gambling trend in the time between the two surveys . Descriptions of the sampling methods have been reported for each study separately . The combined dataset was created by using the ADD FILES command in SPSS allowing for comparable variables from both datasets to be merged. --- METHODS --- Sample and procedure In both the adult and youth surveys, cases were statistically weighted inversely to their probability of selection and weighted to align the sample with the gender, age and race distributions for the U.S. The final weight variable was scaled so that it had a mean of 1, and the weighted N equaled the true N. Weights from the two separate surveys were placed unaltered into the combined dataset, such that the weighted N of the adult survey was equal to the true N of 2,631 and the weighted N of the youth survey was equal to the true N of 2,274. The gender and race distributions of the combined sample are approximately equal to the distributions in the U.S. However, the age distribution of the combined file does not equal the age distribution of the U.S., because young people aged 14 to 21 years are over-represented in the combined file. --- Measures Dependent measures -gambling. Three measures of gambling in the past year were used for the present analysis -any gambling, frequent gambling and problem gambling. Both surveys included questions on the frequency of specific types of gambling during the past year. These types were: raffles, office pools, and charitable gambling; pulltabs; bingo; cards, not in a casino; games of skill, e.g., pool, golf; dice, not in a casino; sports betting; horse or dog track; horses or dogs off-track; gambling machines, not in a casino; casino; lottery; video-keno; internet gambling; and other gambling. The gambling in the past year variable was created by summing the frequency of these types of gambling and recoding the variable to produce dichotomous variables indicating any gambling in the past year and frequent gambling, i.e., 52+ times in the past year. Both the youth and adult surveys also included two problem/pathological gambling scales. The first scale was the Diagnostic Interview Schedule for pathological gambling , which contains 13 items such as preoccupation with gambling and needing to gamble with increasing amounts of money to get the same excitement . Endorsement of five or more items is considered pathological gambling with three or more items considered problem gambling . The second scale was the revised South Oaks Gambling Screen for adults and the comparable version of the South Oaks Gambling Screen, Revised for Adolescents . The 11 items which were common to both the adult and adolescent versions of the SOGS were added to the DIS items. Three or more symptoms denoted problem gambling for the present analysis. [It should be noted that although there is some overlap in subject matter measured by the two scales, no two of the individual questions had a high enough correlation to be considered redundant; most of the correlations between similar items were in the 0.2 to 0.5 range ]. Dependent measures -alcohol. Three measures of alcohol use in the past year were used for the present analysisany alcohol use, heavy drinking and alcohol abuse and/or dependence. Both surveys included quantity-frequency questions for alcohol consumption. A drink of alcohol was defined as a drink of beer, ale, malt liquor, wine, fortified wine, wine coolers, liquor, and flavored malt beverages or any other beverage containing alcohol. Drinkers were defined as those respondents who indicated that they had a drink of any beverage containing alcohol in the past 12 months. A dichotomous measure of heavy drinking was based on whether or not respondents indicated that they drank five or more drinks in one day on 12 or more days in the past 12 months. DSM-IV-based measures of alcohol abuse and dependence also were included in the adult survey and the youth survey , Light . Independent variables. Gender was coded 0 for females and 1 for males. For this report, seven age groupings ranging from 14 to 17 years to 61+ years were derived from the continuous age variable. This report used a collapsed three-level race/ethnicity variable -black, Hispanic and whites plus all others. This classification allows sufficient numbers in each group for meaningful analysis. For the adult survey, socioeconomic status was based on the mean of two equally weighted factors: respondent's years of education and respondent's occupational prestige. When one of these factors was absent , the other was used. Only 129 respondents had any missing data on the SES variable. For the youth survey, the measure of socioeconomic status was based on the mean of four equally weighted factors: father's years of education, mother's years of education, father's occupational prestige and mother's occupational prestige. Occupational prestige was coded based on the census occupational categories . Both of these SES variables were scaled from 1 to 10 and had similar variances, enabling them to be merged for the combined analysis of the two studies. A five-level categorical variable, based on fifths, was also derived for use in the descriptive analysis. Neighborhood disadvantage is an ecological measure based on objective data from the respondent's census block group . Data from each respondent's census block group was attached to her/his case. The four census block-level variables used to create the neighborhood disadvantage scale were: percentage of households on public assistance; percentage of families headed by a female; percentage of adults unemployed; and the percentage of persons in poverty. These variables were standardized and averaged with equal weights. A standardized Z score was used in the logistic regression analyses with a score of 0 representing average neighborhood disadvantage. --- RESULTS Table 1 gives the demographic distributions for the various levels of gambling and alcohol use in this study. Gambling in the past year is more prevalent than drinking any alcohol in the past year ; this pattern is consistent for both females and males . Males have twice the rate of frequent gambling as females ; similarly, males have twice the rate of problem gambling as females . Heavy drinking and alcohol abuse/dependence are likewise twice as common among males as females with 19% of males drinking five or more drinks on 12 or more days compared with a rate of 9% for females; 12.4% of males, as compared with 6.5% of females, have past year alcohol abuse or dependence. Both overall gambling and drinking are highest within the 22-30 year age group. However, the age-related patterns of frequent and problem gambling show differences from the age-related patterns of heavy alcohol use and alcohol abuse/dependence. Frequent gambling is highest in the three middle age groups, 30s through 60s, and problem gambling is highest among the 22-30 age group and the 31-40 age group . In contrast, heavy drinking is highest in the 18 to 21 age group and likewise, alcohol abuse/dependence is highest in this same 18 to 21 young adult age group. As was the case for age patterns, race patterns are different for frequent and problem gambling than for heavy drinking and alcohol abuse/dependence. For example, whereas blacks have higher rates of frequent gambling and problem gambling than whites/others, blacks have lower rates of heavy drinking and alcohol abuse/dependence than whites/others . Frequent gambling and problem gambling are highest in the lowest socioeconomic group and highest in the group with the greatest neighborhood disadvantage . These same socioeconomic and neighborhood patterns do not apply to alcohol use and abuse. Overall alcohol use is highest in the two highest SES groups and highest in the two groups with the best neighborhood conditions. For heavy drinking and alcohol abuse/dependence, there is no consistent relationship to SES or neighborhood disadvantage in this large dataset. Logistic regression analyses were used to predict dichotomous gambling and alcohol variables after entering all of the predictor variables. Table 2 shows the results for the three gambling dependent variables. In predicting any gambling in the past year, being male results in an 84% increase in gambling over being female. The age category 22-30 was set as the reference group in the logistic regression analyses. The age 31-40 is not significantly different from the reference group whereas all of the other younger and older age groups have a significantly lower likelihood of gambling than the reference group. Being black is significantly predictive of a lower odds of being a gambler as compared with all other race/ethnic groups. Being Hispanic versus all other racial/ethnic groups results in no difference in gambling prevalence. Neither socioeconomic status or neighborhood disadvantage are significant predictors of overall gambling in the past year. The logistic regression results for frequent gambling and problem gambling show some striking differences from the 'any gambling' analyses. Males have over twice the odds of being a frequent gambler or a problem gambler as do females. Blacks have a significantly increased odds of being a frequent gambler or being a problem gambler as compared with all others. Lower SES and higher neighborhood disadvantage both are significant risk factors for frequent gambling and problem gambling. The interaction between SES and neighborhood disadvantage was significant in predicting problem gambling after controlling for all main effects and all interactions between SES and neighborhood disadvantage and age, gender, and race/ethnicity. Thus, problem gambling was highest when both SES is low and neighborhood disadvantage is high . [Note: To investigate whether or not neighborhood disadvantage might simply be a proxy for gambling outlet density, we used the best available data from this study. In the adult gambling survey, respondents were asked a series of five questions regarding how convenient it was for them to buy lottery tickets, play bingo, play video gambling or slot machines, visit a horse or dog track, and visit a casino. There were four response choices ranging from very inconvenient to very convenient. The five variables were averaged for a "convenience/density" variable. The correlation between the neighborhood disadvantage score and total gambling symptoms in the past year was .166 . Next, a partial correlation analysis was run. The correlation between neighborhood disadvantage and gambling symptoms, controlling for the "convenience/density" of gambling, had a negligible decrease to .165. Thus, the significant effect of neighborhood disadvantage on gambling symptoms did not diminish with convenience of gambling in the analysis.] Table 3 shows the parallel logistic regressions for the three dichotomous alcohol variables. Males have a 30% increased odds of being a drinker as do females. Respondents 21 and younger, as well as respondents 41 and older, have a decreased likelihood of being a drinker as compared with those in their twenties and thirties. Both blacks and Hispan- ics have a decreased odds of being a drinker as compared with others. Persons with higher socioeconomic status are more likely to be drinkers than those in lower socioeconomic groups. Similarly, those with less neighborhood disadvantage are more likely to be have consumed any alcohol in the past year. Males have over twice the odds of being heavy drinkers and having alcohol abuse or dependence or both in the past year than do females. In general, older adults are significantly less likely to be heavy drinkers and less likely to have alcohol abuse/dependence than are respondents in the twenties reference group. However, young adults aged 18 to 21 years are significantly more likely to be heavy drinkers and to have alcohol abuse/dependence than are their counterparts in the 22 to 30 year reference group. Strikingly, 18 to 21 year olds have over 6 times the odds of having alcohol abuse or dependence in the past year than those aged 22-30 years. Blacks have a significantly lower likelihood of heavy drinking and they have a lower likelihood of having alcohol abuse/dependence than all others. There is no significant relationship between SES and heavy drinking; however, higher SES is associated with a somewhat higher likelihood of having alcohol abuse/dependence. There are no significant relationships between neighborhood disadvantage and heavy drinking or alcohol abuse/dependence. Interactions between neighborhood disadvantage and SES were not significant for all three alcohol variables. --- DISCUSSION Despite decades of theory and research linking neighborhood disorganization to crime and delinquency, virtually no research has been carried out to examine the potential links between neighborhood disadvantage and gambling behaviors and problems among the general U.S. population; and very little research has tested the effects of neighborhood disorganization on alcohol abuse. In this paper, we examined the effects of neighborhood disorganization on gambling behaviors and problem gambling, as well as the relationships between neighborhood disorganization and alcohol use and abuse/dependence. We examined these relationships in a large combined sample of two comparable U.S. surveys with respondents ages ranging from 14 years to 90+ years. Using such a large general population sample allowed for a consideration of a broad range of sociodemographic variables, including gender, age, race/ethnicity and individual socioeconomic status in addition to the ecological variable of focus, i.e., neighborhood disadvantage. The sample also allowed for analyses involving interactions, especially the interaction between neighborhood disadvantage and SES for its predictive effects on alcohol and gambling variables. This approach is a significant methodological advancement for neighborhood effects research. The findings from this study support, in part, the first hypothesis in that neighborhood disadvantage significantly predicts frequent gambling and problem gambling after accounting for the key sociodemographic factors -gender, age, race/ethnicity and respondents' socioeconomic status. However, neighborhood disadvantage does not predict any gambling in the past year. Neighborhood disadvantage appears only to be linked with gambling behaviors which reach frequent or problematic levels. These problem gambling findings are consistent with the early foundations of disorganization theory as developed by Shaw and McKay such that neighborhoods with low economic indicators showed high rates of delinquency and crime. Such socially disorganized neighborhoods were characterized as lacking the capacity to regulate behaviors thereby disrupting community social organization and resulting in elevated rates of delinquency and crime. Our finding linking neighborhood disadvantage with problem gambling is also consistent with the seminal work of Wilson which conceptualizes the effects of living in neighborhoods that are impoverished as "concentration effects". According to Wilson , these impoverished neighborhoods have a lack of access to jobs and job networks, lack of quality schools, and lack of exposure to conventional role models. Sampson and Wilson characterize social disorganization as the inability of a community structure to realize common values and maintain effective social controls. Neighborhoods remain viable if support comes from economically stable families . Such neighborhood disadvantage may provide a likely environmental context for the development of problem gambling. Although social disorganization theory emphasizes the effects of ecological influences on problem behaviors over and above individual characteristics, our analyses show strong effects of individual characteristics on gambling behaviors and problems. Being male, being black and having low socioeconomic status are significant risk factors for problem gambling. In our national sample, there is also evidence of a person-ecological interaction, such that having low socioeconomic status in conjunction with living in a disadvantaged neighborhood is a highly significant condition for the occurrence of problem gambling. Thus, these findings suggest that person-environment effects are important considerations for a fuller understanding of problem gambling. Despite previous findings that problem gambling and alcohol abuse are correlated , all of the predictors of alcohol abuse are not the same as they are for problem gambling in the present study. Thus, the second hypothesis, that neighborhood disadvantage would predict alcohol abuse as it does problem gambling is not upheld. Alcohol abuse is not significantly related to neighborhood disadvantage, nor to the interaction of neighborhood disadvantage and SES. Alcohol abuse is highly prevalent for males and young people, in particular, across neighborhoods of all types. These findings suggest that in spite of the correlation between problem gambling and alcohol use/dependence, there is also something unique in explaining problem gambling. It may be that people who live in disadvantaged neighborhoods do not see many role models of financial success achieved through conventional means, and therefore, gambling may be viewed as one of the few opportunities for financial advancement . Gambling perhaps provides the lure of a means for obtaining immediate money. Although there are a limited number of studies examining neighborhood disadvantage on respondents' substance use, Boardman et al. found a significant effect of neighborhood disadvantage on drug use in a general population adult sample in the Detroit area. Their measure of neighborhood disadvantage is the same one we used in the present study. Boardman et al. also controlled for individual level variables in their logistic regression model. The authors suggest that "if drug use [unlike alcohol use in our study] is more common in highly disadvantaged neighborhoods, then social contact among neighbors may be one of the primary mechanisms through which the increased risk of drug use operates" . Although our measures of neighborhood disadvantage are comparable, the samples are very different -a targeted urban adult sample in Detroit versus a national U.S. sample including adolescents as well as adults. Furthermore, illicit drug use is often perceived as more deviant than alcohol abuse although the devastating consequences are often the same. Also, unlike the present findings, Karriker-Jaffe et al. found that neighborhood disadvantage was related to heavy drinking by African-Americans but not by whites. Further research is needed to better understand the differences in neighborhood effects for problem gambling and alcohol abuse. A limitation of this study and of neighborhood effects research in general is that there may be a differential selection of individuals into certain disadvantaged communities . Thus, while neighborhood disadvantage may influence the development of problem gambling, individuals with gambling problems may move to areas where the cultural context of problem gambling is evident. In addition, a preponderance of problem gambling in a neighborhood may well contribute to further decline in that neighborhood. The present cross-sectional study cannot sort out the bidirectional effects of problem gambling and neighborhood disorganization. Future longitudinal studies are required to address this issue. A further limitation of the present study is that it did not examine social process factors in neighborhoods such as the frequency and types of interactions among neighborhoods. Such interactions are likely to have effects on individual's gambling behaviors. --- CONCLUSIONS The important finding from the present national U.S. survey of 14 to 90+ year olds, is that neighborhood disadvantage has a highly significant effect on problem gambling over and above the significant individual effects of gender, age, race/ethnicity and socioeconomic status. Furthermore, when individual socioeconomic status is low and neighborhood disadvantage is high, problem gambling is at the highest level. This study provides strong evidence that neighborhood ecology as well as the interaction of individual and neighborhood factors are important predictors of problem gambling. Future research should extend neighborhood ecological research by examining social process factors in neighborhoods such as frequency and types of interactions among neighbors and types of social control such as availability of gambling and alcohol in communities.

Background and aims: Based on social disorganization theory, the present study examined the effects of neighborhood disadvantage on gambling behaviors and problems as well as on alcohol use and abuse. Methods: Findings were based on a combined sample of two representative U.S. telephone surveys of gambling and substance use. One survey (n = 2,631) included adults 18 years and older and the second survey (2,274) included young people aged 14-21 years old. Results: Neighborhood disadvantage had a highly significant effect on problem gambling over and above the significant individual effects of gender, age, race/ethnicity and socioeconomic status. Alcohol abuse did not show the same relationship to neighborhood disadvantage as did problem gambling. Furthermore, when neighborhood disadvantage was high and individual socioeconomic status was low, the highest levels of problem gambling were observed. Conclusions: This study provides strong evidence for the effects of neighborhood ecology on the occurrence of problem gambling.

Introduction Globally, over 35 million people are living with HIV, of whom 3.2 million are under the age of 15 years [1]. An estimated 2.1 million adolescents were living with HIV in 2012 in low-and middle-income countries [1]. However, statistical data on young adolescents are limited and this is likely to impact on the HIV response efforts for this age group [1]. Sub-Saharan Africa remains most severely affected by HIV/AIDS [2]. It has been estimated that of the 190,000 children living with HIV in Uganda, 176,948 are below 15 years [1]. Most HIV infections in children occur during the perinatal period, and have resulted from mother-to-child transmission of HIV. This implies a significant burden of care considering that the children under 15 years constituted the majority of HIV positive children, and were much more likely to be entirely reliant on a carer compared to their counterparts who were 15 years or over [2]. Over the last three decades, the HIV epidemic has left many families in sub-Saharan Africa severely affected by the loss of family members, who performed crucial roles and shouldered key responsibilities [3,4]. A key feature during this epidemic has been the death of family members, often parents aged 20 to 40 years, resulting in the emergence of many orphaned and sometimes HIV positive children in the years prior to the roll out of anti-retroviral therapy . The burden of care for these children has often been undertaken by older persons, usually grandparents [5][6][7][8][9][10]. A number of writers have shown that the care for orphans has ranged from financial support for tuition and other scholastic requirements of older children to day-to-day nurturing care for the younger children [10][11][12][13]. In addition to the challenges in meeting children's needs, carers faced a variety of health problems usually associated with their advancing age [9,[14][15][16]. These older foster carers face multiple challenges which impact negatively on the care-relationship and the kind of care given to the foster child [9]. With ART, there is a growing likelihood that HIV positive children can grow into adulthood even in resource limited countries like Uganda. However, their physical, social and emotional development is determined by several factors and by the nature of the care relationship. Other scholars [17,18] have observed that family and peer level factors affecting children and adolescents with HIV mainly relate to adherence, disclosure and social support. Additionally, it has been found that the style of parenting by older carers, which sometimes contrasts with that of younger foster carers, had various social and health outcomes for the children [9]. For instance, older carers were not comfortable talking to children directly about sex, sexual development and HIV. HIV positive children are rendered particularly vulnerable, not only because they require very specific biomedical care, but also because they need considerable psychosocial support as they face the challenge of coming to terms with an illness that has been transmitted to them by a parent [19]. Nevertheless, little is known about the social factors mediating access to ART for HIV positive children [20]. Understanding the role older carers play here is critical to ensuring children are able not only to access health care but also education and other social and community support structures [21]. The challenges older carers experience are often considerable and can undermine the quality of care for HIV positive children. Furthermore, these challenges may ultimately impact on the nature of older carers' relationships with the children in their care. It is important therefore to deepen our understanding of the context-specific family processes that facilitate progress in physical, social and cognitive domains of children's development, and how these processes are influenced by the interplay between parental social support and the changing social, economic and psychological demands placed on the family system [16,22]. Our purpose in this paper is to examine the care relationship between older carers and HIV positive children in order to understand the needs of both carer and child, and the possible challenges of caregiving and how these may influence older persons' relationships with children in their care. We will also identify strategies that might improve care for children and their older carers. --- Theoretical Orientation Our theoretical framework draws from the Family Systems Theory . According to the FST, families-or households , for the purpose of this study-are considered systems because they are made up of interrelated elements/individuals, and have regular interactions among their members who are interdependent on one another [23]. In the context of older carer-child relationship, grandparents and children in their care have different roles and expectations, and needs. But in this particular case, we conceptualise the roles, expectations and needs as revolving around and mediating the provision of social care, that is, both older carer and child have social care duties, and are guided in the provision of such care by roles established normally by the carer, expectations based on these roles, and competing needs of both carer and child. Another concept that is relevant for understanding the older carer -child relationship is that of "circular causality" [24]. Circular causality holds that interactions among family or household members trigger a cycle of behavioural responses; that the behaviour of each member influences and is in turn influenced by that of the other members; and that these mutual influences can be positive or negative, and repeatedly reinforce the behaviour of family or household members, thereby sustaining the cycle of responses [25]. We view circular causality, particularly its key feature of reciprocity in terms of mutual behavioural influences, as a useful conceptual tool for understanding the older carer-child relationship. Thus, we propose that when roles are performed as expected and needs met, the behavioural responses of both the older carer and the child are likely to be positive, resulting in an enduring and mutually beneficial relationship. This may not be the case when the older carer and the child are not performing their roles satisfactorily and therefore not meeting each other's expectations and needs. In this case, the behavioural responses of both parties are likely to be negative, resulting in a deterioration of the relationship. We draw on these theoretical perspectives to interpret and discuss our findings, and to recommend strategies for promoting a mutually beneficial care relationship between older persons and HIV positive children. --- Methods --- Sampling We purposively selected 18 HIV positive children and 22 children whose HIV status was unknown, and their older carers from a clinic run within the General Population Cohort of the Medical Research Council/Uganda Virus Research Institute Research Unit on AIDS, a cohort of 20,000 people based in Kalungu District [26] and a Government Health Centre IV, providing paediatric HIV care in the same district. Our working assumption was that those children whose status was unknown were mostly HIV negative. From each clinic eligible children were selected with the assistance of clinic staff who were asked to select a specified number using stratified random sampling . We recruited 22 children whose HIV status was unknown because we wanted to see if their experiences were different from those of the children who are HIV positive. We obtained a random sample of older carers with children from the GPC. The older carers comprised of seven men and 33 women. These had been drawn in a context where over 2200 older carers of children aged 13-17 years were aged 50 years and above, with a gender distribution of 57% female and 43% male. We included children aged 13-17 years and older carers aged 50 years and above. The age of the carers was amended from 60 to 50 years, on realizing that there were more younger older carers in the area than those aged over 60 years. --- Data Collection Following provision of written informed consent and assent, interviews with participating children were conducted by young interviewers under the age of 25 years, and those with older carers by older interviewers aged over 60 years. Efforts were made to match the participant's gender with that of their interviewer. Interviews lasted between 45 and 60 min. Pilot interviews provided insights into the participants' background, HIV/AIDS awareness, and ART treatment , on the basis of which we modified our interview guides for the main study. For the main study, we conducted three separate semi-structured interviews with 40 children and 40 older carers over a period of one year. To ensure sufficient privacy for the children to freely describe their relationship with their carers, interviews were held mostly at school but when they were held at home, sufficient care was taken so that the carer was not in the vicinity of the interview site. In total, a little over 220 interviews were conducted. The team used a topic guide to explore issues about the participant's background, relationship with carer/child, expectations from carer/child, HIV/AIDS awareness, HIV disclosure, experience with ART and caring practices, among other topics. The interviews were recorded but the interviewers also prepared detailed interview summaries. Data analysis was based on detailed summaries prepared by experienced interviewers and audio recording of participants' narratives. The rationale for use of summaries was to save the enormous time that would have been spent transcribing all the interviews. --- Data Analysis A framework analysis was done. The team read and reread the detailed interview summaries often referring to the respective audio recordings. Reflecting on the concepts highlighted in our theoretical framework, two overarching categories-the care relationship and caregiving challenges-were created. These categories were then used to organise the themes emerging from the data. With the help of MS Excel, key themes and sub-themes were classified on a matrix, capturing participant narratives accordingly to illustrate each theme. The themes emerging from this analysis included reciprocal care relationship, carers' poor physical and psychosocial wellbeing, poverty and reliance on cultivation-based livelihoods, inadequate support from family members and other sources, and carer-child tensions. We describe these themes below. Ethical clearance was obtained from the Uganda Virus Research Institute's Research Ethics Committee and the Uganda National Council for Science and Technology. --- Results The findings are based mainly on interviews with 18 HIV positive children and their older carers. However, for comparison purposes we occasionally make reference to perspectives of children whose HIV status was unknown and their older carers. --- The Care Relationship --- Reciprocal Care Relationship Our findings reveal that even in the context of HIV/AIDS where HIV positive children might be viewed simply through the lens of their status as care recipients, the care relationship between older person and child was more complex, demonstrating that carers and children were simultaneously caregivers and care recipients. In 11 of the 18 cases where the older person was providing care to an HIV positive child, both the child and the older carer had needs that they expected to be met within the care relationship. Indeed in these cases, caregiving was reciprocal. The HIV positive children looked to their older carers for various forms of support such as providing for basic needs including nutrition, shelter, and education, ensuring that drugs are administered appropriately and on time and that clinic appointments are honoured. --- My grandmother cooks for me tea, food, including vegetables, and if she is not at home, she tells the other children at home to get me vegetables and prepare them. If the food I want to eat is not available at home, my grandmother buys and prepares it for me alone when the others are eating the food that was available. I cannot go in the garden before preparing something for him to eat, and we spare something for him from supper for the next morning. Even if it is posho still we spare a piece for him. When he is going to school, I pack something for him to eat. I have sugarcane here and if we are still busy with household work, I tell him to get a piece of sugarcane; I don't want him to feel hungry. At the clinic I was told that the child had to eat something at any time he liked. Several children also talked about the care they provided to their older carers. There are no disadvantages being cared for by an older carer but it is just that my grandmother who would care for me keeps on falling sick and so I take care of her until she recovers and she also does the same. On the other hand, the older carer had a different set of needs, which we categorised as immediate and strategic needs. Immediate needs pertained to instrumental support in executing routine tasks that were necessary for household survival. It is in relation to these needs that the older carer counted on the child for help, particularly in performing household tasks that were too physically demanding for the older carer to perform. I am growing old now...Although he has not yet reached that stage of buying me those items, he is the one who fetches water for me now using a bicycle and our well is far from here. …he is the one taking care of me because he fetches the water, brings the feeds for the animals and cooks the food that we eat. In contrast, older carers' strategic needs pertained to longer term concerns about household income security and viability, and future care of themselves when the children in their care are no longer available to take care of them. Unlike immediate needs that related to routine instrumental support, strategic needs were psychological and related to the older carers' wish to overcome the sense of insecurity regarding their future financial and other forms of social care. The carers did not feel confident that they would have anyone to care for them in the future. There is a Kiganda proverb which says: "better try than never." Me I take him as a cassava stem; it grows wherever you throw it, and in future you can find yourself getting food from it. I am just looking after him . You never know, God can help and I start gaining from him. When you are a parent, you have hope in your children. Notably, this view was not restricted to older carers of HIV positive children. One of the carers of children with unknown status had the following to say: --- Every person looks after a child hoping that if the child grows up, he or she might help…when the old person becomes helpless. When they grow up, they have to help me … Do you think we shall be able to cook food for ourselves in future? We are suffering from different types of illnesses and that is why we have educated this young generation so that they will help us in future. The only worry we have is that what shall we do when they grow up and leave our home? I know God will provide means for us. . --- Caregiving Challenges In the majority of cases, older persons encountered significant challenges in their caregiving role, which undermined their ability to optimally provide for children in their care. In fact, the practice of children reciprocating care to their older carers may be better understood within the context of these challenges, which included poor physical and psychosocial wellbeing, poverty and reliance on cultivation-based livelihoods, and inadequate support from family members and other sources. --- Poor Physical and Psychosocial Wellbeing Poor health was one of the major challenges experienced by older persons in their role as carers, affecting 13 of the 18 older carers of HIV positive children. These older carers reported a range of health problems that were mostly associated with their advancing age, including painful backs and limbs, hypertension and diabetes. These health problems limited their ability to execute caring duties within the household, and to access health care for the child. Another health problem faced by older carers was stress. Stress was usually the result of the carers' concerns about their poor physical health and the attendant failure to effectively provide for their dependents. But another common source of carers' stress was general economic hardship-an issue that is explored in more depth in the subsequent section. Depicting the stressful situation she was going through, one carer observed: --- It hurts my feelings because I cannot provide all the necessary things he needs. The journey from here to the health centre is very long. When I do not have money I go and kneel before the bodaboda riders so that they allow me to pay afterwards. Similarly, an older carer of a child whose status was unknown had the following to say: The father of these children refused to pay school fees for them. He knows I cannot do anything to him since I am an old woman. At times I think that if I was a male person, he would not have refused supporting his children. Now he says that being a woman what I shall do? Had I been…still energetic, he would have feared that I would report him to police. --- Poverty and Reliance on Cultivation-Based Livelihoods Poverty was another major caregiving challenge for older persons, and was reported by 12 of the 18 carers of HIV positive children. In fact, our findings reveal that poverty was closely linked with reliance on insecure small-scale agricultural livelihoods and associated food scarcity problems. A number of older carers were land poor, with others heading landless households. In some cases the cultivated land belonged to the household members. In other cases, where the household did not own land, older carers hired land for cultivation. But some carers, in a determined effort to provide for the children in their care, undertook several concurrent yet physically demanding income generating initiatives. For example, a number of carers undertook cultivation on hired land while also engaging in casual labour: I hire a piece of land where we can dig. We first get somewhere to work and get money, which helps to hire another place in turn. Then we can grow maize, beans and potatoes. We also work in the village and get money for commodities like soap, paraffin and other things. When you came I was going somewhere to work. The children do not have anybody else to help them… Although agriculture provided food for many families, it was vulnerable to serious environmental crises. During the course of our one-year study, participants often complained about food scarcity due to a lack of rainfall in the region, coupled with an inability to purchase food from the market. Indeed older carers of HIV positive children bore the brunt of the food crisis and poverty, as they had to deal with the everyday anxiety about providing an adequate quality and quantity of food for a child on antiretroviral therapy . getting food is not that easy because maize flour is expensive; so we take porridge for lunch and make posho for supper. When we make posho, it is not enough and the boy who is on ART wants more food. He needs a lot of food. I have a big family and so I cannot make posho for two meals each day. There must be enough food for him every day because if he takes three days without enough food, he gets dizzy all the time. That medicine needs a lot of eating and sometimes I get a piece of posho from my share and add it to his. Every time he is looking for something to eat, and so it is a very difficult drug. --- Poverty is the greatest challenge that I have, at times I feel like buying him fish but there is no money. They advised us to feed them properly, what I do is to look for greens so that I feed him on them. Because I stay on the roadside, the fishermen pass by here every day, you see him expressing a desire for fish yet I do not have money to buy it. You cannot buy one for him only, you have to buy for the whole family. At times sugar gets finished and I find myself in a miserable situation where I have nothing I can do. A number of older carers also mentioned the importance of ensuring that the HIV positive child always had something to eat at the point of taking their medication, based on the advice they had received from health workers. However, this was often difficult in a context where the household members did not have food reserves from the previous harvest and the financial means to purchase food from the market. You can also examine the situation at our home, we are living in a poor life, we do not have enough food, and so whatever little food I get is what we eat. It is not that I buy things which are special for her; I do not have that money. When I am going very early to work, I just tell her to prepare some tea for herself even if we do not have sugar, because they told her not to take the drugs on an empty stomach. In another case illustrating the hardships many families faced and, as a result, children's contribution to household income, a carer observed: Here at my home each child works for themselves and they support themselves. …Mostly he fetches water for other people who pay him 150 to 200 shillings a day, after fetching like five jerry cans. At times when he goes to school with that money he spends it on eats. If you look at his shoes, they are torn and he very much needs shoes. Poverty among older persons was a challenge affecting not just those caring for HIV positive children but also carers of children whose status was unknown. This was repeatedly expressed in narratives by both the carers and their children. --- One older carer observed: --- My health status is not bad except this food scarcity which is disturbing us. When you go in the village to look for money and you can't buy food from the money you get only to buy home essentials. So you have to eat what you get, the money I get is not enough for both food and my home essentials. And you can't rely on casual labouring only, you have to work at home. This experience was echoed by a child, as noted by the interviewer: about the challenges she has faced being cared for by her grandmother, she told me that she might be in need of Vaseline, clothes and other things yet her grandmother has no money to provide them…She added that she has not been providing what she needs in time. So she is looking for a job during her vacation because she is still wearing the clothes that were bought for her when she was in primary, she has not received any clothes while in secondary ….. . It should be noted that in Uganda, children especially in the rural areas attend primary school between the ages of six and 15 years, and secondary school between the ages of 13 and 20 years. --- Inadequate Support from Family Members and Other Sources In the context of poor health, poverty and food crises, older carers expected support in the caring role from extended family and other sources. Although the majority reported receiving some form of support from family members, and in other cases, from organisations working in the local area, there were a number of cases where carers and children reported a lack of or inadequate support from these sources. One carer narrated: Before I got that child, we sat in meeting as a family and relatives. It was decided by all to follow his mother's will that I should keep the child and they would be helping me to meet the needs of the child, but none of them responds. In another case, an interviewer records a child's frustration over the lack of family support: I asked her which other relatives can take care of her and she responded that her relatives are there but cannot take care of her because her father died when she was three months and her mother brought her up until 2006 when she was taken by the organization that is taking care of her. She added that their livelihood and the number of children they have cannot enable them take care of her. Likewise, concern about family members not providing adequate support towards caring for HIV positive children was sometimes exacerbated in situations where one of the surviving parents got a new spouse who influenced support to the child, as documented in an interview with one child: He told me that he would like to get sufficient support because his father does not support him enough. For instance, "when I go to my father to ask for something I want to eat, he tells me that is not what I want. And in most cases, it is his wife who says that do not give it to him," he said. His father got another wife and they stay together across the road and it is basically this other wife who mistreats him. When he goes to ask for scholastic materials from his father, he tells him to go back promising to send them but he does not do so. When he asks for school fees of twenty thousand shillings in total, he sends him only ten thousand shillings or five thousand shillings only. He did not pay for his fees the whole of the previous term. He told me he also lacks beddings. . Perceptions of lack of support for older carers were also expressed by those caring for children whose status was unknown: I very much like these children to be educated but I do not have money. I pray to God to see that at least they reach a certain stage. The girl wants to be educated but her father has no responsibility for her. Similarly, most participants perceived a general lack of support from sources other than extended family, with some of the carers expressing pessimism about the prospect of such support. In this community of ours, there are no people who have ever come providing support to us, we are always supporting ourselves. I do not even think I can get anything different because of being an old person. Some of the carers expressed frustration over unfulfilled promises from some individuals or agencies that had initially raised their expectations of much needed support. There is nowhere in the area where assistance is given to me as a carer of an HIV positive child. The government even promised to assist older people but has never done so. This sentiment was also expressed by older carers of children whose status was unknown. I have never heard about organisations that offer support here, and there is no one who can help you. Unless someone is your relative and he or she gives you something but there is no other support you can get from anywhere else. Some time back, they came here and registered us, telling us that they were going to offer us mosquito nets but we haven't yet got them. Another time they called us, the old people, and we went to the trading centre, we waited for the people who were going to register us to get the support for the whole day and they didn't turn up. Now I have decided not to go back again even if they come. --- Tense Relationships In some cases, the care relationship was characterised by a mutual perception that each was not meeting the other's expectations and needs. This was exhibited by six of the 18 carer/child dyads. Older carers complained of the challenges of bringing up children, especially adolescents, who would not do as they were told. In one such case, the adolescent would neglect his household tasks by repeatedly staying away from home. These tensions made relationships difficult between older children and their carers: I used to advise her to behave well but she would not buy my advice and at last she got infected with HIV. She used to move away without my consent and would stay there for some time before coming back home. Whenever I blamed her, she would abuse me and given my age, I just left her for fear that she might attack me and beat me. Carer-child tensions were also reported by carers of children whose status was unknown: The care is not bad but one of the older child gives me a hard time when she comes back late in the evening from school. When I ask her why she becomes annoyed and puts on a tough face. On the other hand, children expressed perceptions of mistreatment, which indicated that the kind of care they were receiving fell short of their expectations and needs. Some children complained of the harshness of carers: I do not quarrel with grandmother like I do with my grandfather who barks at me whenever I am washing utensils, saying that I should stop. Maybe he wants me to leave the utensils there and the flies fly all over them. Grandmother does not bark at me; she does not quarrel or abuse me. Other children complained about numerous day-to-day tasks that they were expected to accomplish: From digging, we are expected to fetch water and relocate cows from one grazing place to another, which sometimes we fail to accomplish because we are tired…. Children's time-consuming domestic workloads sometimes had repercussions beyond the home, as they were often late at school. This was illustrated by school teachers disregarding the children's seemingly extenuating circumstances, and maintaining a zero tolerance policy on late arrival at classes. Thus, these children were also penalised at school. In one instance, the late arriving child was recurrently denied the opportunity to learn lessons already covered in class, in addition to undergoing corporal punishment, as the above participant told us: Then when we get to school late as a result of these domestic chores, we are also beaten. The teachers just tell us that we should not explain anything regarding the late coming; they beat us and do not repeat for us what we have missed. This happens all the days. While this compounded the child's emotional distress, it also might impact the child's school performance. --- Discussion Evidence from this study points to the complexity of the care relationship between older carer and HIV positive child, by illustrating that neither the older carer nor the child could be described as purely a caregiver or care recipient. Because this relationship emerged in the context of HIV/AIDS and the child's seropositive status and need for care, it is easy to perceive it as one in which the care is unidirectional, where the "needy" child is strictly the recipient and the older person the provider of care. However, our findings have shown that HIV positive children in the majority of cases reciprocated care to their older carers. Previous research [27][28][29] has explored reciprocity in parent-child relationships: One of these [28], for example, has described "dyadic reciprocity," with a focus on meanings, behaviours and social relationships between parent and child. Another study [29] has referred to "dyadic mutuality" as entailing parent-child cooperation, shared positive emotion, and mutual responsiveness. Our findings build on those from previous work [9,22] and support the call for recognition of reciprocity of care in older carer-child relationships , lending credence to the argument that it would be inaccurate to view children solely as passive "receivers" in a care relationship [30]. We also observe that because older people play a crucial role in caring for children living with HIV, their needs and general wellbeing are as important as those of the children in their care. This is especially so for those who are suffering from age-related frailty with a heightened need for their own care. We found that a number of older carers, whatever their health status, suffered from stress and anxiety in their role which affected how they looked after the child. This is consistent with other research which has shown that carers of orphaned children, regardless of their age, suffer poor general health and functioning and increased levels of depression, anxiety, and post-traumatic stress [15,31]. On the issue of food scarcity we observe that this is a seasonal problem, although occasionally there are years of more extensive shortages. This lends credence to findings from previous research in the same area which point to this problem as a function of both poverty and environmental crises [32]. Another key finding of our study was the existence of tension within a significant minority of caring dyads, which we attribute to unmet mutual expectations resulting in stress on the part of both carer and child. We observe, for example, that the stress and anxiety that older persons experienced as they provided care to the children in a context of the carer's physical frailty, and poverty and deprivation might negatively influence the way they related to children in their care. This is consistent with findings from previous studies which point to grandparents' concerns about child indiscipline and intergenerational disharmony [33], and the stress that young carers experience as a result of shortage of nutritious food, on-an-off education and preoccupation with work [22]. Tension within the context of the older carer-HIV positive child relationship goes against the common perception in the African cultural setting that children are undoubtedly happy and satisfied when they are in the care of their grandparents [34].Viewed within the framework of the Family Systems Theory, tension within older carer/child relationships is the product of unfulfilled roles by some members of the family unit. Indeed as our findings indicated, this tension arose from an older carer's perception that the child was disobedient and not performing her/his roles as expected, and the child's perception that the carer was overworking them hence not meeting their caring expectations and needs. Study limitations. This was a small exploratory study whose findings may not be generalisable to the wider population although the issues raised may be illustrative of wider structural realities. In addition, interviewing adolescents was challenging since they were sometimes reluctant to discuss challenges especially those related to their carers. However, the participatory nature of the techniques used helped to mitigate against these potential pitfalls and ensured that participants felt comfortable to share information. --- Conclusions The implication of our findings is that programme interventions targeting children living with HIV need to pay attention to the needs of older carers who may struggle to provide emotional and physical support to their children. In particular, there is need for psychosocial support for carers and children, especially where they are living with HIV/AIDS. As well, there is need for social protection initiatives especially for older people in general as they face considerable financial challenges, more so because their livelihoods are largely reliant on subsistence farming that is prone to frequent environmental shocks. These initiatives will benefit those older people caring for children and those who are not. --- --- Conflicts of Interest The authors declare no conflict of interest. The founding sponsors had no role in the design of the study; in the collection, analyses or interpretation of data, in the writing of the manuscript, and in the decision to publish the results.

The care of children orphaned by HIV/AIDS in sub-Saharan Africa is often undertaken by grandparents, yet little is known about the care relationship between grandparent and grandchild. Our aim was to examine this relationship to understand the needs and responsibilities of both the HIV positive child and older carer and the nature of the relationship, and to assess the implications for care for the children and the older carers. A qualitative study was conducted with 40 purposively sampled children (13-17 years) and their older carers (50 years and above). Participants were recruited from two clinics in southwestern Uganda. Up to three semi-structured interviews were held with each participant. Data were analysed using a thematic framework approach. We found that the care relationship was mostly reciprocal: HIV positive children depended on carers for basic and health needs and carers counted on the children for performing tedious household tasks. The relationship was also characterised by challenges, sometimes causing tension between child OPEN ACCESS and carer. We conclude that: (1) interventions targeting HIV positive children need to also address the needs of older carers, and (2) carers and children would benefit from psychosocial support and social protection.

Background Despite increasing interest over the last decade or more in men's views of sexual health services and use of contraception, most published literature focuses on women. This is even more so with young people, where a focus on teenage pregnancy has resulted in young men being, if not invisible in the research, certainly harder to find. [1][2][3] Although, as a result of this policy focus in the UK, resources have gone into research aimed at exploring the needs and behaviour of young men, much of the resulting literature, and associated research 4 5 has been guidance from the Department of Health and Teenage Pregnancy Unit, which informs practice but does not appear in academic literature. Access to sexual health services designed for young people, taking into account their particular needs and fears, is important. 6 Boys have some of the same concerns as girls, such as visibility and the need for confidentiality, 7 8 but boys' sense of responsibility for contraception is not as well developed. 9 For both sexes negotiating condom use can be difficult. 10 11 Boys and girls wish for more, and better, information about sex; 12 boys are also much less likely to obtain information from their families. 13 It is important that the views of boys and young men are better understood, particularly with regard to responsibility for use of contraception. --- Young men's views on sexual health and contraception Second, it aimed to identify themes that may be important to this group, to be explored in a larger study. Focus groups were chosen as they are a useful method for exploring different perspectives within a social network and exploring group norms. 14 It was felt that this methodology was appropriate for this study, as it would allow the investigation of how group norms and expectations inform young men's views about responsibility for contraception. However, the potential disadvantage is that participants may occupy a position that is seen as acceptable within the group, which may obscure individual views. In addition, the facilitator was older and female, which again may have influenced the participants, in that to a large degree they conformed to a desirable positioning of themselves as being responsible. The original intention was to hold four focus groups: two with young fathers and two with non-fathers. Despite repeated efforts, aided by the boys and young mens' sexual health co-ordinator, it was not possible to recruit young fathers. Several reports have noted the lack of engagement of fathers, particularly young fathers, with sexual health services 15 and the difficulties of engaging them in research. 16 17 It was hoped that accessing young fathers via an established service would increase the likelihood of their participation. Unfortunately the Drop-in was not held during the initial course of recruitment for this study, and when it began again, attendance was very low. Despite meeting a number of young fathers at the Drop-in and discussing the project with them informally, they either declined to take part or, when they did agree, did not keep appointments. Two focus groups with non-fathers were held, and the results are presented here. The participants were recruited by youth workers asking for volunteers among young men who attended youth groups. The study was explained to the participants by the youth workers, and they were asked to consent to take part and for the focus groups to be recorded. Each group lasted for approximately 40 minutes, were run by the author, and were recorded and later transcribed. The topic guide included sources of knowledge about sexual health , access to sexual health services and obtaining contraception and responsibility for contraception. Transcripts were analysed using the constant comparative method to build up categories of data. 18 Those non-fathers who did participate were keen to talk about their views and experiences. Pseudonyms are used to ensure the confidentiality of the participants. The first focus group involved Joe aged 15 and Mike aged 16, both at school. They were friends, and the discussion was peppered with a great deal of teasing and joking. This group might more properly be described as a paired interview between friends, as their friendship did appear to have an impact on the way they spoke; they revealed information about each other that might not have been revealed in a more traditional focus group, for example, that one was a virgin and one had been in a steady relationship for almost 3 years. The second group involved Tom aged 14 , Sam aged 17 and Rob aged 18 . In this group, it was up to the individuals to reveal information about themselves, which they chose to do in terms of sexual experience, but unlike FG1, they did not know each other well enough to reveal each other's experiences. Attempts were made by the Young Men's Health Worker to recruit participants from groups of young men not employed or engaged in education, but as with the young fathers, they were not willing to participate. This highlights the difficulties mentioned above of engaging in research with young men who are themselves disengaged from education and employment. --- Results Participants thought that sex education should be introduced at a younger age and made part of the normal curriculum instead of being singled out and presented as something that was "frowned upon". In this way it was felt that it may become easier for young people to talk about sex and contraception. If they had problems that they wanted to talk about, the young men reported that they would consult their general practitioner. None of them mentioned talking to their fathers, and where mothers were mentioned there was a suggestion by Joe, however jokily made, that they might not be reliable sources of information as she had had him when she was 18 years old. In both groups, there was a sense of waiting until the 'right' time as far as having a family was concerned, and it was felt that this was influenced by family background, particularly where having a very young mother might normalise early childbearing. Participants acknowledged that sexual reputations and expectations play a part in determining behaviour, 19 but their views about responsibility and the use of contraception were largely delineated along gender lines, arguing that girls and boys think differently about contraception: Mike: Most boys are prepared not to use [ Mike appears to be more willing to allow for the influence of alcohol, and its loosening effect on people's inhibitions. This discussion highlights the dilemma faced by girls when deciding whether to carry condoms; not carrying condoms increases the risk of unprotected sex, but carrying them increases the risk of a damaged reputation. 10 Rob and Sam did not label girls this way, and felt that responsibility in that setting would still be shared. The differences between FG1 and FG2 may be as a result of the age difference of the participants. Although Sam and Rob discussed how boys would refer to girls as slags, they at no point said that they themselves viewed girls who carried condoms or had one-night stands as such. Asked whether they thought people went out to clubs and parties intending to have one-night stands, they both agreed that this was the case. Asked to think about whether a girl would ask a boy to wear a condom, and if she did, his likely reaction, they responded: Sam: They probably would [ask]. Rob: I don't think they'd say it, but if they did, I think the boy would say OK. The key difference as far as using condoms was concerned was how well one knew the partner: Sam: You probably should still use one… Rob: … if it's someone you don't know properly. Sam: Yeah. Unless you're in a relationship. It appears that a crucial differential in terms of both expected and acceptable behaviour was whether or not people are in a relationship, or having one-night stands. --- Discussion This study found that these young men regard contraception as largely a shared responsibility, or their own responsibility to protect themselves, although this viewpoint is mediated by differences in relationship status. The findings highlight the difficulties young women have in making choices about protecting themselves, essentially between protecting their health and physical wellbeing by carrying a condom, or protecting their reputation by not being seen to be a 'slag'. These findings are based on a small sample, took place in the north of England, and all participants were white and in full-time education. It is likely that attitudes may differ from those who are not in education, training or employment. Nevertheless, although the findings are not generalisable, they highlight areas for further study and considerations for choice of methodologies, particularly recruitment strategies. In a future study, it would be important to put in place recruitment strategies that were sufficiently wide ranging and inclusive to ensure that young men from different class and educational backgrounds are included in the research, as the tentative conclusions from this pilot study suggest that educational status plays a role in beliefs about responsibility. It would also be important to consider the sexual experience of those taking part in focus groups, both in the way that it might influence discussions, and how young men's views might change depending on their experience. Exploration of issues around gender and masculinities, 20 encompassing discussions about views of the life course, would enable deeper understanding of the roles that responsibility and family play for contemporary boys and young men. --- Competing interests None.

Background and methodology Despite increasing interest over the last decade or more in men's views of sexual health services and use of contraception, most published literature focuses on women. It is important that the views of boys and young men are better understood, particularly with regard to responsibility for use of contraception. This pilot study aimed to gain insights into young men's views of sexual health services and contraception; fi ve non-fathers aged between 14 and 18 years took part in two focus groups. The groups were recorded, transcribed and analysed using the constant comparative method to build up categories of data. Results Engaging young men in research is very diffi cult, particularly young men who are not in education or employment. Young fathers proved impossible to recruit. The young men who took part in the study thought responsibility for contraception was shared, although this was partly dependent on relationship status, namely whether sex was with a regular partner or a one-night stand. Discussion and conclusions These fi ndings are based on a small sample and all participants were in full-time education. It is likely that attitudes may differ from those who are not in education, training or employment. In a future study, it would be important to ensure that young men from different class and educational backgrounds are included in the research, as the tentative conclusions from this pilot study suggest that educational status is a factor in beliefs about responsibility.This study aimed to gain insights into young men's views of sexual health services and contraception. It was conducted first as a pilot of the methodology, to explore whether this would be a successful means of engaging with young men of diverse backgrounds in terms of education, social class and experience of fatherhood.

INTRODUCTION The rapid development of China over the past 40 years resulted in economic growth and increased globalization . Advancements in areas such as technology and communication provided an abundance of resources for students to mature more rapidly physically and psychosocially, resulting in earlier sexual maturity . Furthermore, globalization increased the influence of Western populations in China, where sexual experiences in unmarried students are high , creating more liberal attitudes toward sex in Chinese students . Vuong and Napier proposes that this is due to the "mindsponge" mechanism, where the minds of individuals may be reshaped after exposure to different cultures, leading to substantial cultural changes nationally . Sexual and reproductive health issues are therefore further exacerbated , resulting in higher rates of unplanned pregnancies, abortions, and sexually transmitted infections . However, given the conservative nature of the Chinese population, SRH education remains deficient and although modest efforts have been made, a large majority of students in China particularly in rural regions, have not had formal SRH education or access to SRH services. Sexual attitudes and behaviors in the Chinese population are also largely affected by the one-child policy, the ruralurban divide, and traditional beliefs. During the one-child policy implemented between 1979 to 2015 in China, most Chinese families were limited to one child each to reduce China's population growth that was straining available resources. To promote the policy, only-child families were granted concessions such as easier school enrollment, better employment opportunities, and better healthcare benefits . However, as a result of the policy and China's patrilineal family systems, son preference governed much of reproductive behavior in China, resulting in a male-biased sex ratio . A significant change, called the 1.5-child policy, was later made in 1984 to the onechild policy; under the 1.5-child policy, couples residing in rural regions were allowed to have a second child if their first was a female . By the end of 2015, the year that the onechild policy was abolished, there was an estimated 225 million only-child children in China . Studies have shown that these children tend to be given more attention and care from parents , have better educational opportunities and higher school achievements than children with siblings , suggesting healthier development. In addition, despite the modernization of China, China continues to utilize a dual economy, resulting in significant gaps in income, infrastructure, and government services between urban and rural areas . Due to the economic disparity between rural and urban areas, more than 168 million rural residents have migrated internally to cities, resulting in more than one-third of all rural children having had been left behind by either or both parents . Left behind children reportedly have more health deficits and are at greater risk of unhealthy development and as such, may have poorer SRH. Furthermore, traditional beliefs such as Confucianism, ingrained into the Chinese culture, instills the concept of male superiority over females. As such, sexual attitudes and behaviors such as pre-marital sexual activities are more tolerated in males than in females . The physical differences between males and females also influence sexual attitudes and behaviors . Whilst Yan et al. and Li et al. have investigated attitudes toward premarital sex and sexual knowledge, attitudes, and practices among only-child students and students with siblings, these studies were not conducted nationally. Given that residence in rural or urban regions and sex of the child may affect only-child status as well as sexual knowledge, attitudes, behaviors, and SRH outcomes, this study aims to investigate the differences in sexual knowledge, attitudes, behaviors, and SRH outcomes between only-child students and students with siblings in China stratified by region and sex. As students are more likely to seek information online, our findings may provide policy makers with evidence-based findings to guide the design of internet-based learning materials, particularly for female students and students from rural areas. --- MATERIALS AND METHODS --- Study Design and Participants Data from the 2019 National College Student Survey on Sexual and Reproductive Health conducted by the China Family Planning Association and collected by the China Youth Network was utilized for this study. The CYN is the largest youth volunteer organization in China that provides SRH education. A total of 241 institutions of higher education including key universities, ordinary universities, and colleges were selected from all 31 provincial-level administrative regions in China, after balancing for the type of university and vocational college. Although there are more recent survey data, the 2019 survey data was utilized as it was the most recent survey round that contains measures on only-child students. The methods of survey instrument development and study design were previously described in detail elsewhere . Utilizing convenience sampling, the internet-based selfadministered questionnaire was distributed to students through contact points from selected universities and vocational colleges. A total of 55,757 responses from 1,764 universities and vocational colleges were collected. Data collected included sociodemographic, attitudes toward sexual behaviors, knowledge of SRH, sexual history, and SRH outcomes. Respondents were included in final analyses if they provided informed consent, answered all questions and passed the consistency checks and logic verification, and were aged between 17 and 24. A total of 49,569 participants were included in our final analysis. This study was reviewed and approved by the Institutional Review Board of Tsinghua University . --- Exposure The main exposure of interest was whether the participant was an only-child or if they had siblings. The question "How many children are there in your home ?" was used. Participants who answered "one" were considered to be an onlychild and were coded as "1, " while those who reported more than one child in family were considered to have siblings and were coded as "0." --- Outcomes The outcomes of interest are sex-related knowledge, attitudes, behaviors, and SRH outcomes. Sex-related knowledge was measured using 9 questions on contraceptive use, acquired immune deficiency syndrome , abortion/pregnancy, and sexually transmitted diseases . Response options were "yes, " "no, " and "do not know." Each positive answer was coded as "1, " while the negative answer or "do not know" were coded as "0." Sex-related knowledge scores ranged from 0 to 9, with increasing scores indicating higher sex-related knowledge. Attitudes toward sexual behaviors were measured using 5 questions on premarital sexual intercourse, having had sexual intercourse during university, one-night stand, and multiple sexual partners, and guilt of being sexually active. Response options were "strongly disagree, " "disagree, " "unsure, " "agree, " or "strongly agree." Responses "agree" and "strongly agree" were classified as positive attitudes, coded as "1, " while "unsure, " "disagree, " and "strongly disagree" were classified as negative attitudes, coded as "0." Sexual seeking behaviors were measured using 3 questions: "Have you ever read erotic magazines or books?", "Have you ever searched for pornographic information online?", and "Have you ever searched for sexual knowledge online?". Response options were "yes, " coded as "1" or "no, " coded as "0." Sexual seeking behaviors scores ranged from 0 to 3, with increasing scores indicating higher sexual seeking behaviors. Sexual behaviors, including risky sexual behaviors, were measured using 5 questions: "Have you ever practiced masturbation?", "Have you ever had sexual intercourse ?", "Have you ever had casual sexual intercourse ?", "Have you ever had multiple sexual partners at the same time?" and "Did you use any contraceptives during your first sexual intercourse?". Response options were "yes, " coded as "1" or "no, " coded as "0." The frequency of contraception use during sexual intercourse was also measured. Response options were "never, " "seldom, " "almost half of the times, " "mostly, " and "each time." Responses were re-categorized into use of condoms in most sexual intercourse or no use of condoms in most sexual intercourse . Adverse sex-related outcomes were measured using 3 questions on unintended pregnancy, induced abortion, and a STI diagnosis. Response options were "never, " "ever, in the past one year" and "ever, more than one year ago." Response "never" was classified as no and coded as "0, " while the other options were classified as "yes" and coded as "1." Adverse sex-related outcomes scores ranged from 0 to 3, with increasing scores indicating higher adverse sex-related outcomes. --- Other Covariates Data on age, sex, ethnicity, hometown region, school type, average monthly expenditure, sexuality education at school, self-rated parent-child relationship, parents' highest educational qualifications, and parent-child discussion relevant to sexual behaviors/contraception, and tobacco and alcohol use were also collected. These factors have been demonstrated by published literature to have impact on SRH outcomes or to contribute to the difference between the only-child students and students with siblings . Age and self-rated parent-child relationship were analyzed as continuous variables; self-rated parent-child relationship scores ranged from 0 to 10 with higher scores indicating better relationships. The remaining variables were coded as categorical variables. Ethnicity was categorized into Han and ethnic minorities. Hometown region, defined as the place of residence before attending university or vocational college, was categorized into "urban" and "rural." The average monthly expenditure was categorized into 0-999 RMB, 1,000-1,999 RMB and ≥2,000 RMB. School type was categorized into "college" or "university." Highest parental educational attainment was categorized into "primary and below, " "middle school, " "high school, " and "college and above." Sexuality education at school, parentchild discussion relevant to sexual behaviors/contraception, and tobacco and alcohol consumption were categorized into "ever" and "never." --- Statistical Analysis Continuous variables were described as mean ± standard deviation , and categorical variables were described as proportion and percentage. Continuous variables were compared using independent sample t-test and categorical variables were compared using the chi-square test. Linear regression was utilized to examine the association between only-child status with sexrelated knowledge. Logistic regression was utilized to examine the association between only-child status with sexual attitudes, sexual behaviors, and adverse SRH outcomes, stratified by sex and hometown region. The models were adjusted for age, sex, ethnicity, school type, average monthly expenditure, self-rated parent-child relationship, sexuality education at school, parents' highest educational qualifications, and tobacco and alcohol use. All statistical analyses were conducted using STATA/SE version 15.1 . --- RESULTS --- Participants' Characteristics The characteristics of only-child students and students with siblings are presented in Table 1. Only-child students were less likely to be females , of ethnic minority , residing in rural areas , have an average monthly expenditure between 0 to 999 yuan , have ever received sexuality education at school , and to report better self-rated father-child relationship . However, only-child students were more likely to be attending university , have parents with educational qualifications college and above , have had parent-child discussions relevant to sexual behaviors and contraception , have smoked and have consumed alcohol . The associations between only-child status with sexrelated knowledge, attitude, and behaviors, stratified by hometown region and sex, are presented in Table 3. Only-child students have on average, higher sex-related knowledge scores regardless of their hometown region or sex , compared to students with siblings. Only-child students were more likely to agree with premarital sexual intercourse, if their hometown region was in an urban area , and regardless of sex ; sexual intercourse during university, if their hometown region was in an urban area , and regardless of sex ; one-night stand, if their hometown region was in an urban area and if they were female ; and multiple sexual partners, if their hometown region was in an urban area , and regardless of sex . However, only-child students were less likely to feel guilty for being sexually active, if their hometown region was in an urban area , and regardless of sex . Only-child students who were females and whose hometown region was in an urban area were more likely to have ever read erotic magazines or books , and have ever search for pornographic information and sexual knowledge online . Only-child students whose hometown region was in a rural area were more likely to have ever had casual sexual intercourse , and those in urban area were less likely to have no condom use during first sexual intercourse and no condom use during most sexual intercourse compared to students with siblings, especially if they were males. Only-child students whose hometown region is in an urban area were at lower odds of having ever had an unintended pregnancy , while female only-child students were more likely to ever have been diagnosed with STI . --- DISCUSSION In this cohort of students, we found several main findings. Compared to students with siblings, only-child students have higher sexual knowledge, more liberal sexual attitudes, were more likely to seek sexual information, and have higher sexual behaviors but fewer adverse SRH outcomes. The results were influenced by sex and hometown region. Our study found that only-child students have higher sexualrelated knowledge scores compared to students with siblings. A possible reason is that because cultural norms in China inhibits discussion of sexual matters, especially for adolescents and unmarried youth, numerous Chinese students receive limited information on sex . Given that parents of only-child students are more likely to be more financially well-off, only-child students may have greater access to technology and the Internet , which provides them with additional sources of access to sexual knowledge . A study suggested that higher sexual knowledge is also associated with more liberal attitudes toward one-night stands and premarital sex . This is a possible explanation as to why only-child students tend to have more liberal sexual attitudes , be more sexually active, and have greater number of sexual partners . We also found that only-child students were more likely to use condoms during sexual intercourse compared to students with siblings. Given that in this cohort, only-child students were more likely to have had parent-child discussion related to contraceptive use compared to those with siblings, they may be more aware of the need for contraceptive use during intercourse. Furthermore, only-child students were found to be more likely to actively search for sex-related information online . A survey conducted in Malaysia reported that respondents who searched for sex-related information online had greater intention to use condoms during intercourse compared to those who did not . A study conducted in Chinese college students also reported a positive relationship between exposure to porn and condom use . Given that a higher proportion of only-child students in this cohort of students, reported having had searched for sexual information and pornographic materials online, this may explain the greater use of condoms during sexual intercourse. Compared to males, the differences between only-child students and students with siblings in sexual knowledge and attitudes were more significant in females. Females tend to have better communication skills and may therefore obtain greater sexual knowledge from friends and family . Previous research has indicated that the sexual attitudes and behaviors of children are associated with their mothers, as mothers were more likely to discuss sex-related topics with daughters . Besides, as a predominantly patriarchal society, and despite social revolutions over the recent decades, son preference still exists in China today. Therefore, male-biased parents with multiple children may allocate more resources to sons than daughters . However, in one-child families, females may be more similarly treated as to males because they are an only-child . Several studies have supported this notion; only-child females reported better educational opportunities compared to females with siblings . Compared to females, the difference between only-child status with condom use during sexual intercourse were more significant in males. This is because in China, due to the concept of male superiority, females hold less power when negotiating safe sex . Studies have suggests that lack of control over sexual behaviors is associated with a lower likelihood of condom use among women aged 15-24 . Furthermore, our findings indicate that only-child students were more likely to have parent-child discussion relevant to sexual behaviors and parentchild discussion relevant to contraception than students with siblings. This suggests that only-child students are more likely to be aware of the importance of condom use during sexual intercourse. Therefore, the difference in condom use during sexual intercourse is more significant between male only-child students and students with siblings. Compared to rural areas, the difference between onlychild students and students with siblings in sexual knowledge, attitudes, behaviors and SRH outcomes were more significant in urban areas. Only-child students living in urban areas may have better access to sexuality education as urban regions may have better educational opportunities and facilities. Furthermore, because the phenomenon of being left-behind is prevalent in rural regions, students from rural regions may lack social and economic support from parents . Studies have also consistently reported rural-urban inequalities in access to education due to the household registration system in China , which excludes rural residents from access to state-allocated resources . Therefore, students from rural areas tend to have fewer education opportunities, regardless of their only-child status. Given the low quality of sexuality education in rural regions, only-child may not be able to obtain good sexuality education, much less those with siblings. Urban regions were also found to have better sexuality education programs compared to sexuality education programs in rural areas . Therefore, the difference in sexual knowledge, attitudes, behaviors and SRH outcomes is smaller between onlychild students and students with siblings in rural areas. Overall, our study has several strengths. First, our study has a large sample size and given that our data was collected across various areas in China, it is nationally representative and allows for better generalizability of results. Second, our study reports on the difference in sex-related knowledge, attitudes, behaviors and SRH outcomes between only-child students and students with siblings stratified for sex and region in China, with the control of important covariates such as sexuality education from both school and family. Lastly, compared to existing literature, measures on only child status with seeking behaviors for sexrelated information were also collected and analyzed, providing greater comprehensiveness. Our study also has several limitations. First, causal inference cannot be drawn as this was a cross-sectional study. Second, there might be response bias as sex-related topics remains a taboo topic in China. However, we utilized an internet-based approach to minimize response bias. Third, as the study utilizes self-reporting, there may be recall bias although logic correction was employed to reduce the number of incorrect answers. Fourth, we were unable to control for covariates such as sibling sex and age composition, which were not collected in this survey. Subsequent studies should investigate the relationship between sibling sex, number of siblings, and age gaps between siblings with sexual and reproductive health in students. It would be beneficial also, to study the effects of only-child status and SRH in later years, such as during adulthood given that the prevalence of sexual behaviors and therefore, adverse reproductive health outcomes, are relatively low in college students. --- CONCLUSION Only-child students demonstrated higher sexual knowledge, more liberal sexual attitudes, and fewer risky sexual behaviors. Female students and students who resided in rural areas were more likely to demonstrate seeking behaviors for sex-related information both offline and online. Comprehensive sexual education for students should aim to better include females and students from rural areas by ensuring that sexuality education is accessible and available both offline and online. Public healthcare may also provide sexual health clinics offering subsidized consultations and contraceptives. --- DATA AVAILABILITY STATEMENT The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. --- --- SUPPLEMENTARY MATERIAL The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpubh. --- 2022.925626/full#supplementary-material --- Conflict of Interest: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Publisher's Note: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Copyright © 2022 Zhao, Liang, Hee, Qi and Tang. This is an open-access article distributed under the terms of the Creative Commons Attribution License . The use, distribution or reproduction in other forums is permitted, provided the original author and the copyright owner are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

Objective: The differences in sexual knowledge, attitudes, behaviors, seeking behaviors for sex-related knowledge, and sexual and reproductive health (SRH) outcomes among only-child students and students with siblings in China, was examined for sex-and region-specific effects. Research Design and Methods: Data on 49,569 students from the 2019 National College Student Survey on Sexual and Reproductive Health, conducted across 31 provinces in mainland China was utilized. Multivariable regression and stratified analyses were employed to analyze the differences in sexual and reproductive health between only-child students and students with siblings. Results: Only-child students reported higher sexual knowledge, more liberal sexual attitudes, and fewer adverse SRH outcomes compared to those with siblings. Results were found to be influenced by sex and hometown region after controlling for socioeconomic factors, parent-child relationship, and sexuality education. Conclusions: Female students with siblings who resided in rural regions were more likely to have poorer SRH compared to male only-child students who resided in urban regions. Comprehensive sexual education for students should aim to better include females and students from rural areas both offline and online, and public healthcare should offer subsidized consultations and contraceptives.

Introduction The latest manifestation of "all connected world" is IoT [1]. Following "Internet of-" idiom, Internet of Vehicles ) is ready to be visible soon, which in addition to having strong connection with IoT builds on already deployed technologies, such as Vehicular Ad hoc NETworks [2] and Vehicular Social Networks [3]. However, the conception of VIoT is well beyond the state-of-the-art and ready to leap into a chasm of possibilities, recently termed as Social Internet of Vehicles [4], whereby vehicles build and manage their own social network. Just like social networking opening up a plethora of new recommendation applications, SIoV has an unlimited potential of changing what we do and how we live. It can be used to recommend location-and profile-based services, including finding trustworthy services [5,6], vehicles navigation and monitoring [7][8][9], managing network access [10], targeted advertising [11,12], drivers behavior modeling [13], and more. Although, these examples are mostly related to the vehicular networking, but we can easily imagine a broader perspective wherever recommendation services in a dynamic and evolving network of objects is involved, such as social networking, IoT, Wireless Sensor Network , microrobotics, etc. Hence, the findings of this paper can be applied to any other related domain equally well. All technologies designed to be subservient to humanity always have an intrinsic attachment with its users, drivers and passengers in case of SIoV. Therefore, when a vehicle is part of SIoV, it must build and deploy its network in order to accomplish users' goals. But then the resolution of the dichotomy of vehicular and human social network turns out to be a real challenge. The model proposed in this paper builds on belief that the evolution of intervehicle social network can be realized by examining the characteristics of human social network. In this paper, we investigate interfacing of these two kinds of networking and the extent of users' satisfaction with respect to their goal achievement, while making use of vehicular networking. In a nutshell, the focus of the paper is on investigating the potential of vehicular social networking towards achieving human goals. We perform this research by designing an agent-based model and simulating it in various conditions. The study revolves around researching about competence of SIoV for providing accurate recommendations to the users in a dynamically changing environment. The changes occur in resources' quality and network dynamics. The concept behind the model makes use of basic assumptions about IoT connectivity, principles of social network evolution, and users profiling. Some interesting what-if questions are asked against a couple of intuitive hypotheses. The remainder of the paper is structured as follows. In Section 2, we present the related work in detail, followed by scenario section . The proposed model is explained in Section 4. Simulation settings and the analysis of the simulation results are presented in Section 5 followed by Section 6, which provides discussion about hypotheses and conclusion of the paper. --- Related Work With the increase in vehicles coupled with IoT [14], the traditional VANETs are transforming into the IoV. IoV is an advanced shape of modern vehicles equipped with sensors thereby constituting a communication system which can receive and transmit information useful for navigation and traffic management [15]. Another technology is VSN where passengers can exchange data related to entertainment, social networks, and situations [3]. Further the evolutionary pathway is the SIoV, which originated [16] as an application, engaging human social behavior into physical vehicular networks. Later, it attained a more general form in which it is considered to be a network of vehicles in which vehicles build and manage their own social network [17]. The first social networking framework for the commuters called RoadSpeak was presented by the Smaldone et al. [16] and named as a VSN. The people traveling along the same road can dynamically form interest groups or social communities based on spatial and temporal preferences/interests to share messages using a chat application. SocialDrive [18] is a crowdsourcing-based VSN for the socialization of the drivers to be aware of their driving status/behavior, which may lead to improve their safety and reduce fuel consumption. This framework leverages traditional social networks and cloud computing. Other VSNs based on traditional social networks includes [19], where drivers may share their driving experiences with other drivers in social groups. These experiences of drivers are further aggregated based on temporal and spatial context. In Roadcast [20], vehicles share content based on the query. A vehicle can query other vehicles in a network. The system returns the highly matched content based on keywords; therefore, the result may not always be relevant or useful, but it may be useful in the future for the other vehicles. However, in [21] vehicles spread the queries in a VSN, which pushes the potentially useful data in a bounded time towards the vehicles. Yasr et al. [22] studied information sharing in the VSNs through social relations that ensure relevant and reliable information has to be shared. Furthermore, information can be reliably shared in social groups in terms of trust [23]. VSNs have been modeled and evaluated based on agentbased methodology. For example, a semantic-based multiagent platform was presented in [24], which integrates semantic techniques and software agents to provide a framework for VSN applications. This work was extended by the authors in [25] which provides a service-oriented VSN platform that enables commuters to improve transportation efficiency by collaboration, including context-aware mechanism to predict traffic congestion. An agent-based architecture for a swarm of socially aware IoV called social vehicle swarms is proposed in [26], which leverages the agent-based modeling, big data, and cloud computing to analyze and reveal hidden patterns in social relationships. The Social Internet of Things [27] enables IoT device to establish a social-like relationship to create social groups/communities based on common goals and interests. Most of the research work described above does not comply with SIoT philosophy in which objects are main social entities. However, SIoV is an application of SIoT in vehicular networks, where vehicles are social objects [28]. A middleware for SIoV was proposed by [29], which extends the intelligent transportation system station for SIoT. Moreover, [15] presents a detailed architecture for the SIoV, which outlines the main components, interrelations objects, and defines interactions. TNote [30] is a social network of vehicles, which defines the key aspects of the social structure of vehicles, their interactions, and relationships. This framework helps the vehicles to share efficiency and safety-related notes in communities. In [31], authors have proposed an architecture that incorporates moral and ethical rules in SIoV. Traditional Social Network has been linked to SIoV using big data analytics by the [32]; furthermore, it addresses the privacy and security issues including accessibility attacks, privacy attacks, data genuineness attacks, and routing attacks. In [33], authors have proposed a social big-data-based content dissemination approach, which combines information of both social layer and a physical layer for optimized information sharing. S-Aframe [34] is an agent-based multilayer framework with context-aware semantic service that utilizes the context information and provides a high-level platform for the development of applications for SIoV of drivers, passengers, and pedestrians. Butt et al. [35] proposed an architecture, which provides a foundation for the development of SIoV applications using Restful web technologies. Currently, SIoV is in its evolution stage; thus, enabling technologies and protocols, e.g., sensors, On-Board Unit , machine learning [36,37], image processing [38], communication technologies, etc., require extensive research to develop a comprehensive architecture for it. SIoV has brought many Complexity 3 challenges, such as to establish right social connectivity to right vehicle because the vast amount of data is being generated by these social networks and moreover issue it to provide right service the users and to disperse contextaware relevant information [34]. To this end, it should enable people to collaborate seamlessly and read social content, e.g., recommendations, that enable travelers to experience the locale in a social context. A survey of the recent advancement, architectures, and challenges in a social network of vehicles is presented in [39]. However, all the presented papers have not addressed the incentive mechanism in SIoV to ensure cooperative involvement of users. It is linked to the selfishness of the users [40]; therefore, we argue that users cooperate for common interests and useful recommendations; furthermore, it works as positive feedback in a whole system in sense of user's goal achievement and improvements in productivity that stabilize the SIoV. In this paper we have studied information dissemination in terms of recommendation system which is entangled to the user's goals and quality of experience that leads to successful SIoV strategies. This work is the continuation of our previously published work [41,42]. --- Scenario and Research Hypothesis Most of our time during the weekdays is consumed in traveling between home, work, school, and shopping. Drivers use ample applications such as navigators [43] to reduce the travel time or to switch between various possible choices. Recently, Location Based Social Networks [44] such as Google maps and Facebook Places have gained significant attention, where people can share their physical locations, experiences, ratings, etc. In the sequel, in recent years recommendation systems have been a subject of interest for the researchers, where big data is appropriating towards provisioning of context-aware recommendations. Realization of a recommendation system in collaboration with the capabilities of SIoV is an interesting topic. It is possible to accomplish this due to the recent industrial efforts. It is predicted that, by 2020, 50 billion objects will be linked to the Internet, and the considerable number of objects will be vehicles [45]. In the Carestream project [46], technological integration is performed to provide data-driven services. The system collects a variety of information such as "vehicle status, driver activity, and passenger-trip information" from a network of approximately 30,000 chauffeured driven cars. The parameters such as pickup point, pickup time, arrive time, and destination are collectively used to initiate the user demand. These data fields motivated us to use the concepts of 𝐼𝐷, 𝑇𝑖𝑚𝑒, and 𝐷𝑢𝑟𝑎𝑡𝑖𝑜𝑛, to characterize a resource. Then, a vehicle plan is created making use of user profiling , in which a user has to visit some resources every week. A vehicle serves its user. The social networking of users is not considered in this model; only a network of vehicles is established. Therefore, a vehicle and its user are dealt with as a single entity or an agent. The plan of an agent consists of a number of activities. An activity is constituted by a source, a destination, a time, and duration. Agent's home is the source of an activity. A resource is the destination of an activity represented by identity of a resource, time is the time to reach to the resource, and duration is the time to stay at a resource. The initial plan is generated randomly choosing a number of resources and assigning an hour to visit a resource and duration of a few hours. The objective of the model is to investigate the competence of SIoV in providing resource recommendation to the agents, particularly when there is no obvious/natural choice. This can happen if resources known to an agent are no more an attractive choice due to the poor quality of service or the agent individually having a much higher expectation. It is assumed that the interface between a vehicle and its owner is seamless so that a vehicle does not only have information about its owner's plan but also if the current experience regarding a resource was good or bad. Figure 1 shows the overall scenario in a simple way. Each agent has a plan to visit a certain resource ; for example, agent 20 has a planned activity to visit resource 501 at 𝑡𝑖𝑚𝑒 = 38 and the duration of the visit will be 4 hours. Therefore, the return time would be 𝑡𝑖𝑚𝑒 = 42. During this visit, the agent experiences a quality represented by * * * , which is a good quality . The agent 20 will visit the same resource again in week 2 as per plan, quality, and user expectation. But, in case of quality not as per agent's expectation, the agent's plan will be updated for the next week based on the information exchange between the agents present at the same resource. The model proposed in this paper is used to explore some hypotheses. These hypotheses are based on following generalizations of the scenario presented above. Consider that the users of a social informational system repeatedly execute a plan-initially stochastic in nature -and that a planned activity is associated with a resource-which gradually increases or decreases its quality of service. If the information shared-as a consequence of information sharing by vehicles when parked at a resource at the same time-augments the plan of a user, can we say that more interactions lead to more awareness . Specifically, the following hypothesis are formulated: Hypothesis 1. Given that a long-term and consistent connectivity transforms into interactions , the prospect of the spread of useful information in the network increases with evolution of the network, occurring solely due to internal dynamics of a regular network . Hypothesis 2. More information sharing is capable of ending into a fair distribution of resources across the users, resisting against a skewed distribution. Hypothesis 3. Since the information sharing is expected to improve the user experience due to the availability of more useful information , we hypothesize that the early the information sharing starts, the better would be the state of the system in terms of users' experience in a longer run. --- Model Each vehicle helps its user execute its weekly plan, even when a user is short of resources due to bad experiences. From this point onwards, a vehicle would be treated synonym to a user, and both termed as an agent. The model is based on discrete-time agents' rules. The purpose of the agents is to visit resources given in the plan at their prescribed time. The purpose of proposed socially driven recommendation system is to assist agents, when there is no choice for an agent due to mismatch in expectation and experience. Further, the model provides a means to evaluate the extent of success of the recommendation system characterized by different strategies of information sharing. Towards this, important hypotheses are tested under reasonable connectivity and data constraints, culminating into important "what-if" questions. --- Overall Model Flow. The plan matrix contains three components , which are randomly set at the start of the simulation. Each activity has an associated experience value. Each agent has an expectation, which is a static personal trait. If the quality of service provided by a resource at the time of a visit is not up to the agent's expectation , the resource is considered practically excluded/suspended for that particular agent. The recommendations are enabled through an evolving network. The information flows through a physically and temporally constraint network of vehicles. Vehicles connect with each other if they are at the same resource at the same time, thus building a social network of vehicles evolving with repeated encounters. The system transits between four states. If 𝑠𝑡𝑎𝑡𝑒 = 0, the agent is located at source . An agent will transit to the next state if it is 𝑡𝑖𝑚𝑒 to execute an activity of the plan. However, it first prepares itself based on the strategy being used. The procedure prepare to move to resource would be explained when we talk about four strategies, each practically fine-tuning an agent to move to the next state based on strategy description. If 𝑠𝑡𝑎𝑡𝑒 = 1, the agent moves to the desired resource along with transiting to the next state . This move is a one-step jump without any trajectory following. Additionally, in this procedure, the current 𝑞𝑢𝑎𝑙𝑖𝑡𝑦 offered by the resource with a random variation of ±25% is stored in the plan matrix as experience of the agent against that resource. If 𝑠𝑡𝑎𝑡𝑒 = 2 and the 𝑑𝑢𝑟𝑎𝑡𝑖𝑜𝑛 to stay at the current resource expires, the agent transits to the next state . The procedure prepare to move to source checks for completion of this duration. If 𝑠𝑡𝑎𝑡𝑒 = 3, the agent communicates with its neighbors . This is done through interact procedure. After interacting, the agent moves back to home and resets its state to 0 to start the cycle again. This is achieved through move to source procedure. Again, the move is a one-step jump without any trajectory following. In the following, the interaction process is given in more detail. interact. The agents colocated at the same resource network with each other to form ties. Initially, it is a weak tie. With repeated encounters, a weak tie becomes a strong tie. Each agent has a data structure, which contains this information. This is a tuple constituted as {time of the latest encounter, how many encounters? strong tie?}. If a number of encounters with the same agent reach a threshold, a weak tie changes into a strong tie. Figure 2 presents a graphical view of agents' state transitions. In the following, a description of each of the strategies used in the system is given. --- Plan Execution Strategies. The purpose of the agents is to perform activities given in the plan at their prescribed time. The four strategies adopted are strategy 1 , strategy 2 , strategy 3 , and strategy 4 . --- Strategy 1 . s1 is static, just executing the plan as it is. The algorithm is given in Algorithm 1. The process is invoked at each iteration and when an agent is in state 0. 𝑖 is the loop counter for plan table, iterating through 3 activities of the plan. If 𝑡 is equal to the current time then the current activity is selected and the agent moves into state 1. At each iteration, only one activity can possibly be selected due to different execution times of the 3 activities. --- Strategy 2 . The strategy s2 blacklists a subquality resource and the agent never visits it again, hence performing exclusion. The algorithm is given in Algorithm 2. 𝑡 is the time to execute an activity, 𝑞 is the previous experience of the agent about that resource, and 𝑠 is indicating if the resource attached with the activity is already excluded or not. If 𝑡 is equal to the current time and 𝑠 is 1: 𝑖 ←󳨀 0 2: while do 3: 𝑡 ←󳨀 plan[i, 1] 4: if then 5: 𝑠𝑡𝑎𝑡𝑒 ←󳨀 1 6: 𝑠𝑒𝑙𝑒𝑐𝑡 𝐼𝐷 𝑜𝑓 𝑡ℎ𝑖𝑠 𝑟𝑜𝑤 𝑎𝑠 𝑑𝑒𝑠𝑡𝑖𝑛𝑎𝑡𝑖𝑜𝑛 7: end if 8: 𝑖 ←󳨀 i + + 9: end while Algorithm 1: Prepare to move to resource algorithm. false, then the current activity is selected. When an activity is selected, if 𝑞 is greater than agent's expectation, then the agent moves into state 1, ready to perform the activity; otherwise, the resource attached with the activity is excluded. An activity previously excluded would never be selected in s2. that were not excluded. The information about those resources would come from "friends" of an agent. The whole excluded row in plan matrix of the agent would be replaced by a not-excluded row attained from a friend. After replacement, Algorithm 2 would be executed to select a resource. --- Strategy 3 (s3 Related to this strategy is replace procedure given in Algorithm 3. 𝑡 is the time to execute an activity, and 𝑠 is indicating if the resource attached to the activity is already excluded or not. Out of three activities, this strategy will choose the first one which is excluded . The chosen excluded activity will be replaced as follows. A possible friend 𝑓 with a strong tie that is appearing first in the friends' list is selected. 𝑓's plan is copied to 𝑓𝑝𝑙𝑎𝑛. If there is an activity which is not excluded in 𝑓𝑝𝑙𝑎𝑛, the friend's activity would replace current excluded activity of the vehicle. --- Strategy 4 . It is worth noting that the replacement of the excluded resources can be performed without considering the internal social networking dynamics, as in s3. Alternatively, it may also consider internal dynamical feature-triadic closure of strong ties-of an evolving network. Formally, the triadic closure property states that if a node 𝑎 in a social network has strong ties with two nodes 𝑏 and c, then, there is a likelihood that 𝑏 and 𝑐 would become friends sometime in the future. Algorithm 4 provides pseudocode performing triadic closure just before the execution of Algorithm 3. The closure is performed as follows. If there are more than one common friends of an agent, two of them are randomly chosen and assigned to 𝑓V and 𝑠V. If 𝑓V has a strong tie with the agent, and either 𝑓V is not a friend of 𝑠V or 𝑠V is not a friend of 𝑓V, then they are made a friend of each other with a strong tie. --- Simulation and Results The model is created in NetLogo [48], an agent-based modeling environment. A grid of cells of size 75 × 75 is used to populate agents. There are three types of agents, two static and one mobile. The static resource and source agents are randomly placed in the environment. The count of resources is 15, available for 500 sources. Vehicles are mobile agents which are 10% more than sources; a vehicle attached randomly to a source. The users are abstract entities encapsulated within vehicles, termed as agents from this point onwards. One simulation iteration is equal to one hour. Initially, a random weekly plan, constituted by three scheduled activities, is generated , which executes on weekly basis. Figure 3 depicts visual of simulation space when initialized. The four strategies adopted are evaluated based on three parameters: Quality index: when an agent visits a resource, it experiences the quality offered by that resource at that time. In one week, an agent has to visit three resources that is why the quality index is the average of three experiences of the agent at each hour. The quality index of the whole simulation space is then an aggregation of all these experiences divided by the agents count. Connectivity index: when two agents are at some resource at the same time, they connect with each other, irrespective of the type of tie they may have . The connectivity index of the whole simulation space is an aggregation of all current connections divided by the agents count. Resource utilization: the population of agents at a resource at an hour determines its current utilization. Derived from it is the standard deviation of utilization of all resources. Figure 4 presents the Mean of the quality of resources experienced by all the agents for a timespan equal to 800 hours and averaged across 100 simulation runs. A comparison between the four strategies reveals the following: s1: it represents an extreme case and it is basically taken for the comparison purposes only. This case is not real because constantly visiting the same resources-which may be not up to the agents expectations-is less likely to happen. However, due to randomness in the progression of the quality parameter, a resource which does not perform as per the expectations of the agents may become qualitywise acceptable in near or far future. This is exactly what is revealed in the simulation results; the quality index oscillates around the middle and has fewer variations as shown in Figure 4. s2: once a resource is excluded, no agent will visit it anymore, hence resulting in the least average quality index. s3: the replace strategy replaces an excluded resource that an agent wants to visit based on a friend's recommendation. Since a friend is a strong tie, the value of threshold is important. A higher value of threshold , such as 5 in this case, would delay the replacement process. That is the reason the quality index, in spite of being considerably better than s2, still is less than s1. However, the variations towards positive quarterlies are higher than the negative quarterlies, which is evidence of improvement. s4: the triadic closure provides novel information about unknown resources. Due to this, it was expected that the quality index would be a bit higher than s3. But it turned out to be on the lower side. It was a surprising result because it does not correspond to the connectivity index and resource utilization data presented next. The only possible reason can be that friends engaged in triadic closure share the same resources and incidentally, the novel information turns out to be of lesser quality. However, we need to explore it further. The graph shown in Figure 5 verifies the above findings related to the connectivity index with the average degree of agents. As expected, the connectivity of strategy replace with closure is by far higher than the other three strategies. Another exciting aspect is the study of the utilization of resources. Although there is no safeguard against it within the model itself, it is undesirable to have high asymmetry in resources' utilization, which is analyzed through the standard deviation of resource utilization . Figure 6 presents the comparison of the four strategies. It is evident that the SDU of s2 is the least, whereas the highest is in the case of s4. No doubt, high SDU is due to extreme asymmetry in resources utilization. Sensitivity Analysis. In the following, we provide sensitivity analysis of four strategies when executed for three thresholds, namely, 𝑡ℎ 𝑠𝑡𝑟𝑜𝑛𝑔-𝑡𝑖𝑒 = 2, 𝑡ℎ 𝑠𝑡𝑟𝑜𝑛𝑔-𝑡𝑖𝑒 = 4, and 𝑡ℎ 𝑠𝑡𝑟𝑜𝑛𝑔-𝑡𝑖𝑒 = 8. Figures 7,8, and 9 provide this information. An interesting aspect of these graphs is that s3 and s4 show some phase shifts in quality index. These phase shifts are due to variations in resources utilization; the more the variation is the more are chances of these tipping points. It is also evident that these shows that all of a sudden the resources are reduced, which turns out to be a tipping point. The things get smoother after second tipping point . Interactions between users are also evident in all three simulation views. tipping points occur more prominently when 𝑡ℎ 𝑠𝑡𝑟𝑜𝑛𝑔-𝑡𝑖𝑒 = 4 as compared to 𝑡ℎ 𝑠𝑡𝑟𝑜𝑛𝑔-𝑡𝑖𝑒 = 2. Interestingly these vanish in case of 𝑡ℎ 𝑠𝑡𝑟𝑜𝑛𝑔-𝑡𝑖𝑒 = 8, and the curve acquires a half parabolic shape smoothing into a constant attractor. It is also evident that the tipping points correspond to resource utilization. Also, it seems that tipping points have no relationship with the connectivity index. Overall the simulation results revealed that the best value for phase shifts is 𝑡ℎ 𝑠𝑡𝑟𝑜𝑛𝑔-𝑡𝑖𝑒 = 6. The tipping points in this case are strong and more frequent ). We can still produce tipping points for higher threshold value, but for that we need to either increase the population or decrease resources, or both. For example, we may get delayed tipping points in case of 𝑡ℎ 𝑠𝑡𝑟𝑜𝑛𝑔-𝑡𝑖𝑒 = 8 ) and even 𝑡ℎ 𝑠𝑡𝑟𝑜𝑛𝑔-𝑡𝑖𝑒 = 10 ) but with number of resources equal to 10 not 15. Lastly, the relationship between the quality index and SDU is explained with the help of Figure 10. --- Discussion and Conclusion 6.1. Discussion on Results. In a dynamically changing quality of service provisioning by resources, the vehicles with excluded resources in their plan can utilize their strong ties to have novel information about some resources those are not excluded at a particular time. Is this information turns out to be useful? We analyzed this question with the aid of agent-based simulation. Hypothesis 2 is validated straight away as with information sharing the resources are distributed more fairly. With the increase in thresholds, the effect of information sharing reduces. This is evidence of delaying the information reduces the benefits of information sharing, hence, validating hypothesis 3. However, there are two aspects that provide a little variation. The first one is that, with very quick start of information sharing , we have some improvements, but the real improvements happens when 𝑡ℎ 𝑠𝑡𝑟𝑜𝑛𝑔-𝑡𝑖𝑒 = 6. Hence, it can be concluded that with moderate delay, the best results are achieved. Increase in connectivity is responsible for increasing the quality index as depicted in Figures 3 and4. Hypothesis 1 is valid as long as we perform interaction and information exchange. But, triadic closure decreases the quality index of the system, in spite of improving the connectivity and resources distribution. --- Conclusion. In this paper, an agent-based model of a recommendation system powered by expected capabilities of the futuristic SIoV is proposed. A number of simulations were performed by selecting appropriate parameters to test three hypotheses. Hypothesis 1-internal dynamics of an evolving friendship network helps in increasing the appropriateness of the recommendations-is partially verified. Triadic closure improves connectivity and resources distribution of the system; however, it does not help in increasing the quality index of the system when compared with networking without triadic closure. Hypothesis 2-information sharing results in fair distribution of resources across all the agents-is convincingly verified. Hypothesis 3-the early the information sharing starts, the more appropriate the recommendation would be-is also mostly verified; however, the best results are achieved at moderate level of delays. These results reveal that closure of social ties and its timing impacts dispersion of novel information substantially. These insights provide useful guidelines for designing a SIoV recommendation system and would help practitioners to rationalize their expectations. --- Data Availability The data used to support the findings of this study are available from the corresponding author upon request. --- Abbreviations IoT: Internet of Things. --- Conflicts of Interest The authors declare that there are no conflicts of interest regarding the publication of this paper.

Internet of Vehicles (IoV) is turning out to be one of the first impressive examples of Internet of Things (IoT). In IoV, the factors of connectivity and interaction/information dispersion are equally important as sensing/actuating, context-awareness, services provisioning, etc. However, most of the researches related to connectivity and interaction are constrained to physics of signaling and data science (semantics/contents), respectively. Very rapidly, the meanings of these factors are changing due to evolution of technologies from physical to social domain. For example, Social IoV (SIoV) is a term used to represent when vehicles build and manage their own social network. Hence, in addition to physical aspects, the social aspects of connectivity and information dispersion towards these systems of future should also be researched, a domain so far ignored in this particular context. In this paper, an agent-based model of information sharing (for context-based recommendations) of a hypothetical population of smart vehicles is presented. Some important hypotheses are tested under reasonable connectivity and data constraints. The simulation results reveal that closure of social ties and its timing impacts the dispersion of novel information (necessary for a recommender system) substantially. It is also observed that as the network evolves due to incremental interactions, the recommendations guaranteeing a fair distribution of vehicles across equally good competitors is not possible.

Introduction Previous theories and research suggest that children's postdisaster stress reactions are determined by multiple and complex processes. Most conceptual models include preexisting conditions, characteristics of the stressor, and the child's post-disaster environment . Out of these factors the role of the stressor has been the most highly examined. These studies suggest that the degree of actual threat in terms of children's proximity to the disaster, physical injury, and witnessed experiences is proportional to their risk of developing Posttraumatic Stress Disorder . In addition, several studies have found children's immediate subjective responses to the event to be predictive of later reactions . The study of pre-trauma conditions has been primarily focused on characteristics of the child such as age and gender, rendering inconclusive results . Also, previous trauma has been found to affect outcome . Although several researchers have emphasized the role that the post-disaster environment may play in the development of post-trauma symptoms, this subject has been far less studied . This article will examine one aspect of children's posttrauma recovery environment, namely parents' efforts to aid their children to cope with severe trauma. The child trauma field has had a main focus on identifying markers of risk. Markers of risk typically include preexisting conditions, demographic characteristics, etc. Although these aspects are significant, it is important to distinguish between passive risk markers and active operating processes that can contribute to the maintenance of posttrauma responses. Passive risk markers include little intrinsic information concerning what processes contribute to alleviating or aggravating the development of posttraumatic stress . The present study's focus on naturally occurring processes of parenting serves to bridge these bases of knowledge. One aspect of the child's post-trauma environment that has been examined is the relationship between parents' post-trauma symptoms and those of the child. Research has consistently found a strong positive association between parental and child PTSD . While parents' stress reactions may increase the risk of distress in their children, a supportive family environment, on the other hand, may contribute to a better adjustment in children. The buffering effect that parental support and positive family functioning have on children's reactions to trauma has also received empirical support . Thus it is suggested in the literature that one of the mediating pathways by which disasters can harm children is via their effects on parents and the quality of parenting . There may be many ways in which parenting practices can be affected after a disaster. As mentioned, parents' own exposure and reactions to trauma may affect their parenting behaviors, and subsequently may impact the quality of care and support they provide . However, parents may vary in their abilities to provide children with sensitive and supportive parenting, whether they themselves have been directly exposed to trauma or not. Children's reactions after traumatic incidents may differ from what parents are accustomed to, and this change may lead to uncertainty regarding how their children can be helped. Cohen noted that children's unfamiliar reactions, as well as parents' fears of causing harm to the children by inappropriately reacting to their behavior, may influence the parents' capacities to provide the appropriate care. Parents can assist their children in coping with their experiences after a disaster in numerous ways. They may facilitate their children's adjustment by providing them with suggestions for how to cope with what happened , and by listening to their fears and concerns . Recently, a set of evidence-informed recommendations have been put forward, suggesting how parents should care for their children after surviving high-impact disasters . These recommendations include a number of suggested parenting strategies, for example, helping children feel safe, helping them talk about the distressing experience, soothing children by serving as role models, avoid further exposure, and try to maintain a family life as normal as possible . The amount and type of coping advice parents provide for their children may depend on the severity of their children's symptoms , which suggests that parents may help their children by being sensitive to their specific needs following their exposure to trauma. However, one study conducted after the 2001 terrorist attack on New York City showed that the coping assistance mothers provided was more closely connected to the mothers' own traumatic experiences rather than to what their children had experienced in the attacks . Also, parents' views on what constitutes good parenting practices may change after exposure to a traumatic event. Another study of parents living close to ground zero in New York following the 2001 terrorist attacks demonstrated that parents had changed perspectives as to what they perceived as important in their roles as parents. They became more focused on bonding with their children, as well as loving, protecting, and providing for them . Despite an abundance of literature claiming that parental responsiveness is important in post-trauma coping in children few studies have actually focused on parenting practices in the aftermath of trauma. Given the hypothesized role of these relationships in post trauma functioning, and an increasing body of research on the impact of traumatic events on children, the lack of studies is surprising. Hence, the focus of the present study is to fill in gaps in the literature by addressing the nature of post-trauma parenting: How do parents understand the needs of their children and what do they do to help their children cope in the aftermath of trauma? The answers to these questions are important. First of all, insight into these processes may enhance our understanding of how to assist parents in helping to facilitate their children's recovery after exposure to disasters. Secondly, the answers can help us to further develop models for early intervention. We cannot prevent disasters from happening, but understanding more of what we can do to prevent the development of severe post-trauma reactions is of great importance. Increasing our knowledge of children's posttrauma functioning through analyzing children's naturally existing coping resources is a perspective that has been long-awaited to be studied . --- Method --- --- Procedure The study was approved by the National Committee for Research Ethics in the Social Sciences and in the Humanities in Norway. The parents were asked to sign a consent form prior to participation, and informed that they could withdraw from the study at any time. Face-to-face interviews with the parents were conducted approximately 10 months after the tsunami, in the participants' homes, by experienced psychologists and psychiatrists, who had received training in the use of the interview protocol. The training entailed a particular focus on techniques for facilitating the telling of trauma narratives without leading or interfering in the story. In addition critical aspects related to interviewing potentially traumatized individuals were emphasized during the training. Audio-taped interviews were transcribed verbatim, including minimal phrases, pauses and emotional expressions. --- Interviews The interviews were semi-structured. To capture the specific experiences of the families, the parents were asked to provide a trauma narrative describing their experiences during the tsunami. All participants were presented with the following introduction: ''I know that you and your family were in Thailand at Christmas. While you were there something happened. Please tell me about that.'' Emphasis was put on having the participants narrate as freely as possible. However, a number of prompts were also provided in order to help the participants elaborate on events that seemed significant in the narrative. Subsequently, the following open-ended questions were asked: ''How would you describe your child's reactions after the disaster?'', ''What did you think your child needed during the time following the disaster?'', and ''How did you adapt to your child's needs?''. --- Analyses The analysis was inspired by the Consensual Qualitative Research framework . This method emphasizes cooperation among researchers in order to strengthen the credibility of the analyses, ensure multiple perspectives, and reduce subjective bias. First all interviews were read and reread by the researchers to establish domains, which are topics used to cluster or group the data. Two domains were established: the parental process of interpretation and parents' support strategies. The parental process of interpretation refers to how the parents go about identifying and interpreting signs of distress in their children. Parenting support strategies refers to what the parents do to aid their child in the recovery process. The interviews were then reread and blocks of data were assigned to the domains. In the next step of the analysis core ideas were established within each domain and each individual case. Through this process we sought to capture the main essence of what each parent had expressed within the theme of each domain. The core ideas reflected the parent's perspective and meaning with minimal interpretation. In the third and last step in the analysis we created categories across cases. The categories were based on the core ideas through cross analysis, where the core ideas that could be grouped together were transformed into broader categories. This step brought the analysis to a higher abstraction level, with a search for similarities and differences across cases. These are the presented results. If any coding diverged throughout this process, the codes were discussed with reference to the text excerpts until a consensus could be reached. Finally, the first author read all interviews again to make sure the original information was actually represented in the final categories created. --- Results The parents in this study provided long and rich descriptions in response to the question about how they perceived their children's needs and how they proceeded to provide support. Two main themes emerged from the analyses which described their efforts to observe and interpret possible signs of discomfort in their children. These were a heightened awareness towards their children's reactions, and, their efforts to interpret children's behavioral changes. The second part of the analyses, where we examined parental strategies to provide support, revealed two main categories: preventing symptoms and reducing symptoms, which again were comprised of three subcategories, namely reestablishing safety, resuming normal routines, and coping assistance. The findings are presented in further detail below and illustrated with quotes from the interviews. --- Parental Process of Interpretation Heightened Awareness: Looking for Signs A general tendency in this sample was, with very few exceptions, that parents told about a heightened awareness that their children could display negative reactions due to their experiences. That is, the vast majority of the parents voluntarily reported an increased tendency to follow and observe their children, looking for signs indicating that they were upset. The mother of a 13-year-old boy who nearly drowned in the tsunami said: ''I was extremely aware that he could react in some way. I kept a close eye on him, and asked him every now and then whether he was feeling ok. '' In trying to manage the balance between not inducing distressing emotions on the one hand, and not doing enough to support their children on the other, these parents monitored their children closely and waited to see what would happen. One father said about his 11-year-old son: I didn't want to nag him the first few weeks.… I just tried to observe him, make sure he wasn't just sitting there being depressed … and I made sure he was still going out with his friends and that kind of thing. I guess I was just observing him for a while, maybe for a month or so after returning home. Yet another father focused on following his 15-year-old daughter's own pace of adjustment: ''We let her handle it in her own way … so we kept an eye on her just to make sure she was coping alright.'' In this way the parents observed their children and monitored the progression of reactions or symptoms. Their hesitation to intervene should not be confused with a reluctance to provide support or the idea that certain reactions would cease more easily if they are not brought up or mentioned; rather, it seems to represent the idea that the children's emotional reactions to a stressful event will eventually cease if care is given in the usual way. --- Interpreting Signs of Discomfort When parents observed and paid attention to some specific reactions from their child, they then had to interpret the meaning of these reactions and try to understand the underlying cause. Through this process of interpretation, they made assessments both according to the existing cultural norms and expectations of child behavior after disasters, and according to their own knowledge about their child's personal characteristics and developmental progress. For instance one mother focused on her children's different reactions, and understood this discrepancy as being a function of age: So, I have actually realized that there are some important differences in an eight-year-old and a tenyear-old when it comes to simply realizing the consequences of what happened. John seems to have grasped the gravity of such an event. Roger doesn't seem to have grasped that at all. In these interpretations the child's age is referred to as an explanation for their differing behavior, Another common attribution was based on the children's personal characteristics as explained by this mother: I think our 16-year-old has more vivid fantasy than his older brother, and I think he has been dreaming more as well. He tends to create a little drama because he is quite a dramatic person. The other one doesn't make much fuss about it. Thus, this boy's dramatic reactions were considered normal, and did not warrant concern. Attributing his reactions to his dramatic nature seemed to function as an aid for the parents understanding of their child's behavior. These ways of interpreting behavioral signs helped the parents to inquire into what caused them, and helped them understand the extent to which a particular behavior ought to cause concern and subsequently require more intervention on their part. Within this frame of cultural and personal attributions two categories of behavioral signs emerged and were labeled: analogue signs and contingent signs. Analogue signs were comprised of reactions or behavioral changes that were attributed to the disaster because of their thematic resemblance to the tsunami-related exposure. Such reactions were activated by reminders of trauma, or they bore a clear resemblance to what the child had experienced during the disaster or in its immediate aftermath. Typical reactions that parents had observed in their children were being afraid of water or having nightmares where the content was closely related to experiences of death or fear of losing parents or siblings. One father said: ''She dreams about death. And she has these compulsive thoughts about funerals. Her thoughts circle around death and funerals.'' His daughter, who was eight at the time of the tsunami, was evacuated during the disaster and was accidentally taken into a church where the bodies of deceased children were being kept. Contingent signs referred to reactions that were more general, and the interpretation of such behavioral signs relied more on situational cues. The contingent signs included diverse behaviors, mood states or symptoms indicating that things were awry, but where the connection to the traumatic incident is more unclear. When the parents had attributed these signs to the tsunami it was because they occurred shortly afterward. The most frequently mentioned contingent signs were sleep difficulties, moodiness, irritability, separation anxiety, and social withdrawal. Despite the nonspecific nature of these reactions, parents generally tended to relate these to the disaster, mostly because of their temporal closeness/proximity to the tsunami. Both the analogue and contingent signs were thus interpreted as being post-trauma reactions and were viewed as normal and understandable. Taken together the findings suggest that a vast majority of the parents could give nuanced and detailed descriptions of their interpretational efforts. Attributing the child's reactions to understandable post-trauma reactions and therefore as something to be expected, reduced the alarming impact of the observed signs. Because these reactions made sense, they thereby had the potential to reduce parents' worry and concern. The findings also suggest that the parents adjusted their expectations and practices according to several factors, and thus exhibited flexible expectations of their children's behaviors. --- Parental Support Strategies The parents mentioned a range of actions taken with the intention to support their children's post-trauma coping. In general, these made up three main categories. The first two, re-establishing safety and resuming normal routines, represent parental efforts to adjust and prevent distress and the development of symptoms in their children, while the third, coping assistance, describes how the parents in different ways made active efforts to help children cope with symptoms. The parents often reported more than one supportive strategy, and some of them described using all the different types of support. Below follows a description of the support strategies. --- Re-Establishing a Sense of Safety Twenty-nine of the parents said that they put an extra effort into making their children feel safe and secure after returning home. This involved spending more time with their children, not leaving them home alone, and generally creating a family atmosphere in which their children could feel safe. A frequently mentioned change in routines was a reduction of their own workload and working hours, or a shift in their work schedule in order to be able to stay home with their children. Many parents also spent less time engaging in their own leisure activities for a certain period in order to be able to spend more time at home. They put a considerable amount of focus on being available if their children needed someone to talk to: We spent a lot of time together…and made sure that one adult was always home in the morning. And that there was at least one of us at home in the afternoon … that kind of things. So, we had, like, a careful transition, in order to get back to normal life. The mother of two teenage girls said: ''We all slept in the same bed for at least a week after returning home. And then, after a few days, we rearranged this and let the girls share a bedroom. I actually think this was very important at that point.'' Parents also put considerable effort into protecting their children from stimuli that could induce distress. Many parents tried to hide their negative emotions in the presence of their children. They also tried to protect their children from people continually asking about the disaster, as they thought this type of exposure could serve as a trauma reminder. Even though parents emphasized the importance of protecting their children, some of them retrospectively expressed concern that they might have been overprotective. --- Resume Normal Roles and Routines Thirty-five of the parents provided statements that in various ways reflected efforts to follow daily routines and getting back to normal family life as soon as possible. In particular, parents focused on re-gaining normal family functioning and helping their children continue with their normal activities. One family provided increased support for a period of time, in order to let their children, aged nine and eleven, focus on their daily routines and activities: We put a high priority on helping the kids with their homework. They needed a little extra at home.… It was nice being able to provide a little extra help, and in that way enable them to go on with their other routines and activities as usual. Hardly any of the parents in this sample expressed a concern that the special adjustments made in the aftermath of the tsunami would imply a permanent change in routines. They seemed to accept that certain routines could not be followed as strictly as they would be under normal circumstances. --- Coping Assistance Thirty-nine of the parents also tried to help their children cope with the trauma by engaging in supportive actions toward them. Such action was often initiated when the parents noticed specific psychological reactions in their children. There was a wide variety in the strategies parents used to facilitate their children's recovery. For instance, some children developed a fear of water after the tsunami,and many of the parents said they had taken their children to the swimming pool in order to help them overcome this fear. Other children struggled with nightmares and had difficulties falling asleep at night. In these cases, parents adopted different routines in an attempt to enhance their children's sleep. The importance of dialogues and supportive talk was mentioned by more than half of the parents in this sample, in particular talking to their children about what had happened. The parents mentioned that helping their children talk about their experiences and feelings was one of most important strategies they employed to help their children cope. In most of the cases, parents themselves found opportunities to facilitate conversations about the event, either by initiating such dialogues or by encouraging the child to ask or tell when he or she felt like discussing it. One mother emphasized the importance of retelling the trauma narrative, and gave her seven-year-old daughter a task that was intended to help. After returning home I gave her the task of retelling her story three times every day, and one of the times she was supposed to tell the story to a new person.… We had a lot of people coming by to see us.… And after 12 days she said, ''Mommy, I'm done telling the story now''. Some parents also adopted a psycho-educative approach to talking, in this way teaching their children about normal psychological reactions after a traumatic experience and how to cope with distressing thoughts. One mother said: So I have talked to them and told them that, that if they feel bad or scared or whatever, it may not always be easy to know why they feel that way, but it could…I mean, it could of course have to do with what they experienced down there. And then I have explained them a little about ''flash-backs'' and that kinds of things…and that it is normal to have these reactions. A few parents also emphasized the positive aspects of the situation with their children. Typical themes were talking about positive memories of the vacation before the disaster occurred, and suggesting that they had been lucky to survive the disaster and been given a new chance in life. Such reframing might serve to foster positive thinking in a family setting. Furthermore, parents tried to explain to their children that the world is still mostly safe despite the fact that disasters do happen. Hence, supportive talk seemed to serve the function of communicating about and addressing confusion, fears and anxieties, helping the children process the traumatic event, and correct misconceptions. However, a small subsample expressed concerns about their ability to provide adequate care. Their capacity to assess their child's reactions seemed to be closely connected with their own well-being. The few parents who stressed this issue had themselves been severely affected by the disaster, through loss, serious physical injury, or severe posttraumatic reactions after returning home. Thus, the impact of secondary stressors may have been of particular importance for these families. In spite of this the parents could explain how they tried to compensate for their own shortcomings by involving their social network in the child's post-trauma environment. For instance, one of the fathers who expressed a concern that he had not sufficiently tended to his ten-year-old daughter's problems, had been dealing with a long process of grief after the loss of close family members. In the interview, he emphasized that he had taken compensatory precautions by bringing other key persons into the household. So, I have used others as support … so that Siri could also be able to use others, and not just me. Just to ensure she got what she needed. Because I have not been able to give her 100% of my attention. But I made sure that others could give her what I couldn't. Made sure there was always someone there for her. This suggests that their increased psychological vulnerability made parts of their parenting more difficult than they would have been the case in a normal situation. --- Discussion Although the importance of supportive parenting is acknowledged in the field no studies have actually asked parents what they do in order to support their children. This paper has addressed this important gap in the literature by focusing on the naturally occurring parenting practices as they are perceived by the parents themselves. We thereby shift the focus of attention from the passive markers of risk that have been typically studied in the literature on children and disasters to a focus on the process of recovery and how parents try to assist in providing an optimal post-trauma recovery environment. There are two results in particular we wish to draw attention to. The first is related to the parental process of interpretation and the second is related to parents' actual coping assistance. The findings highlight the ways in which the parents' sensitivity to their children's levels of post-traumatic stress enables them to adjust their parenting strategies to encompass their child's needs and thus contribute to a favorable post-trauma recovery environment. The parents' support strategies are closely connected to interpretations of child behavior and situational characteristics after a traumatic event. When considering how parents perceived and interpreted the post-trauma behavior of their children, it is essential to take into account what kind of trauma they were exposed to. Totally unprepared, these families found themselves in a life threatening situation in a foreign country. This experience was, however dangerous and painful, shared among the surviving members in the family. The fact that this was an experience shared by family members seems to have been an important prerequisite allowing the parents to create a nuanced and well grounded understanding of their children's needs. Having access to and knowledge about the children's actual experiences may have facilitated the parents' capability to make probable associations between observations and attribution, and thereby contribute to their understanding of their children's needs. Other studies have found that when parents are unaware of the trauma their children were exposed to, the process of interpretation becomes much more difficult. Parents then make use of a wide repertoire of possible interpretations, where more culturally accepted interpretations are preferred . The consequence in such instances is that the parents' efforts to help their child to cope with the trauma may fail. The second finding we wish to underline is related to the parents' attempts to help their children to cope. The parents emphasized re-establishing a sense of safety and emotional support, and sought a return to normality as soon as possible, including resuming their usual roles and routines. Reluctant to interfere with their children's own ways of coping, the parents adjusted their support to let the children use their own strategies as much as possible. This parental strategy may be referred to as ''scaffolding'', or, raising a metaphorical scaffold around the children in order to support their development . Inspired by Vygotsky's descriptions of the ''zone of proximal development,'' scaffolding has been described as an interactional process by which parents adjust or modify the amount and type of support they offer to the child that is best suited to his or her level of development. These parents' ways of providing care after the tsunami mirrors parenting practices that in previous studies have been associated with better outcomes in children as well as findings on how parents' focus on parenting has changed after their children's trauma exposure . These studies have documented that warm, supportive and loving parenting is associated with better outcomes after disasters. Moreover the way parents observed and monitored their children's actions and reactions, along with their focus on being available and supportive could be referred to as ''watchful waiting''. This concept refers to a way of monitoring the progression of potential reactions over a period, in order to determine whether the child needs extra care or treatment. This way of ''keeping an eye'' on their child while at the same time providing a feeling of safety are quite intuitive strategies that they had not necessarily learned. Interestingly, this way of caring, closely resembles the care strategies outlined in the recently developed guidelines for parents after terror and disasters and National Center for PTSD ). In this protocol the focus is on promoting a sense of safety; calming procedures; promoting a sense of self efficacy and connectedness; and lastly promoting hope. This striking similarity between the recommended care, and what parents described doing in order to best help their children to cope following the tsunami, could be interpreted in at least two ways. First, given the character of this particular event, as outlined above, it may have left the parents in the present sample particularly fit and suited to care for their children in the best possible way. That is, the shared experience and their safe surrounding upon returning home may have expanded their ability to provide the warm and sensitive support that has been associated with positive child adjustment in several studies . It is worth noticing that these children reported fewer symptoms of PTSD compared to children in other disaster studies . However, whether low levels of symptoms in the children eased the parenting, or whether the support from the parents reduced the level of symptoms in these children could not be determined within the frames of this paper. Second, it might be that these findings simply reflect how ordinary parenting strategies may apply to more extreme situations as well. The basic argument that we would like to pursue is that what parents do to support their children, will be based on their observations and subsequent interpretations of the child's conduct when ordinary and exceptional conditions are compared. Parents will accumulate specific knowledge about their children from just sharing the events of everyday life with them. The finely tuned interpretation of the child's state of mind makes a difference for what strategies to apply in each case . Some limitations need mentioning. First, the analyses were exclusively based on interviews with parents, and the children's perspectives are not represented. Interviews with the children may have added important perspectives on the quality of care, particularly the extent to which they perceived that the care provided and attention given was appropriate and sufficient. Also, examining how these parenting practices relate to children's post-trauma adjustment and well-being could have added useful information, but was beyond the scope of this article. Yet, as previously noted, these children had, despite their trauma exposure, low levels of posttraumatic stress symptoms. Moreover, we only have information from one parent in each family, most of whom were mothers. Interviewing both parents may have provided us with a richer understanding of how discrepancies between parenting practices within families, as well as spousal support might influence post-trauma caretaking. It also bears mentioning that, on average, the families in this sample were privileged with regard to socioeconomic status and education . This may also have assisted the families in reestablishing a safe and secure everyday life more than what might have been the case in other samples. The families were also removed from the location of the disaster, as opposed to families examined in comparable studies , a fact that could compromise the generalizability of some of the findings to populations living in areas severely affected by disaster. On the other hand this makes the findings unique for studying the role parents may play in children's posttrauma recovery since there were few secondary stressors that could play a role in maintaining posttrauma symptoms. The aim of this study was to understand more of the pathways for children's trauma recovery and how parents can contribute to the recovery. Models of post traumatic stress emphasize pre-, peri-and post-trauma conditions as important contributors to our understanding of the development and maintenance of post-traumatic stress reactions. Although these processes are highly complex and intertwined, this study contributes to the field by highlighting one certain aspect of children's post-trauma environment. By studying how parents naturally adjust their parenting skills to encompass new challenges that emerge after serious traumas we may be able to understand why many children actually do cope well despite experiencing high impact traumatic incidents. In the literature much emphasis has been put on understanding passive markers of risk in the development of post-trauma reactions . This study contributes to the field by studying ongoing processes of parenting as they naturally occur after a serious disaster. The results highlight the importance parents can serve in creating a post-trauma environment aimed at alleviating post-trauma reactions in their children. Inferences must however be made with caution. This study's design does not allow us to conclude that the parents' post-trauma parenting practices actually contributed to less post-trauma stress in their children even though such a connection may seem warranted. In any case, the parents themselves make this connection and their strategies had this specific aim. The strength in this study rests first of all in its design. The in-depth and open interviews allowed the parents to elaborate and reflect on their efforts to help their children to cope. The large number of interviews allowed us to discover patterns of post-trauma parenting. In the analysis we were struck not so much by the differences in parenting practices, but by the similarities. The results lend support to the already established guidelines for early intervention and, at the same time, pave the way for a more careful and individualized monitoring of the clinical work that is conducted with children after trauma. First, the parents' awareness and ability to make use of their usual parenting practices represent valuable resources for assessing and interpreting distress in a child. Early intervention may initially focus on supporting some parents' existing developmental supportive strategies when handling mild and expected symptoms in their children. Second, being able to understand and support one's children seems to be connected to the extent to which the parent has been impaired by the trauma. Hence, severely traumatized parents may need extra support to give optimal care to their children. This could include psychological help for their own distress, or temporary support outside of the family in order to optimize their child's post trauma recovery environment. Open Access This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author and source are credited.

How do parents support their children after a high-impact disaster? To answer this question, face-to-face interviews were conducted with 51 Norwegian parents. These parents and children were all severely exposed to the trauma of the tsunami disaster. The analyses show how parents interpret their children's signs of distress, as well as their own strategies of support in the aftermath. The main strategies described by the parents were watchful waiting, careful monitoring of the children's reactions and a sensitive timing when providing support. Such monitoring, and interpretation of signs of distress, served as an aid for the parents in determining what needs their children had and what support they therefore needed to provide. A range of support strategies were employed, including re-establishing a sense of safety, resuming normal roles and routines, and talking to their children. Parents who were themselves severely impacted by the disaster reported a reduced ability to assess their children's reactions and thereby were unable to provide optimal care in the aftermath. Interestingly, the parents' support strategies mirror the early intervention recommendations put forward in the NICE guidelines and in the Psychological First Aid guidelines which is a well accepted and promising practice for helping children after disasters.

Introduction News media use and participation in political discussion with family and friends -along with sociopolitical interest, internal political efficacy, civic self-efficacy, etc.-are all considered to be indicators of adolescent civic engagement . Despite discrepancies in the definition of the term, the majority of research underscores how civic engagement fosters positive youth development and civic participation in adulthood . Studies have shown that adolescents who engage in sociopolitical debates with family and friends develop a higher sense of belonging , increased social responsibility, and high self-efficacy . However, recent studies have voiced concern about low levels of youth social and political participation , In light of this, our principal research objective is to explore youth engagement in communication about political and social issues and news media use, and examine its predictive capacity on other important dimensions of adolescent development: sense of unity and perceived social competence . In addition, both inter-and intra-country analyses are conducted in two neighboring countries: Spain and Portugal. National research in both countries has found relationships between indicators of adolescent civic engagement and sense of belonging or sense of unity , in addition to other variables related to social skills . --- Sociopolitical communication with parents and friends and information seeking as indicators of adolescent civic engagement Civic engagement is considered to be a complex and multifaceted phenomenon, containing certain problems in its conceptualization . There is a lack of consensus regarding what defines a civically engaged adolescent. Barret and Pachi offer two concepts to define civic engagement: civic participation and psychological commitment . Ekman and Amnå also differentiate between civic engagement -when youth are involved in individual or collective activities outside the political sphere-and political participation -such as voting, protesting, affiliation to political parties, etc. The present research adopts the definition by Schultz et al. , and more specifically one of their indicators: student engagement in communication about political and social issues, which includes debates about sociopolitical issues with family and friends and seeking out information about these issues in newspapers, on the internet, or through national or international televised news. Student self-beliefs : indicative of psychological involvement; Student engagement in communication about political and social issues : indicative of individual civic engagement; Student participation in civic activities outside of school: reflects student involvement in collective civic engagement that is not part of the formal learning context; Student participation in civic activities at their schools: reflects student involvement in collective civic engagement that is related to education; and Students' expected political participation in the future: refers to behavioral intentions with regard to legal and illegal forms as well as individual or collective forms of formal participation . Adolescents' political interest is influenced by cognitive changes as well as new social experiences . Likewise, social interest is the basis of future action and participation in society . Awareness about social problems, discussed within the nuclear family, with friends, or at school fosters the adolescents' sociopolitical interests . For example, research by Lee et al. demonstrated that youth who discuss public issues with peers and family members, as we as read and reflect upon news, have more interest in and knowledge about sociopolitical problems. Furthermore, communication skills are important for effective civic participation . An adolescent's communication skills -developed in interpersonal communication with other people through debating sociopolitical issues-constitute the foundation of their civic competence . Accordingly, youth whose parents are interested in sociopolitical issues will have higher levels of civic knowledge , and those who frequently discuss politics with a family member will have a higher probability of volunteering or voting as adults . Communicating with parents about sociopolitical issues has been demonstrated to mediate between consumption of sociopolitical information and civic engagement . Likewise, adolescents who discuss social problems with their classmates -expressing and listening to different opinions-have shown higher levels of civic knowledge and a higher probability of voting in the future . Furthermore, discussing political issues with friends usually contributes to the youth's interest in politics . Similarly, López et al. observed that 15-25-year-old North Americans who regularly followed the news -either on television or through other means-had a higher probability of participating in political groups, voting, or addressing community problems. In addition, according to the IEA Civic Education Study conducted in 29 countries in 1999 and 2000 , expected adult voter participation in national elections was higher among 14-17-year-olds who frequently read the newspaper compared to those who did not. Accordingly, awareness and understanding the aforementioned issues leads to long-lasting civic attitudes, values, and behaviors during adulthood, contributing to the development of prosocial reasoning . Political socialization in different contexts throughout the youth's formative years will foster civic awareness and the development of their first political and social ideas . Along the same lines, Bandura highlighted the plasticity of the developmental process, citing systemic and probabilistic relationships between a developing organism and a changing context. The author encourages theories on personality development to consider an adolescent's social network and the changes and interactions between its members in order to better understand phenomena such as personality, cognition, or behavior. According to Hayden , behaviors have an influence on and are influenced by value systems, social structures, governmental regulations, etc. Thus, adolescent civic engagement, in its different forms, is reinforced by developmental contexts and experiences . Consequently, following the theory of social capital, the social connections and trust amongst people produced through these experiences will benefit both the individual as well as society . Therefore, social contexts exert an important influence on an adolescent's values and civic behavior . Moreover, given the demonstrated relevance of news media use and discussions about sociopolitical issues as indicators of active and engaged citizens, it leads us to explore whether this indicator can predict other variables of adolescent development such as sense of unity and social self-efficacy. --- Adolescent engagement in communication about political and social issues as a predictor of social self-efficacy and sense of unity As previously mentioned, indicators of civic engagement are related to aspects of adolescent development such as sense of belonging or self-efficacy . Likewise, positive development models consider the family, school, or friends to be assets that promote skills for healthy youth development . This data highlights the importance of considering adolescent social and political communication in their contexts, as well as seeking out sociopolitical information, as predictors of variables such as social self-efficacy and sense of unity. Civic self-efficacy is considered to be a predictor of adolescent participation in different types of civic activity in the future , and political self-efficacy is associated with democratic school environments which are open to dialogue . However, research conducted to date has not yet clarified the relationship between social self-efficacy, as a specific subdimension of the general self-efficacy scale proposed by Bandura et al. -understood as an individual's perception of their ability to develop healthy and supportive social relationships -and student engagement in communication about political and social issues. In response to the lack of research on the relationship between these two variables, the authors believe that there could be a positive relationship between news media use and political discussions and social self-efficacy. Similar to adolescents' political self-efficacy -developed through vicarious experiences and knowledge gained through conversation about political and social issues , it is likely that political debate fosters an individual's perception of their ability to express opinions, relate appropriately with others, etc. Accordingly, the specific characteristics of the debate could foster a positive attitude towards one's own ability to participate . In addition, certain aspects measured in social self-efficacy related to communication with others are also present in communicative self-efficacy, the latter understood as an individual's beliefs about their ability to express themselves using their own words, thoughts, and feelings . For that matter, various studies confirm that discussing controversial issues can improve communicative self-efficacy . Sense of unity is conceptualized as the positive feeling derived from forming part of a larger social structure that provides a common good . Indicators such as seeking information or speaking about sociopolitical issues with family and friends can predict a sense of interdependence with others . For example, an open style of communication with the family fosters the youth's connection to their family unit, and consequently, their connection with others . Our research addresses sense of unity as a broader and psychological concept, examining beyond the connection and interdependence of specific contexts. According to the theory of social capital, youth with a stronger connection to and trust of others will develop more social responsibility as a means of maintaining their network of social connections . In addition to considering the independent associations of these variables with student engagement in communication about political and social issues, we believe it is important to explore the association between this indicator of civic engagement and social self-efficacy through sense of unity, given the current relevance of the latter in other research . Following 1 3 Smetana et al. , adolescents who participate socially increase their interdependence with others, social responsibility, and the perception of self-efficacy. Lastly, we would like to explore this complex network of relationships according to country of origin. The role of country in student engagement in communication about political and social issues, sense of unity, social self-efficacy, and the associations between them Both Portugal and Spain foster sociopolitical participation in and out of school. For example, the Dream Teens project in Portugal aims to increase adolescents' social participation through a participative methodology of research-action . In Spain, the Barómetro Opina aims to understand youths' concerns regarding national or international social and political issues. Furthermore, the Ciudades Amigas de la Infancia program -sponsored by UNICEF -has been implemented in both Spain and Portugal, aiming to design effective policies in any city, town, community, or local government agency, to foster youth social participation and respect for their rights. These projects and activities are indicative of these countries' consideration of the youth's opinions in order to improve their wellbeing and foster positive youth development , as well as promote different forms of adolescent civic engagement . For example, youth participants of Dream Teams report a positive impact in their personal development, actions, relationships, and feeling of support . Thus, it can be hypothesized that implementing these types of programs in both countries can help increase different forms of participation, such as that proposed in this study: seek sociopolitical information and participate in debates or discussions with family and friends regarding political and social issues, and simultaneously foster characteristics such as sense of unity and social self-efficacy. Torney-Purta summarized the analyses conducted in a study with 28 European countries about 14-year-old students' knowledge and civic skills. In this summary, Torney-Purta affirms that adolescents from Southern European countries show a high sense of self-efficacy and were considered capable of collaborating with their peers to resolve problems. Schools in these countries promoted student councils, ensured an open-classroom environment for discussing problems, and considered voting and elections to be important. Likewise, perceiving similarity with others and feeling a sense of belonging to a stable and reliable society may serve as a catalyst for youth social participation . A Portuguese focus group study attributed low adolescent participation in civic and political issues to feeling generally excluded by adults, despite parents and media providing information and making them aware of social and political issues . Accordingly, in a study conducted by Sloam comparing fifteen European countries it was found that Portuguese adolescents were less likely than their Spanish peers to engage in active forms of social participation . The main aim of this research is to analyze the influence of Spanish and Portuguese adolescent's participation in communication about sociopolitical issues on social selfefficacy, both directly as well as indirectly through the sense of unity, considering the country of origin. Each scale was examined according to sex, age, and country of origin. It was first hypothesized that there would be differences in the mean scores according to sex and age. The authors believe that boys will have higher scores than girls in civic engagement, but lower in sense of unity and social self-efficacy. In addition, it is hypothesized that age could be key to the development of these indicators, i.e., more civic engagement, sense of unity, and social self-efficacy as age increases. Likewise, the authors argue that the adolescents' dialogue at home or with peers will increase connections between members, generating a sense of unity which will in turn foster social self-efficacy. Likewise, it is believed that the quality of social and political communication will impact the adolescents' participation and will be generalized as a phenomenon for developing social and psychological instruments such as social self-efficacy and sense of unity. Lastly, given the similarities between the countries explored, the authors hypothesize that there will be no significant differences in the relationships established in the proposed model. --- Method --- Study design and participants Population data comes from the 2017/18 edition of the Health Behaviour School-aged Children study, a WHO collaborative cross-national study which has collected data from 11-18-year-old adolescents every 4 years since 1983/1984. This research includes 36,992 adolescents 13-18years old from Portugal and Spain , with 18,715 girls and 18,277 boys . Participants were selected using stratified random multistage sampling by conglomerates and data was collected through an online questionnaire, answered anonymously by the adolescents during school hours. Data collection was carried out according to the standardized international protocol and ethical consent was granted by the competent institution of each country: the Comité de Ética de Andalucía in Spain, and the X Comissão de Ética da Faculdade de Medicina da Universidade de Lisboa in Portugal. --- Instruments The instrument used was the Health Behaviour in Schoolaged Children questionnaire . This instrument has three types of questions: mandatory questions , optional questions , and national questions according to each country's specific interests. For the present research, the following variables were selected from the optional "Positive health" optional package , in addition to sex, age, and country: -Student engagement in communication about political and social issues is one of the indicators used to define civic engagement proposed by Schulz et al. and measured through 7 items that asked, "How often are you involved in each of the following activities outside of school?", with four responses options . Some of the items were: "talking with your parent about political or social issues", "watching television to inform yourself about national and international news", "talking with friends about political and social issues", etc. Flanagan et al. uses the 7 items on three different scales, calculating the reliability coefficient for each of them. Thus, item 1 is part of a scale called "Communication with parents about politics", with an alpha coefficient of .86; item 4 is part of the "Communication with friends about politics" scale with an internal consistency of .86; and items 2, 3 and 5 are part of the "Overall media consumption" scale along with other items, with a Cronbach alpha of .81. For items 6 and 7, no validation studies have been found. The present study combines all the items on a single scale with an internal consistency of .85. -Social self-efficacy is an 8-item subscale inspired by the self-efficacy scale developed by Muris . Some of the items were "How well can you express your opinions when other classmates disagree with you?", "How well can you tell other children that they are doing something that you don't like?", "How well do you succeed in preventing quarrels with other children?", etc. The response options oscillated between 1 and 5 . The Cronbach alpha of the original sub-scale was .85 . In this study, the Cronbach alpha was .86. -Sense of unity was evaluated through an 8-item measure developed by SALUD project and included in the HBSC study . Some of the items were: "I feel that I contribute without expecting anything in return", "I feel that I mean a lot to others", "I feel that others mean a lot to me", etc. This scale has been used in other research with alpha values between .76 and .89 . The Cronbach alpha coefficient for this study was .90, showing adequate internal consistency. --- Data analysis Firstly, descriptive statistics were used for the three variables, including the mean and standard deviation. Secondly, Student t test, ANOVA test and Tukey HSD test was employed to examine mean differences in the examined variables across sex, age and country of origin. Effect size was estimated using Cohen's d, considering values around .40 as small effect, values between .50 to .70 as intermediate effect, and values equal to or higher than .80 as strong or large effect . In addition, Pearson's correlations were calculated to examine the associations between the studied variables. Thirdly, the influence of SECPS on SSE through SU was analyzed in a structural equation model with unweighted least squares . Model fit was evaluated using different adjustment indices: Chi-square , -its value should not be significant, although Cheung and Rensvold argue that sample size can influence its value-; Comparative Fit Index , with values above .90 considered acceptable; Root Mean Square Error of Approximation ; and Standardized Root Mean Squared Residual . For RMSA and SRMR values near or below .08 and .05 are considered indicators of acceptable model fit, respectively. The intensity of the direct and indirect effects were examined considering the standardized coefficients and their effect sizes. Following Peterson and Brown , the standardized β coefficients were transformed into r and this indicator into eta-square , following the procedure of Dunlap . The etasquare values were interpreted as follows: small effect for values around .05, moderate effect for values from 0.06 to 0.11, and a large effect when the values were equal to or greater than .14 . Lastly, configurational invariance analysis was performed by country of origin and without parameter restrictions. An increase of .01 in CFI was considered to be an indicator of significant change in the model by country . In addition, Fisher's Z test was employed to examine significant differences in the indicators by country. Only significant differences are reported. IBM SPSS Statistics 22 was employed for descriptive statistics, correlations and mean comparison, and JASP 0.14.1 based on R, was used for the model of structural equations and invariance analysis. 17-18 Spanish - 13-14 Portuguese: p tukey < .001; d = .11 17-18 Spanish - 15-16 Portuguese: p tukey > .999; d = .00 17-18 Spanish - 17-18 Portuguese: p tukey < .001; d = .25 13-14 Portuguese - 15-16 Portuguese: p tukey = .013; d = .11 13-14 Portuguese - 17-18 Portuguese: p tukey < .001; d = .33 15-16 Portuguese - Tukey HSD test 13-14 Spanish -15-16 Spanish: p tukey = .030; d = .04 13-14 Spanish -17-18 Spanish: p tukey = .453; d = .04 13-14 Spanish - 13-14 Portuguese: p tukey < .001; d = .63 13-14 Spanish - Tukey HSD test 13-14 Spanish -15-16 Spanish: p tukey < .001; d = .15 13-14 Spanish -17-18 Spanish: p tukey < .001; d = .17 13-14 Spanish - 13-14 Portuguese: p tukey < .001; d = .65 13-14 Spanish - showed significant differences across countries. Portuguese adolescents showed higher SECPS scores with small effect size, while Spanish adolescents showed higher SSE and SU with intermediate effect size. Regarding sex differences, boys have higher scores than girls in SECPS in both the global sample and in Spain, however these differences were not observed in Portugal. Concerning SSE and SU, girls scored higher than boys in the global score, but no sex differences were identified in SSE and SU when comparing Spanish boys and girls. Analysis by age showed that as age increases so does SECPS -as well as a slight increase in SSE-both in the global sample and by country. However, there are no significant differences between Spanish 17-18-year-olds and the other two age groups, and between Portuguese 17-18 and 15-16-year-olds. Lastly, SU is higher in 13-14-yearold boys, with significant differences compared to the other groups in both the global sample and in Portugal. However, 17-18-year-old Portuguese youth have higher SU than the rest of the groups, despite no significant differences between 15-16 and 17-18-year-old Portuguese youth. Lastly, Pearson's correlation coefficient values showed that SSE and SU have positive and significant relationships with SECPS scores, with a low intensity. In addition, SSE and SU are positively associated, with moderate intensity. --- Results --- Descriptive statistics and correlations --- Testing a mediation model of the influence of student engagement in communication about political and social issues on social self-efficacy through sense of unity and invariance across countries As presented in Table 2, goodness of fit indices for the hypothesized mediation model showed excellent data fit despite Chisquare being significant . Figure 1 presents the standardized coefficients of the global model estimating the direct association between SECPS and SSE, as well as the indirect effects of SECPS on SEE through SU. The model explained 34.2% of the variance of SSE and 6.2% of the variance of SU. The direct path from SECPS to SSE showed low and significant factor loading Likewise, mediation effects suggested that the effect of SECPS on SEE through SU operates significantly and with low intensity . In both cases, the effect size was small. However, the directs effects of SECPS on SU and of SU on SEE showed greater intensity, with intermediate and large effect sizes. Configurational invariance analyses were performed with samples segmented by country. Figure 2 presents the standardized coefficients for Spain and Portugal. The models explained 34.1% and 29.0% of SEE, and 7.2%% and 12.4% of SU. The configurational invariance model fit was excellent for country with a .003 decrease in CFI . In addition, small variations in the standardized coefficients were examined using Fisher's Z, showing that changes were not significant , except in the association between SECPS and SU which was significantly more intense in Portugal and between SU and SEE which was significantly greater in Spain . However, these results only show relationship strength, without affecting the rest of the indicators that reveal the model's configurational invariance across the country. --- Discussion The main objective of this research was to analyze whether adequate socio-political and informative communication influences social self-efficacy -either directly or through sense of unity-promoted by the interactions with others, in a representative sample of adolescents residing in Portugal and Spain. Consistent with previous findings, results showed that establishing debates with family and friends is associated with youth self-efficacy and promotes a recognized interdependence with others . Likewise, the influence of sociopolitical debate with family and friends on sense of unity, as observed in this research, has been confirmed by other studies which highlight the importance of political debate at home for reinforcing active civic engagement in the future . Prior to examining the relationships in the proposed model, each of the variables was explored individually and according to different sociodemographic variables. Thus, results showed country differences in the scores of the three variables . Regarding social media and political communication, Portuguese adolescents scored higher in engagement in communication about political and social issues than their Spanish peers. These results contrast with other research. For example, Magalhães and Moral characterized Portuguese youth as having low exposure to political information and therefore lower levels of political knowledge due to their notorious discontent with the functioning of their democratic system. In another study, Sloam compared five forms of political participation -voting in national elections, displaying a badge, signing a petition, joining a boycott, and participating in a demonstration-across fifteen European countries. Results showed Portuguese adolescents participating at comparatively lower levels than Spaniards, as well as presenting lower participation than adolescents from the other countries such as Sweden and France or those that are part of continental European democracies such as Germany, Belgium, Austria, and Luxembourg. In this sense, and considering our results, it is possible that countries 'specialize' in certain forms of participation according to their cultural or social characteristics, with Portuguese youth exercising more passive forms of socio-political participation such as social and political debates and searching for information. This coincides with Menezes et al. , who found in the International Study of Civic Education that Portuguese youth -compared to other European youth-engaged more in passive-conventional or social activities such as the tendency to discuss politics mainly with parents and very rarely with teachers. Secondly, and regarding social self-efficacy and sense of unity, Spanish adolescents showed higher scores than their Portuguese peers. These results must be interpreted with caution due to the lack of prior studies comparing the two countries in this subdimension of general self-efficacy by Bandura et al. , nor with the new dimension proposed by Samdal et al. : sense of unity. In fact, previous research conducted by Verdugo et al. using the general self-efficacy scale of Bandura et al. have found no differences between Spanish and Portuguese adolescents. Likewise, research about the social skills of native and immigrant adolescents from Spain and Portugal found high mean scores in both groups and in both countries . On the other hand, Spain and Portugal are considered to be two very culturally similar countries , in which their citizens function interdependently and synchronously in relation to other members . Therefore, as the 2012 Report on Youth in Spain points out, we can hypothesize that Spanish youth tend to have a more-positive self-evaluation than their peers in other European countries. Analyses by country, sex, and age produced interesting results. When asked if they participated in sociopolitical discussions with family and friends, or if they would seek out sociopolitical information in the media, Spanish boys scored higher than girls. This data coincides with prior research on civic engagement and gender, reporting that boys are more interested in following news or have a higher intention of voting in the future and that girls are more interested in volunteering . Sex differences in sense of unity and social self-efficacy were also found in favor of girls, except when comparing Spanish youth. Research focusing on these variables have shown discrepancies. It seems that girls have higher social self-efficacy than boys at a younger age , however these differences dissipate as they get older . Little research has been conducted about sense of unity, as it is a relatively new construct. However, a study by Wahlström et al. found no sex differences in a sample of Swedish adolescents. From a developmental perspective, the results showed an increase in civic engagement as the youth aged. Older adolescents have higher levels of political interest due to the frequency of political debates with peers and family . In addition, the development of cognitive skills affords them more complex thoughts and reasoning regarding different issues . Likewise, country comparison of social self-efficacy produced significant results at a later age, however not within countries. This contrasts with sense of unity, which showed higher scores for younger adolescents in both countries, although without significant differences between 15-16 and 17-18-year-old Portuguese adolescents. Data fit to the hypothesized structural model suggested that student engagement in communication about political and social issues influences adolescents' social selfefficacy, both directly and indirectly through sense of unity. According to Russo and Stattin , young people interested in social and political issues are more likely to participate in debates with family members, stimulate their civic interest with friends, and seek out relevant sociopolitical information from the media in order to stay connected. Therefore, the need to relate to others through these debates, feel connected with them, and share the same social and political interests will promote a greater affinity to these groups, providing learning experiences which increase the youth's perceived social competence. Research by Oosterhoff et al. supports these results, demonstrating that social connection between peers fosters civic engagement, and conversely, this civic engagement cultivates social connection. Despite country differences, the structural model demonstrated invariance, thus confirming similar relationships between the variables in both countries. However, slight differences were found. Firstly, a stronger association between student engagement in communication about political and social issues and sense of unity was found for Portuguese adolescents. This may be due to the fact that, although Spain and Portugal are culturally similar, Portuguese society has for centuries been considered to be a stable structure where the extended family and kinship relations are of great importance and based on mutual obligations . In this sense, it is possible that communication between Portuguese adolescents' family members creates stronger bonds of union and interdependence than in the case of Spanish adolescents. Secondly, the relationship between sense of unity and social self-efficacy was stronger for Spanish adolescents. This result suggests that feelings of similarity and togetherness with others generates a greater social self-competence in Spanish than in Portuguese adolescents, promoting a more positive perception of their abilities to establish satisfactory and healthy social relationships. As has been stated since the first Eurobarometer surveys between 1982and 1990 and in previous research , Portuguese youth show low attendance to associations or participation in organizations where they can develop social skills or abilities. It is possible that this decrease in social contacts influences the lower intensity of their feelings of belonging to a broader social structure and, therefore, decreasing opportunities to develop higher social self-efficacy. Despite the mentioned differences in the intensity of some associations, the model representing the direct effect of communication about social and political issues on social self-efficacy and its indirect effect through sense of unity demonstrated to be adequate in both countries, as confirmed by the fit indices. Portugal and Spain are two countries that share similar cultural and historical traits as well as educational policies for social participation , thus, it is not surprising that the development of adequate social skills is promoted by both sociopolitical debates established in social relations and the feelings of belonging to these social groups, reinforced by those interactions. The research findings highlight developmental contexts and media as principal agents of adolescent political socialization. The ties between members of each context foster the youth's development of adequate social selfefficacy and relationships beneficial to others. Thus, this study contributes from a theoretical standpoint to the existing literature on adolescent civic engagement, emphasizing its importance for the youth's individual and social development. In addition, family, school, and friends are fundamental for fostering critical thinking and social commitment in adolescents. Likewise, from a macrosystemic perspective, educational and governmental policies aimed at promoting positive social skills amongst youth as well as a healthy adolescent social and political participation are necessary. This requires that procedures be promoted from within the political and public sphere towards news media and formal education, as mentioned later in the strengths of the study. Therefore, the theoretical, practical, and social implications of this research can be summarized in one fundamental aspect: understanding the influence of youth sociopolitical communication in their developmental contexts and the use of media contribute to developing adequate political, educational, and social measures based on improving individual and social wellbeing. Finally, some limitations should be considered when interpreting the results of this study. Among them, the inability to establish causal relationships due to the crosssectional research design and the self-reported assessment of the participants. In addition, the novelty of the concept sense of unity made it difficult to contrast our findings due to the lack of research comparing this construct in Spain and Portugal. Similarly, this research focused on the social subdimension of the general scale of self-efficacy proposed by Bandura et al. whereas most of the previous studies explore the general score. On the other hand, some strengths should also be highlighted. For example, the large sample size, the use of valid instruments to assess the different constructs, and the comparison between two Mediterranean countries. Although causation could not be established, one strength of the mediation analyses used is that the inference of the relationships between variables offers a more complete view of the data. Furthermore, this research contributes to understanding how to generate social and political interest, knowledge, and participation in democratic societies by exploring the influence of news media use and political discussions on sense of unity, and through it, on adolescents' social self-efficacy. These findings suggest the benefits of implementing educational policies and practices that reinforce news media use and communication through debates with family and friends about sociopolitical issues. For example, school curricula could include information about national and international social and political issues provoking classroom discussion and debate. This dialogue could foster the adolescents' sociopolitical interest and concerns, and be complemented by other civic activities such as denouncing and advocating against inequalities and injustices on social media, collaborate with human rights associations, etc. These programs should foster critical thinking and socially active citizens, which may in turn contribute to increasing a sense of belonging and to developing adequate social skills. Future research could examine the impact of these educational interventions on the future behavior of the youth as adults. Conducting longitudinal studies may help to better understand if the relationships established in this study change with age and if interventions have generated more civic engagement in different social and political areas. --- Conclusions Despite studying different aspects and dimensions of civic engagement, researchers have yet to arrive to a common definition . However, it has been found that participating in discussions about political and social issues, as well as seeking information and news in social media is a good predictor of civic knowledge and political interest , both factors key for the development of civic engagement. During adolescence, adequate social engagement and participation in actions for the benefit of others is necessary and effective for the youth's positive development . In this sense, this research has highlighted the importance of young people's interactions in sociopolitical debates with family and friends and seeking out information to adequately develop their social skills, as well as for the feelings of unity and the recognition of interdependence with others. In addition, despite slight differences in some associations between variables, it has been found that the development of both Portuguese and Spaniards' civic commitment fosters a positive sense of unity and social self-efficacy. Lastly, school interventions promoting classroom debates among peers on current and past, national and international, political and social issues, as well as programs that promote political-social dialogue between family members and children, contribute to developing greater and more effective citizen engagement and developing social skills among adolescents and future adults. Therefore, these are favorable contexts for promoting skills for youth citizen-participation -in terms of social sustainability-, fostering greater solidarity and inter-generational respect. It has been demonstrated that adolescents equipped with skills and offered opportunities to participate in their society become more capable of solving problems and being agents of change . Funding Funding for open access publishing: Universidad de Sevilla/ CBUA Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. --- Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Discussing sociopolitical issues with family or friends during adolescence, as well as staying informed through media outlets, is key to developing active civic engagement for both youth and future adults. Likewise, these communicative exchanges promote a sense of belonging and social skills which foster adolescents' wellbeing. This research aims to analyze the influence of participation in communication about political and social issues on adolescents' social self-efficacy, both directly and as well as indirectly through sense of unity, in two Mediterranean countries (Portugal and Spain). The sample was selected through random multistage sampling by conglomerates. The 36,992 adolescents (50.6% girls and 49.4% boys) participated in the Health Behaviour in School-aged Children (HBSC) study in Portugal and Spain, and the age groups were distributed as following: 39.6% 13-14-years-old; 35.8% 15-16-years-old; and 25.6% 17-18-years-old. Analyses showed positive associations between the proposed indicator of civic engagement -news media use and political discussions-and social self-efficacy, both directly as well as through sense of unity, with similar results for adolescents in Portugal and Spain. Accordingly, educational programs promoting communication about sociopolitical issues could foster adolescents' sense of unity and social skills, contributing to their positive development, wellbeing, and civic engagement. Curricular materials on current politics and social affairs could increase classroom dialogue based on respect, healthy peer relationships, and sense of belonging, fostering effective political socialization amongst youth and the development of democratic behaviors beneficial to society.

INTRODUCTION It is a worrisome development that educational awareness does not translate to community development in this part of the World. This is as a result of skewed development approaches; one which keeps the people aloof from the development process. Education is meant to grow people into productive citizens who use their knowledge, talents, and learned skills to sustain themselves as members of their respective communities and generally that of the whole human race. However, the reality today is that educational awareness in Ogoni communities does not correspond with the reality on ground development wise. Ogoni is one of the nineteen ethnic groups in Rivers State, Nigeria . By population ratio, Ogoni is a major ethnic group in Rivers State accounting for 1/5 of the total population of Rivers State. The 2006 Population Census puts the population of Ogoni and Rivers State at 832, 000 and 5,198,717 respectively . The estimated population presently according to public records stands at over 2 million people from the initial 500,000 people in 1963 . The people live in clustered settlement measuring 1,050 square kilometre . There are over 124 villages and towns in Ogoni like Bori, Bane, Bodo, Kono, Norwa, Kpor, Ogale-Nchia, Onne, Bera, Deeyor, among others. Politically, the people are spread across six kingdoms of Ken-Khana, Nyo-Khana, Babbe, Gokana, Tai and Eleme while administratively the area is calibrated into four Local Government Areas. Ogoni is situated in Niger Delta region, the third largest mangrove ecosystem in the World . The people are educationally advanced today with higher school enrollment ratio into primary, secondary up to tertiary levels vis-à-vis her neighbours. Toward the end of the 20 th century, Ogoni record as it affects education and development was a far-cry from community development. Though, there is an improvement in the scenarios pictured in Ken Saro-Wiwa's literature on the people, the transitional effects of education on development is still a far-cry. Wiwa decries: …At the moment, the number of our people in the junior and senior cadre of the federal public service and the corporations can be counted on the fingers of one hand…Our children are largely out of school and university, many families have lost their breadwinners and our economy has collapsed completely. Our position is certainly not enviable, not even comparable with other nationalities in Rivers State. There are still more general indicators of under-development in Ogoni that do not set us at par with some nationalities in Rivers State. For instance, some Ogoni LGAs still have no Bank; Ogoni meteorological data or information are not captured among other areas on local radio stations; local radio and television signals have remained poor in Ogoni, most GSM networks in most Ogoni communities are poor or nonexistent; with a parochial subsistent economy; there are fewer organized registered cooperatives that can access loans and grants from public agencies and banks for common projects; electricity still has not reached 90% of rural communities in the area. This study tasks educated people of Ogoni extraction to pilot knowledge based development in their communities. A functionalist approach is advocated where subsets drive community development unit by unit. Knowledge based development can be achieved once education is chosen as the vehicle and development as its destination. Thus, it is important to do an appraisal of the functionalist theory of development on the tripod concepts of education, community and development. --- II. THEORETICAL AND CONCEPTUAL FRAMEWORK A. Functionalist theory of Development Functionalism was developed as a social science theory in the first half of the 20 th century by Talcott Parsons. This contribution was a synthesis of the insights of Durkheim, Webber, and others. Parsons' functionalism takes the form of a schema or descriptive framework of society, its component parts, and the interactions between them. After the Second World War, the new framework was widely embraced by the growing army of researchers conducting empirical research into social phenomena for which they were fiercely berated by social theorists. This led to new variants of functionalism as existing programmes were re-examined. New schools of functionalism also evolved. Functionalists view education as one of the more important social institutions in a society. They contend that education contributes to two kinds of functions: manifest functions, which are the intended and visible functions of education; and latent functions, which are the hidden and unintended functions. There are several major manifest functions associated with education. The first is socialization. Beginning in the preschool and kindergarten, students are taught to practice various societal roles that extend beyond the school setting. The French sociologist Emile Durkheim ), who is regarded as one of the founders of the academic discipline of sociology, characterized schools as "socialization agencies that teach children how to get along with others and prepare them for adult economic roles . Another role of education which is of interest is that schools teach students conformity to law and respect for authority. By organizing schools as bureaucracies, much like what is found in the labour market and in other social institutions, schools teach children what is commonly referred to as "the hidden curriculum". Thus, educational awareness should be the precursor of community development. --- B. CONCEPTUALISING EDUCATION "Plants are shaped by cultivation and men by education… We are born weak, we need strength; we are born totally unprovided, we need aid; we are born stupid, we need judgment. Everything we do not have at birth and which we need when we are grown is given by education" -Jean Jacques Rousseau Etymologically, "education" is derived from the Latin word "Educare" . It is the process of facilitating learning, or the acquisition of knowledge, skills, values, beliefs, and habits . From all intent and purpose, education is meant to grow people into productive citizens to use their knowledge, talents, and learned skills to sustain themselves as members of the human community and generally that of the whole human race. However, this does not match the reality and impact of education in most communities today. That means, education has categories; we should be concerned here with the typology that develops our communities with a focus on the "student community"; this applies to all categories of persons or subsets so long as they are teachable and willing to use learned skills for the good of the community. The question before us is, "what is the true meaning of education in context?" Education should be measured by the growth of a student into a true human being and a "civil being". A civil being is someone who believes in and obeys the rule of conduct within a subset and the larger system subsuming all other held moral and parochial beliefs to the former. That is to say, an educated person is not arrived at merely by the marks scored. True education tries to develop creativity and inspires relationship; it gives life a meaning, purpose, wholeness and peace. True education therefore entails the learning and application of learned skills to human and community development. It is worth noting that 'true education' operates a life-based curriculum, providing people of all ages the opportunity to learn and apply learned skills in their community. Thus, there are categories of learners and education systems. The three main types of education are formal, informal and non-formal. There is no assignable age limit that restricts anyone from outsourcing information for community development and so no one is excused. --- C. CONCEPTUALISING COMMUNITY "if children have a problem, it is said to be society that is at fault…there is no such thing as society, there is a living tapestry of men and women and people and the beauty of that tapestry and the quality of our lives will depend upon how much each of us is prepared to take responsibility for ourselves" --- -Margaret Thatcher Sometimes, one would have to go off convention; "locked in ideas" of what a "community" is to what it should be. One of the hallmarks of effective learning for learners is going tabler-rasa on subject-matter. The conception of "community" here encourages collectivization of actions, which discourages organization that seeks to provide shade for people who maintain some sort of social distance from their social ugly, caused by their actions or inactions. Therefore, imbued in one's sense of community is membership of organizations which imposes on him or her a duty or sense of responsibility. It is in this wise that one is persuaded to admit in argument that society for want of a better taxonomy lies at the heart of an ideal community organizing: Society includes every kind and degree of relationship entered into by men, whether organized or unorganized, direct or indirect, conscious or unconscious, cooperative or antagonistic…the whole issue of human relations …without a boundary or assignable limit. Society encompasses all of humanity, that is, the varied and multiform relationships into which men necessarily enter in the course of group life. Society could include any of the multiplicity of groups, units or organizations found among men A community should connote a system of subsets with each subset working for the functionality and overall performance of the former. Systems thus exist for the wellbeing of the subsets within them. This shows a symbiotic rather than a parasitic relationship evident in most systems. An ideal subset does not work at cross purposes with another in a system. Subsets should be able to identify purposes which are injurious to the interests of the larger whole. That's the community spirit the people should carry in their varying subsets as community leaders, politicians, faith-based organizations, social entrepreneurs, youth groups, educationists, students groups, professionals and other informed socio-political and economic appellations under which people operate. A community of subsets should endeavour to engender communication within and also between subsets because the meeting of minds or dearth of it explains the developmental differences between societies which practice them. It is imperative to take a clue from the "Commune" system practiced in ancient China during the reign of Chairman Mao Tse Tung around AD 1500. This is necessary in building a new perspective for the emergence of a new informed community: A 'commune' expresses the institutions, structures and organizations of the system of communalism. Communes were usually very large; perhaps as large as our local government areas …each plans and manages all aspects of its government, economy, education and development. And each elects representatives into its Governing Council, and the People's Assembly at the national headquarters . As practiced, there were ample qualities that made it a reference point, some of which are remedial to the social issues in human communities today. In discussing development, one notes it pays when various subsets participate in decision-making from cradle rather than ostracizing them; this was one of the cultures of the commune system. Where effective communication is conspicuously absent, perceived marginalized subsets may react. This is most avoidable if communication among subsets is frequent. One may ask why opportune subsets and individuals marginalized others from participating in the 'just' allocation and ownership of community resources. Fear and greed most often underscore this anti-community spirit and this happens if one feels insecure, literally social insecurity. There is need to put in place some measure of social security for not just "the Ogoni-Poor" but also persons with record of selfless service to the community. People who serve and feel insecure will go into primitive accumulation. --- D. CONCEPTUALIZING DEVELOPMENT There are several meanings ascribed to the concept of development. Prof. Naanen describes development as "the incremental improvement in the living standard of a people or society" . It is the macro effect or the improved performance of a system different from economic growth. An instance of development is when a society begins to produce more products and increase it overall wealth . Development is equally seen as the process by which a nation improves the economic, political, and social wellbeing of its people as opposed to the abstraction of economic growth based on increased inflation-adjusted market value of goods and services by an economy over time and conventionally measured as the percent rate of increase in real gross domestic product . The difference between economic growth and development lies in the latter as a policy of intervention which aims to improve the economic and social wellbeing of the people as opposed to the former which is a phenomenon of market productivity and rise in GDP. Economic growth deals with increase in the level of output, but economic development is related to increase in output coupled with improvement in social and political welfare of people….therefore economic development encompasses both growth and welfare values . Social scientists believe when local quality of life is improved, there is more development . And 'improving living standards' to most social scientists have come to mean among other indicators higher level of education and literacy, improved workers' income, health, and lifespans . Other social indicators of development of interest are "birth and death rates, Human Development Index , infant mortality rate, literacy rate and life expectancy" . Interestingly, development thrives on improved infrastructure, education, health and wellness, justice, safety, human rights, consumer protection and fair competition . In a nutshell, development is a complex and difficult term to define that requires unpicking and deconstructing, not simply taking for granted its meaning at face value. However, what underlies the nature of development is the process of economic and social advancement in terms of quality of human life . It must be a process that creates growth, progress, positive change or the addition of physical, economic, environmental, social and demographic components. --- III. EDUCATION AND COMMUNITY DEVELOPMENT IN CONTEMPORARY OGONI SOCIETY There are issues raised in course of x-raying the tripod concepts above. There is need to raise and address the following questions: Can contemporary Ogoni society be described Contemporary Ogoni society is an educated one; she has people in all fields of human endeavours and cannot be said to be unaware of the trend of human development. Home and in the Diaspora there are people of Ogoni extraction who have risen to the peak of their career and have made important contributions in their fields; professors, of no mean fit. Beyond the academia, Ogoni people are making useful contributions in the corporate world. Some of them have attained selfsufficiency and could introduce interventionist schemes to assist others. There is also a sea of bureaucrats who are stuck in public service with less means to contribute to the development of their immediate environment. There is a student community peopled by Ogoni students who are leading lights in various institutions of learning home and abroad yearning for scholarship and grants; most have ended up losing out on good grades because they combined studies and work. There are the rural dwellers some less educated but exposed who work and earn; unfortunately, the Ogoni environment promises so little at the moment. The socio-political organizations like the Movement for the Survival of Ogoni People , KAGOTE, GBO KABAALI, and other socio-cultural groups like the Amanikpo are points of social cohesion. The people are organized and governed under both traditional and political leadership. These are subsets in contemporary Ogoni society. A question has been posed above over who takes responsibility for community development in contemporary Ogoni society? It is a valid question to end the blame game. Community development like security is everyone's business because the benefits of community development like crime reduction and unpolluted environment benefit all. If community development is everyone's responsibility proper communication and community spirit should be engaged. Traditional leadership in contemporary Ogoni communities should foster consensus opinion among kit and kin. Baala identifies divisive tendencies accruing from self-embraced westernization and its associative ideals as a challenge to community development in contemporary Ogoni society. More so, Ntete-Nna traces this challenge to lack of common culture in elite recruitment where Ogoni depended on British modernization efforts to raise its elites . It is pertinent to identify development indicators absent in contemporary Ogoni society and dearth of basic modern infrastructure is first. Ogoni lacks modern basic infrastructure from good roads, health facility, post primary education, adequate electrification, and good drinking water, among others. The trunk and feeder roads crisscrossing the community are mainly ancient footpaths yearning for improved design. There is an opinion that road constructions are government preoccupation not community but one may add that development is more organized when it comes from the people. As expected every community should revisit the composition of their respective Community Development Committee as an important governance structure of the communities. Subsets like the Community Development Committee corroborating with Indigenous Civil Engineers of Ogoni, should identify and design projects beforehand that meet the needs and taste of the people; to control external designs that may take little or no cognizance of such needs and taste. Some of the roads in Ogoni communities are terribly constructed. Development must be controlled. It must be noted at this juncture that informed sons and daughters of Ogoni communities are not in short supply. Also, there is the need to address knowledge gaps and development devils. No community develops or makes significant changes using the same method. Change comes at great cost. Sometimes, it is driven by people least expected. Other times, the old order is challenged, overthrown, deconstructed and rebuilt. The student organisations in Ogoni communities should rise up and demand their knit in the governance structure of their communities. The umbrella body of Ogoni students has a lot to contribute to the wellbeing of contemporary Ogoni society. First, the current Executive of National Union of Ogoni Students should begin to collate details of its members across tertiary institutions in the country, and according to their fields to serve as "Think-Tanks" for quality advice on developments par fields which may help in the advancement of Ogoni society. The Diaspora is another very important subset of contemporary Ogoni society. The vision of contemporary Ogoni society is safe with the Diaspora subset. More functionally, there is no community in Ogoni without a handful of her people abroad. Unfortunately, the kind of feedback expected is still farfetched. One is often lost in thought over how someone perceived to be learned in a community of unlearned people by their rating with all their knowledge takes the backseat and rather assumed the role of a community armchair critic. Such persons need to be reminded that "…knowledge exists in two forms -lifeless, stored in books, and alive, in the consciousness of men. The second form of existence is after all the essential one; the first, indispensable as it may be, occupies only an inferior position" . The Diasporas Ogoni is being reminded wherever they are that Ogoni needs them, their knowledge, and resources for community advancement. One is yet to see a public Library or private one accessible to Ogoni people many years after formal education was embraced as a way of life in Ogoni society. Ogoni scholars are doing well in the social sciences and humanities, some of them studied History and related discipline with utilitarian impacts but if one may ask, where are aspects of Ogoni History? Reconstructing the History of Ogoni may usher in an era of peaceful coexistence within Ogoni communities and between Ogoni and her neighbours and as such put paid to the endemic destruction of lives and property experienced over the years. There is need to open up Ogoni community. One of the challenges the society face as a people is living in denial. Ogoni communities are a hub of cultural varieties. As a potential source of revenue, these are tourist bit-coin that has given the State of Israel billion worth of dollar in revenue. Quite amusing that even though Israelis do not believe in the reason for Christian pilgrimage they are tolerant and wise enough to allow millions of people and their money pour into their country to fulfill multi-religious obligations. . It has become a case of producing leaders who are "fellow victims" who are mere administrators of circumstance. One queries the kind of leadership provided by contemporary Ogoni elites where it is easier to count on the fingers of one hand the number of communities in Ogoni which are not presently at war with another sister community. The carnage and fraternal conflicts in Ogoni communities must end and all shades of opinion leaders and organization must rise up to condemn it. IV. CONCLUSION Community development is never an event but a process. It begins where people who are conscious and desirous of a change do not just identify a cause but tenaciously follows its course to maturity. By education, most people have a copious understanding of the kind of community development fashionable today but do not know how to engage their kits and kin to buy-in. Education is good but technology is better. Knowledge is good but wisdom is better. Education should sharpen people's skills and abilities to identify challenges and solve them as subsets in a system. Education should stimulate people's instinct and improve their human conditions. Education should bring about ease of doing things. All the issues raised above could transform the contemporary Ogoni society once it forms the basis of discussion in subsets with practical steps and commitment to enforce them.

There is a missing link between education and community development. This becomes glaring when considering the educational awareness vis-a-vis the level of development in Ogoni communities today. Education should come with the know-how and consequently the ease with which communities evolve, notwithstanding, there continues to be a downward slope in community development in the face of growing educational awareness. This study uses the functionalist theory of development to evaluate how subsets within communities can functionally impress on the quality of development in their locales. The study argues that development is more organized where the people are involved concluding that by education, most people have a copious understanding of the kind of community development fashionable today but do not know how to engage their kits and kin to buy-in. It recommends the kind of education which sharpens skills and abilities of the people to identify challenges and solve them as subsets in their communities.

Introduction Reducing newborn mortality remains a global priority [1]. The majority of the 2.76 million deaths that occur among neonates each year happen in low and middleincome countries and progress in reducing neonatal mortality in these settings has been slow [2,3]. Although Kenya is successfully reducing child mortality, little progress has so far been made in reducing deaths among newborns . Improving the quality and access to care for sick newborns is central to reducing this burden [4,5]. While ensuring effective clinical management is vital to the survival of sick newborns, there is a growing body of evidence suggesting that social factors and caregiving processes are also important in shaping neonatal health outcomes [6,7]. The recently published standards for improving the quality of newborn care in health facilities published by the World Health Organisation [8] include new quality statements regarding the need to ensure that all carers receive counselling and education about the baby's illness , that families are recognised as partners in care and that all small and sick newborns have access to a sufficient health care staff at all times according to standard levels of care . Involving parents in the care of hospitalized sick newborns can help in avoiding poor outcomes and reduce stress among parents [9][10][11]. There is also increasing evidence that maternally delivered interventions such as emollient therapy [12], kangaroo mother care [13][14][15] and preterm infant massage [16] have positive effects on the survival and long term development of low birth weight and premature neonates [12][13][14][15]. Parental involvement in the care of their sick newborn has been highlighted in several studies from high income countries as an important element of the effective implementation of family centred care [17][18][19][20]. These studies also note that involvement in care results in better mother-infant bonding and attachment and, as such, improved parental experiences. Subsequently, with increased involvement, there are better rates of breastfeeding, kangaroo mother care , and ultimately shorter hospital stays [21,22]. Most recently, a multi-country randomised control trial implemented by the WHO in Ghana, India, Malawi, Nigeria and Tanzania found that immediate KMC for sick newborns with mothers present with their babies in the hospital newborn unit 24 h per day aiming for 20 h a day of KMC from within 2 h of birth resulted in a 25% reduction in mortality in the first 28 days of life in the iKMC arm compared with standard KMC [23]. In addition to the potential for enhancing newborn survival, involving parents in newborn care has the potential to enhance parenting skills and to improve their perceptions of the care their child receives. Perceptions of the appropriateness of care is central to encouraging uptake of services, increasing compliance, and potentially enhancing recovery and later development outcomes [24,25]. Currently however, the formal norms of care provision in new born units in most settings are that the tasks involved in caring for hospitalised sick newborns are handed over to nurses and other health professionals [26,27]. While nurses are central to the provision of care in NBUs, many low-and middle-income countries struggle to meet the ratios of staff to babies recommended by national standards bodies [28,29]. In the UK, the minimum standards for nurse staffing levels for each category of neonatal care are 1:1 for neonatal intensive care, 1:2 for neonatal high dependency care and 1:4 for neonatal special care [30]. Similar standards are found in the European standards of care for newborn health, with a recommended ratio of 1:1 for the very sick/critical babies and a ratio of 1:2 for relatively stable babies [31]. India recommends 1:3 or 1:4 for special care; and South Africa recommends 1:1 or 1:2 for intensive care, 1:2 or 1:3 for high-dependency care and 1:6 for standard inpatient or kangaroo mother care units [8]. These ratios are in stark contrast to the ratios of witnessed in many LMICs and a recent analysis of factors influencing the delivery of quality newborn care in 12 countries in Africa and Asia found that shortages of neonatal nurses was a key problem [32]. Kenya has no published standards for staff to baby ratios. A recent study of neonatal nursing conducted across 33 hospitals in Nairobi found a low nurse to baby ratios of 7-15 babies per nurse [29]. Studies have documented the effects of these high baby to nurse ratios both on the ability of nurses to provide care and the types of care that are prioritised or missed with the NBU [23,[33][34][35]. There is a growing body of evidence that as a way of coping with the high workloads exacerbated by understaffing, nurses prioritise clinical tasks over other tasks within the newborn units [36]. This raises the question as to what happens to the missed clinical care? For example, who undertakes the primary tasks not completed by the nurses, such as washing the baby, changing the diapers, and more specialised tasks such as overseeing the placement of eye masks for babies in phototherapy and even nasogastric tube feeding? Studies in NBUs in Iran where the nurse to baby ratio ranged from 1:5 to 1:8 found that unplanned and informal delegation of care by the nurses to the parents was common with little or no parental training [33]. These studies also found that the parents' needs were neglected by the clinical staff with support being provided by the informal parent to parent support networks that developed in the NBUs. Studies that examine how mothers are involved in providing care for their hospitalised newborns, and could address these gaps in care, are scarce. To the best of our knowledge, there is no systematically collected and published information on who provides the day-to-day non-clinical or 'primary' caring roles for the particularly vulnerable babies in neonatal units in the highly resource limited settings found in many NBUs in sub-Saharan Africa where the baby to nurse ratio far exceeds international recommendations. Understanding the tasks that mothers undertake and the support they receive to carry out these tasks is central to identifying interventions to improve sick newborn care and enhance staff and the families' ability to partner in delivering quality care. The focus of this paper is to: provide insight into the roles that mothers play in caring for their sick newborns in two contrasting NBUs in Nairobi, Kenya; to critically examine how the structural, economic and social context shape the nature of mothers' participation in care; identify key gaps and discuss opportunities for improving NBU care in line with the WHO standards in such highly resource constrained settings. --- Methods --- Study design and setting Data for this paper are drawn from a qualitative study that employed an ethnographic approach [37,38] to examine the context of care inside the NBU in one public and one faith-based hospital in Nairobi, Kenya. --- Participants and sampling The participants were mothers of hospitalised small and sick newborns. The two hospitals were purposively selected based on the provision of inpatient neonatal care 24 h 7 days a week, their patient volumes, ease of access, and willingness to allow observers on the ward for extended periods. Mothers taking part in the discharge interviews were purposively selected. --- Data collection Data were collected through non-participant observations in the NBUs of the two hospitals and in-depth interviews with purposively selected mothers at the point of the discharge of their baby from the hospital. The focus of the observations was the roles that mothers played in caring for their baby born in the hospital and immediately admitted to the newborn unit. The approach to data collection, analysis, and interpretation was informed by critical medical anthropology [39,40]. Through observations, our attention increasingly focused on examining the context of care provision and the social relationships that shaped newborn care in the two hospitals. The discharge interviews were guided by a semi-structured interview guide developed through the observations and designed to follow up on issues that arose during the non-participant observations and informal conversations. The interviews were conducted either in English or Swahili depending on the choice of the participant. The data were collected between March 2017 and August 2018. During the study, the first author and her research assistant spent 4 months in each hospital conducting approximately 627 h of non-participant observations [41]. Observations covered day and night shifts and were spread over weekdays and weekends. Alongside the observations, informal conversations with the mothers present in the newborn units were a critical source of information in this study. These day-to-day conversations were unstructured and enabled follow-up on issues that were observed while in the NBU [42]. The non-participant observations focused on the context of care, tasks, and roles that the mothers played, and the nature of the relationships that existed within the ward. A total of 40 discharge interviews were undertaken with purposively selected mothers. Selection of the mothers was based on the condition of the baby at discharge and the willingness of the mother to participate in a formal interview. The demographic details of these interview participants are provided in Annex 1. Data were in the form of field notes which were kept daily and reviewed weekly by DO and TO. --- Data analysis A modified grounded theory approach [43] was used in the analysis of these data with DO, LH, and CJ meeting monthly to review emerging themes and plan for subsequent data collection. Additionally, in the analysis and interpretation of the data, we adopted a Critical Medical Anthropology lens, an approach to understanding how health inequities are shaped by social and economic structures and institutions that create, enforce and perpetuate observable disparities in health [44]. The approach emphasizes "the importance of political and economic forces, including the exercise of power, in shaping health, disease, illness experience, and health care" [40]. --- Results The results describe the contrasting economic and structural contexts of the two NBUs, variations in the participation by mothers in the care of their baby and the views of the mothers on their participation in their baby's care. --- The economic context of newborn care The two hospitals, located in different neighbourhoods in Nairobi, served very different populations. The faithbased hospital was in an affluent neighbourhood while the government funded hospital was located next to a slum. Most of the women seeking care from the faithbased hospital were in full-time employment and had attained post-secondary level schooling. Many of these women had access to medical health insurance which catered for hospital admission costs for themselves and their babies. At admission to the faith-based hospital NBU, a mandatory deposit of approximately USD 300 was required. In addition to this mandatory deposit, the daily cost of admission for a bed for the mother, and a cot and treatment for the baby ranged from a minimum of USD 35 to a maximum of USD 150. There was also the daily doctor's fee which ranged from USD 20 to 35 per day, depending on the doctor treating the baby. The average monthly minimum wage for Nairobi is approximately 188 USD . By contrast, many of the mothers in the government funded hospital were either unemployed or worked in the informal job sector and lacked any form of medical insurance cover. The care in the government funded hospital was supposed to be free, but most mothers incurred some costs for necessary medication and tests which were not available in the hospital. In the public hospital, mothers were expected to stay throughout their baby's admission. Mothers in the faith-based hospital had the option of early voluntary discharge and due to the daily bed costs, many mothers chose this option, leaving behind their babies under the care of the nurses. --- Nurses to baby ratios and technologies of care The essential biomedical technologies for neonatal care were available in both NBUs. Both had functioning incubators, radiant warmers, phototherapy machines, and other essential basic equipment. A recent study of the availability of equipment supporting care across NBUs in Nairobi found that public hospitals were better equipped than faith-based hospitals, but due to patient pressure, much of the equipment in government funded hospitals had to be shared [29]. This was mirrored in our observations with babies in the government funded hospital often sharing incubators and phototherapy machines, a practice never observed in the faith-based hospital. There were also stark differences in nurse to baby ratios. In the government funded hospital, we observed that often one nurse could be caring for 15-40 newborns during any given shift. This compares to our observation in the faith-based hospital of a nurse to newborn ratio of 1:1 for the Neonatal Intensive Care Unit and a maximum of 1:4 for stable babies. Despite the observations of inadequate nursing cover in the government funded hospitals, the mothers themselves did not appear to be particularly concerned about a lack of staff during the day. This might be because during the day there were often several professional staff in the ward including nutritionists, intern clinical officers, and nursing students who were present attending to the babies and the mothers. However, the mothers were very concerned about the lack of staff at night . As highlighted by one participant from the government funded hospital: "I feel that quality of care in the facility was good, especially during the day, because they are always there, however at night when there is only one nurse, it is just you and your God".. --- Diverse environments of care --- Chaos versus calm Entering the NBUs of each hospital was like stepping in to two different worlds. The entrance to the maternity ward and the NBU in the government funded hospital was crowded, noisy, and to an outside observer, appeared chaotic. In the corridors and public waiting areas, the sights and sounds of childbirth were clearly visible and audible. The NBU was located on the first floor, above the maternity ward, and women in labour were often found sitting or pacing around in the corridor and stairs leading up to the NBU. The air had a slightly fetid, unpleasant odour; a combination of blood, body odour, and the damp mops used in cleaning the floors. While access to the NBU was controlled and visitors were restricted, during the day it was always very busy with the mothers sitting by their baby's incubator or cot and nurses, clinical officers and nursing students on attachment carrying out their work. The postnatal ward where mothers were supposed to stay while their baby was an in-patient was on a different floor and some way from the NBU and often there were insufficient beds for all the mothers. Consequently, mothers often spent their time in the NBU on chairs next to their baby. The space was stuffy, hot, and filled with the beeping sounds of machines. The smell inside was not much better than outside the NBU and the floor was mopped only once a day. Outside the NBU there was a waiting area with two wooden benches where relatives would sit and wait for the mothers to come out of the ward, or on occasion, be let into the ward to visit a mother and baby. This bench was constantly full with relatives waiting to visit the mothers. By contrast, the single waiting area for maternity and the NBU in the faith-based hospital was calm, quiet, organised, clean and fresh smelling. Mothers in labour were kept inside the maternity ward and there were few signs or sounds of the pains of childbirth. The joint waiting area outside the maternity ward and NBU of this hospital was rarely occupied as the visitors and family members could sit in the private rooms where the mothers slept. These rooms were along the corridor opposite the NBU. Inside the NBU, there was an air of quiet organised calm; the floor was covered in white tiles; the care spaces were separated by glass partitions covered with pinkish light curtains that were decorated with a warm coloured cartoon print. The floors were mopped thrice a day and the air smelt fresh and clean. Mothers were not permitted to be continuously present in the NBU and during allotted visits to their baby they would usually sit in the breastfeeding room, tucked away from the ward. Here, there were no students on attachment or clinical officers, just the nurses caring for the babies. --- Participation by mothers in care To an observer, the roles that a mother undertook, and her involvement in the care of her baby, were very different in each NBU. In the government funded hospital, mothers who had experienced a trouble-free delivery and were able to walk up to the NBU, on their own or accompanied by a nurse, immediately became involved in providing care for their babies. As is described in Fig. 1; Table 1, within the first 24 h of their baby's admission, mothers, even those with babies that needed to be placed in a resuscitator, became involved in providing some level of care. The main task performed by the mothers during the first 24 h of admission in the government funded hospital was diaper changing, top tailing and, for those who were not on the resuscitator, feeding. A few mothers with full-term babies with jaundice or an infection were able to breastfeed their baby. Mothers of premature babies who were not on oxygen therapy expressed milk and fed their babies through a nasogastric tube once that had been inserted by a nurse. The following description of the NG tube feeding from our observations provides an illustration of how the mothers in the government funded hospital learnt how to provide care for their babies and from where they received support. --- NG tube feeding After the initial fixing of the NG tube by anurse, the mother would be given syringes for measuring the milk that she wasexpected to express and given brief instructions by a nurse or a nutritionintern on how to administer the milk through the tube. The following is an illustrative explanationgiven to one of the mothers by a nutritionist intern on how to feed her baby: Get a cup, express some milk then measure with this syringe. Connect this syringe to this tube and transfer the measured milk. When done remove the syringe and cover the pipe like this… During such explanations the mothers would rarely ask any questions. In some instances, the staff fixed the NG tubes but offered no explanation about how to feed the baby. In such cases the mother would ask a mother sitting next to them for help or they would turn to a nurse. In many instances the nurses would ask the mother to ask their neighbours, referring to other mothers, for help. To provide an example, Table 2 contains a description of a mother's experience with NG tube feeding. Over the course of the observation period, there were days when, during NG tube feeding, a baby would choke and vomit which caused panic with the concerned mother screaming for help from the nurses. Mothers had developed two main ways which they adopted to prevent this choking, one practical, and the other very much influenced by cultural beliefs. The practical trick some mothers used involved devising a way of controlling the flow of milk in the tube. They would pour some milk in the syringe and then pinch, fold and hold the pipe. Many mothers were seen to be doing this and upon asking they told DO that they did this to regulate the flow of milk. If they felt the milk was flowing fast, they pinched the pipe, slowly releasing from time to time until they had finished feeding. Many of the mothers would also at times tear off a small piece of paper from the feeding chart and place it on the baby's forehead. This they said they did to prevent the baby from getting hiccups which they believed could lead to reflux and potentially choking. Throughout the first 24 h of admission the nurses on the ward concentrated on caring for the very sick babies, giving medication while the mothers took up the bedside monitoring and care of their babies often with little support from the nurses. In the government funded hospital, in the days following the admission the mothers would settle into a routine. During the hours that followed the daily ward rounds by the medical team the mothers would sit by their baby's incubator observing the baby, delivering primary care, such as diaper changing and 3 hourly feeding as well as some more specialized tasks such as the NG tube feeding, fixing oxygen masks in place for babies on oxygen and eye masks for babies on phototherapy, and observations of their babies while receiving phototherapy or on a respirator. The ways that most mothers learnt how to undertake these tasks was either through instruction from other mothers or by observing how the other Mother takes on caring tasks within 24 h of admission. Mother takes on caring tasks > 24 h after admission. Mother NG tube feed baby. Nurse NG tube feeds or bottle-feeds baby. Mother top tails baby. Nurse bathes baby. Mothers present throughout the NBU and directly observe their babies. . Mothers are absent from the NBU and therefore no direct observation of their baby. Mothers breastfeed and some continue to NG tube feed. --- Stable phase Mothers breastfeed. --- Mothers top tail . Mothers bathe their babies. Mothers place baby on phototherapy after breastfeeding. Mothers place baby on phototherapy after breastfeeding. --- Mothers change diapers. Mothers change diapers. Mother practices continuous KMC and mothers continue to observe their own baby in the KMC room. Mother practices intermittent KMC. Table 2 Description of NG tube feeding of a premature baby, First 24hrs in Kijani -room A Among the three other mothers whom nurse Mwajuma had shown where to place their babies was Hellen. Hellen is a first time mother and she was told to place her baby in an incubator that already contained another premature baby. For the first one hour, she mostly just stares at what is happening around her. Next to her are other mothers including the mother of the baby that is in the incubator with her baby, all of whom are busy either expressing milk or feeding. Nurse Mwajuma is still attending to the baby in the radiant warmer and Hellen appears to have no idea what is expected of her or what she is to do. She keenly observes what the other mothers are doing and strikes up a conversation with the mother of the baby who shares the incubator with her baby. Her neighbour had been there for a couple of weeks. Under each incubator are drawers and her neighbour informs her that she can keep her babies diapers and feeding cups in the drawer. After checking, Hellen realises that her baby needs a diaper change and goes down to the postnatal ward to bring some diapers, she had brought with her essential baby items such as clothes and diapers which she had left in the postnatal ward. Upon her return, she turns to one of the mothers next to her and asks how it is done, she asks how to open the incubator and how to change the diaper. Having been there for some days now, her neighbour knows how to open and close the incubator and Hellen observes and learns from her neighbour. She is instructed by the mothers next to her on how to go about diaper change and how to close the incubator when done. Mwajuma passes by and tells her to express milk and feed the baby. "She doesn't have this pipe", she informs Mwajuma who then tells her and asks her to wait a bit. As nurse Mwajuma goes away, Hellen turns to the mother next to her and asks, "Where do I get the cup and how do I know how much I am to feed?" she asks her neighbour. "The nurse will fix for you this pipe and tell you how much to feed", she is told. She sits and waits for Mwajuma to get to her baby. When the nurse gets to her, she stands behind the nurse wanting to observe what the nurse was doing. However, immediately she notices the NG tube being inserted in her baby's nose, she moves back and looks the other way, with both hands on her cheek. She can't bring herself to observe the procedure. One of the mothers next to her smiles and asks her what was wrong to which she responds that she can't watch that pipe being fixed, she imagines that the baby is feeling pain. After fixing the NG tube, the nurse provides her with feeding syringes and tells her "do you have a cup, get a cup and express milk, then measure 5 mls and then feed the baby, I will get you the syringe, so you measure and then pour into this pipe". She is handed the syringes which she places on the drawer on her baby's incubator and proceeds to get a cup. She is advised by the mother she is sharing the incubator with to go and buy a cup downstairs by the gate. Hellen leaves the NBU and after a few minutes she walks back in with a cup, sits back on the chair and begins expressing This process of peer learning enhanced the friendships that developed amongst the mothers in the government funded hospital as they were together both in the wards and their sleeping areas, often even having to share a bed. By contrast, in the faith-based hospital, care of the very sick babies and premature babies during the initial hospital admission hours was undertaken by nurses. Mothers played very little role in the care of their babies, with any access and activities being closely monitored by a nurse. Mothers were rarely present during, and did not participate in, the key tasks such as feeding, changing, and bathing of such babies. --- "Because like I said, when I came the first the day, I didn't... I'm not the one who cleaned. She's the one who cleaned the baby. She did it herself ". These mothers only started becoming actively involved in the care of their babies, taking on tasks such as feeding and bathing, several days after their baby's admission and only after a baby had stabilised and been moved to the general ward. While the nurses acted as gatekeepers to the babies, the mothers did receive considerable support from them as the babies started to get better and the mothers became more involved in helping with care . The process started with one-onone learning sessions where mothers were progressively introduced to and taught how to perform tasks such as bathing by the nurses before eventually performing these tasks on their own. --- "The first time I washed the baby they asked me whether I'm comfortable washing the baby, if I have another baby, so washing the baby and I tried it, the nurse guided me, they were there and help me when I was unable to".. Since the mothers were rarely continuously present in the faith-based hospital NBU and as many opted for voluntary discharge, the opportunities for getting to know the other mothers and develop peer support groups were limited. Consequently, the mothers in the faith-based hospital tended to rely on the nurses for support rather than each other. A summary of the tasks that the mothers undertook in each hospital is provided in Figs. 1 and2; Table 1. --- Mothers' views on participation in care In both hospitals, many of the mothers faced challenges in taking up the caring roles they were required to undertake, especially within the first few days. These challenges were related to their fears about the size of the baby as well as their ability to perform these tasks. In both hospitals, a key task expected of the mothers was expressing milk to feed their baby. But most of the mothers were not familiar with expressing milk and many struggled, feeling that they had insufficient milk for their babies. From our observations and even in their interviews, many of the mothers reported the challenges they faced with this task. --- "I even did not….they would insist that I express, but I did not express because I did not feel like…. I knew I had no milk, so I wondered what they expected me to express yet I did not have, I didn't have!". Neither hospital had adequate breastfeeding support interventions in place to address these challenges. Only on few occasions would a member of staff take time to sit and talk with the mothers and show them how to express or reassure them that their milk would come. In the government funded hospital mothers who were unable to express milk were sometimes scolded by the nurses or nutritionist without being given any suggestions as to how to do better. Support for these mothers was often provided by the other mothers in the NBU who offered advice to try and avoid stress or showed how to transfer the expressed milk into the NG feeding tube. Expressing and feeding were particularly challenging for mothers who had delivered through C-section and those who had twins. The postnatal ward in the government funded hospital was on a different floor and some distance from the NBU which either meant walking long distances every 2 to 3 h for feeding or just spending most time sitting on a chair in the NBU with little opportunity for appropriate rest: "It was stressful, because my legs were still swollen and my stomach was still painful, when you are still breast feeding this one you are called to breast feed the other one, and when you walk you get tired, and at the same time you dont have milk, you just feel overwhlemed. You lack sleep, most of the time you are just sat on the chair. It was stressful".. In the faith-based hospital mothers who were struggling with expressing milk had an alternative feeding option as they had access to the funds to purchase formula feeds and mixed feeding was allowed and encouraged. Many of the mothers in the faith-based hospital were discharged home while their baby was still in the NBU and even though they were able to express and store their milk in a fridge either in the hospital or at home, formula feeding was often logistically easier. However, some of the doctors in the faithbased hospital were keen to ensure that the mothers did breastfeed. For some of the faith-based hospital mothers, there were elements of struggle with transitioning from bottled feeding to breastfeeding when the nurses had told them to start on formula milk but the doctors then instructed that the baby should be receiving breastmilk: --- Breastfeeding? Breastfeeding, okay, first of all uhh, I think first of all my milk did not come immediately but it didn't take long. So my baby I think was not getting enough milk and I remember first of all what they do-[the nurse], they tell you to buy the 'Nan' [formula milk] just to introduce the baby so that so that the baby doesn't stay hungry. Yeah, so you find that the baby becomes comfortable because with the bottle is faster and then now you want to change to this one that he has to use so much energy pulling. So I went and innocently told the doctor that that uhh the baby drinks formula milk but he doesn't breastfeed. "No mun. This baby has to be, has to become very hungry. No formula milk, just breastfeed" the doctor responded. I think that is why I stayed for Fig. 2 After 24 h long hours because he was waking up every other time. So, yes, we tried breastfeeding the whole night. And I remember when I was cold [breast feeding room was cold at night] and I was thinking oh God, let them not call me because I can't handle the baby. Let them not… I am just thinking I pray they don't call me, they don't call because I can't carry the baby.. Despite the struggles, the mothers across the two hospitals appreciated their involvement in care as it enabled them to bond with their babies. Mothers reported that getting involved in the daily care and being able to hold their babies and breastfeed enabled them to feel attached to their newborns: I have been involved in things like breastfeeding, at least you feel good when breastfeeding you feel good when you wash her, when you change diapers at least you start to bond. You hold her she keeps quiet, when she cries you hold her and she keeps quiet. I think this is the time now you become a mother. . However, in contrast to the mothers in the government funded hospital, many of the mothers in the faith-based hospital reported feeling powerless in the wards. As one of them said to DO in an informal conversation during the observations: 'I tend to come in only during visiting hours just to bring milk because even if I stay here, there is nothing much I can do anyway'. In the discharge interviews with mothers, we asked about how confident they felt about their ability to care for their babies post-discharge. Unlike the mothers in the government funded hospital, the mothers discharged from the faith-based hospital expressed their concerns about coping with the small baby at home without the support of the health care staff. This was less commonly reported by mothers from the government funded hospital who were present throughout the baby's admission and much involved in the bedside care of their babies. --- Discussion In this paper, we have described the contrasting roles that mothers play in caring for their sick newborns in two NBUs in Nairobi, Kenya. Geographically close, the two NBUs are embedded in strikingly different structural, economic and social environments and the roles that the mothers play in the care of their babies are directly shaped by these different environments. In general, the government funded hospital catered to women who were economically disadvantaged and this disadvantage was echoed in the much inferior staffing ratios and the neglected and crowded environment. The understaffing in the government funded hospital contributed to the unplanned and informal high level of involvement of mothers in their children's care, providing both primary care as well as some more specialised neonatal care such as NG tube feeding. It also impeded the level of support that nurses were able to offer mothers. Understaffing within the public sector has been documented as a bottleneck to service provision in LMICs [32]. A recent study by McKnight and colleagues [36], undertaken in Nairobi, found that nurses working within the government funded hospital setting are often overstretched, having to provide care in very challenging conditions. Working in overstretched and constrained environments can lead to the development of collective coping strategies and prioritization of care, emphasizing clinical care to the detriment of non-clinical aspects of care [36]. These findings echo those found in the studies in Iran where in understaffed NBUs the nurses delegated primary care to parents [33]. Having a child admitted to a hospital newborn unit is a stressful event for a parent, often mixed with fear, grief, and uncertainty [9,45,46]. Admission to a newborn unit disrupts the maternal-infant bonding and attachment process, an important factor in infant development and maternal wellbeing [47] as well a potential precursor to the consolidation of parenting skills and confidence, and future emotional, developmental and social milestones [7]. A recently published review of the literature on maternal-infant bonding reported that mothers who participated in immediate skin-to-skin contact and initiated breastfeeding within two hours following childbirth were more sensitive to the infant's needs and the child seemed more content at one year; while poor interaction affects the child's cognitive and socio-emotional development, physical health and personal relationships [7]. Paradoxically, given the widespread benefits to mothers and babies of maternal involvement in newborn care, it was the mothers in the government funded hospital who participated far more in the routine daily primary care of their babies. However, as was found in the Iranian study [48], this participation was frequently without adequate training and often with minimal support and supervision from health care staff in the NBU. Thus, although mothers in the government funded hospital gained experiential and lay knowledge that enabled them to develop expertise and confidence in the care of their babies [49], they had little choice in when and how they participated in care and the skills they gained were not always most appropriate. The lack of professional support and supervision by the health care staff had the potential for serious mismanagement of care with dangerous and possibly life-threatening consequences for the baby. Staffing was not a challenge in the faith-based hospital and many mothers, for cost saving purposes, felt obliged to discharge themselves before their baby was ready to be discharged, leaving their babies under the care of nurses. This separation and limited initial participation of mothers in the care of babies may have contributed to stress and affected their well-being [11] as well as having potential consequences for their confidence in their ability to care for their baby post-discharge [40,50]. In both hospitals the focus of the health care staff was on the immediate clinical needs of the babies with attention rarely given to the physical or psychological needs of the mother. This neglect of the mothers was visible both in the day-to-day interactions of the health care staff with the mothers and in the structural systems level in the physical separation of the mother and baby spaces. The day-to-day interactions between the staff and the mothers focused on the needs of the baby with little or no support for the mothers in their struggles to express milk or breast feed, and a lack of attention to their social and emotional well-being. At the structural systems level in the faith-based hospital no financial support was provided to help the mothers stay in the hospital near to their baby and in the government funded hospital the accommodation for mothers was on a different floor and distant from the NBU, constraining their ability to both get adequate rest and participate effectively in the care of their babies. Ward lay out and consideration of the mother-infant dyad are significant concerns influencing the ability of mothers to effectively participate in the care of their hospitalised sick newborns [51,52]. In the recent iKMC trial the participating hospitals modified their existing neonatal intensive care units , or purpose-built new units, to create Mother-NICUs where each mother-baby dyad was provided with a bed to allow for the practice of iKMC. In addition, the provision of care within these units was provided by teams of obstetricians and neonatologist, addressing the needs of both mother and baby [23]. This strategy not only facilitates the implementation of iKMC but also provides the structure through which a more holistic approach to the delivery of neonatal care could be implemented. Taking into account the needs of the mother is the first step towards family centred care and the concepts of parental involvement and team work [41,[44][45][46]. LMIC settings, including Kenya struggle with numerous health system challenges such as understaffing and inadequate infrastructure which impact the implementation and uptake of interventions such as FCC. However, we need to begin to critically think about contextually and culturally appropriate interventions that can build on the opportunities that exist to improve care within the NBU in line with the WHO standards in such highly resource constrained settings. We argue that health providers and mothers need to work together and develop a strategic focus on the clinical, emotional and social needs of mother-baby dyad. The iKMC approach clearly has potential but requires significant system changes and resourcing. As a first step, the skills of nurses to better identify and support the needs of mothers and their families need to be developed and implemented. In particular, mother infant bonding within the first 24 h should be encouraged across all hospitals and strategies to better equip mothers and supervise their participation in care need to be established. Central to any approach needs to be consideration of strategies to address the chronic understaffing seen in many government funded hospitals in LMICs. --- Strengths and limitations The key strength of this study was in the ability to triangulate findings across two different hospital contexts across 2 sectors. Non-participant observations provided a more nuanced understanding of the context of newborn care in an inpatient setting. This study also begins to shed some light on how hospital norms and contextual factors shape the support mothers receive and influence their participation in care which has implications for FCC. However, this study only focussed on understanding mothers' experiences and did not document the voices of the nurses. The data reported in this manuscript was collected between 2017 and 2018. However, ongoing work within public hospitals in Kenya suggests that staffing and the structural contexts of care have not changed considerably with the most government funded facilities still experiencing the reported bottlenecks. --- Conclusion This qualitative study of care provision in the neonatal units of a government funded and a faith-based hospital in Kenya points to stark differences between the care environment available for women of different socio-economic classes. Our findings suggest understaffing, and less formal support for mothers in the government funded sector, which necessitated the development of stronger peer support bonds between mothers than was observed in the faith-based hospital. While higher levels of involvement and participation in bedside care can be beneficial for mothers, there is a need for approaches to improving care within neonatal units that take account of the perspectives of mothers and staff, as well as advocacy for structural changes to support respectful care and reduce inequities in neonatal care. Focusing on technological improvements risks ignoring structural inequities and further embedding a biomedical paradigm that ignores the relationships and needs of mothers, families, and staff. Advances in medical care are necessary but not enough to address the disrespect and persistent structural inequities affecting access to and quality of care. A first step is recognising the importance of the mother/infant dyad in the NBU; understanding who is providing 'care' in the NBU and drawing on the experiences of mothers, families, and staff to develop interventions that address their needs. This involves treating mothers with respect through a recognition of their role as key partners in care provision; leveraging their agency and developing their cultural competence in providing care for their sick newborns. • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year --- • At BMC, research is always in progress. --- Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: --- --- Abbreviations --- --- Additional file 1. Authors' contributions DO, CJ, ME, GI and LH conceived and designed the study. DO, LH, CJ and TO analysed the data and drafted the manuscript. All the authors reviewed the manuscript. The author read and approved the final manuscript. --- --- --- Competing interests The authors declare no competing interests. ---

Introduction There is growing evidence that parental participation in the care of small and sick newborns benefits both babies and parents. While studies have investigated the roles that mothers play in newborn units in high income contexts (HIC), there is little exploration of how contextual factors interplay to influence the ways in which mothers participate in the care of their small and sick newborn babies in very resource constrained settings such as those found in many countries in sub-Saharan Africa. Methods Ethnographic methods (observations, informal conversations and formal interviews) were used to collect data during 627 h of fieldwork between March 2017 and August 2018 in the neonatal units of one government and one faith-based hospital in Kenya. Data were analysed using a modified grounded theory approach. Results There were marked differences between the hospitals in the participation by mothers in the care of their sick newborn babies. The timing and types of caring task that the mothers undertook were shaped by the structural, economic and social context of the hospitals. In the resource constrained government funded hospital, the immediate informal and unplanned delegation of care to mothers was routine. In the faith-based hospital mothers were initially separated from their babies and introduced to bathing and diaper change tasks slowly under the close supervision of nurses. In both hospitals appropriate breast-feeding support was lacking, and the needs of the mothers were largely ignored.In highly resource constrained hospitals with low nurse to baby ratios, mothers are required to provide primary and some specialised care to their sick newborns with little information or support on how undertake the necessary tasks. In better resourced hospital settings, most caring tasks are initially performed by nurses leaving mothers feeling powerless and worried about their capacity to care for their babies after discharge. Interventions need to focus on how to better equip hospitals and nurses to support mothers in caring for their sick newborns, promoting family centred care.

Introduction Although there is controversy over its definition, empathy refers to the variety of reactions of one person to another's observed experiences, and it has four components: perspective taking, empathic concern, personal distress, and fantasy [1]. This study uses the multidimensional approach to empathy and incorporates the union of the cognitive and affective dimensions. Affective empathy is the ability to experience users' emotions and perspectives, while cognitive empathy is the ability to identify and understand, from one's perspective, the emotions of others. The cognitive dimension comprises subconstructs, such as perspective taking and fantasy, while the emotional dimension contains empathic concern and personal distress as sub-components. Empathy has been highly studied, and it has been associated with spirituality [2] and prosocial behaviour, a voluntary behaviour intended to benefit others. A study identified that empathy and moral identity were the potential mechanisms underlying the association between intelligence and prosocial behavior [3]. Research also suggests empathy can be nourished and developed [4]. There is a growing body of work focusing on empathy as a multidimensional construct in the health professions. Recent research [5] conducted with Chinese medical students showed that women displayed higher empathy levels than their male peers. This same study noted that empathy declined as students progressed in their studies. Schoenfeld-Tacher, Shaw, Meyer-Parsons, and Kogan, [6] used the Interpersonal Reactivity Index in a sample of veterinarians. Their findings revealed that with increasing age and therefore, expertise, fantasy and personal distress decreased. These results are consistent with research carried out by Bratek, Bulska, Bonk, Seweryn, and Krysta [7] on medical students and physicians, which found a negative correlation between age and fantasy as well as personal distress. However, in this same study, a positive correlation was found between age and perspective taking. Moreover, empathy plays a vital role for people in health related professions [8]. For instance, Hunt [9] found decreased compassion fatigue in cancer healthcare professionals with higher levels of empathy while Mc Farland [10] noted greater resilience in an oncology setting associated with empathy. Besides, Shanafelt [11] found increased personal wellbeing in medical residents with higher levels of empathy. Empirical research shows that it is essential to work effectively with others in suffering, contribute to the development of a strong therapeutic relationship, and correlate with better clinical outcomes [4,12]. Furthermore, developing empathy is necessary to become professionally competent and have higher professional fulfilment [13]. Given what we already know about empathy and the consideration that empathic communication and response during a healthcare crisis could lead to better support [14], this report aims to explore it in students and professionals in health related occupations in Ecuador during the COVID-19 pandemic. This research is of particular interest, because the global health crisis outbreak could and likely did generate remarkable emotional changes in healthcare providers who usually do not have specific training in empathy or dealing with their own emotions. In addition, this study is relevant because of the abrupt changes to the way people relate to each other and perform their tasks brought about by the significant measures imposed in Ecuador to protect people's physical health, which included a national lockdown, curfews, and the cancellation of all non-essential in-person activities for over five months. --- Method --- --- Instruments Data collection took place through an online questionnaire that included sociodemographic variables, such as age, gender, marital status, and professional/occupation speciality. In the case of professionals, they were asked about the number of years of professional experience, the number of patients they attended weekly, hours of work per day, percentage of cases of high emotional burden, and self-care practices. The Interpersonal Reactivity Index questionnaire [1], in its Spanish translation [15], was also administered to measure the total empathy levels of the participants, as well as its four sub-dimensions . Fantasy is the tendency to identify with characters in movies, novels, plays, or fictional situations. In the questionnaire, some of the items that measure this subdimension are: "I daydream and fantasize, with some regularity, about things that might happen to me" or "I really get involved with the feelings of the characters in a novel." Empathic concern involves feelings of compassion, concern, and affection for the discomfort of others . Perspective taking is the tendency to adopt the psychological point of view of others spontaneously. Example statements of perspective taking from the IRI include "I try to look at everybody's side of a disagreement before I make a decision" and "I sometimes try to understand my friends better by imagining how things look from their perspective." Finally, personal distress is the emotional reactions, discomfort, and anxiety to others' negative experiences. Items such as "being in a tense emotional situation scares me" or "when I see someone who badly needs help in an emergency, I go to pieces" represent personal distress in the scale. The IRI has been widely used in the health field's scientific area since its creation in 1983 [1]. It is currently one of the most frequently used questionnaires to measure empathy levels in health students and professionals. This instrument has shown satisfactory psychometric properties [16]. The Cronbach's Alphas in this study were between 0.70 and 0.78 in each of the dimensions. --- Procedure Health centers and universities were contacted to participate in the study. After approval by the Ethics Committee of the Universitat de les Illes Balears and Universidad de Las Américas , an email explaining the study including a link to the questionnaire, developed through Google forms, was sent to the health centres and their members, as well as published on their websites and social networks . Data collection took place from April to July 2020. During this period, Quito was in lockdown due to COVID-19. --- Analyses All statistical analyses were carried out using SPSS version 20. Reliability and normality tests were performed, followed by descriptive and frequency analysis of the sample. Pearson's correlations, student t-tests, and ANOVAs were conducted for univariate analysis. A linear regression model was also performed to consider the effect of multiple variables on empathy levels. --- Results The empathy scale had a Cronbach's Alpha of 0.762; its subscales also showed right internal consistency . The Kolmogorov-Smirnov test was greater than 0.05; thus, all the performed tests were parametric. The main sociodemographic characteristics of the sample are shown in Table 1. There were no significant differences in gender between the groups; there was a greater female to male ratio in both cases. However, professionals were significantly older = 244.65, p < 0.01). The average age of the full sample was 30.80 years , that of the professionals was 40.65 years , while that of the students was 23.99 . Professionals were also more likely to be married = 98.31, p < 0.001) than students and had significantly more representation in the physical health field versus the emotional health field in the students = 56.98, p < 0.001). Lastly, more than 70% of the students were about to finish their careers and had already carried out internships in work centres. In the case of professionals, they had an average of 8.07 years of experience at their workplaces and attended an average of 74.97 patients weekly . They also indicated that 44% of the number of weekly cases were of high emotional distress. Table 2 shows the average levels of empathy, its dimensions for professionals and students, and the combined sample. The t-test results between professionals and students show statistically significant differences between professionals and students in all the categories. Students show higher levels of empathy, fantasy, empathic concern and personal distress, while professionals have higher levels of perspective taking. When considering the sample as a whole , effects of gender were noted on the total level of empathy, fantasy, empathic concern, and personal distress, in which women had higher scores than men in all the analyses. The marital status differences were also noted in total empathy, with singles having the highest scores over the other groups. However, post hoc analyses did not reveal significant differences between the groups. Singles also had significantly higher levels of personal distress compared to divorced participants after post hoc comparisons. Empathic concern, perspective taking, and fantasy levels were also significantly different between the groups. Post hoc analyses revealed that singles had higher levels of empathic concern than common law married participants, lower levels of perspective taking than married people, and higher fantasy levels than divorced participants. Age was significant and negatively correlated with all empathy variables = -0.203, p = 0.001; fantasy r = -0.300, p < 0.001; empathic concern r = -0.205, p = 0.001; perspective taking r = -0.241, p < 0.001; personal distress r = -0.214, p < 0.001). Differences were also noted according to health field, but only for perspective taking and empathic concern. Post hoc analyses revealed participants in the emotional health field had higher empathic concern levels than all other groups. Still, no post hoc significant differences were noted on perspective taking. According to sociodemographic and work variables within each subsample, different patterns emerged when analyzing differences in empathy; age was associated with students' empathy levels but not professionals. Besides, gender was associated with most empathy variables in students, but only one in professionals. Differences in health related fields were significant only in one variable in the professional's group and none in students. The Pearson's correlation results show that age was significantly and negatively associated with total empathy levels in students. As the age of the students increased, the total levels of empathy decreased = -0.245, p = 0.001). Similarly, the dimensions of fantasy = -0.295, p < 0.001), empathic concern = -0.175, p = 0.023), and personal distress = -0.167, p = 0.030) also decreased as age advanced in students. In addition, empathy levels were not associated with the number of patients seen weekly in professionals. As seen in Table 4, when comparing total empathy by gender, only perspective taking was significantly different in the sample of professionals. On the other hand, in the student sample, women had higher mean levels of empathy, perspective taking, empathic concern, and fantasy than men. Effects of the health field, whether professions were related to physical, mental, or rehabilitation health, on total empathy were noted only in the professionals' sample. Those working in the physical health field had significantly higher personal distress values than those working in emotional health. Lastly, a linear regression model was performed to see if occupation , age, cohabitation status , health field, or gender variables could predict empathy and its dimensions. For each model, we first included all the above-mentioned variables and then removed the non-significant ones. For total empathy, we removed health field , occupation , and cohabitation status . The model with age and gender was significant = 14.456, p < 0.000, R 2 adjusted = 0.087). For fantasy, the non-significant variables were health field , occupation , and cohabitation status . Again, the model with age and gender was significant = 16.831, p < 0.000, R 2 adjusted = 0.101). In the case of empathic concern, the nonsignificant variables were age and cohabitation status . The model with the remaining variables, occupation , gender , and health field was significant = 13.593, p < 0.000, R 2 adjusted = 0.118). Given that occupation was a significant factor in this model, we performed the regression for the students and professionals' subsamples separately. Only the model for students was significant. Gender was the only variable with an effect = 17.471, p < 0.000, R 2 adjusted = 0.089). For perspective taking, the non-significant variables removed from the model were age and health field . The remaining variables, gender , occupation , and cohabitation status , were significant for the model = 15.249, p < 0.000, R 2 adjusted = 0.131). As with empathic concern, the model was run separately for students and professionals, but they were not significant. Lastly, in the regression model for personal distress, the removed variables were health field , age , gender , and cohabitation status . Only occupation remained in the model = 17.825, p < 0.000, R 2 adjusted = 0.056). --- Discussion The impact of the COVID-19 health and social crisis in Ecuador, during which the present data was collected, seems to be reflected in the professionals' responses, who reported that almost half of the weekly cases they attended presented a high emotional load. Despite this, the high levels of empathy obtained were in line with those collected from other contexts and methods [17]. The fact that the professionals who attend to physical health care needs have presented the highest level of personal anguish is consistent with other research [18] and is understandable since they are the ones in direct contact with death overflowing in the health system. In this study, the students show greater distress and concern than experienced professionals. The instability related to the current global socio-health panorama also disrupted university educational environments. Students in health related fields have had to substantially change their study routines and access to learning due to the imposed confinement measures. Despite all the changes, a high level of total empathy was observed in students. This could be explained by having access to information about the situation of those affected by the disease and being privy to the difficulties of the social health professionals who were taking care of them and by whom they were being trained [19]. On the other hand, the high levels of personal distress could be explained by their efforts to adapt to the new reality, both academically and in their future profession. Likewise, many of them have carried out their clinical practices. As Taylor, Thomas-Gregory, and Hofmeyer [20] point out, they may have experienced a deficit in professional supervision, a lack of personal protective equipment, and fear of personal contagion or exposure to it for their families. In this study, professionals have greater perspective taking, probably due to professional experience. One of the most relevant results of this research is the relationship between perspective taking and age. This confirms that the greater the professional experience in the health field, the greater the person's ability to understand others while they taking a healthy distance. These data are consistent with the findings of Bratek, et al. [7]. Interestingly Richards, Petty, and Zelenski [21] found results contrary to the present study. Their sample of new professionals and students in genetic counselling found that first year students had greater perspective taking than sophomores or graduates. However, this study does not take into account professionals with more years of professional practice. If more veteran professionals had been included, perhaps, results more in line with those proposed in the present study could be found. As Harrison and Westwood [22] point out, emotional distancing allows more effective help to patients. Health professionals mark the necessary limits for the correct development of their professional empathy. It should be noted that too much distance can lead to inattention to the patient and dehumanisation of care. On the other hand, excessive bonding with the patient can lead to resonances and an overexpansion that makes it difficult to take perspective. The regulation of said emotional distance can be a preventive mechanism against burnout and secondary traumatic stress derived from professional practice. Furthermore, the present research shows that age and gender have an impact on total empathy levels. As in previous studies [5,23,24], female students have higher empathy levels than male students. In contrast to what was presented in Extremera Pacheco and Fernández Berrocal's research [23], the students decreased their level of empathy as their age increased in the present study. This data should be the object of further research since it may be derived from the exceptional situation in which the participants responded since the previous literature are not in line with these results. COVID-19 has raised many moral dilemmas [18]. However, the high levels of empathy obtained in this study likely reflect the determination of healthcare providers fighting the pandemic and accompanying those who suffer, despite technical and material difficulties or shortages. Therefore, attention to empathy, as well as personal and organisational self-care should not be forgotten [20] since greater emotional management is positively related to greater empathic concern, better perspective for students and professionals, and less personal distress for professionals [23]. This research is not without limitations. We do not have at our disposal measurements before the onset of COVID-19, and as a result, the pandemic's impact on the levels of empathy of professionals and students cannot be ascertained. Having previous measurements would have made it possible to determine the fluctuations in levels concerning the stress and tension experienced in the social health area. Future studies should measure the changes in empathy as the pandemic advances. Additionally, future research should include a bigger sample size. Although, considering the extraordinary limitations and difficulties of confinement, this study has representative sample results and opens the door to future studies on the subject. Empathy is a double-edged sword. It is essential to establish an adequate helping relationship, but at the same time, it may induce an emotional overinvolvement with those who suffer. This lack of empathy balance can lead to negative professional practice consequences, such as burnout, depression, or post-traumatic stress. Empathy scores can serve as traffic lights to warn of occupational hazards. The high levels of empathy obtained in this study should alert healthcare managers to increase their staff's care to avoid the abovementioned consequences. The COVID-19 pandemic has shown the importance that health professionals have in fostering community wellbeing. Understandably, the lion's share of the responsibility has fallen onto them to provide ongoing support to address this global emergency. The number of stressors to which they have been subjected may have increased their burnout levels, secondary traumatic stress, loss of quality in professional life, and satisfaction with life [25]. Studies such as this one help to support professionals in their practice by allowing us to explore how the current global crisis has impacted them. Based on these results, responses aimed at improving their wellbeing can be proposed. This study is a first step that focuses on empathy as a skill, in which the optimal relationship between professionals and patients is sustained. Ultimately, empathy is an essential skill for health professionals as it promotes clinical competence, patients and professionals' wellbeing, and trust and emotional intelligence [26]. Work from a humanistic model, in which both the professional and the patient are essential, contribute to building a more efficient health model in these critical moments. --- Data Availability Statement: The data presented in this study are available on request from the corresponding author. The data are not publicly available due to ongoing longitudinal analysis. --- Declaration of Helsinki, and approved by the Ethics Committee of Universitat de les Illes Balears and Universidad de Las Américas . Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. ---

Empathy plays a fundamental role in health related occupations. In this study, we analysed empathy levels in professionals (117) and students (170) from various healthcare fields in Ecuador during the COVID-19 pandemic. The Interpersonal Reactivity Index was used in an online survey. The results show high levels of empathy in both groups, influenced by age and gender. The students presented higher levels of personal distress, and their age was negatively correlated to empathy. Additionally, professionals working in physical health scored higher levels of personal distress compared to those in the field of emotional health. COVID-19 has placed social health systems in great stress. Despite this, the personal capacities for empathy of both students and health professionals have not been diminished.

Background How a community views, understands and responds to disability is important to the everyday life of people who have disabilities. Cognitive maturity [1][2][3], cultural beliefs and practices [4], information availability [5,6] and exposure to people who have disabilities [7] are influential factors [8] . Childhood representations of disability tend to be negative and typically associated with an undesirable condition [9], resistant to parental influence and closely aligned to an individual or medical model, with attention given particularly to physical and biological factors [9,10]. Federici et al. proposed that this was evidence of a 'cognitive mechanism' underlying the conceptualisation of disability [2]. As cognitive maturation occurs, there is engagement with alternative perspectives on disability [1][2][3]. However, it was observed that parents resort to early mental representations in response to open-ended questions, as opposed to closed question responses that were aligned to the social model [1] . Cultural learning is also important. Traditional beliefs have tended to associate disability with negative images and experiences [4], although some variations have been reported in East Africa, e.g., people with physical disabilities were perceived as pacifiers of evil spirits [4], while autistic spectrum conditions were attributed to evil spirits, witchcraft and curses [11]. Some of the distaste associated with disability in sub-Saharan Africa may lie in the cultural explanations that imply: breach of social conventions, e.g. in Botswana [12], Ghana [13], Kenya [14][15][16], Tanzania [17], Malawi [18]; external, preternatural forces, e.g. in Kenya [14][15][16], Namibia [17],; and the will of God [19][20][21][22]. Negative and sometimes fearful images associated with disability, not only hinder acceptance of the person who has a disability, but also undermine possibilities and opportunities for the person [23]. Corrigan [24] and Corrigan & Watson [25] asserted the cognitive, affective and behavioural components of stigma, whereby people who share a particular characteristic that marks them out as different, e.g. an inability to walk, are cognitively appraised by others, leading to a possible negative emotional response and discrimination [26]. The question of how such stigma spreads across a community remains. Dipple et al. considered the relevance of 'stigmergy' , a term originating from the French language, to the human world [27]. Thus stigmergy provides an explanation for the tendencies that exist within a community in their responses to persons with disabilities and their close family members. At its most extreme point, community responses to people with disabilities may descend into abuse, neglect and exploitation. For example, discriminatory practices reported in Nigeria [28] and Uganda [29] included physical, sexual and emotional violence, human trafficking, ritual killing and alms begging. As a counter point, purposeful encounters with people who have disabilities appear to be critical to the formation of positive attitudes. Allport's early 'contact hypothesis' proposed that increased interactions by one social group with another would promote more positive attitudes and decrease prejudice [30]. This has been borne out in studies of children's attitudes towards disability [31,32]. A systematic review of 35 studies revealed that children's attitudes were positively associated with the social contact experienced with people who have disabilities [33]. This was corroborated in a more recent cross-sectional survey of over 1800 children, in which empathy for and underlying anxiety about interacting with people with disabilities were reported mediators of the contact-attitude link [34]. Disability awareness training often involves interacting with people who have disabilities, combined with activities that have a more cognitive focus, e.g. engaging with new information and participating in discussions that challenge stereotypes. By increasing exposure to the people who are the very objects of devaluation in a positive and informed way, previous evaluations may be challenged [35][36][37]. Krahè & Altwasser [38] found that combining cognitive and behavioural activities, , improved attitudes toward persons with physical disabilities. In direct contrast, no change was associated with the cognitive condition only and the control condition. Similarly, Papaioannou et al., used disability simulation in sports activities, some lectures and video presentations, to bring about positive change in attitudes [37]. A systematic review of interventions targeted at schools in Canada revealed shifts in attitudes and behaviours of the children associated with five basic approaches to disability awareness: social contact; simulation; curriculum; multi-media curriculum; and multiple components [39]. Documentary evidence of living with disability and discursive interactions were found to improve disability awareness in teenage boys in Australia [37]. In an epilepsy education programme conducted on the Kenyan coast to reduce the treatment gap, traditional songs with positive depictions of epilepsy were used to change community perceptions and to promote new understanding [40]. Disability awareness training features in the 'empowerment' component of the Community-Based Rehabilitation /Inclusive Development guidelines, which cuts across the other four components of health, education, livelihood and social [41]. However, there is a dearth of evidence for CBR initiatives generally, with published studies being criticised for a lack of research rigour, and for being overly descriptive or theory-based [42]. In their recent Systematic Review, Lemmi et al. [43] focused on assessed impact. Modest benefits were suggested for people with mental disabilities and their caregivers, although the authors acknowledge 'methodological constraints' in the studies. More recently, consultations on how best to address the evaluation of CBR have favoured a flexible and iterative approach [44]. Based on the evidence that personal encounters with people living with disability have positive influence, and in harmony with the recommendations of the CBR guidelines, a project was set up to engage neighbourhood communities in a process of reflection and education. The first aspect -reflection, focused on community beliefs about disability causation and the attendant challenges [14]. The current paper reports on the second aspecteducation. The aim was to develop and evaluate the effects of a disability awareness intervention on the values, perceptions and understanding of community groups in a rural part of Kenya. The research question was: How do personal narratives of 'experts by experience' , i.e. persons with disabilities, affect the perceptions, values and understanding of community groups? --- Methods With reference to Creswell et al.'s description, a phenomenological approach was adopted [45]. Thematic analysis of focus group discussions data was conducted before and after a disability awareness intervention. The setting was Kilifi County, one of the poorest areas in Kenya, with a poverty level of 71% [46]. Situated on the Indian Ocean coast it covers a large rural area where subsistence farming dominates. Characteristics of the region include low levels of nutrition, inadequate control of infectious diseases, poor enrolment in schools, limited literacy amongst adults generally and unreliable rainfall. Typical dwellings are of mud construction with a thatched or iron sheet roof, no power supply or running water. Per capita, the average income for a household is Ksh1, 000 per monthless than $13 USD [46]. Languages spoken include Mijikenda and Swahili. Christianity is observed by about 70% of the people, traditional religious practices by 20% and Islam by about 10%. Based on a county-wide population of 1,109,735 and using a 15% prevalence of disability [46], it is estimated there were 166,460 people with disabilities in Kilifi County. --- Sample The study was conducted across five constituencies of Kilifi County, Kenya. They included Magarini, Malindi, Kilifi North, Kilifi South and Ganze. The focus was on two types of established community groups, namely community health worker groups , who are volunteers brought together to support the work of health dispensaries and health centres [47], and women's groups , who are registered with the Kenya Women Finance Trust [48]. The two types of community groups were targeted for their convenience, being readily accessed, formally constituted and meeting at regular intervals, for their ongoing community engagement, and for their stable membership. The latter point was considered critical to the facilitation of focus group discussions. Inclusion criteria used in the study required that the group should: Be formally constituted as a CHW or WG; Be active, having regular meetings, at least once a month; Have a consistent membership as evidenced by attendance of meetings; Have contact with people with disabilities. Recruitment of the groups was facilitated by Pambazuko Disability Initiative, a community-based organization involved in providing support for children with disabilities in Takaungu, a sub-location in Kilifi County. Building on their existing links with WGs and CHWs, they identified 20 groups who met the criteria in ten sub-locations. Each community group had around 15 members making an estimated total of 300 participants. The representative from Pambazuko approached each community group and arranged for the Project Researcher to attend a meeting for purposes of providing information about the project and soliciting their consent to participate. Each member gave their consent separately, recorded by signature, personal mark of thumb print on a consent form designed for the purpose. Fig. 1 shows the sub-locations of the groups. --- The intervention The intervention aimed to promote disability awareness in the community. Based on Allport's contact hypothesis [31], the rationale for the intervention was concerned with exposing members of the community groups to people living with disabilities. Disabled Persons and Parent Organisations are not common in this rural part of Kenya and therefore it was decided to recruit adults with disabilities via local networks A retired special educational needs teacher, familiar with local organisations for people with disabilities, and conversant in the local languages, as well as Kenyan Sign Language, recruited five adults living with disabilities, termed 'experts-by-experience' , to the group. Individuals were identified via special schools and units and invited to take part in the development and delivery of a disability awareness training initiative. The group was comprised of: V, male, aged 45 years with cerebral palsy; W, female, aged 36 years with hearing impairment; X, male, aged 20 years with intellectual disability and traits of autism spectrum condition ; Y, female, aged 20 years with hearing impairment; Z, male, aged 25 years with intellectual disability The group met for three hours every day over a sixweek period to develop their personal narratives of living with a disability for sharing with the community groups. The first author facilitated the training in partnership with the special needs education teacher. JKG organised for the training facility, and made sure the expert group reached the training venue on time. The special needs teacher rehearsed with the group and did sign language translation. 1. Each member of the expert group was first given an opportunity to narrate his or her life story. 2. Then JKG and the special needs teacher assisted in putting their stories in organised sequences. 3. Each member was given a topic to talk about to minimise repetition: V was to talk about social life experiences, W about challenges of doing business, being independent, X to talk about challenges of being in school and living in a homestead. A homestead in this context means a structured cluster of houses belonging to relatives of the same grandparents. After the six weeks, two members of the expert group, Y and Z, opted to pull out of the study. Y did not have basic knowledge of Kenyan Sign Language and Z relocated to another place with his parents. The special needs teacher interpreted the sign language to members of the community groups who participated in this study. The experts-by-experience were invited to share their stories with the community groups. Meetings started with prayers. Then the persons with disabilities proceeded to relate their stories one at a time. V was first to give his story about when he was in school, after school and his struggle in trying to get a wife. W was next in giving her story. She had a hearing impairment so the special needs teacher translated her sign language to the participants in the community groups. She based her presentation on personal experiences of starting a business within a community and being discriminated against within a family setup. X talked about experiences in school, and the challenges experienced at home. After the narratives, the community group members were given the opportunity to discuss issues and pose questions to the experts-by experience. The expert group was given financial compensation for their time, both for the training and the visits to the community groups during the intervention period. --- Data collection and management Focus group discussions were conducted with each of the twenty community groups, pre-and post-intervention. The meetings were arranged by the Pambazuko representative in communication with group's chairperson. The meetings took place in usual meeting place for each group, and refreshments were provided. The Chairperson opened the meeting and led the group in a prayer. The discussion was then introduced by the first author who followed an interview schedule consisting of topics and a questioning route. At baseline, the groups were also asked to reflect on the causes and challenges of: having a disability; and caregiving for someone who has a disability. This is reported in a separate paper [14]. They were encouraged to talk about their experiences of children and adults with disabilities in their own communities and to consider the role of the community in the lives of such individuals. Post-intervention visits were conducted approximately 2 months after the group had been exposed to the personal narratives of the experts. The groups were asked to consider the narratives shared by the experts by experience and what they had learned. The schedule of questions is provided in Additional file 1. Table 1 shows the attendance of the focus group discussions for the different group types -CHWs and WGs. The WGs had a higher membership attendance rate compared to that of the CHW with a higher median attendance . The WG were comprised of all women and the CHWs had female to male ratio of 6:9. Discussions were carried out in the preferred language of the group and recorded on a digital audio device. With reference to the questioning route, questions were posed to the whole group initially to invite spontaneous responses. In order to establish a fair distribution of turns, this was followed by encouraging others to add to or comment on the views previously expressed. The duration of the FGDs was about two hours. The recorded data were uploaded to a computer and transcribed in the language used by the group, before translation into English; imported to the data management programme NVivo 10 for storage and analysis. To ensure accurate representation of meanings in the translated copies, checks were carried out during the analysis period. This involved the first author and the last author -based visitor to Kenya) reviewing the transcripts, querying emergent concepts and their meanings, and using back translation as appropriate. --- Data analysis 'Framework' was the method of analysis used to evaluate impacts associated with the intervention [49]. The entire data set was analysed following a structured process of: familiarisation with the transcripts initially, through repeated readings; identification of a thematic framework by reviewing the data, by noting key concepts and establishing a first generation of nodes using NVivo-10; 'indexing' whereby the data were explored and assigned to nodes and sub-nodes as appropriate; application of the 'index' whereby transcript excerpts were assigned to the different node levels; and finally, mapping the range and nature of phenomena, creating typologies and finding associations between them, and interpreting the data as a whole. The first and last authors analysed the data separately, coming together on several occasions during the process to compare their separate analyses. The first author, a native of this area of Kenya, applied his local knowledge and experience to the analysis, while the last author, a UK-based, regular visitor to Kenya, provided a more remote and impartial stance from which to analyse the data. Differences in frameworks were discussed, nodes and their relationships were reviewed, until agreement was achieved. --- Results The emergent framework consisted of four organising themes: Burden; Agency; Sub-human; Human. They were arranged as pairs of opposing constructs along a continuum , i.e. burden vs agency; sub-human vs human as shown in Fig. 2 which captured the possibility of movement between them. Subordinate constructs were identified at lower levels of each main construct pair, as indicated by filled bullet points. Only the sub-human vs human construct was further subdivided into opposing pairs of basic themes, as indicated by unfilled bullet points. With reference to the framework in Fig. 2, the results are presented in a series of tables. Each table focuses on a different pair of subordinate constructs, labelled in the side margins of the table, from within an organising pair of constructs . Basic themes are shown in Tables 3 and4 only indicated by unfilled bullet points within a column, where data from pre-and post-intervention are presented. Immediately after each quote, the source is given in square brackets, i.e. the group type and participant number, and the pre-or post-intervention status indicated by -1 or -2 respectively. 'Burden vs agency' construct The 'burden vs agency' construct was composed of two pairs of opposing themes, referred to as sub-constructs: Inability-ability; Hopelessness-opportunity. 'Burden' was defined as a problem that affected other people around the individual with a disability, with particular mention of the mothers as the primary caregivers. In the context of limited resources, both economic and psychological, there was the expressed view that having a person with a disability in the family, and in the wider community, was a source of difficulty. Reference was made to negative images of disability and disturbance to the well-being of others, both caregivers and members of the community. It was the case that disability was associated with 'burden' defined by 'hopelessness' and 'inability' , almost exclusively at the pre-intervention stage, with 81 references made pre-intervention and only 3 references post-intervention. However, there was also recognition of the 'agency' of people with disabilities, defined as the capacity of an individual or group of individuals to act in the world. At post-intervention, 'agency' , defined by 'ability' and 'opportunities, emerged more emphatically with 193 references pre-intervention and 230 references post-intervention. --- Inability-ability On one side of this sub-construct was inability, which captured a lack of capacity or what the person could not do, and what was perceived to be missing in the person. This was associated with dependence on others for care and support needs. The other side of the construct was ability, which referred to the individual having skills. It focused on the things the person could do, e.g. walking, learning, drawing water, earning money. ' Ability' was recognised both pre-and post-intervention. Pre-intervention discussions tended to focus on persons with disabilities who were known to the respondents. For example, references were made to individuals who used manual signing for communication, whose work involved digging the land and carrying water. Post-intervention the discussions focused more on the abilities of the experts who had presented their personal narratives. However, the opposing construct of 'inability' was expressed almost exclusively in the pre-intervention groups, with the exception of just 1 reference made post-intervention:" …having limitations and there are things they cannot do." [CHW10-2]. However, the use of 'things' implies a more measured response that is different from the more exacting responses in the pre-intervention discussions. Here the emphasis was on what the person could not dotheir lack of capacity, as exemplified by the use of intensifiers as in "totally disabled", and qualifiers as in "…cannot even eat". Negative structures, such as "not", were used in the pre-intervention references to people with disabilities. There was a shared perception of them being incomplete. Post-intervention, disability and ability were not mutually exclusive. That is, the respondents cited examples from the experts-by-experience, concluding that disability was not inability. Table 2 provides supporting quotes from the transcripts. --- Hopelessness-opportunity Hopelessness captured the sense of giving up and the desire to have the individual with a disability gone. Associated with this was a lack of ideas or momentum to act in a different way. The antithesis of this was opportunity, which encapsulated the idea of supporting the person with a disability to take part in activities, e.g. attend school, regardless of their particular challenges. This implied putting help in place, and helping to circumvent some of the observed difficulties. Pre-intervention, the focus group discussions told of situations that implied hopelessness. A sense of despondency was conveyed in the stories shared, and even of individuals with disabilities being rejected. There was some notion of supporting the child with a disability to be "in the world" at the pre-intervention stage, but this was augmented considerably in the post-intervention discussions. Opportunity was defined as support and calls to action by the community. Greater awareness of opportunities was shown post-intervention, with consideration of health, education and employment, which underpinned the groups' growing sense of the 'agency' of people with disabilities. Calls to action included having a community meeting as "…one voice..." and getting others to listen to the stories of people with disabilities. Table 3 provides supporting quotes. --- 'Sub-human vs human' construct The theme of 'Sub-human' denotes a focus on the disability before the person, which dominated group discussions before the intervention .. In contrast, the theme 'Human' refers to recognizing the person before the disability. This included attributing positive value to the individual and viewing him/her as a member of the community . The 'sub-human' vs 'human' construct was composed of two pairs of sub-constructs: Dehumanisation-recognition; Discrimination-inclusion. --- Dehumanisation-recognition Dehumanisation included references to people with disabilities that rendered them as negatively valued. The words used for people with disabilities referred to the type of disability as a collective noun, with neglect of the person. Practices were described that were devoid of humanitarian effort, e.g. chained up, tied, food passed with stick. The individual was rendered as an object, without human feelings or potential. The antithesis of this was recognition, which acknowledged the person in his or her own right. There was notice of the person's struggles and difficulties, with expressions of empathy for their experiences. The terms used in the groups before the intervention, depicted persons with disabilities as 'things' belonging to the ki-vi class of nouns in the Kiswahili language. Individuals were referred to as "kiwete" , "kiziwi" and "bubu" , or "akili punguani" . This had the effect of classifying the person with a disability as sub-human, viewed as an 'object' or "grown like that of an animal". However, there was also 'recognition' of the 'person' as stories of familiar individuals were told, e.g. "…raise my hand to greet him…", '"the first person to get the news". Post-intervention, the terms used previously in relation to people with disabilities were revised towards greater 'recognition' of the person, before the disability, e.g. "we are all human beings…" Realisation of the sameness of persons with and without disabilities was expressed in relation to their religious convictions, e.g. "created by the same God" also. The basic theme 'shame' referred to the indignities suffered by people with disabilities and some of terrible Opportunity treatment meted out by others. The stories were told both at pre-and post-intervention; however, greater 'empathy' characterised the stories at post-intervention, although not exclusively, e.g. "…you will sympathise with him…" said one respondent describing someone in her local community before the intervention. However, after hearing the narratives of the experts-by-experience, the group members expressed affect as well as awareness of the person's emotions, e.g. "…saddened by their stor-ies…" and "…he also has emotional feelings". Table 4 gives supporting quotes. --- Discrimination-inclusion Discrimination described the ways in which the person was treated differently to their peers, being 'discounted' or 'abused'. Opportunities were constrained and exclusions were commonplace, rendering the person disempowered and disenfranchised. The opposite of discrimination was 'inclusion' , which captured being 'valued' and establishing 'acceptance' in everyday life. The discussions pre-intervention revealed tales of 'discounted' individuals who were acknowledged to be "… valueless in the family" and discriminated against when earning a living, e.g. "…the fish will not be bought…" Other stories told of 'abuse' which included exploitation also. Post-intervention there was deliberate consideration of the actions needed to counter the 'discounted' ways of responding to persons with disabilities. Actions that valued the individual included "...showing them love…", "…care for them…", and "…know about their health, their wellbeing…"all expressions of human need. Finally, there was 'recognition' of a need for 'acceptance' such that persons with disabilities should be treated in a similar way to those without disabilities, with a call to "…sensitise the community…" Table 5 gives details. --- Discussion Occurring mainly in pre-intervention group discussions, dominance of negative images and the use of language that tended to focus on the physical or visible characteristics of disability were suggestive of the individual or medical model [9]. The open-ended questions asked of the groups may have triggered access to early mental representations of disability that were formed in childhood as suggested by Federici, Meloni and colleagues [1,2]. This would imply the presence of an underlying cognitive process at work [2]. In seeking to formulate a clear and consistent response to the disability focused questions, the respondents resorted to the most immediate explanation available in their mental lexicon [1]. The models act as reference points, which inform the development of social responses and relations [1]. In the absence of a more social representation of disability, the respondents have attributed disability to deviation from more typical biomedical structures and functions [9,10]. Cultural beliefs and traditional practices have been commonly associated with negative images [12,[15][16][17][18] 'I have seen one and he was "kiwete" and had to use a wheelchair to move from place to place.' [WG5-1] 'There are the "bubu" and "kiziwi" .' --- [CHW2-1] 'There is a boy who has been nicknamed "Kahindi Kadzitswa" .' [CHW9-1] 'You find that his body has grown like that of an animal.' [CHW10-1] 'when I pass by there I raise my hand to greet him and he responds by doing that also. and concepts that inspire fear in others [4]. Thus the explanations of disability that were espoused by members of this East African community, e.g. preternatural forces [12,[15][16][17][18][19], may have shaped the images of abnormality and dysfunction that featured in the groups' discussions. As suggested by Meloni et al. [1], the polarisation of the individual/medical and social models is more a product of culture than cognition, which infers a learning process. Inadequate coverage of information about disability across rural communities [5,6] means that alternative models or ways of understanding disability may not have been available. Thus, it could be said that the cultural construction of disability upheld the idea of the individual model, at least in part, by viewing disability as a direct consequence of a supernatural event [15][16][17][18]. In contrast, breach of social conventions [14,17] placed responsibility on the actions of other people, and therefore deflected the focus from the child to the setting. Ultimately, however, the individual was still viewed as different to most people. Many of the stories of people with disabilities spoken about in the focus groups, particularly pre-intervention, yielded images that portrayed children and adults with disabilities as undesirable or associated with an unpleasant encounter, which affected the extent to which the person was accepted by the community [24][25][26]. There was some distancing of the person with a disability through the language used to describe individuals. Use of inanimate linguistic objects to describe people with disabilities, e.g. words belonging to the KI-VI class of nouns in the Swahili language such as "kiwete" which has a plural of "viwete" , rendered the person as an object. Thus disabilities were viewed as entities, devoid of human life. There was the added tendency to describe a lack, or total absence of ability. This finds resonance in the reification/recognition dichotomy explored by the German philosopher Axel Honneth [50]. He highlighted the human tendency to treat other people as mere objects instead of responding to them as individuals who are more than just things . The propensity to abuse vulnerable individuals springs partly from this tendency. However, we also have the capacity to recognize others, which was evident in some of the accounts of persons with disabilities who were known to the respondents, and emphasized in the post-intervention data. Encountering experts-by-experience who shared their personal narratives, appears to have triggered a reassessment of the meanings associated with disability [32][33][34][35][36][37]. Confronted with the experiences of people living with disability, the groups developed stronger recognition of the person who has a disability. According to Honneth, this is a prerequisite for empathy [50], and certainly the post-intervention discussions revealed the expression of empathetic views. There was also greater emphasis placed on the agency of persons with disabilities in terms of abilities and facilitating actions by others, and on being human in terms of recognition and inclusion. Allport's 'contact hypothesis' attributes these changes to the deliberate interactions with people living with disabilities [32], which corroborates the findings of Armstrong et al. [34,35], MacMillan et al. [32], and Schwab [34]. The deliberate exposure of the groups to people with disabilities, who were the frequent subjects of negative images and devaluation, challenged the stereotypical views of persons perceived to be a burden and the 'objects' of dehumanizing practice [36][37][38]. This resonates the findings of other studies that have promoted encounters with people with disabilities in pursuit of attitudinal change [36][37][38][39]. Post-intervention, there was a general realization that people with disabilities needed love and respect, just like other human beings. The stories by the experts-by-experience caused many participants to shed tears, and to express new insights into the experiences of people with disabilities, which were commensurate with empathy. The call to act on this new knowledge was shared by many of the groups, who sought to sensitize the community and employ experts-by-experience themselves to affect positive change in their communities. --- Limitations The sample was comprised of established groups of community health workers and women's groups. Whilst the value of formally constituted memberships brought mutual familiarity and shared local knowledge to the focus group discussions, gender bias in the women's groups may be a factor in their discussions, just as the more health focus of the CHWs may have influenced their particular views on disability. Future research could involve other community groups considered to be opinion leaders, such as heads of ten homesteads, village leaders and even traditional healers to give diversity of perceived views. The experts-by-experience originally recruited to deliver the intervention were representative of a range of the most common developmental disabilities . However, 2 individuals withdrew their services during the preparation period, which reduced the diversity of personal narratives presented to the groups. This may have affected opportunities to consider the broad spectrum of disability in the groups. Saturation checks on the data for possible follow up in terms of data collection were not made due to financial and time constraints. The questions and prompts used to guide the focus group discussions post-intervention, might have usefully included reflections on the pre-intervention discussions. By asking participants to comment on the views expressed prior to meeting the experts-by-experience, direct capture of any transformations in attitude may have been possible. Whilst the study was able to capture community group responses to disability, and to examine the potentially mediating effect of exposure to experts-by-experience, further research is needed to evaluate intervention potency and effect size. --- Conclusions The research contributes to the evidence on CBR, demonstrating the potential of a low-cost, relatively brief intervention, which may be applicable in other lowincome settings. The very act of employing people with disabilities assigns them a higher status as 'experts-byexperience' , which is in opposition to the more familiar 'discounted' role seen in many communities. The participants wanted the experts to have the same opportunities as others, which extended to being married and having their own home. The sharing of life experiences through organized community engagement, appears to have facilitated community groups to reflect on, examine and adjust their attitudes towards people with disabilities. The personal narratives served to authenticate the process of learning by the community groups. Listening to and interacting with the experts-by-experience invited the participants to reflect on their own values, against a backdrop of deeply held cultural beliefs and practices. The personal stories guided the participants in viewing the 'person' before the 'disability'. This appeared to be a catalyst for change, both in the attitudes espoused, and in the motivations expressed by the groups. For example, the groups articulated a shared quest to achieve better understanding of disability and to reach more people in their communities. Ultimately, the findings from the current study have implications for mobilizing DPOs, parent or caregiver groups and other rights-based organizations as agents for change in local communities. In this way, a new type of 'stygmergy' [28] may be triggered by engaging with the lived experiences of people with disabilitiesone that revises the view from the community bringing about recognition of the person -who just happens to have a disability. --- Additional file --- --- --- Competing interests The authors declare that they have no competing interests. ---

The last decade has seen improved public awareness of disability in sub-Saharan Africa. However, negative and stereotypical views of disability still persist in many communities. We conducted a study to promote awareness of disability in rural Kenya, using a process of reflection and education. This paper reports on the second aspecteducation. The research question was: How can personal narratives of living with disability affect community attitudes and responses to disability? Methods: A qualitative phenomenological approach was adopted. Twenty community-based groups involving 249 participants took part. Each group participated in one focus group discussion at baseline, to explore the members' personal experiences and views of disability. The intervention involved three adults with disabilities sharing their personal narratives with each group. After the intervention, repeat focus group discussions were conducted with each group. Thematic analysis was carried out according to the framework method. Results: The emergent framework consisted of four main themes, organised as opposing constructs: 'burden' and 'agency', 'sub-human' and 'human'. 'Burden' focused on the perceived hopelessness of the situation. Post-intervention revealed greater support for the 'agency' of persons with disabilities, evidenced by what the person could do, rather than their inability, and the relevance of support. The 'sub-human' to 'human' construct captured dehumanising and discriminating practice towards persons with disabilities on one side, and recognition of the person and inclusion in the community on the other. Whilst support and empathy were evident at the pre-intervention stage, postintervention revealed greater recognition of people with disabilities as fellow human beings.This study provides a proof of concept regarding the deployment of persons with disabilities as agents for change. Exposure to experts-by-experience provided community groups with opportunities to reflect on, examine and adjust their views on disability in this rural part of Kenya. The sharing of personal narratives appeared to resonate with group members, to encourage recognition of the person and not just the disability, and to move their resolve toward ideas for collective action. Further research is needed to assess the effects of such interventions.

Introduction In Norway, almost all preschool children attend kindergarten, and parents and staff are expected to cooperate in the best interest of the child . Parent-teacher conferences, usually offered twice a year, are one arena in which this kind of cooperation is realised. In Markström's ethnographic study, such conferences are described as constructing a "picture of the child, the institution and to some degree even the home environment" ). In parent-teacher conferences at kindergartens, parents receive information about their child, but information also goes the other way around, i.e., family information about the child and/or their home situation is provided to an early childhood education and care teacher , which is the topic of this paper. In an interview study, 60% of nursery teachers reported that they had sufficient information about a child's home situation, while around 40% were less certain . In this study, an informant asked: "How much should we know? If there are problems at home, we should know about them, so we can take good care of the child. Otherwise, I'm not sure that we need to know very much" ). From a moral and legal perspective, this citation displays an ambivalence which can be formulated as a potential conflict between the family's right to privacy and a child's right to proper care . According to Article 8 of the European Convention on Human Rights , privacy is a human right: "Everyone has the right to respect for his private and family life, his home and his correspondence". In accordance with the integrity perspective, individuals have a right to manage themselves without outside interference, . However, privacy is not an unconditional right, and the family can be viewed as both a unit and as conflicting parts . The second paragraph in Article 8 states exceptions that are "necessary in a democratic society," for instance, "for the protection of health or morals, or for the protection of the rights and freedoms of others". Thus, in an ECEC setting, children's rights, and especially the principle of the best interest of the child , may conflict with the family's right to privacy as a group. In ECEC research, the dichotomisation between "public" and "private" has been understood as a threat to children's citizenship by discouraging teachers who suspect child neglect from contacting the child welfare service . In Norway, the number of child welfare notices has increased significantly in recent years, and the findings from a transnational research project indicate that Norwegian child welfare workers are more willing to intervene in the family . Seen from this angle, information about a child's home situation is also a display of parenting skills, living conditions, etc. that welfare state agencies can monitor and police . Questions about a child's home situation may, at least to their parents, have a potential power dimension, and there is a lack of research about how teachers manage such topics during parent-teacher conferences. The concept of privacy has various dimensions ), and this study mainly investigates the social dimension of the concept, addressing participation in interaction . Inspired by ethnomethodological conversation analysis , the paper investigates how family information is occasioned or "talkedinto-being" in ten audio-taped parent-teacher conferences by asking the following questions: What interactional strategies for eliciting family information do ECEC teachers adopt in parent-teacher conferences and how do parents respond? What embedded understandings regarding the norm of privacy are thereby displayed? Not unexpectedly, the analysis shows that the teachers use the available interactional resources, mostly questions. After briefly outlining the broader educational context, major findings from conversation analytic research on inquiries will be presented. In the Results section, the range of interactional practices at work in the material will be demonstrated, and some elements will be reflected upon in more detail in the Discussion section. While the historical backdrop of the growing institutionalisation of childhood can be taken to suggest that there has been a dissolution of the boundary between the private and the public, including boundary work in micro-sociological settings, this analysis suggests otherwise. --- ECEC at the Intersection of the Private and the Public The right to privacy seems to presume a private domain separate from other people and state authorities, which could be easily associated with the nuclear family and the "housewife family" ideal which developed during the 1900s. This ideal, pinpointed by Lasch as "the family as a haven in a heartless world" , was founded on a strong separation between public and private life . However, as a social democratic society, Norway has followed a defamiliarizing family policy, nurturing women's economic independency , in line with general individualising tendencies in Western societies. These tendencies have also affected both children and upbringing practices, and the UN Convention on the Rights of the Child in particular manifested an ideological shift from the family to the child as an independent third party . The institutionalisation of childhood was increasingly intensified, gradually including toddlers as well as young people . In this 'public family' policy children underwent what Dencik called "double socialization" in private and public arenas. Consequently, parent-child relations have become more transparent to state agencies, suggesting a norm of visibility . On the micro-level, this development suggests a normative shift in how professionals and parents are supposed to relate to one another in educational settings . Instead of the idea of separate domains, an understanding of a so-called "shared responsibility" has developed in ECEC . This means that teachers have a broader responsibility for children's well-being while enhancing parental involvement . In the OECD report Starting Strong II: Early Childhood Education and Care , ECEC staff are advised "to form a partnership with parents, which implies a two-way process of knowledge and information flowing freely both ways". The Norwegian ECEC framework plan outlines the parent-staff relationship using somewhat similar wordings: The kindergarten shall facilitate co-operation and good dialogue with the parents. At an individual level, the kindergarten shall ensure that the parents and the kindergarten can regularly exchange observations and evaluations concerning every child's health, well-being, experiences, development and learning . Questioning is not specifically mentioned in the framework plan . In Markström's study, teachers tended to avoid questions about the home, though they had different personal styles in this respect. Some teachers adopted a more telling style, while others adopted a question-answer style . Fairclough describes a development whereby the interaction in more and more institutional settings is characterised by a hybrid genre, mixing everyday and institutional discourse. Within a Foucauldian framework, welfare state professionals represent contemporary versions of the Christian pastor figure , and dialogue is a technology that is used to govern individuals and families . It is more likely that privacy is easier to maintain in a formal institutional context where people are in a "frontstage" mode than in an informal setting . An ECEC parent-teacher conference, characterised by much laughter , represents the latter. Thus, the interaction style is not designed to maintain a public-private boundary. The discourse analytic perspective has been much adopted in Nordic research on parent-teacher conferences , included in Markström's study. Markström describes how teachers' "innocent questions" can be understood as a soft way of governing the conduct of the family , in this case, guiding parents to facilitate reading. However, Markström's ethnographic approach is primarily targeted at describing the overall context of the conferences , and the data extracts were not accurately transcribed. As a consequence, the very processes of "questions and answers" and how the visibility norm is managed in social interaction remain opaque. In contrast, the ethnomethodological tradition of conversation analysis , introduced in the next section, can provide an in-depth account of the inherent power aspect of questioning. --- The Power of Questioning In the ten parent-teacher conferences examined in this paper, the most common way of seeking information was, not unexpectedly, asking questions. A question may be understood as: "a form of social action, designed to seek information and accomplished in a turn of talk by means of interrogative syntax" ). However, as will be demonstrated in the Results section, there are also less direct ways of seeking information, for example, fishing for information without using interrogative syntax . Across institutional settings, professional inquiring is a recognisable activity and a primary means to "determine truth and amass facts" . In addition, professionals' questions may be "drenched with implicit moral judgements, claims, and obligations" , effectuated through the apparatus of talk. In ethnomethodological conversation analysis, questions and answers are prime examples of 'type-connected' two-unit sequences in talk . After a 'first-pair part' , a 'second-pair part' can be expected. Since the question, so to speak, induces addressed respondents to produce the expected second-pair part , the "adjacency-pair" apparatus gives the questioner a momentary power . While the inherent power or control dimension of questioning is seldom addressed in CA studies, Harvey Sacks, a central founder of the discipline, did in one of his studies "As long as one is in the position of doing the question, then in part they have control of the conversation" . This means that the questioner is in a position to control the turn-taking as well as the topic/ agenda of the next turn . Moreover, the power of the questioner can be intensified through the use of multiple questions, for instance, a "questions cascade" producing different versions of the same question or "follow-up questions" pursuing a more substantial answer . However, the power can also be softened by adopting prefaces and accounts in the question design . Obviously, power is not referred to here in its absolute sense, literally "forcing" respondents to provide information, yet accountability as "good parents" might nevertheless be at stake. From an ethnomethodological perspective, "not answering the question" is less of an option. Of course, we can imagine parents who refuse to answer questions in a conference, but given the adjacency-pair apparatus , where a question establishes a normative expectation of an answer, a deviance from expected conduct represents a "noticeable action," calling for an account of why the second-pair part is not produced . Regardless of role, the normative pressure, embedded in the adjacency-pair structure, represents a situated power on behalf of the questioner and, as mentioned, parents may also ask questions in an ECEC conference. Ultimately, interactional roles are deeply intertwined with institutional roles. As previously mentioned, teachers have a legal obligation to monitor a child's well-being and contact the child welfare service if child neglect is suspected. This structural power might, at least indirectly, affect how parents behave in ECEC conferences. --- Method The data comprise 10 audio-recorded parent-teacher conferences collected from two kindergartens in Southern Norway in a qualitative research project dealing with parent cooperation and diversity in ECEC. The project received funding from the University of Southeast-Norway and the Oslo Fjord Fund. The study was approved by the Norwegian Centre for Research Data and was conducted in accordance with the Norwegian National Guidelines for Research Ethics in the Social Sciences, Humanities, Law and Theology . The two kindergartens differed in terms of size and cultural diversity. In kindergarten 1 , located in an urban environment, two ECEC teachers and six parents participated in the study. The parents in kindergarten 1 were from an immigrant background, and in three of the conferences, a multilingual assistant who worked at the kindergarten was present to interpret and assist . Kindergarten 2 was a small kindergarten located outside the city centre and all the parents were from a majority background. In kindergarten 1, the researcher was present at the conferences but only played a minor role. In all the conferences, the teachers followed a form, structuring their reporting about the child in question , but the analysis in this paper does not focus on how the form contributes to organising the interaction, which would have required video data. The data extracts were transcribed according to the conversation analysis convention, originally developed by Gail Jefferson . Information about institutional roles and parents' minority/majority status will be provided in the transcripts , which is a controversial issue in conversation analysis . However, the sociocultural background and reflections on privacy will mainly be discussed after the conversation analysis. Moreover, in order to preserve the anonymity of the informants, some background information may have been changed. In the analysis process, the teachers' eliciting actions were identified and coded to identify patterns. As can be seen from the table above, the material had a total of 109 instances of inquiries, 69 of which dealt with the home situation. Regarding content, the questions addressed the child's eating, sleeping, toilet, playing and dressing habits at home, language code and reading, crying and the use of a dummy, older siblings, and other persons living in the household. The table shows that teacher A asked the vast majority of the 69 questions addressing the home situation. The teachers also asked 40 questions relating to other topics than the home situation. For instance, the parents' views about the contact with the staff, their child's well-being at kindergarten, whether they had received certain information or permission to perform language tests, etc. All teachers asked these kinds of questions, dealing with various aspects of the services. Inquiries work both ways, and some of the parents in this study also asked questions . The parents asked a total of 33 questions, many of which were about their child's conduct in kindergarten. Thus, Markström's finding that parents asked more questions than teachers could not be confirmed. In addition to questioning, the practice of fishing for information as well as parents' unelicited reports from the home situation, were also noted and coded. In terms of generalisation, the study cannot confirm how usual it is to ask for family information in this setting. After all, in her study, Markström stated that direct questions about the home were "rare" . The same applies to two of the three teachers in this study, whereas one of the teachers asked a lot of questions. Regardless of the distribution of eliciting practices, this paper will demonstrate, in Peräkylä's wording: "Social practices that are possible, i.e., possibilities of language use " in this setting. Thus, rather than focusing on frequency, the paper will demonstrate a range of practices for managing family information. --- Results This section examines various ways of managing home information in ECEC parent-teacher conferences. The analysis is organised in such a way as to demonstrate a continuum of teachers' eliciting actions and their ability to affect parental accountability in the talk, i.e., the analysis will demonstrate increasing interactional pressure on the parents to provide family information. The first extract shows a practice called "Parents' spontaneous comparisons" in which parents volunteer family information that is not being sought by the teacher . Extract 2 demonstrates the rarer strategy of fishing for information, and extracts 3 and 4 investigate the teachers' questioning practices. --- Parents' Spontaneous Comparisons Parents' "spontaneous comparisons," revealing family information not pursued by the teacher, were a common resource for bringing information about the home into the ECEC conference. In this practice, the parent sticks to the topic under discussion but makes a topic transition into the home setting on the same topic . . In this extract, the teacher talks about the kindergarten's reading group, before the mother self-selects: The turn allocation system described in conversation analysis suggests that usually, one party speaks at a time . In line 2, the mother self-selects, and there is a moment of overlapping talk where the mother recycles part of the teacher's turn . Rather than completing the teacher's turn , the mother's recycling starts a new project, describing their reading practices at home before the child's baby sister was born. While momentary overlapping around anticipated turn endings does occur in talk, second speakers rarely persevere after the drop point , which can be interpreted as competition for the floor . The mother's recycling in line 2 can thus be heard as claiming independent access to knowing what her daughter loves, but in line 6 she is also making an excuse about why they do not read as much at home anymore, indicating a certain defensive orientation on the part of their reading habits. In this extract, privacy is not disturbed as the mother is not required to provide family information, but she does anyway. The example is representative of the overall material, and the practice of making a spontaneous comparison is likely to be seen as a symptom of the visibility norm associated with modern parenthood . This norm might induce parents to display appropriate/reflected parenthood and/or invoke excuses and accounts when the highest standards cannot be met . It is possible that the defensive orientation can also be related to what previous research on minority parents has described as "enhanced awareness" of parenting practices . With language code and reading as a topic, the mother would perhaps expect to be asked about reading at home. Thus, anticipating the upcoming topic could be an interactional strategy to portray herself as a good parent. --- Fishing for Family Information The next extract demonstrates an indirect information-seeking strategy, described in Anita Pomerantz's seminal paper: "Telling my side: "Limited Access" as a "Fishing" Device". In her example, a caller with "limited" knowledge can fish information about an event from a person with authoritative access. The line: "I wz trying you all day.en the line wz busy fer like hours" might be perceived as a report of an experience or as a careful request to volunteer information . If the missing counterpart is not produced, this might be perceived as a form of "withholding" . In extract 2, below, the teacher adopts the fishing device to elicit family information, more specifically, information about why the child, Adrian, was absent from kindergarten the week before. On this particular day, the teacher says that she had walked with a group of children through the garden at Adrian's house: In the middle of line 1, the teacher switches over to direct reported speech , but it is not clear whom she is animating. She is most likely animating her own thinking at this moment. This animating voice prevails in line 3, but the talk now refers to a talk with Adrian , where the child had mentioned reasons for his whereabouts on this day No, grand-"). Again, it is not clear whether the confusion about the child's whereabouts is part of the reported scene , or whether the teacher has problems recalling what the child had said. In either case, the teacher has, in Pomerantz's terms, "limited access" and the parents are the best source of authoritative information. In this extract, both parents align to bring up "the missing counterpart" and in line 4, the mother takes a leading role and addresses the father who, in line 5, gives a tentative answer. However, in line 9, the mother provides the correct information herself: they had been at the eye specialist in the neighbouring city. In line 11, she says, "So that's why we were off-= Took some time,", and this phrase explicates the teacher's project of gaining an account of their whereabouts on this day. Compared to the direct questions described in the two next sections, the fishing strategy makes the parents less accountable for providing the "requested" information. In another instance in the same conversation , the parents did not volunteer family information. Thus, this strategy gives the parents a choice: whether to provide or withhold the pursued information. In terms of privacy, this instance likely represents only a minor disturbance, which the parents could have avoided reacting to. Nevertheless, their willingness to fill in the missing pieces supports the presence of a visibility norm in child-rearing . The parents managed to convey that they had very good reasons for why Adrian did not attend kindergarten on this day, precluding alternative scenarios that may have occurred to the teacher. --- Asking for Family Information In extract 3, below, the common sequential structure "contextualizing talk + question + answer" will be demonstrated. "Context" is understood here in Heritage's two-fold account of actions as both shaped by preceding actions and as context-renewing, producing new actions . Regarding the first dimension , questions are seldom asked "out of the blue". Rather, professionals tend to produce some kind of "prefatory statements" , which contribute to "contextualizing" the upcoming question and thus "explain" the reasons for the question. In ECEC, this often takes the form of the teacher accounting for how the child is doing in the ECEC along various parameters, and this kind of reporting/ describing is the dominant activity throughout the conferences. The teacher then makes a transition into the home setting on the same topic. . Earlier in the talk, the teacher in this extract had been describing the child's playing skills. In line 19, she switches to talking about the child's motor skills in kindergarten, which makes up the context for the question in line 21 : The change in topic is basically indexed through the keywords "motor skills, sensing" in line 19, which possibly reflects topics in the form structuring the meeting agenda. An evaluating description of the child follows, ending in the assessment "She really loves walking". In ordinary conversations, assessments of this kind make a subsequent assessment relevant , but through the minimal response in line 20 , the mother positions herself as the receiver of a report rather than an assessment. In line 21, the teacher continues the accounting, and the turn ends with a question: "I don't know whether you hike and such like?". The preface "I don't know" signals a low epistemic gradient on behalf of the teacher, framing the matter as outside her domain of knowledge. The tag "and such like" performs the same function. The mother responds to the question in lines 22-24 but does not elaborate much and speaks in a very low voice . Compared to extracts 1 and 2, the mother in this extract did not appear to be very talkative, though her conduct can easily be interpreted as an example of what Burgoon called reserve, which "implies some degree of interaction with others and refers to the manner in which one communicates as to limit others' knowledge of oneself . The structure is nevertheless representative of the overall material. In the rest of the extract , the teacher uses the fishing strategy to get a more substantial answer to whether the family hikes. Another strategy for the teacher to map the family's lifestyle on this matter would be to pursue the topic more directly by asking elaborating questions, which takes place in the last extract in the next section. It is perhaps not so easy to argue that a question about hiking habits represents a potential disturbance of privacy. However, a major survey revealed that many patients considered a question about their use of leisure time during medical consultations to be a violation of their privacy . Thus, regardless of the setting, lifestyle questions about hiking habits might invoke moral feelings, but they are probably more expected and justified in a medical setting. Information about the family's hiking habits contributes to filling in "the picture of the child" . However, it is hard to see the necessity of the question. As in extract 2, the teacher may have other reasons for asking the question rather than a professional need to know. --- Asking Multiple Questions Professionals often ask clustered or "multi-unit questions" . Overall, in these data, there were not many questions, but one of the three teachers asked many questions about the home situation. In a sequence from one of the conferences, the teacher asked four questions about the child's sleeping habits at home: Ending the account of the child's sleeping habits in kindergarten, the teacher asks about the child's sleeping at home. In line 1, she asks an open-ended question "How does he sleep at home?" which is immediately reformulated into a candidate question "does he sleep well at home?". In designing candidate answers, speakers tend to propose legitimate actions , in this case, "sleeping well" instead of "sleeping badly". According to Svennevig, the practice of reformulating questions with candidate answers is usual in conversations between native and non-native speakers, and immediate reformulations are designed to pave the way for both minimal and extended answers . The mother in this extract aligns to both options by confirming the candidate question as well as reporting problems about putting her child to bed . The account that the child has followed this sleeping pattern "all his life" has a certain excusing quality, possibly suggesting that the child's sleeping habits are due to innate traits, rather than caused by parenting practices. In line 3, the teacher asks another candidate a question following up on the subject: "does he fall asleep late at night?". In response, the mother confirms and repeats the glossed account already given . In line 5, the teacher asks a question that narrows the subject "When does he fall asleep at night?". Again, the mother aligns and provides an approximate time . In line 7, part of the teacher's comment that the child does not appear to be tired at kindergarten is stressed . This report can be heard as a momentary mending and retreat from the investigator role, calming the mother . By speaking from the ECEC perspective, the teacher implicitly acknowledges bedtime as being the mother's expert domain. Thus, a certain awareness of the public-private boundary being negotiated could be detected in this line. The rising intonation in the mother's "No?" may be heard as a cautious request for more information about the child in ECEC, but instead of elaborating, the teacher poses a new question about when the child gets up in the morning . Once more, the mother aligns and provides an exact time and, upon receiving this information, the teacher assesses that the child gets enough sleep . As mentioned, the use of reformulations can be a device for pre-empting potential problems in communication . However, the move back from the candidate question in line 3 to asking about the exact time in lines 5 and 9 suggests otherwise. Based on the teacher's direct when question in line 5, it is more likely that in line 3 she pursued new information about the child's sleeping habits. This may suggest that speakers who adopt candidate answer designs do not always pursue a yes-no response, simply confirming their own model of the world. The use of multiple questions and the explicit assessment of the child's sleeping habits give the sequence a certain investigative touch, examining in detail whether the child gets enough sleep, though a certain awareness of the delicacy of boundary work is displayed in line 7. In terms of privacy, the multiple questions made the mother accountable for providing family information about her child's sleeping habits. Moreover, the assessment in line 14 represents an even deeper move into the parents' private domain as the teacher in this turn topicalizes the parents' need for "quality time in the evening". The mother did not pick up on the quality time topic and thus did not align with the teacher's quite intimate style of communication . At best, the instance demonstrates a more hybrid orientation to the public-private boundary on behalf of the teacher. Given the information in extract 4 , it is hard to identify an obvious professional concern for the child's well-being. --- Discussion To the family, bedtime is likely a more sensitive topic than language code at home , as this may provide pointers to the intimate relationship between the parents and lifestyle in general . Rather than focusing on content, the paper has examined ECEC teachers' interactional strategies for eliciting family information in parent-teacher conferences, and the parents' responses to them. The analysis demonstrated a continuum of practices in terms of how and the extent to which parents were made accountable for providing family information in the conferences: Parents' spontaneous comparisons , fishing and asking questions . In the example of parents' spontaneous comparisons , parental accountability was very low, as the teacher had not requested information about family affairs. The fishing device strategy represented a middle position, as this practice gave the parents a choice as to whether or not to provide the missing information. As could be expected, the teachers' inquiries made the parents accountable for providing family information in a stronger sense, especially when the teacher asked multiple questions . What embedded understandings of privacy were thereby displayed in the eliciting practices? Undoubtedly, direct questions are more likely to challenge the family's social integrity compared to more indirect strategies. In asking questions, the teacher positions herself as being entitled to know the matter in question, but the use of prefatory statements or accounts, explicating the teacher's reasons for asking , as well as mending actions , can be seen as paying some attention to the issue of parents' privacy. An extensive example of boundary crossing was demonstrated in extract 4. In addition to the number of questions and the private content already commented on, the forms of inquiry adopted in the extract are also interesting. The shift from asking candidate questions to asking open when questions represented a shift from a less obliging question design to a more obliging question design. In CA research, open-ended question designs are considered to give speakers more freedom to formulate independent answers, compared to candidate questions where speakers are restricted to choose between ready alternatives or ready models of the world . However, this theorising presumes that professionals and clients have a shared project in obtaining accurate information about the client. If this premise does not hold , the shift to open-ended questions can be seen as dominating, rather than emancipating . While the less obliging candidate question design adopted in the first part of the extract allowed the mother to provide rather glossed information about the family's sleeping habits, the following open-ended question did not. Undoubtedly, the fishing strategy described in extract 2 represents a less intrusive action as the parents could choose whether or not to provide the missing information. When the fishing strategy is adopted, the implicit action is treated "as preferably not said" and, in her paper, Pomerantz relates this phenomenon to the management of privacy: We would argue that with respect to an orienting to privacy , this design lies somewhere in between absolutely respecting that right -for example, not considering asking, probing, or "fishing" and orienting to sharing . The "my side" tellings display an orientation to and acknowledgement of your right to privacy while not fully respecting it to the extent of no recourse . Thus, on the one hand, the ECEC teacher does not disturb privacy by asking them directly why the child was not present in kindergarten. On the other hand, the fishing design also suggests telling as an optional action . While exerting milder pressure, even the fishing strategy may be seen as having a control dimension. In this version, the teacher can be heard as orienting to "check" whether the parents had a legitimate reason for taking their child out of kindergarten on a particular day. Thus, exerting social control, or pursuing a concern, could be the implicit action "preferably not said" in this example. Other equally relevant candidates are asking out of sheer curiosity, small talk, snooping into other people's affairs, etc. The parental practice of volunteering family information and providing the pursued information indicates that parents actively orient to the norm of visibility in child rearing , rather than accentuating the negotiation of privacy. However, although complying with the expectations in talk, there are also subtle signs of ambivalence in terms of providing short or glossed answers and accounts that defend/excuse practices in the home . This ambivalence may indicate that parents have privacy needs but that they lack the discursive resources to defend the private-public boundary in educational settings. It is possible that the discourse of the best interest of the child, which suggests that parents are accountable for how they present themselves and the home environment, provides poor conditions for the negotiation of privacy. Another interpretation could be that parents orient to ECEC professionals as members of their primary group and that this kind of "kinning" may be functional in a post-traditional society. Thus, more qualitative research from the parents' perspective is needed in order to get a more in-depth account of parents' emotions and conceptions of privacy and boundary negotiations, for instance, examining whether there is a gap between the "desired" and the "achieved" level of privacy in ECEC settings. Moreover, interactional studies involving more informants are needed to establish what constitutes normal and what constitutes deviant conduct in parent-teacher conferences. It is hard not to perceive the multiple inquiries in extract 4, mapping and assessing the child's need for sleep, as a kind of policing . From an intersectional perspective, research should be conducted as to whether parents' ethnic, sociocultural background, gender or age may impact the way that teachers negotiate privacy in educational settings. Given the size of this study, we cannot confirm this claim at this point. However, the phenomenon of asking minority parents multiple questions has been observed in another small ECEC study , and immigrant parents have reported a sense of being evaluated . --- Conclusion While it may seem trivial to study talk in parent-teacher conferences, interaction in everyday encounters may nevertheless reflect larger issues with a moral and even a political dimension. The OECD's proposal to form a partnership with parents , which this study sheds light on, can ideally be portrayed in at least two ways: First, understanding ECEC staff and parents as an "integrated team" suggests that a high degree of sharing is taking place between the participants , and there is therefore little need to engage in public-private boundary work. In this scenario, staff and parents orient to each other in a hybrid and extended family fashion, and questions are normal and quite direct. . However, the relationship between ECEC staff and parents can also be understood as a "complementary team," in which parents and staff play different roles and have different expert domains . In this scenario, which suggests that there is some distance between the parties, there is a certain need for the negotiation of public-private boundaries. Thus, to take care of the family's privacy needs, inquiries should be properly accounted for, or alternatively managed in more indirect ways. In neither of the cases examined in this paper did the teachers' eliciting actions appear to be necessary to take care of the child's well-being, or at least the teacher did not explicate such reasons. Nevertheless, there were also signs of an awareness of the public-private negotiations: Questions were contextualised, fishing strategies were sometimes adopted, and two of the three teachers rarely asked for family information. While present ECEC practices might be slightly closer to a "complementary" than an "integrated" understanding of the partnership between parents and staff, there are probably voices in the transnational political landscape who do not welcome a development into the latter in which childcare is a shared responsibility in a much stronger sense. Ultimately, the management of privacy in educational settings deals with a nexus of related ideological and political questions: What is the value of family? Who is responsible for bringing up children ? What is freedom? What is citizenship? What is the good life?, etc. These issues are intertwined with social class and cultural diversity, suggesting different ways of relating to educational institutions. Both Norwegian and US class research suggests that middle-class parents are more oriented to a close homeschool cooperation than working-class parents. For the latter group, the family may still be more of a haven in a heartless world, meaning there is a greater need to keep ECECF staff at a distance . As referred to in the introduction of this paper, privacy is a human right, and it must be skilfully managed by ECEC staff. Thus, the competence and values of staff are of major importance in terms of realising "the real living law" in educational settings. The findings in this paper may have implications for teacher education which, not surprisingly, is primarily engaged with children's rights and children's citizenship . As previously mentioned, privacy presumes a segregation between the family's "frontstage" and "backstage" , which may be at odds with children's rights . From this perspective, the very issue of family privacy may appear out of place or old-fashioned, especially for younger people . In-depth education on human rights and ethics can help prepare teachers for the complexity of the field, where conflicting norms need to be identified and managed on a professional basis. Thus, practitioners and student teachers need knowledge about the law, the principle of integrity and its exceptions, both theoretically and practically in an every day/communication context. Moreover, the findings in this study suggest that ECEC teachers' knowledge of the social aspects of privacy needs to be strengthened. The hybrid border between the private and the public, and the power of questioning described in this paper, indicate a need to develop professional standards about when to ask for family information and how this can be accomplished in everyday action. Working with authentic interaction data from the field could be one route to enhance professional reflection in kindergarten and in ECEC teacher education. ---

Families have a right to privacy, but we know little about how the public-private boundary is negotiated at the micro level in educational settings. Adopting ethnomethodology, the paper examines how talk about the home situation was occasioned and managed in ten parent-teacher conferences in early childhood education and care (ECEC), with a special focus on the ECEC teacher's strategies for eliciting family information. The paper demonstrates a continuum of interactional practices which, in various degrees, make parents accountable for providing family information. The analysis shows that parents both volunteer and provide the pursued information, thus actively orienting to the norm of visibility in child rearing. However, although both parties orient to questioning and fishing as business as usual, the parents' accounts sometimes had an excusing quality or they adopted a reserved communication style, suggesting a certain ambivalence as well. The paper outlines different ways of understanding the present partnership ideal in parent-teacher teacher cooperation with implications for the negotiation of privacy. The paper also addresses training, which can contribute to staff and student reflections on the management of the public-private boundary.The public-private boundary • The public family • Visibility norm • Eliciting practices • The power of questions • Telling my side • Fishing * Janne Solberg

Polarization is never being that easy. So, a single person can polarize millions easily. The main objective of study is to measure the Political Polarization on Social Media. This study is specified in Pakistan. I also study how Political Parties use Social Media to There are two hypotheses that we are studying in this research. First is Social Media causes polarization in society. My first hypothesis discusses that Polarization is a threat to democracy. The real problem is that social media polarization is going unaddressed. The division which is done by Social Media is in favor of the Politicians in elections. So, it directly influences democracy. This Polarization is now worse. In hypothesis, we measure Political polarization. In which we discuss the Social Media Algorithms. And different social media Algorithms which are filtering content and creating echo chambers. The boundary between different websites is becoming thick and thick day by day, which should be thin. So respondents can easily jump into other walls and expose to diverse ideas. The second hypothesis discussed that social media's new algorithms are killing diversity. That diversity is the main characteristic of social media which is fading day by day. Our second hypothesis also study is that really we want that new social media algorithm or these are just there to facilitate the big companies. The next question is that if Political Polarization exists how we can reduce or minimize its effects. My studies focus on just Social Media. --- Social Media and polarization Social Media is the new world with new dynamics. The main feature of Social Media is the diversity and plenty of mix bag information is available on the Social Media. When you are using Social Media you were interacting with new ideas. But in the last few years, lots of things are changed. Now internet filters are common things and Social Media platforms itself changing their Algorithm. At first, there is a porous boundary between websites. So users can easily jump into the other side of the opinion. But now that wall is going dense day by day. So, it kills the main feature of Social Media which is diversity. So, result in the user interacting with the world which is not new, it just reinforces their preexisting beliefs. Also, it further results in different blocks in the society. These different blocks have no common ground to stand. Social Media dividing the society day by day. Social Media fuels Polarization. There is also another side of Social Media that binds the world together. They argue that Social Media is the Platform in which people debate about different topics and issues. The world is a global village because of Social Media. We are not socially isolated even when we are at our homes. You feel connected to the world when you are on Social Media. So there are different opinions about Social Media and its Polarization. Many studies explain polarization exists on Social Media. And some also argues that Social Media is connecting the ideological Partisan peoples. They conduct different researches on different topics and explain the Polarization between Republicans and democrats. They measure Polarization on the bases of the following topics. And there are many more topics like that which explain polarization among these two Political Parties. They have different opinions just because they are following different parties. The above studies suggest that Social Media creates Polarization by filtering Algorithm and creating Echo chambers. We still can't ignore the fact that Social Media connects the People. Founder of the world phenomenon Facebook, said, "More than 175 million people use Facebook. If it were a country, it would be the sixth most populated country in the world." There are also studies available which talk about which kind of the content cause Polarization. The content which causes polarization is the emotional and fake content. This is also fact that emotional fake content is more popular on the Social Media as compared to the other content. There also a question rise why Social Media Algorithm promote fake and emotional content. The above journals study different aspects relating to Social Media polarization. --- Theoretical Foundations 3.1. Mediated Deliberation In Pakistani society, it has been noticed that because of social media in Pakistan the deliberative quality of media has increased. Pakistani mainstream media is analyzing and discussing on different political issues. Keeping in view above mediated discussion of deliberation, it may be expected that for strengthening the democracy in Pakistan media professionals should move forward towards deliberation. This dissertation provides a platform to gauge whether Pakistani media is deliberative or not. So, the deliberative democratic theory may be observed by placing the news media in middle of public and media professionals. Page & Shapiro, 1992;Habermas, 2006; discussed process of voters based upon the discussion of the media professionals. For rational thinking of voters, media discussions quality should fulfill the key features of mediated deliberation and it paved the way for the right voting decisions and choices of voters . Interpersonal discussions reduce uncertainty about political issues. News media encourages mediation and discourages it for others public forums .Mutz concluded that deliberative democracy, which is realized by political discussion with heterogeneous others, might not coexist with participatory democracy. Nevertheless, it is recommended that the discussion measure, which only taps into similar others, is still a useful predictor of political participation because voters are inherently prone to engage in a political conversation with homogenous others, such as friends and family . Contrary to deliberative theory assumption, discussion increases the voters' belief that is motivated by self-interest. Political deliberation and mediation is major part of dissertation which argues that interpersonal discussion process with peers and relatives regarding politics is very important to understand political issues. Several scientific studies have reported that public discuss about politics to colleagues, families and friends. During elections these discussions mediate with political news that affect voters' behavior regarding knowledge about political issues, interest in political activities, and participation in vote casting process. With the facilitation of news media exposure, the process of deliberation and political discussion can be elaborated by analyzing the media content and surveying the audience. Interpersonal political discussion is regarded very natural, instantaneous phenomenon. People communicate with peers and friends, colleagues at job place. Political mediation occurs as routine activity. This dissertation argued that voters discussed political happenings in a daily life. They also get exposure with political information and knowledge from interpersonal political discussion and also through news media exposure. Deliberation of the political news exposure may provide them awareness regarding political issuesand vote casting. RQ1: How doessocial media polarized the society? As media consumers have more choices to expose themselves to selected medium and media contents with which they agree, they tend to select the content that confirms their own ideas and avoid information that argues against their opinion. People don't want to be told that they are wrong and they do not want their ideas to be challenged either. Therefore, they select different media outlets that agree with their opinions so they do not come in contact with this form of dissonance. Furthermore, these people will select the media sources that agree with their opinions and attitudes on different subjects and then only follow those programs. --- H1. Social media causes polarization in society Most of the studies are conducted in America and Europe. There are very few studies that are conducted in Asia. There are hardly any studies conducted on Social Media in Pakistan. My study is related to the previous studies and measures Political Polarization in Pakistan, where politics is a hot word and people have very diverse and interesting opinions. My study not just measure Political Polarization caused by Social Media but also educate how we eradicate the differences which cause by Social Media. Every political party publicized their achievements and highlighted the failures of others from television channels, radio, social media and newspapers. People were encouraged to cast their vote as every single vote is critical to change their fate. Media played a crucial role to mobilize voters by inducing among them political awareness. Envisioned the scope and influence of mass media several scholars in them researches incorporated it among the various factors impacting voter's choices. The assumption regarding media's influence strengthened after finding a positive correlation between political news and voting patterns. --- Method and Measures Survey was conducted both in Urdu and English language for easiness of respondents. The researcher draws a set of sample on the basis of gender and from the rural and urban combination. --- Findings and Discussion We found that social media causes polarization in society. In recent years Social Media is new norm and it is now part of life for many. Propaganda on Social Media is common know. This main purpose of this propaganda is to polarize the society. This division is always existed but the emergence of Social Media it is now very easy to divide. So, social media fuels Polarization.

The majority of People are Polarize online in Muslim countries of world. It is very easy to create polarize groups with the use of Social Media. The purpose of the study to measures Polarization and the echo chamber feature of Social Media in Islamic Republic of Pakistan. The results suggested that Social Media Polarization extreme form is existed and also confirmed that Social Media new Algorithms creating echo chambers. The interesting results suggested that our respondent wants diversity not echo chamber. People are politically Polarize because of fake and emotional content. It is the responsibility of the Social Media companies to watch for this type of content which makes people divide socially, religiously, and politically. Keywords: Social media, political polarization, religiously, fake and emotional contentSocial Media connecting the People and it also made people hate each other. The condition is now worse because now it is very easy to propagate polarize views with Social Media. Now there are a lot of Social Media tools are available at the market. So, they can use many accounts at once. The main purpose of these groups is to spread fake and junk news. These fake and junk news acts as fuel to the Polarize groups. Social Media promote that content. It is because of the Algorithm of social Media. The Algorithm of Social Media provides individualized experience to the users. So they never get a chance to know the other side of the story. We are all of us enveloped inside our own personalized bubbles (Lau & Akkaraju, 2020). Many news media affect voters' behavior and voters' turnout (Teixeira, 1992). Rogers ( 2002) discussed role of news media in agenda setting and effects on voting choice. Social Media has tons of mixed bag information. Information is overloaded on the internet. As we above discussed Social Media Algorithms are biased. We have to know which political party or company is doing this. And they divide the whole society. Polarization is not limited to the individual or some distinct group. It can affect the whole country's political system. What if people have no harmony and common ground for people to stand? So, this Polarization can be a threat to democracy. In today's perspective when the situation of coronavirus and people are angry against the government. The state should have to consider that social media data which polarizes society. There also many other questions are raised by many researchers like "Can Democracy Survive the Internet?" (Persily 2017). Social Media provides its user an opportunity to publish and communicate in whatever way they like, or whatever they want to, quite easily. Due to this ease of access, the internet users are able to create a huge amount of data. You Tube users upload Seventy-Two (72) hours of videos every minute. 340 million tweets are created on Twitter, by the users. Bloggers submit 500,000 new posts, on wordpress.com, and these posts receives 400,000 comments on daily basis (Jeffbullas). Social Media is not just used by young and agitated, it is now global and used in every corner of the web. 72% of the internet users are active on social media, 89% of the users are of 18-29 years old, 71% internet users access social media from the mobile. 1.15 billion Internet users use Face Book, 70% of marketers uses Face Book to approach customers. Over 1 billion users uses Google+, with 359 million monthly active users, and on Twitter, there are over 550 million registered users, with 215 million active users (Jeffbullas).71% adults use Face Book, 17% use Instagram, 21% use Pinterest, and 22% of the online adults use Linkedin. 74% of women and 62% of men use social networking sites. It is also revealed that Face Book users are more politically engaged than most people (Pewinternet). Despite of low penetration rate against the advance countries, Social Media is becoming a powerful tool for information sharing in Pakistan. According to the statistics, the number of Social Network Users in Pakistan is 8007460, and total Internet users are 291298970. 8.5 million Users in Pakistan use Face Book, 28% of the Internet users use Linkedin (1.5 million), Instagram, and Twitter. Nokia, OLX and Ufone have over 1 million fans on Face Book (Reader). Social Networking sites have become an important tool for Politics. This medium is an important source for news, finding likeminded people, information, and also to approach voters in case of elections (Pewinternet). According to a Report, 22% of registered voters have let other know, on a social networking site, how they voted. 30% of the registered voters have been encouraged to vote for their Favorite Candidate on Social Media, in American General Election (Rainie, 2012). In United States Election 2008, Presidential Candidate Barrack Obama focused mainly on social media, and after that successful campaign, public has a new way to participate in the political process. The use of Social Media for political purpose is not restricted to the United States only, but also in other movements, like the Arab Springs, and the protest in Turkey (Baig, 2013). Polarization is not limited to a single person. So, it is very easy to create polarize groups with the use of Social Media. Now we can easily create many social media accounts for free and promote polarize opinions easily. These many accounts can handle easily with one account. They are different bots sharing fake and emotional content on their accounts. Most viral content on social media is that fake and emotional content.

INTRODUCTION Injuries cause significant mortality and morbidity worldwide, accounting for almost 10% of world deaths, and are the leading cause of death among men aged 15-49 years. 1 Injuries carry considerable social and economic ramification, particularly in low and middle income countries. [2][3][4] This makes them highly relevant to the development agenda and to poverty-related issues such as maternal mortality and childhood illnesses that for long have been the focus of public health initiatives in those countries. Effective injury prevention requires the identification of vulnerable groups and specific settings where maximum benefit can be reaped of targeted prevention interventions, healthcare provision and research. Non-fatal injury patterns in particular point to groups that are at higher risk of spending time with disability, with all the ensuing social and economic consequences. Injury prevention principles are universal, and the effectiveness of many interventions has been established. [5][6][7][8] However interventions are delivered at national and subnational level, and cross-country as well as withincountry variations in vulnerability to injury are expected. For example, while road injuries in Tanzania were more common in urban areas, 9 there was no evidence of urban-rural differences in Nigeria after accounting for other sociodemographic variables. 10 Methodological differences could partly account for cross-country differences. However, genuine differences in social and environmental conditions that largely determine the risk of injury are expected to contribute considerably to such variation, which warrants investigating country-specific patterns. Sudan is a middle income African country where population-based epidemiological injury research is scarce. A study in Khartoum state, which hosts the capital city and the major urban centre of Sudan, revealed gender, socioeconomic and educational differentials in non-fatal all-cause injury, 11 yet, little is known about cause-specific injury patterns. These are critical because vulnerability patterns are expected to vary by injury cause. Thus, this study sought to identify the broad population groups that are vulnerable to non-fatal injury in Sudan by exploring the association between key sociodemographic variables with injury from various causes using national survey data. --- METHODS Design This is secondary exploratory analysis of existing data from the Sudan Household Health Survey 2010, a national cross-sectional survey. --- Settings Sudan is a middle income African country with a population of more than 30 million, with 70% living in rural areas. Of the population 43% is aged less than 15 years and only 5% is aged 60 years or more; 46.5% of the population are under the national poverty line. 12 Overall, men have higher labour force participation than women across all ages. 13 The main occupations in urban areas are professional and managerial occupations , elementary occupations; for example, street vendors and cleaners , and crafts and related trades; for example, construction workers and mechanics . In rural areas, 40% of the economically active population are agricultural, forestry or fishery workers, two-thirds of whom are men and a third are women. 15 Political and tribal conflict had plagued Sudan for decades, with relatively recent unrest in the west and south. --- Data Sudan Household Health Survey 2010 had a sample size of 83 510 respondents, drawn from the 15 states of Sudan using a two stage cluster sampling design. In each state, 40 clusters were randomly selected with probability proportionate to size and 25 households were randomly selected from each cluster. 16 We developed a short injury module with four questions administered in Arabic to the main household respondent, asking whether an injury was experienced over the 12 months preceding the survey, and for the most recent injury, the time, cause, type of healthcare received in the 1st week and whether injury resulted in disability. Causes were mutually exclusive categories defined according to the tenth revision of the International Classification of Diseases and Related Health Problems . 17 . The study was approved by the Federal Ministry of Health Research Ethics Committee and informed consent was obtained from the respondents. Field work took place from March to May 2010, with a response rate of 99% of households. --- Variables Independent variables used to define sociodemographic groups were age, gender, area of residence at time of interview, socioeconomic status and highest education attained. Age was aggregated into five groups , with potentially varying risks of the different causes of injury. 1 Further age aggregation for injury by fire or hot substance, poisoning and mechanical causes was necessary. Socioeconomic status was represented by the wealth index score based on household assets and characteristics at the time of the interview. It was classified into tertiles separately for urban and rural areas to ensure that poverty was defined relative to the local context and to avoid misclassification of some households in rural areas as poor, the potential of which was indicated by different wealth score distributions in the two areas. 18 Highest education attained was aggregated into none, primary and secondary or higher, for those aged 15 years and over. To ensure that the construct of inappropriately low educational achievement was properly captured, the mother's highest education level, a known determinant of injury, 7 was assigned to those aged less than 15 years, as their low education levels would be appropriate for their age. The highest education of the head of the household was used where the mother was absent. Dichotomous indicator variables for the cause of the most recent injury; RTC, poisoning, fall, mechanical , fire or hot substance , animal bite or venom and assault were used as dependent variables. Causes with small numbers; electric shock , near-drowning and intentional self-harm , as well as injury from unspecified cause , were not considered. Complications of medical or surgical care were also excluded due to uncertainty in the performance of self-reports in this category. --- Analysis Frequencies and percentages were used to describe the sociodemographic characteristics of the survey respondents and the injury cause distribution. The independent association of sociodemographic variables with each cause of injury was examined by multivariable Poisson regression analyses with robust variance, which estimates the directly interpretable prevalence ratio . This method was recommended for the analysis of crosssectional surveys as an alternative to the estimation of the OR with logistic regression. 19 One set of models included injuries that resulted in any form of healthcare in the 1st week. Presumptively non-significant injuries that did not lead to healthcare utilisation in the 1st week were excluded. There are considerable urban-rural differences in the social and physical environment, and urban-rural differences in injury vulnerability have been previously reported. 9 11 20-22 Thus, interactions between the area of residence and each independent variable in the model were tested. As socioeconomic differentials in childhood injury are well documented, 23 interaction between age group and socioeconomic status was also tested. Statistically significant interactions were retained. Another set of main effects multivariable models was limited to injuries that resulted in hospitalisation . For the latter, two age groups were used because of the smaller numbers of injured people at that level. All analyses excluded records with missing values for at least one variable. PR was statistically significant at the 0.05 level if its 95% CI excluded one. The analyses being exploratory rather than confirmatory, adjusting the cut-off p value for multiple hypotheses was not attempted. The data were prepared in PASW V. 18 --- RESULTS Table 1 shows the socio-demographic characteristics of the survey sample. Out of 83 482 who responded to the injury question, 1626 respondents had an injury over the 12 months preceding the survey. A total of 1444 were injured from the causes included in the analysis; 1220 received any form of healthcare, 204 of whom received inpatient care. Table 2 displays the distribution of injury causes in the survey population. Falls were the most common cause of injury that received any form of healthcare , followed by animal bite or venom and RTC . RTC was the most common cause of injury that received inpatient care followed by falls . The sociodemographic distribution of injuries that received any form of healthcare varied by cause of injury . Adjusted PR, with 95% CIs for each cause leading to any form of healthcare, are displayed in table 4. Falls were more likely to affect men ) and those aged 5-14 years, 45-54 years and 65 years and over than children under 5 years, with the highest PR among those aged 65 years and over ). Falls were associated with socioeconomic status in rural areas but not in urban areas. Animal bite or venom injuries were more likely to be reported in rural areas than in urban areas , and in all other age groups than in children under 5 years. Men and urban residents were more likely to report road traffic injuries than women and rural residents respectively. People aged 15-44 years, 45-64 years and 65 years were at higher risk than children under 5 years, as were those from households in the middle wealth tertile than those in the poorest tertile . Men were three times as likely as women to suffer injury from assault, while the richest third of the population were about 60% less likely to experience such injury. Assault was also associated with age, with PR up to 11 in those aged 65 years and over compared with children under 5 years. Men were also more likely to have injury from mechanical forces which were also independently associated with age and highest education attained. Risk of injury by fire or hot substance was 70-80% lower in older age groups than children under 5 years. Poisoning was more likely to be reported in people with primary education compared with those with no education. There were no statistically significant interactions between age and socioeconomic status for any of the injury causes considered. Table 5 shows adjusted PR with 95% CIs for each cause of injury that received inpatient care in the 1st week. Those due to RTC were associated with gender , area of residence and age --- DISCUSSION --- Summary of findings There were discernible vulnerability patterns of non-fatal injury, fairly generalisable to the national level in Sudan, given the high household response rate. Men and elderly people were the most vulnerable to injury from RTCs, falls, assault and mechanical forces. Children under 5 years were the most likely to suffer injury due to fire or hot substance. Assault was concentrated in people of low socioeconomic status. RTCs and falls affected urban residents more than rural residents, while the latter were more affected by injury from animal bites or venom. Socioeconomic status was associated with falls in rural areas only. Most of the differentials in RTIs, falls and assault persisted at the level of injuries that required hospitalisation. --- Limitations of the study The study used self-reported data which are subject to reporting errors. Another major limitation of self-reported injury data is recall bias. An in-depth analysis of recall bias showed injury among children under 5 years to be particularly under-represented. Thus, lower risk in children under 5 years should be interpreted cautiously. The area of residence was at the time of interview and could be different from that at the time of occurrence of the injury. This is particularly relevant to RTIs, as †Because interactions between wealth index tertiles and area of residence were statistically significant they were kept in the falls regression model. Thus the coefficients for these main effects represent the coefficients in the reference group of the interacting variable; the coefficient for area is that of the group in the poorest wealth tertile in rural areas, while those for wealth index tertile represent those in urban areas. In a subgroup analysis, prevalence ratio for the richest tertile of the population in rural areas was 0.6 95% CI 0.4 to 0.8, while there were no statistically significant differences between wealth tertiles in urban areas. they are linked to mobility, and people are more likely to have been injured far from their area of residence than for other injury causes. Nevertheless, the finding of higher vulnerability in urban settings concurs with well-documented link between road injury and urbanisation. 5 Wealth index was measured after occurrence of the injury, which should be considered when interpreting socioeconomic associations. Using hospitalisation as a proxy for the severity or impact of injury may limit the conclusions drawn for some of the associations due to differences in healthcare access and healthcare seeking behaviour. --- DISCUSSION OF FINDINGS Road traffic crashes RTCs were the leading cause of injury resulting in hospital admission, and similar to findings in Tanzania and Nigeria, men and urban residents were more vulnerable. 9 10 Age differentials were also similar in Nigeria, but the socioeconomic differentials observed here were not evident in the other countries. The findings for RTIs mirror what is expected from their known risk factors. 5 Men in Sudan are more economically active than women. 15 Their higher mobility therefore makes them at higher risk of RTIs. Men are also more likely to drive and to engage in risky driving and pedestrian behaviour. 25 Higher mobility could also explain the higher vulnerability of those aged more than 15 years. The higher risk among higher socioeconomic group, diminishing at hospitalised injury level, conflicts with the suggested link between poverty and RTI mortality and morbidity. 26 Whether a poverty-mortality link exists in Sudan, generating this cross-sectionally observed finding through selective survival of those of higher socioeconomic status, remains to be confirmed. --- Falls Age differentials in falls risk are not surprising; older people are more prone to falls due to biological factors and chronic conditions, 27 28 and similar age and gender patterns were observed in Tanzania. 9 A survey with similar methodology in Mongolia revealed similar predominance of falls among school aged children as observed here. 29 Urban-rural variations in fall circumstances were reported in other low and middle income countries where falling from trees or in farms and roads was more common in rural areas. 9 22 In the rural outskirts of Khartoum state, the majority of non-fatal falls were from the same level or heights lower than 2 m. 11 Whether this is also the case in other rural parts of Sudan and whether a link with agricultural activities may explain the association of falls with poverty only in rural areas needs to be investigated. --- Animal bites and venom The link of snake bites with agricultural activities and with rural areas in developing countries has been previously highlighted. 30 31 Thus, predominance of agricultural activities in rural areas and the different natural and built environment compared with urban areas may explain the higher risk of injury by this group of causes there. Differentials in care-seeking behaviour may explain the higher risk of hospitalised injury from that cause among those with higher education achievement. --- Assault Assault is more subject to under-reporting than other injury causes, particularly among women, 32 33 which could partly explain the observed gender difference. The study revealed a socioeconomic differential that concurs with the welldocumented violence-poverty link attributable to a range of factors such as income inequality, unemployment and material deprivation, operating at individual and contextual levels. 6 34 35 The situation in Sudan is complicated by continuing conflict, with substantial internal displacement that perpetuates poverty. 36 It also tops a pattern of inequitable spending on healthcare where low income families spend the same proportion of their income on healthcare as higher income families. 37 --- Mechanical injuries Unintentional injury from mechanical forces may include a considerable proportion of occupational injuries. For example, data from Tanzania suggests that workplace is one of the most common place of mechanical injury in urban areas, while farms were the most common place in rural areas. 9 There are no similar details for this group of injuries in Sudan, and an investigation of the place of such injuries is required to confirm if occupational risk could explain why men and adolescents/adults of working age were particularly vulnerable. --- Poisoning Acute poisoning is more likely to have been captured by the survey, being more recognisable than chronic poisoning. Heterogeneity in the range of possible causative agents, such as household cleaners, agricultural chemicals and hair dye hinders the interpretation of the observed educational differential, but could also explain the lack of association with the other sociodemographic variables. Clearer vulnerability patterns may exist at causative agent level, and could be revealed in a study that identifies those agents. --- Fire or hot substance Similar to the situation in other parts of the world, 7 children under 5 years in Sudan were the most likely to suffer burns with fire or hot substance. This may point to lack of child-proofing of cooking amenities or of the handling of hot or flammable material in the home, where the majority of childhood burns occur. 38 39 For example, in rural Bangladesh, housing with easy access to cooking areas and use of the traditional kerosene lamp was associated with childhood burns. 40 Evidence from Sudan suggests that hot liquid or steam is the main cause of burns, which were also a major form of home injuries. 11 An investigation to confirm if a similar association with housing conditions exists in Sudan is thus warranted. --- CONCLUSIONS Men in Sudan were vulnerable to more than one injury cause, requiring special consideration in a country where they contribute the most to families' financial stability but where their health receives little attention. Elderly people, children under 5 years and people of low socioeconomic status are universally vulnerable groups that are already affected by communicable and noncommunicable diseases but were also differentially affected by injury from various causes. Thus, existing disease prevention and health promotion programmes in Sudan should expand to include safety elements targeting those groups, within an overarching multisectoral injury prevention policy. Longitudinal studies are needed in Sudan to confirm the vulnerability patterns observed and whether they also apply to fatal injury, and to investigate the context-specific behavioural, social, environmental and occupational risk factors and injury circumstances that shape them. What is already known on the subject ▸ Various patterns of sociodemographic differentials in injury morbidity were reported from low and middle income countries. ▸ Limited research from Sudan revealed gender and socioeconomic differentials in all-cause non-fatal injury. What this study adds ▸ In Sudan, men and elderly people were at higher risk of a multitude of injury causes including road traffic injuries and falls which were major causes of non-fatal injury. ▸ Children under 5 years were at higher risk of non-fatal injury due to fire or hot substance and interpersonal violence was concentrated in people of low socioeconomic status. Urban residence was associated with road traffic crashes and falls, while rural residence was associated with injury from animal bites or venom. ▸ Heterogeneity in vulnerability at subnational level was evidenced by an association of falls with socioeconomic status that was limited to rural areas. --- Collaborators Sudan Household Health Survey National Management Team: Dr Mohammed Ali Yahia Elabassi, Dr Igbal Ahmed Bashir, Dr Muna Hassan Mustafa, Miss Nahed Abdelgadeir, Miss Amal Khedir, Miss Rania Moawia, Dr Khalda Khalid, Dr Lyla Mahdi, Miss Alawia Ali, Mr Ali Ibrahim Ali, Mr Adil Mohammed Ali, Miss Raja Babiker, Miss Nisrin Elamin, Mr Mohammed Ahmed, Ms Nuha Khogali. Contributors SA contributed to the development of the injury module, conceived the idea of this paper, conducted the analysis and interpretation and wrote the manuscript. --- Competing interests None. Ethics approval Federal Ministry of Health Research Ethics Committee. Provenance and peer review Not commissioned; externally peer reviewed. Data sharing statement The Sudan Household Health Survey 2010 data will be deposited by the National Management Team at the Sudan Central Bureau of Statistics from where it can be requested. --- Women win in UN negotiations on guns The 5th Meeting on States on Small Arms has committed all UN members to promote the participation of women in policy, planning and other projects related to guns. Source: Rebecca Peters, The International Action Network on Small Arms, 20 June 2014 .

Background Successful injury prevention requires identification and targeting of particularly vulnerable groups. Little is known about injury vulnerability patterns in Sudan. This paper aimed to fill this gap using survey data. Methods Data from the Sudan Household Health Survey were used. This was a national cross-sectional interview survey of 83 510 individuals selected by multistage cluster random sampling. Multivariable Poisson regression was used to investigate the association of cause-specific injury that received care by traditional healers, outpatient care and inpatient care, and those that received only inpatient care, with age, gender, area of residence (urban or rural), socioeconomic status and education. Relevant interactions were tested. Results Independent of other sociodemographic variables, men were at higher risk of road traffic injury ( prevalence ratio (PR): 3.3 95% CI 2.4 to 4.7), falls (PR: 1.5, 95% CI 1.3 to 1.9), assault (PR: 3.0 95% CI 1.8-5) and mechanical injury (PR: 2.0 95% CI 1.2 to 3.1) that received any form of healthcare. Those aged 65 years and over also had the highest risk of those injury causes, while children under 5 years were the most likely to suffer burn injuries. Socioeconomic status was associated with assault (PR for the richest group 0.4 95% CI 0.2 to 0.8). Vulnerability patterns for injury that received inpatient care were fairly similar for some causes. Conclusions In Sudan, existing disease prevention and health promotion programmes should expand to target men, children under 5 years, elderly people and those of low socioeconomic status with injury prevention interventions. Further research is needed to investigate the context-specific proximal risk factors that shape the various vulnerability patterns observed.

Introduction Whilst not always acknowledged as such, video games are a pervasive and significant aspect of modern culture. According to the Interactive Software Federation of Europe, 32% of all adults in the UK describe themselves as 'gamers' , while one child in two plays games every day . As an industry, games generate as much revenue asor more thanthe film or music business, with blockbuster releases such as 2011's Call of Duty: Modern Warfare 3 out-grossing the last four Harry Potter films combined . The proliferation of internet-enabled smartphones has fuelled growth in social gaming, and expanded the 'gamer' demographic. Further, this demographic group is more diverse than ever before, and is no longer dominated by young males. In their annual survey of 2000 US households, carried out over a period of several weeks in early 2013, the Entertainment Software Association reports that women aged 18 or older represent a greater portion of the game-playing population than boys aged 17 or younger . In the UK, The Telegraph reported in July 2013 that the typical video game player is 35, has one child and earns more than £23,000 per annum, based on a survey of 2000 self-identified video game players carried out on behalf of the video games social network Pixwoo.com . The study of games, and their effects on the broad demographic group who play them, has never been so pertinent. The learning potential of video games has, in recent years, received some considerable academic attention . Researchers including Jenkins , Gee and Squire have been particularly influential in establishing the role video games might play in our educational and cognitive development, with a recurring theme that modern video games are complex experiences: to play them, players must comprehend this complexity, and develop an understanding of the in-game systems. The work described here sought to examine how players work together to learn from one another about the systems found in their favourite games and how they engaged with social mediawikis, in particularto facilitate this collaborative learning. However, in engaging with social media and game-based research, a number of challenges and concerns were presented: these, too, are discussed in some detail. --- Gaming and social media Social media enjoys a close, almost symbiotic, relationship with video game culture. In fact, one might argue that the networks that facilitate competition, cooperation and communication between playerssuch as the PlayStation Network i , Xbox Live ii or Steam iiiare examples of niche or special interest social networks. Outside of these relatively closed communities, however, there exists an abundance of player-generated, moderated and curated gaming information on the Web, much of it exposed via social media channels including wikis, blogs, Facebook and Twitter. Video games increasingly feature direct integration with social networks, with games such as Infinity Ward's Call of Duty: Modern Warfare 3 allowing players to log into their Facebook accounts from within the game and invite their Facebook friends to play along, to organise clans of players and to post updates on their in-game progress . Many other games post game progress updates to external social networks, with Relentless Software's Buzz Quiz World, posting 'stories' to the player's Facebook account, for example, and Naughty Dog's The Last of Us providing players with the option to populate their in-game factions with Facebook friends, as well as simply checking on their current activity. Further, Facebook itself has developed into a platform for games with a social dimension, with titles such as Farmville, Candy Crush Saga and Words with Friends, tapping into Facebook users' existing social connections to drive their rapid proliferation . In 2012, Facebook released data that suggested their platform boasted 235 million game players . On its own, such a figure comfortably eclipses the number of subscribers to more established gaming networks, such as Microsoft's Xbox Live service, which in April 2013 reported 46 million subscribers worldwide . While the Facebook figures provide little detail it is difficult to ascertain how regularly these 235 million users actually play games, for examplethere is little doubt that gaming on social media platforms is a significant phenomenon. --- Video game wikis The wiki, ranging from tailored installations of the MediaWiki software that powers Wikipedia, to third party-hosted sites such as wikia.com and wikidot.com, has become the de facto online reference medium for game players. While less obviously social than Facebook, Twitter, or online discussion forums, wikis nonetheless fall under the banner of social media, and typically support an active core of contributors, with a larger proportion of the community consulting the material created and curated by this small central team. On video game wikis, community-spirited gamers meticulously document their gaming experiences for the benefit of others, from simple guides to complex theories and strategies . While gaming-related wikis are generally considered to be non-academic in terms of their content, those who contribute to and make use of such websites might be described as engaging in inherently scholarly behaviour. It could certainly be argued that in collating, synthesizing and disseminating the often complex behaviours observed in a modern video game, the wiki author is displaying academic proficiency in a nonacademic field. It is the practices of those engaged in populating and using gaming wikis that are discussed here. Such wikis are, it may be argued, some of the most tangible manifestations of gamers' pseudo-academic engagement with the gaming world, made all the more interesting by the collaborative and social processes that underpin the wikis' creation. For non-gamers or, indeed, casual gamers who have never consulted these online resources, it might appear somewhat improbable to suggest that these wikis or, rather, the endeavours of those who contribute to them, are of academic value. The quality and scale of these wikis may be illustrated by means of the following examples, which demonstrate the lengths to which gamers will go to document their activities for the benefit of their peers. Bay 12 Games' Dwarf Fortress, for PC, is not a mainstream title. Lauded for its "randomly-generated worlds , gruesome combat mechanics and ubiquitous alcohol dependency", it is designed and developed by a two-man team, relying on player donations for support. What the game lacks, subjectively, in terms of aesthetic nuance and marketing muscle, however, it more than makes up for with its innovative take on the role-playing genre and its steep learning curve. The game's challenging difficulty and somewhat impenetrable user interface have resulted in a loyal cult following which maintains an exhaustive wiki devoted to the title. Take, for example, a flowchart which encapsulates one possible approach to attaining self-sufficiency for the player's in-game fort iv . The subject matter might be considered trivial, but the chart is meticulous in its construction and suggests that its creator has thought carefully about the aspect of the game system he wishes to model, and how to present that model to a wider audience. The Dwarf Fortress wiki v currently comprises 1812 articles, a comprehensive system for quality assurance, and a 'Centralized Discussion' area, where consensus is reachedapparently with successon a range of content-related issues. In many ways, this seems the very model of a collaborative academic project: the data are carefully curated, articles are properly debatedpeer reviewed, one might sayand the information is presented in a lucid and coherent style. From Software's notoriously unforgiving Demon's Souls and its spiritual sequel, Dark Souls , are games seemingly designed to invite users to document and share their ingame findings and winning strategies. The games feature scant guidance to new players in the form of tutorials or training levels , precious little exposition and virtually no explanation of the somewhat arcane in-game systems that govern player progression. The games are also widely considered to be some of the best titles that the current generation of gaming hardware has to offer vi . Much of the satisfaction derived from playing these titles is found in overcoming a series of hugely challenging foes, but the games' many fans also revel in the need to consult the wealth of player-curated data and game world lore that is collated in the titles' dedicated wikis. A Demon's Souls or Dark Souls player with the relevant wiki open on a laptop next to them as they play may appear to be cheating, but, especially given that the games cannot be paused, this arrangement might be reasonably likened to sitting an open book examinationanother 'academic' analogy. Dark Souls was chosen as the focus for this project precisely because its more opaque qualities encourage players to consult or contribute to online resources such as wikis, as they seek to make sense of the world presented by the game. Collaboratively-edited, wiki-based guides now exist for most modern games and, as the examples here illustrate, represent a significant and perhaps overlooked application of a social media platform that warrants closer inspection, particularly if those engaged in creating and consulting the wikis are engaged in a form of learning. Incorporating features that one might expect to find in more academic environs, the best gaming wikis are highly structured, and provide multiple routes through the information, a variety of appropriate media and a lively discussion element. From the point of view of researchers and academics, increasingly expected to engage with a wider audience, it might be that we can learn from these 'non-academic' projects and the means by which they manage the collaborative process. Certainly, the Dwarf Fortress wiki users' ability to reach consensus is a quality some academic communities might envy. --- Research questions This study's primary research question was: given the near-academic quality of much of the work, and the effort required to produce it, what motivates players to contribute to gaming wikis? And, related to this, what can we learn from how players interact and collaborate on gaming wikis that might be applied in more formal learning environments? While not central to the study, the interviews also provided an opportunity to explore how the collaborative process was managed in the gaming wiki context, and to ask those who contribute to such resources whether they felt there was any academic value to their efforts. --- Methods At the core of the project was a series of online interviews with gaming wiki users to determine how they collaborated and learned from the wiki, as well as from one another. Interviewees were drawn from wiki readers and wiki contributors, founders and administrators, in order to gain their unique perspectives. While the majority of questions related to how and what interviewees learned about the game on which the wiki focussed, questions about how the experience of contributing to the wiki affected them were also asked of those who did so. --- One-to-one interview Interviews conducted with contributors to, and users of, the Dark Souls wiki at http://darksouls.wikia.com and were carried out in a number of ways. First, a one-to-one interview was carried out, in real-time, using the 'chat' or 'instant messaging' system built into the wiki. Figure 1: One-to-one interview using the Dark Souls wiki's built-in chat system Aside from finding a mutually suitable time at which to conduct the interview , this method of interview was straightforward to administer. Allowing the interviewer to copy and paste in questions from a prepared interview script while still responding to the interviewee's answers and the natural flow of the interview, it also provided an opportunity to hone the questions to be used in subsequent interviews. In addition, it provided the interviewer with a ready-made transcript. --- Group interviews The chat-based one-to-one interview was followed by a more novel group interview, conducted again within the confines of the wiki in question, but this time using its blogging feature. In this scenario, the interviewer posted small groups of questions as blog entries and asked the wiki's most active contributors and administrators to answer the questions by leaving comments beneath the relevant blog entry. This approach permitted participants to answer in their own time and, not unlike how a focus group can generate additional answers through respondents' interaction, allowed interviewees to reflect on one another's responses. As with the chat-based approach, blog posts and comment-based responses had the advantage of resulting in a ready-made transcript. While one might have hoped for some wider participation , the asynchronous group interview approach was arguably the most useful. On reflection, a degree of good fortune may have prevailed with regards the engagement of the Dark Souls wiki contributors, who remained committed to the interview process until its completion. While providing respondents with the flexibility to respond in their own time had potential benefits for engagement, the converse is also true, as no participant is obliged to complete the interview. Indeed, attempts to conduct a similar interview with another Dark Souls wiki were ultimately unfruitful, and a pre-arranged Skype interview failed to materialise. Such problems are common to research methods that involve remote participants, particularly where no incentive to participatesuch as cash reward or voucheris offered. Another issue that must affect online research is that of knowing with whom you are conversing. For the purposes of this project, for example, ethical approval was sought from the appropriate ethics committee at the researcher's host institution. The ethics application stated that no participant would be under the age of 18, and that all participants would be provided with a digital copy of an agreed consent form. The consent form, which detailed the purposes for which participant responses could be used and who to contact in the event that some issue should arise, was posted online, and linked to from the blog or chat page, as appropriate. However, it is difficult to ascertain that respondents consulted the form or, if they did so, fully understood its implications. Further, it is also impossible to know with certainty that respondents are over 18, which is somewhat more problematic. Both of these issues are exacerbated by the open, online nature of a blog-based interview: any registered wiki user can comment on the questions posed, and there is no guarantee that they have even viewed the initial post which contained the consent form link. Certainly, one unidentified commenter actually stated in one of their posts that they were in high school, so their responses cannot be used directly in any subsequent accounts of the project. There is a solution to this problem, and that is to seek ethical approval for all age groups, but this can be a more difficult and timeconsuming process, depending upon the nature of the relevant institution's ethics policies. More difficult to address, however, is the issue of knowing with whom you are engaging: without proof of identity, it is impossible to determine with any confidence just who your respondents are. --- In-game interview A third research method was also trialled, interviewing a wiki user within a game of their choosing. The interviewee opted to be interviewed via his PlayStation 3 console, and suggested Media Molecule's LittleBigPlanet 2 as the title to play as we conducted our interview, largely because the game was available to both interviewer and interviewee through Sony's PlayStation Plus subscription scheme. The interviewer had access to a digital recording device, designed specifically to capture the audio and video output from a games console such as the PS3. However, the process of recording the interviewa necessary step for later transcription, especially as both interviewer and interviewee were expecting to play the game as the interview progressedwas not straightforward. First, when engaged in in-game audio chat, only the sound from the game itself and the sound of the remote player's voice are actually played through the games console's audio output. In order to eliminate unnecessary feedback, the local player's voicethat of the interviewer, in this instanceis not played back through their game. Therefore, the interviewer's questions and prompts were not present in the audio mix that was being captured, requiring that it be recorded separately using a digital audio recorder and subsequently dubbed over the rest of the interview. While this approach was more time consuming than anticipated, the end results were acceptable vii . A second issue lay with the challenge of playing the game while attempting to conduct an interview. It was anticipated that the unusual circumstances might affect the interview, but one of the motivations for conducting it in this way was to ascertain what these effects might be. On the one hand, it was hypothesised that the interviewee, engrossed in the game, might enter a 'flow' state and effortlessly progress through the game while providing honest and insightful answers to the questions posed. However, the opposite outcome was observed: the game was no more than a distraction. A number of important factors appear to have been at work here. First, neither the interviewer nor interviewee was familiar with the chosen game, so both participants' ability to concentrate on the interview was diminished as they simultaneously attempted to fathom the game's mechanics. However, the end result was that the game suffered more than the interview: at times, particularly in the second half of the interview, game play virtually ceased while interviewer and interviewee instead focussed on asking and answering questions. Second, the game was selected for practical reasons aloneit offered voice-based chat, and both parties could access it at the time of the interviewrather than any attempt to align the game being played with the questions being asked. The obvious choice of game might have been Dark Souls, as the interview touched upon that title on numerous occasions, and it was the focus of the other wiki-based interviews. However, in keeping with that game's opaque constructs, it does not permit voice-based communication, and is not a multiplayer game in the conventional sense of allowing one to play with one's specified friends. Even if the interviewer had contrived to 'summon' the interviewee to battle collaboratively through the gameand this process is difficult to control in terms of who may be summonedplayers may communicate solely through a small set of physical gestures. Other issues with the ingame interview process, such as the interviewer being forced to play the game through a small window on the laptop being used to record the footage, rather than on a conventional TV screen, and the relatively low quality of the video captured by the device were not, on reflection, critical. Technical challenges related primarily to the recording of footage from a games console, in this case a PS3. Recording footage of games played on a Windows PC, for example, can be achieved without the need for additional hardware, using a software solution such as Fraps viii . While the in-game interview process described here was not entirely successful, a serviceable interview was conducted despite the game's distractions. Further, the approach may not be without merit if the questions asked are more closely associated with the game being played. Drawing on the think aloud protocols used by researchers in the fields of psychology and human-computer interaction , interviewees might be asked to describe and explain their in-game actions, to help understand the cognitive and learning processes at work. However, asking questions about an almost unrelated ancillary resource, such as a wiki, appears less useful. --- Results & Discussion --- The motivation to contribute to gaming wikis One of the questions this study sought to answer was what motivates certain players to voluntarily contribute material to wikis in the first place, especially considering the academic quality of the much of the work: could we learn anything from how players interact and collaborate on gaming wikis that might be applied to more academic endeavours? Interview questions were designed to uncover the motivational factors that have inspired those who contribute to gaming resources such as the Dark Souls wiki at wikia.com to devote time and effort to these sites. Broadly speaking, the psychology and education literature refer to two forms of motivation: intrinsic motivation, where the task at hand provides its own reward, and extrinsic, where the motivation is driven by the desire for external rewards such as money or prizes, or recognition from one's peers . For the most part, as these interview excerpts illustrate, the motivation described by these more dedicated wiki contributors is almost entirely intrinsic: they do it for the love of the game itself, and out of a desire to support the community that has grown up around it. The first interview to be conducted was the chat-based one-to-one interview with a gaming wiki contributor and admin, Balagog gro-Nolob , from darksouls.wikia.com. The primary focus of the interview was on gathering Balagog's opinions on the use of the wiki, and initial impressions were that those involved in running a video game wikiat least, those involved in the Dark Souls Wiki at wikia.comare thoughtful and dedicated in their approach. Standards are of paramount importance to these wiki contributors, not just in terms of the completeness and accuracy of the information contained within the wiki, but also with respect to grammar and writing style. Indeed, Balagog indicated that his work on the Dark Souls Wiki had improved his writing ability, as well as encouraging him to learn about more technical aspects of web authoring, such as Cascading Style Sheets . The blog-based asynchronous group interview confirmed the emphasis placed on standards. Wiki contributor Brainwasher5 started work on the wikia.com Dark Souls wiki after discovering that he could not correct erroneous information on other wikis devoted to the game: "I came across this wiki, which was at the time even more woefully undeveloped than it is now. But the only important thing for me was that I could edit it. Dark Souls is a very meticulously crafted game, so it only makes sense that all of its information sources be held up to the same standards, and that's why I'm here." In short, Brainwasher5's motivation was "to help make this wiki the most accurate and concise information source of this game on the internet". His love of the game itself is a clear driving force behind these endeavours, as this exchangewhich includes a sly dig at the oeuvre of film director Michael Baydescribes: "One reason I contribute to Dark Souls wiki that I hadn't brought up before is the fact that there is so much in the game that you can miss at first glance. This is not a Michael Bay film; things are not shoved into your face. Dark Souls is much more coy when it comes to facts and lore, and it'd be a shame if busier or less attentive players play through this amazing game and miss stuff. So I like to think that when I'm writing here, particularly about elusive or esoteric lore or item facts, I am helping enlighten fellow gamers regarding this jewel of a game." Fellow contributor Ltpliers agrees that the quality of the game is what motivates him to devote time to the wiki: "I contribute to this wiki simply because I enjoy the game and, with the information I have, I feel that it's good to give back to the community for such a great game." Ltpliers continues: "A wiki I find is a very convenient source of information that saves time and the more extensive we can make our wiki and the more convenient we can sort it, the better off it will be for standard users to allow them to enjoy the game as much as we have." So, while Ltpliers doesn't consider himself an expert on the game, he does introduce the notion of the "standard user", highlighting the imbalance in wiki user behaviour: a very small minority contribute to and maintain these wikis, while the standard users reap the benefitsa phenomenon sometimes referred to at the '1% rule' or 'participation inequality' . Ltpliers understandably considers his contributions to be of value to the wiki: "I feel that writing up the wikia is just as important as using it as not only am I learning information myself but also giving back to the community for such an excellent game. Any time I have some free time I try to get some information updated or try to fix up problems or formatting that we may have missed so that others may find information more conveniently and more thoroughly written." When asked to describe his motivation for contributing to the wiki, interviewee Balagog gro-Nolob states simply "to tell the absolute truth, I don't really know. For some reason I just enjoy doing it." His motivation, while simply stated, is also clearly intrinsic. By contrast, the in-game interviewee, a user of gaming wikis but not a contributor, found it somewhat "baffling" that anyone would sit down with a game and attempt to determine its underlying logic and statistical characteristics. He could only assume that, in lieu of more tangible remuneration, such wiki contributors were motivated by the acquisition of "community kudos". Such an assumption suggests that more extrinsic motivational forces are at work. However, evidence gathered directly from wiki contributors does not indicate that the extrinsic reward of receiving admiration from one's peers is actually a motivating factor. Returning to those who choose to contribute to gaming wikis, Brainwasher5 offered some final thoughts which underline the general consensus that working on a wikieven one devoted to something as apparently trivial as a video gameis a worthwhile experience: "I think that editing wikis can be a very rewarding experience. Even if you don't learn anything yourself, enlightening others is, to me, highly fulfilling. Hours spent editing and creating pages on this wiki are worth it, as long as people read them and learn something new. I highly recommend ANYONE, not just gamers, to try out editing a wiki and see what they can do to improve it." --- Academic engagement with gaming wikis Brainwasher5's comments hint at the learning potential that working on a wiki, even one devoted to a video game, can offer. While none of the interviewees could recall contributing to a wiki in an academic context such as school or college, such practice is common . However, the collaborative nature of wikis makes attributing work, and therefore credit or grades, to individuals' contributions problematic . A solution suggested by these interviews might be to rely on peer assessment; a technique which, if administered appropriately, can have positive formative effects on the students involved . Driven by their interest in the topic and a desire for high standards, the wiki contributors described herefrom the formal peer discussion area used by those who administer the Dwarf Fortress wiki to the less formal but by no means less rigorous approaches taken by those on the Dark Souls wikiare effectively engaged in peer assessment. In fact, peer assessment, or peer review, is central to the quality control processes on which these wikis rely. On the Dark Souls wiki, for example, the page relating to the Knight Artorias ixone of the game's 'bosses'has been composed by means of over 100 edits by 25 contributors x . The page also features over 200 comments which, in addition to musings on the Artorias character and his backstory, also comprise suggestions on how the page might be improved: what the title of the article should be, which sections need further work, and whether aspects of the character's lore are accurate. The Dwarf Fortress wiki's 'Centralized Discussion' area xi is organised into three areas of peer discussion: 'Content', 'Administration' and 'Article-specific'. Discussions which have been inactive for more than one month are automatically archived, but remain available for consultation. Under 'Content', for example, there are sections that document the discussion around topics such as redundancy, article consolidation, and the quality labelling system used by authors' peers to review articles xii . Supporting evidence for positive academic engagement with social media can also be found in the existing literature. Achterman identifies a number of features of wikis that make their use in education appealing. These features include the ease with which student contributors may create and edit articles, and the in-built facilities for collaboration, reflection and metacognition through comment and discussion. The study here demonstrates that these features are just as important even when learning outcomes are not expressly identified. Contributors to the Dark Souls wiki, for example, rely upon these facilities in order to produce a better standard of work: that the work is not linked to their formal studies is irrelevantthey are engaging in 'academic' activity, regardless. That the work carried out by contributors to gaming wikis is self-regulated is another important consideration for educators. Social media or 'web 2.0' tools such as wikisincluding gaming wikismight be considered personal learning environments , in which learners may create, organise, and share content. A wiki provides scaffolding for the learning experience , allowing the contributor, or learner, to carry out their work with autonomy. However, somewhat contrary to the prevailing notion in the literature that a teacher must play some part in the construction of such scaffolding and the development of a collaborative ethos , video game wikis are entirely self-regulated; the requisite collaboration is a spontaneous outcome of a shared interest in the subject matter. --- A learning "meta-game" Another emerging aspect of video game wikis with particular implications for game-based learning was that of the meta-game which such wikis facilitate. This idea was perhaps hinted at in the wiki-based interviewsthese respondents clearly enjoyed their work on the wiki, which, for them, was an extension of the game itselfbut it was during the in-game interview that the term "meta-game" was coined. The in-game interviewee suggested that the associated processes of documenting and consulting online resources such as wikis and Facebook groups made playing the games a "collective experience" or multiplayer meta-game. Rather than spoiling the game, consulting such wikis could "enhance players' enjoyment", as they "appeal to the 'completionist'". In common with those who took part in the group interview, the in-game respondent reported that he consulted gaming wikis before, during and after game play to find out more about aspects of the game that might otherwise elude him. However, the interviewee did concede that it might seem counterintuitive to derive pleasure from learning about a game through a "separate text" and that one must tread carefully when consulting such resources for help with more story-driven games, as the story itself may be spoiled by "reading ahead". Games, he stated, support a multimodal approach to understanding and enjoying them, in much the same way that one might wish to read reviews of a film before or after watching it, or view a documentary about the making of the film. Continuing with the film analogy, the respondent noted that there were basic, functional requirements for enjoying a film but that these functional requirements, and the necessary understanding of the underlying systems they imply, were much more complicated for interactive entertainment such as modern video games wherein control schemes, rules, conventions and even software bugs must be understood before the player can advance. The learning meta-game is about working to understand these systems and requirements, collaboratively. --- Limitations of this study and further work While small in scope, the project outlined here elicited a range of interesting responses from a limited number of wiki contributors, primarily related to the game Dark Souls. However, a more broad analysis of the collaborative forces at work in gaming wikis might be carried out by means of a larger study, comprising interviewees from a wider range of wiki communities. The Dark Souls site was chosen as the focus for this work, as noted above, due to the game's apparent reliance on an associated wiki: further studies should consider wikis related to games from a range of genres and platforms to ascertain if the high quality of contributions and adherence to standards is replicated across other titles. Further, wikis relating to interests and pursuits other than video games must be examined: it is likely that the intrinsically motivated and self-regulated efforts that games elicit may be replicated in other communities of practice. Another limitation of the study was in the implementation of the in-game interview, documented here in order to share an account of less successful practice. The choice of game in which to conduct the interview was left to the interviewee, and constrained by certain practicalities, detailed above. The interviewee chose a game which was unfamiliar, and, as a result, the context in which the interview was conducted was not entirely suitable for this study, which otherwise focused on activity motivated by an intrinsic interest in a particular game. As an experiment in the use of novel interview techniques, and the practicalities of conducting an interview via a games console, the endeavor was of some value. However, any further work involving in-game interviews should be embedded within a game in which the interviewee is personally invested. In terms of further work, a logical next step would be to pilot a gaming wiki-based project within an academic context; that is, to attempt to determine the learning outcomes that working collaboratively on a gamerelated wiki might produce in a group of students, in a more formal learning environment. --- Conclusions The intersection between games and social media provides a rich seam for research. Games are important, or, at least, significant, to those who play them, and they are inherently social: it could be argued that it is their sociability that makes them ideal conduits for collaborative learning. Social media is increasingly integrated into modern video games, in many different ways, from the simple posting of high scores on a player's Facebook page, to the complex collaborative meta-game that is played out around popular games on social media platforms such as wikis. In the case of some titles, such as Dark Souls, learning from one's fellow players is very much part of the game. The work here was founded on the premise that through studying the means by which non-academic users interact with social media such as wikis, students, educators and researchers can obtain a fresh perspective on their own academic activities: fastidiously maintained gaming wikis and their social media adjuncts currently thrive while more academic efforts to engage with social media are often hampered by concerns that such engagement might have a detrimental effect on how educators are perceived or that the privacy of students or the integrity of their work might somehow be jeopardised . This study suggests that while such concerns may be valid, there is tremendous potential for educators to engage with their students via social media tools such as wikis, under suitable circumstances. Achterman also acknowledges the privacy concerns that educators, and learners, may have about working in a web-based environment as suggests that the wiki software be installed on a school server, behind the institutional firewall. While this approach requires some resource and technical expertise, it offers and effective solution to the problems of privacy, and security. What motivates gaming wiki contributors is a love of the topic, while wiki users and those who contribute to more private gaming resources, such as Facebook groups, derive enjoyment from the meta-game that these social interactions comprise. If educators could harness or emulate some of the enthusiasm players possess for documenting, curating and discussing video games on social media, opportunities for real engagement in collaborative academic projects would surely arise. Such projects, even if their topic, such as video games, is one not yet routinely taught in schools and colleges might see students work to improve their writing skills, hone their ability to critically analyse one another's contributions, and collaborate with their peers to produce better work. The implication for educators is, perhaps, that by allowing students the autonomy to work with social media in their own way there is much they can learn about working collaboratively, while immersed in a tangible piece of work that is relevant to them. However, wikis aloneeven well-maintained, content-rich video game wikisdo not elicit contributions from the masses to the extent that other forms of social media can. Educators must be cognizant of the effects of participation inequality: a small number of key contributors effectively run such wikis, with others choosing to engage in game-related discussions elsewhere, such as in private Facebook groups. Others, still, may choose not to discuss even their favourite games with their peers. Video game wikis, then, offer an interesting insight into how suitably motivated players voluntarily commit to collaborative, high-quality work that might be considered academic in nature. Nevertheless, despite the popular appeal of the topic being worked upon here, there is an imbalance in players' engagement with such wikis: for educators seeking to emulate the learning meta-game that surrounds video game wikis, a broader approach, encompassing a range of alternative opportunities to contribute and collaboratesuch as private discussion groups or forums for personal reflectionshould be considered.

He lectures on various video game-related topics, and on multimedia systems and web technologies. His PhD research examines the relationship between video games and learning, with a particular focus on how commercial video games may be used to develop graduate attributes.

Introduction Globally, there is evidence to show that there is a significant decline in total fertility rates although some sub-Saharan African countries continue to have high fertility rates . Ghana is among the countries in sub-Saharan Africa performing remarkably in relation to reducing high fertility . This is evident in the reduction in TFR from 6.4 in 1988 to 4.2 in 2014 . Notwithstanding, the desire for more children continues to be a major demographic issue in sub-Saharan African countries including Ghana . While men in most sub-Saharan African countries prefer to have a higher number of children than their partners, those in developed countries prefer to have fewer children compared to their partners . This situation is what is referred to as fertility desire discordance . Fertility desire discordance refers to the situation whereby partners have different fertility goals and expectations . In sub-Saharan Africa, men play a dominant role regarding the reproductive decisions of women . Furthermore, FDD can happen as a result factors that may include age, relationship duration, and presence of children from previous relationship . Unintended and mistimed pregnancies can happen within unions and marriages ; therefore, understanding the factors associated with FDD underscores a great opportunity for developing countries like Ghana to reduce the likelihood of unintended and mistimed pregnancies in marital unions. In Ghana, although the TFR declined from 6.4 in 1988 to 4.2 in 2014, women and men desire large families: 4.3 children for all women and 4.5 children for all men . The preference among married women and men is for 4.7 and 5.1 children, respectively . Extant studies on fertility desires have focused mainly on women and persons living with HIV , with few studies having been conducted on men . Sarnak and Becker examined the accuracy of wives' proxy reports of husbands' fertility preferences in SSA and found that wives across a number of countries either inaccurately perceive or are uncertain of their husband's fertility preferences. Nevertheless, most studies measuring the concordance of couple's fertility preferences rely on the report of women as proxies regarding the fertility desire of their partners . As such, little is known about the factors associated with fertility desires' discordance between husband and wife from men's perspective. This paucity in empirical evidence limits the comprehensive understanding of men's fertility desires. The number of children an individual wants to have during his/her reproductive life is not fixed but changes due to changing circumstances at the individual level . Hence, factors influencing fertility preferences may not be the same across different age cohorts. Gaps in age, education and economic circumstances among couples predispose their marital relationship to an unequal sexual behaviour . Men usually have greater control over decisions regarding family size and contraceptive use. This study examines the association of age difference and education difference between marriage couples with men's fertility behaviour. Specifically, this study seeks to determine whether age difference, education difference and other covariates between marital partners are negatively associated with men's fertility desire. To the best of our knowledge, not many studies have undertaken a quantitative assessment using nationally representative data from Ghana on the relationship between men's characteristics and fertility desire, and whether it is similar across different age cohorts. Using nationally representative data would aid our understanding of Ghanaian men's fertility desires. We assess three synthetic cohorts of men, 20-35-year-olds, 36-50-year-olds and 51-59-year-olds, to examine differences in individual characteristics according to fertility desire across the three age groups. Drawing on the husband-and-wife reports on fertility desires, this study examined characteristics associated with men who desire a higher number of children than their partners. This paper also seeks to ascertain the contribution of the fertility behaviour, socio-economic and cultural characteristics of married men as well as couple characteristics on their fertility desires in Ghana using the 2014 Ghana Demographic and Health Surveys . --- Methods --- Data Source Data for this study come from the 2014 Ghana Demographic and Health Surveys . The GDHS is a cross-sectional, comparable and nationally representative survey that collects data on key populations and health indicators from women aged 15-49 years and men aged 15-59 years. The DHS uses a three-stage sampling technique, through a multistage, stratified cluster design. The GDHS uses three different questionnaires: household, women's and men's questionnaires. We relied on data from the couple datasets file. The couple data were derived by linking eligible interviewed men and women from the same households who are in a union. The sample size was restricted to couples in monogamous marriages because we were able to match the husbands' responses to any specific wife or partner within a monogamous marriage. Therefore, the results from this analysis are generalized only to men in monogamous marriages. Among married men aged 15-59 , 83.6% were in a monogamous marriage, while 16.4% were in polygamous marriage. The number of couples in monogamous marriages was 1528. Of this sample, we further restricted the data to men and women who provided a numerical response to questions on ideal family size. Thus, couples with non-numerical response were excluded from the study. Finally, sterilized and infecund men were excluded from the study since they do not have any fertility desire. We applied weighting to the dataset to obtain unbiased estimates according to the DHS guidelines. The survey command in Stata was used to adjust for the complex sampling structure of the data in all of the analyses. A weighted sample size of 1431 was used for the analysis. --- Measurement of Variables Dependent variable: Fertility desire was the dependent variable. We constructed two indicators of fertility desire. The first indicator is based on the ideal family size question: "If you could go back to the time, you did not have any children and could choose exactly the number of children to have in your whole life, how many would that be?" This question yielded both numerical and non-numerical responses; however, the data were restricted to men and women who provided numeric responses. This variable was used to compare men's ideal family size with women's ideal family size. The difference yielded three outcomes: 'wives desire more', 'same couple desire' and 'husbands desire more'. Since the study's focus was on husbands' or partner's desire for more children, we excluded 'wives desire more' and reduced the multivariate analysis to two categories . The second indicator was a prospective measure of fertility desire derived from the question "Would you like to have a child with your partner, or would you prefer not to have any more children with her?". Men who responded they want another child were considered as having a desire for more children and coded as 1. On the contrary, those who responded that they want no more were considered as not having a desire for more children and coded as 0. For our multivariate analysis, we relied on using the second indicator since research has confirmed it as more reliable and valid than the first indicator . Nevertheless, the first indicator was also used for comparison purposes . Independent variables: The study used nine independent variables, grouped into individual-and couple-level factors. The individual characteristics were age, current contraceptive use, parity , place of residence, religion, wealth quintile and age at first marriage/cohabitation. The couple characteristics are age difference and difference in education. Details of how each of these variables were coded can be found in Table 1. --- Statistical Analysis Analyses of data were carried out at three levels-univariate, bivariate and multivariate-with the use of Stata software. Frequencies and percentages were used at the univariate level to describe the socio-demographic and other characteristics of respondents, while one-way ANOVA was carried out at the bivariate level of analysis to examine the differences in mean ideal family size among couples across the three age cohorts . Multivariate binary logistic regression models were conducted to assess the relationships between men's characteristics and fertility desire. Models were run for all men aged 20-59 years, as well as across the three age cohorts using the two different measures of the outcome variable and the fertility desire. The statistical software package, STATA version 16, was used for the analyses. For all models, an adjusted odds ratio with its respective 95% confidence interval was computed and reported. --- Ethical Approval The DHS survey obtained ethical clearance from the Ethics Committee of ORC Macro Inc. as well as the Ethics Boards of the partner organizations, such as Ministry of Health and Ghana Health Service. During the survey, written or verbal consent was provided by the women. In this study, we sought permission from MEASURE DHS website for access to the data. Hence, the data analysed are available in the public domain "www.measuredhs.com" after obtaining the necessary approval. The data do not contain any identifying information. --- Results The distributions of the two indicators are displayed in Figures 1 and2. The results indicate that both Figures 1 and2 show differences in fertility desire, and age is statistically significant. Comparing the two distributions, the first indicator shows that 41.1% desire to have more children . According to the second indicator, 39.9% desire to have more children . This is a clear indication that the first indicator is upwardly biased, while the second indicator is downwardly biased. Both indicators show that desire for more children increases with age. Genealogy 2023, 7, x FOR PEER REVIEW 5 of 11 --- Ethical Approval The DHS survey obtained ethical clearance from the Ethics Committee of ORC Macro Inc. as well as the Ethics Boards of the partner organizations, such as Ministry of Health and Ghana Health Service. During the survey, written or verbal consent was provided by the women. In this study, we sought permission from MEASURE DHS website for access to the data. Hence, the data analysed are available in the public domain "www.measuredhs.com" after obtaining the necessary approval. The data do not contain any identifying information. --- Results The distributions of the two indicators are displayed in Figures 1 and2. The results indicate that both Figures 1 and2 show differences in fertility desire, and age is statistically significant. Comparing the two distributions, the first indicator shows that 41.1% desire to have more children . According to the second indicator, 39.9% desire to have more children . This is a clear indication that the first indicator is upwardly biased, while the second indicator is downwardly biased. Both indicators show that desire for more children increases with age. The individual and couple characteristics of all married men across the age groups in our sample are depicted in Table 1. The majority of the married men were aged 36-50 years , were not using any contraceptive methods , lived in urban areas , were adherents of the Christian religion and were first married at 20 years and over . A little over 90% of the men in this study's sample were older than their spouses; specifically, half of men were 6 or more years older than their partners. Half of the men were more educated than their spouses. Table 2 displays the mean of the ideal family size of couples. We categorized the ages into three cohorts for both men and women. For men, we were focused on young men , young male adults and older male adults . For women, we were focused on young women , young female adults and women ending their reproductive life . Generally, the mean ideal number of children for men and women was 5.4 and 4.7, respectively. The mean fertility desire of both men and women increases with age. Table 3 shows the multivariate level of analysis where age, religion, wealth quintile, contraceptive use, parity and differences in education were the significant predictors of a desire for more children . The model revealed the adjusted odds ratios of the selected independent variables on fertility desire. Men belonging to the Islam religion were more likely to desire more children compared with those adhering to Christianity. The likelihood of desire for more children was lower among those aged 36-50 years , those aged 51-59 years , those using any contraceptive methods , those with a parity of 2-3 children and those with a parity of 4+ , and those who were more educated than their spouse . In addition, we found that as the wealth quintile increased, the desire for more children significantly decreased. Across the three age groups, we observed similar patterns in the relationship between religion and fertility desire, wealth quintile and fertility desire, and parity and fertility desire, as well as differences in education and fertility desire. We observed similar patterns in the relationship between contraceptive use and fertility desire, except for the 20-to-35-year-olds, as those using any contraceptive methods were more likely to desire more children compared with those not using contraceptive methods. To complement the multivariate findings, using the first indicator, we found that correlates of fertility desire included religion, wealth, parity and age difference. There was consistency with the first indicator of fertility desire regarding the direction of coefficients The individual and couple characteristics of all married men across the age groups in our sample are depicted in Table 1. The majority of the married men were aged 36-50 years , were not using any contraceptive methods , lived in urban areas , were adherents of the Christian religion and were first married at 20 years and over . A little over 90% of the men in this study's sample were older than their spouses; specifically, half of men were 6 or more years older than their partners. Half of the men were more educated than their spouses. Table 2 displays the mean of the ideal family size of couples. We categorized the ages into three cohorts for both men and women. For men, we were focused on young men , young male adults and older male adults . For women, we were focused on young women , young female adults and women ending their reproductive life . Generally, the mean ideal number of children for men and women was 5.4 and 4.7, respectively. The mean fertility desire of both men and women increases with age. Table 3 shows the multivariate level of analysis where age, religion, wealth quintile, contraceptive use, parity and differences in education were the significant predictors of a desire for more children . The model revealed the adjusted odds ratios of the selected independent variables on fertility desire. Men belonging to the Islam religion were more likely to desire more children compared with those adhering to Christianity. The likelihood of desire for more children was lower among those aged 36-50 years , those aged 51-59 years , those using any contraceptive methods , those with a parity of 2-3 children and those with a parity of 4+ , and those who were more educated than their spouse . In addition, we found that as the wealth quintile increased, the desire for more children significantly decreased. Across the three age groups, we observed similar patterns in the relationship between religion and fertility desire, wealth quintile and fertility desire, and parity and fertility desire, as well as differences in education and fertility desire. We observed similar patterns in the relationship between contraceptive use and fertility desire, except for the 20-to-35-year-olds, as those using any contraceptive methods were more likely to desire more children compared with those not using contraceptive methods. To complement the multivariate findings, using the first indicator, we found that correlates of fertility desire included religion, wealth, parity and age difference. There was consistency with the first indicator of fertility desire regarding the direction of coefficients of religion and wealth . However, men with 4 or more living children were more likely to desire more children compared with those with a 0-1 parity. We observed the same pattern across the age cohorts of 20-35 years and 36-50 years. In addition, men older than their spouse by 1-5 years were more likely to desire more children compared to those couples with same age. Men older than their spouses by 6 or more years were more likely to desire more children compared to couples of the same age. This pattern was similar across those aged 20-35 years and 36-50 years . --- Discussion This study investigated the discordance in the fertility desires among monogamous married men in Ghana. Generally, men are observed to have higher fertility desires compared to their partners/wives. This was true in all the different age categories of married men. This finding means that married couples have to deal with the perceived differences in fertility preferences. The high fertility desire of husbands compared to that of wives showed the pronatalist tendency of men in sub-Saharan Africa, including Ghana. It is interesting to note that men are likely to wish for more children compared to their spouses because they do not primarily go through the physical and psychological demands of pregnancy. Women, who primarily undergo the stress associated with pregnancy and child birth, seem to be modest in their fertility desires compared to their partners. The gendered discordance in fertility desires has been previously described in other studies carried out in Africa . Within the West African context, women are often coerced to meet the higher fertility desires of the men; however, persuasion is often used when the women have higher fertility desires . This differential resolution of conflicts arising from discordance in fertility desires is underpinned by cultural values that assert a male dominance in fertility decision making . Men in sub-Saharan Africa are the heads of most families, are usually older than their partners and are expected to make decisions affecting their wives. This study found the following factors to be significant independent predictors of discordance in fertility desires between husbands and wives in monogamous marital unions: age, religion, wealth quintile, contraceptive use, parity, and differences in education. This study showed an inverse relationship between age and the desire for more children. As the age of men increases, the less likely it is for them to desire more children. This is similar to the results of a previous study carried out in China . The study found that men who had attained a higher education than their partners were less likely to have a desire for more children compared to that of couples with the same level of education. This observation was similar in the age-stratified analysis carried out for married men aged 20-35 years and those aged 51-59 years. It appears that after receiving some level of formal education, men are able to have a more informed expectation regarding their fertility desires matching those of their female partners. There is a mixed relationship between educational attainment and fertility desires in men . However, higher education attainment and aspirations may reduce fertility desires in men as they would not want childbirth to interfere with their educational attainment, especially when career growth and fertility desires are difficult to attain at the same time . As seen in this study, this was especially the case for married men aged 20-35 years, who are more likely to be focused on career advancement than higher fertility desires. Our finding is in agreement with those of a recent study in Nigeria, which reported that higher education was associated with decreased fertility desires among men . The relationship between wealth and fertility is generally reported to be positive among men but negative among women . This study found that those who were poorer were less likely to have higher fertility desires than their female partners. Therefore, household wealth appears to be protective against higher fertility desires. This finding is in contrast with the tenets of microeconomic theory on fertility, which assume a positive association between income and fertility desires . This study found that the number of living children is negatively associated with having higher fertility desires. This was not surprising given that one would have expected men to have decreased fertility desires as the number of living children increases . In addition, men with a large family size in Ghana may not desire to have additional number of children due to economic and social problems. This finding contradict the results of a study in Europe which found that fertility expectation of parents increases after the birth of additional children . Our finding may be attributable to the sense of satisfaction in family life in the presence of few living children, which reinforces the desire of the men for more children. Nevertheless, using the first indicator of fertility desire, we found that high parity increases with men's desire for more children. This suggests that the measurement of fertility desire influences the direction of the independent variables used in a model. Interestingly, we found that as compared to men not using contraceptive methods, men using any contraceptive method were less likely to have higher fertility desires. We find similar patterns across the age groups 36-50 years and 51-59 years. The only exception observed was with those aged 20-35 years. Among young women using any contraceptive method, they were more likely to desire more children. At a younger age, men may be using contraceptives with the intention of spacing out childbirth and not necessarily for limiting childbirth. This finding is similar to those of the study carried out by Yeboah et al. who explained that modern contraceptives used by women were used to space out high parity rather than with the intention of limiting child births. However, at older ages, men might use contraceptive methods to prevent having additional children. Further studies, including the use of a qualitative methodology, are recommended to explore and better understand this finding. The odds of a large family size desire were higher among Muslims than among Christians. This finding corroborates those of studies conducted in Nigeria . The desire of Muslim men for more children could be due to the practice of polygamy, which is mostly affiliated with the Muslim religion. From a policy standpoint, our findings highlight the discordance in fertility desires between Ghanaian men and their female partners. Our study uncovers some socio-economic factors that underpin such a discordance. Public health professionals can use our findings to guide their public health practices by taking into account the socio-economic complexities of the gendered discordance regarding fertility desires. --- Limitations of the Study Despite the merits of our study, this study used secondary data from a cross-sectional survey; hence, the associations observed in this study do not imply a causal relationship. Our study was also restricted to only variables available in the dataset. --- Conclusions We investigated the discordance in fertility desires among couples with reference to a high fertility desire in men and its associated factors among married men. The independent predictors of discordance in fertility desires between husbands and wives in marital unions included education, wealth status, religion and parity. The findings reflect the pervasive role of education, wealth and socio-cultural norms on the complexities of the gendered discordance regarding fertility desires. --- Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/genealogy7030048/s1, Table S1: Multivariate analysis of factors associated with married men desire for more children . Institutional Review Board Statement: We did not seek ethical clearance for this study because the dataset used is freely available in the public domain. However, we sought permission from the MEASURE DHS, and approval was given before using the data for this study. We ensured that all the ethical guidelines concerning the use of secondary datasets in the publication were strictly adhered to. Informed Consent Statement: Detailed information about DHS data usage and ethical standards is available at http://goo.gl/ny8T6X, accessed on 3 April 2023. --- Data Availability Statement: The dataset freely accessible at www.measuredhs.com, accessed on 3 April 2023. ---

Generally, men in sub-Saharan Africa make reproductive decisions that affect their partners. We examined the predictors of fertility desires among married men across three age cohorts: 20-35 years, 36-50 year, and 51-59 years. Using the 2014 Ghana Demographic and Health Survey dataset, we conducted ANOVA and multivariate binary logistic regressions on 1431 monogamous married men aged 20-59 years. Two indicators of fertility desire are constructed: (i) the comparison of men's ideal versus women's ideal family size, and (ii) the desire for more children. The results indicate that the fertility desire of men is stronger than that of women. The predictors of fertility desire are age, parity, religion, contraceptive use, wealth quintile, couples' age difference and couples' difference in education. At ages 20-35 years, men using modern contraceptives were more likely to desire more children compared with those not using any modern contraceptives. However, at ages 36-50 years, men using modern contraceptives were less likely to desire more children. This finding suggests that men change their fertility desires in response to changes in their ages.

INTRODUCTION Pregnancy can lead to changes in a woman's body that can affect her oral health, making it a time of particular vulnerability in oral care. Most pregnant women experience toothache and periodontal diseases during pregnancy due to tooth decay. 1 In addition to the final cost of oral and dental disease, pregnant women bear the potential for adverse pregnancy outcomes, including premature birth, low birth weight babies, 2 3 and increased dental caries in children after birth. 4 Previous studies have pointed out the difference between the oral health needs of pregnant women and the general population. Several studies have shown that oral and dental diseases harm the quality of life and health perception of pregnant women. 1 5 The WHO recognises the importance of oral and dental hygiene as part of preventive measures in healthcare for pregnant women and newborns. 6 Despite guidelines and institutions recommending dental care during pregnancy, many pregnant women of different ages do not seek treatment, and this serious health problem is often ignored. 2 The results of a population-based study during 2014-2015 among 4071 Iranian pregnant women in 10 different provinces showed that about 55% of pregnant women are not visited by a dentist during pregnancy. 7 More than half of Iranian pregnant women brush their teeth once a day, and the average DMFT score of pregnant women was reported as 3.93±5.64. 8 Pregnant women may not have enough knowledge about the effects of their oral health on the fetus and the consequences of their pregnancy. 9 Some pregnant women may refuse to go to the dentist despite knowing the clinical condition of their oral health due --- STRENGTHS AND LIMITATIONS OF THIS STUDY ⇒ Semistructured in-depth interviews allowed a rich exploration of the participants' perspectives to explore new and emerging concepts in oral healthcare in pregnancy. ⇒ A multistakeholder perspective on prenatal oral health to obtain a comprehensive understanding of insights into the perspectives of pregnant women, midwives/healthcare workers, supervisors of prenatal care services and dentists in a broad and diverse range of different education levels and work experiences. ⇒ Triangular analysis was performed by three researchers who clarified their assumptions before starting the study. ⇒ Due to the restrictions caused by the COVID-19 pandemic, the use of group discussions to better understand the findings was not possible. ⇒ Views of pregnant women who do not visit government health centres were not explored. Open access to incorrect knowledge. 10 Others may think that poor oral hygiene during pregnancy is normal. 11 Pregnant women may have a negative attitude towards oral care during pregnancy due to incorrect information and fear of the adverse effects of dental procedures on the mother and fetus. This negative attitude can lead to poor oral health during pregnancy and is associated with an increased risk of adverse pregnancy outcomes. 12 However, evidence shows that oral healthcare during pregnancy is safe and should be recommended to improve the oral and general health of the woman. 13 14 Many pregnant women in Iran do not go to health centres for oral care during pregnancy due to limited knowledge, attitude and lack of a coherent structure for oral care. 15 16 However, it is crucial to pay attention to the importance of oral health during pregnancy for those involved in the oral care of pregnant women and newborns. 17 Prenatal care providers play a crucial role in supporting and providing preventive oral care during pregnancy. By working together, physicians and dentists can help pregnant women initiate and maintain oral healthcare during pregnancy and through the life span, improving the oral and overall health of both the mother and baby. 18 Since mothers are responsible for establishing healthy habits in their children at an early age, public health programmes can help prevent adverse health outcomes for both the mother and baby and establish healthy habits in children at an early age by prioritising maternal oral health during pregnancy. While various studies have examined the factors affecting oral care during pregnancy, few studies have comprehensively examined these factors at all individual and social levels. Therefore, more research is needed to discover, describe and explain these factors in the oral healthcare of pregnant women from a qualitative perspective. Qualitative research is a suitable method to gather people's opinions about their experiences and clarify concepts and relationships between them. This study benefits from using qualitative content analysis, which enables the exploration of pregnant women's experiences in an open and non-judgemental way, simulation of individual experiences and objective data analysis without being influenced by prior judgements or biases. Qualitative content analysis is a valuable research method that can help identify factors that may be missed by quantitative research. 19 Therefore, the content analysis approach is an appropriate method to attain a comprehensive and deeper understanding of the factors affecting oral health in oral care, especially considering the sociocultural and organisational conditions governing society in pregnant women. 20 The study aimed to answer the question, 'How do pregnant women and prenatal oral healthcare providers understand oral healthcare during pregnancy'. --- MATERIALS AND METHODS Study design and participants This study used a content analysis approach based on a guided approach. 21 This study was conducted and reported according to the Consolidated Criteria for Reporting Qualitative Research 22 . Pregnant women were sampled through comprehensive health service centres. Oral health providers who were working there were also included in the sample. The sampling method was purposeful with maximum diversity. Pregnant women in different age groups, gestational age, education and economic status of the family were invited. Also, oral health providers were selected from all levels of the health system in terms of education, work experience, and different health centres. The inclusion criteria for pregnant women to enter the study were as follows: over 18 years old, a pregnancy age of at least 20 weeks, being Iranian and living in Mashhad City. The oral health workers in the study were dependent on two criteria: having at least 1 year of work experience and the ability to express their experiences, opinions and views. Participants who did not want to continue cooperation or were not satisfied with recording the conversation were excluded from the study. Finally, after interviewing 18 pregnant women, 7 midwives/healthcare providers, 3 supervisors of prenatal care services and 3 dentists, the data reached saturation. --- Patient and public involvement The public and patients were never involved in study design or management. --- Data collection Data were collected using a guide questionnaire and semistructured interviews employing individual interview techniques. The guiding questions were tested through four interviews with pregnant women to determine the time required, people's reactions and the need for additional questions. The interview questions were modified and new questions were added as the study progressed. At first, 17 questions for pregnant women and 15 questions for oral health workers were included, which finally changed to 18 and 11 questions , respectively; all the interviews were conducted by a female doctoral student of health education and health promotion who received comprehensive practical training in the field of conducting qualitative studies and was interested in conducting studies in the field of improving oral care. Participants were identified by visiting selected health service centres, and the interviews were conducted at a time and place convenient for the participants. There was no other person present at the interview place. The interviews began with an introduction of the researcher and the purpose of the research to establish communication and gain the participants' trust. The interviews were conducted face-to-face based on guided Open access questions, and all conversations were recorded by a recording device. Probing questions were used during the interviews to further reveal the answers to ambiguous questions. Examples of probing questions include 'Can you give an example?' and 'I do not understand what you mean, please explain more'. The duration of each interview varied between 15 min and 45 min, and interviews continued until data saturation was reached. Efforts were made to avoid giving feedback, inducing opinions to the participant, and correcting their statements, according to the purpose of the interview. --- Data analysis The process of content analysis is carried out with a guided approach using SE and Shannon's method during three stages of preparation, organisation and reporting of findings. 21 In the preparation stage, the researchers use the intervention mapping model to create an interview guide and resolve ambiguities in the text. The intervention mapping model is a theory-based and evidence-based health promotion programme planning model that takes an ecological approach to understand health problems and to intervene at multiple levels, including individual, interpersonal, organisational and community levels. 23 In the organisation phase, the standard method of analysis is used to determine semantic units and primary codes, which are then compressed and summarised with appropriate labels. The next step is to form categories from subcategories, and as the researcher's understanding of the primary codes increases, more abstract codes develop. In the reporting of findings phase, the researchers provide a rich background of the phenomenon under investigation and create an understandable description with sufficient interpretation for the reader. The data are organised using MAXQDA V.10 software. To confirm the research, triangular analysis was performed by three researchers who clarified their assumptions before starting the study. The full text of all interviews and extracted classes approved by supervisors, consultants and experienced individuals in qualitative studies. Additional comments from these people were used in all stages of implementation, coding and extraction in primary categories. The transferability of this study was ensured by selecting samples from different socioeconomic classes and educational levels as well as diverse groups. In addition, findings were compared with those of people who were in similar situations to the participants. --- RESULTS --- Sample characteristics In this qualitative research, 32 pregnant women and 15 oral health workers were invited to participate in the study, 31 participants were interviewed, and the findings reached saturation. The main reason for refusing to participate was due to the restrictions caused by the COVID-19 pandemic. Pregnant women with an average age of 24.5±29.83 were in different months of pregnancy. Of 50% had previous pregnancy experience, and about 72% were housewives. The average age and work experience of oral health workers were 55.84±7.03 and 17.61±7.26, respectively. Other demographic characteristics of the participants are given in table 1. --- Qualitative results The 1220 primary codes extracted from the data analysis were organised into 11 subcategories, which were categorised into three main categories as presented in table 2. --- Individual and physiological determinants The first category was divided into five subcategories, and it includes determinants related to the pregnant woman herself. --- Oral health literacy The participants reported that certain factors such as awareness, education and shaping behaviour can increase oral care in pregnant women. However, some participants regarded oral issues during pregnancy as a regular part of the process. Every mother believes a tooth must be lost with each pregnancy and breastfeeding. This is a natural occurrence for them, but they are unaware that it can be prevented . If we had more information about oral health, we would be more likely to take care of our teeth regularly . Also, the participants recommended institutionalising oral care from childhood and strengthening the culture of prevention and timely treatment. --- Open access All study participants expressed that they needed training on how to provide care, especially for pregnant women. It would be helpful if someone could teach which type of toothpaste is suitable for use during pregnancy, and which type of toothbrush is good if it is taught, that is good . --- Perceived importance Several oral health providers reported that it is crucial to raise awareness among pregnant women about the significance of oral health and sensitivity toward oral diseases. However, they stated that pregnant women do not prioritise oral hygiene. It doesn't matter to them; they don't think about it. My tooth is a part of their body, like the leg, like the head. They say very easily, well, now it is broken, throw it away . Pregnant mothers don't care about their teeth as much as they care about their other physical issues, which means they haven't developed the belief that their dental health is as important as the health of their other body parts . --- Care needs Participants' statements suggest that practising behavioural skills such as using toothpicks, dental floss, brushing and attending dental examinations can be effective in prenatal oral care. The most important and main way is that in the prepregnancy period when she wants to get pregnant, Also, Healthcare providers and midwives recommend that due to pregnancy conditions, pregnant women should observe their oral hygiene more than before pregnancy. --- Motivation Several factors that increase prenatal oral care include application incentives, reducing fears, motivating patients and prior experience. A few participants raised the fear of complications for the fetus as a reason why pregnant women may be less motivated to perform medical visits during pregnancy. With the spread of the COVID-19 epidemic, one of the concerns of pregnant women when visiting the dentist was the fear of contracting the coronavirus. Currently, because of the coronavirus situation, the pregnant mother herself does not like to take off her mask or, for example, take care of her mouth and teeth . --- Restriction of the nature of pregnancy Participants reported that pregnancy-related physical discomfort and fatigue can hinder oral hygiene maintenance. Focusing on prenatal oral care can be improved by managing drug use, stress, morning sickness and changing habits. Boredom and lethargy in this period; It means that the mother has been lethargic from the beginning. She is not bored with anything… She does not like to do anything . --- Organisational determinants The next category, which environmental-organisational factors that facilitate oral healthcare behaviour, was divided into four subcategories. --- Costs The participants frequently reported barriers to oral care for pregnant women, such as high dental costs and inefficient insurance. Most people believe that the cost of dental services in Iran is very high, which is one of the main obstacles to not visiting a dentist. Pregnant mothers never go directly to the dentist because of the high costs of dentistry . Iran has made significant progress in improving its healthcare system over the past few decades, but there are still concerns about the effectiveness of health insurance in protecting against high costs. I think work insurance can be useful. It should be insurance that can pay part of the expenses … most dentists do not accept insurance . --- Access to equipment and services Respondents in this study stated that they do not have access to a written educational package for oral care during pregnancy. The oral health providers have expressed concerns about the ever-increasing tariffs and lack of access to dental services for pregnant women. They suggest that revised care guidelines should place greater emphasis on prenatal oral health. Unfortunately, now the tariffs have increased. That is, our dental tariffs have increased recently at the beginning of the week, and this makes the number of visits to our centers much less . --- Review the service delivery process Organisational support, referral system importance, health centre limitations and teamwork can improve prenatal oral care. Most participants, including midwives, and dentists, considered collaboration and coordination among the care team to be significant to ensure the oral health of pregnant women. The study also found that most participants mentioned the need to pay more attention to the referral system and monitor the process more closely. For this, a group must work together. The team includes healthcare workers, midwives, center doctors, nutritionists, mental health experts, husbands, and everyone involved. The midwife's work is not alone, but this group must all work together . --- Professional behaviour Pregnant women have reported that their willingness to perform oral care is affected by factors such as inappropriate behaviour of oral health service providers and a lack of commitment from dentists to work in comprehensive health service centres. They argued with me badly and told me to go, take back the money you paid, and leave until this time was over . The dentist does not work… The dentist says I have no more energy. Dentists at the health center are not very busy… most of them are unemployed . --- Social-cultural determinants The third category of factors affecting oral healthcare was social-cultural determinants, which was divided into two subcategories. --- Open access --- Educational services Participants stated that factors such as the use of media and educational resources can increase prenatal oral care. They believed that most of the education provided in the health system is more traditional and does not have the necessary efficiency. New educational techniques incorporating social media and written content can be effective, especially in virtual spaces. From the point of view of the participants, social media can be an effective tool for oral healthcare. Other internet channels, such as WhatsApp and Telegram, are much better for pregnant mothers. For example, to increase general information about pregnant mothers, those who are members of the health service, and who are pregnant, should be on the same channel . --- Support from family and friends Participants often reported the supportive role of family and the importance of drawing on the experiences of the entourage. They also stated that husbands can support family oral health by prioritising their wives' prenatal oral health. This indicates low social support for oral care, resulting in decreased continuity and consistency of care. On the other hand, it is the support of their wives, because many people say, for example, my husband says that he does not want you to fix it, now you will pull your teeth later. In my opinion, awareness should be created for men in this field that the risk is higher . --- DISCUSSION --- Main findings The study aimed to provide a comprehensive understanding of the determinants of oral healthcare among Iranian pregnant women. The study examined oral healthcare behaviour in pregnant women from multiple perspectives, including midwives/healthcare workers, supervisors of prenatal care services and dentists. The results provide a coherent and complete picture of the determinants of oral healthcare behaviour during pregnancy. This study suggests that oral health literacy and understanding the importance of oral healthcare are crucial factors in preventing tooth decay during pregnancy. The results of a 2020 study on oral health literacy among 169 pregnant women in Kerman, Iran, indicated insufficient literacy levels. The study highlights the necessity of dental consultation before pregnancy to improve oral health literacy among pregnant women. 24 Another study on the oral health literacy of 2263 pregnant women suggests that it is necessary to improve the oral health literacy of pregnant women with the help of education. 25 This study found that pregnant women often neglect their oral and dental hygiene due to a lack of attention given by oral healthcare providers and insufficient education and sensitivity provided to pregnant women. As a result, pregnant women tend to rely on unprofessional oral care. Most of the participating pregnant women stated that they need to receive the necessary training from prenatal care providers in this field, especially in the first level of prevention. Oral healthcare providers need to prioritise and educate pregnant women about the importance of maintaining good oral and dental hygiene during pregnancy. However, it is possible that pregnant women may not receive professional training due to a lack of understanding of the importance of oral care during pregnancy and may disregard the training provided by oral healthcare providers. 26 According to a 2019 study, one of the most significant reasons for not using oral healthcare is the lack of awareness of its importance. This issue was identified as an intergenerational responsibility, meaning that the health status of one generation can influence that of the next. 27 The participants suggested that the habit of performing oral care should be institutionalised in the individual from childhood, and the culture of prevention and timely treatment should be strengthened. The results of previous studies emphasised the comprehensive education of pregnant women about the importance of oral health during pregnancy. 28 29 From the participants' point of view, performing examinations before and during pregnancy plays a vital role in maintaining oral health. However, access, attitude and awareness of women before pregnancy should be improved. 13 The oral health providers in the study suggested reinforcing positive behaviours and correcting negative behaviours in mothers. Pregnant women can benefit hugely from education that promotes caring behaviours and minimises the limitations of pregnancy. Improving oral health outcomes during pregnancy can be achieved through regular dental care, education, broader oral health coverage, routine oral health maintenance and access to resources. The results of various studies in Iran also indicate the effectiveness of education on the awareness of pregnant women in this area and the promotion of oral care behaviours. 30 31 Several factors, including lack of time, high cost and fear of complications for the fetus, can cause reluctance to receive dental services during pregnancy, as highlighted by Mousa et al's study in 2019; However, encouraging regular dental care, providing education, advocating for broader oral health coverage, reinforcing routine oral health maintenance and informing pregnant women about the importance of oral health can all contribute to improving oral health outcomes during pregnancy. 32 Bao et al's study evaluated and summarised clinical practice guidelines on the prevention, diagnosis and treatment of dental diseases during pregnancy. The study found that published clinical guidelines provide clear messages and guidance to dentists for providing timely and safe care during pregnancy. 33 --- Open access Most participants found dental services too expensive, hindering dental visits. Pregnant women may face significant financial constraints when it comes to receiving oral care. According to a population-based study conducted among Iranian women in 2019, the main reasons for not receiving oral services during pregnancy were lack of perceived need and financial barriers. In Iran, oral care services are not completely covered by health insurance, which means households have to pay high out-of-pocket expenses to receive oral care. 34 Evidence suggests that having insurance coverage can significantly improve access to oral health services, enhance service quality and reduce financial barriers to utilisation. 35 However, lack of adequate insurance coverage, incomplete oral services in public sectors, people's lack of trust in their quality, poor access to services and customer dissatisfaction are among the problems of oral services provided in public health centres in Iran. Health professionals are concerned about the everincreasing tariffs and the lack of access for all pregnant women to oral services in Iran. Oral care guidelines in pregnancy should be updated to prioritise oral health. Also, it is essential to have seamless cooperation and coordination between healthcare providers, such as midwives, obstetricians and dentists, to ensure optimal oral healthcare for pregnant women. A study conducted in 2021 suggests that interprofessional collaboration can significantly improve oral healthcare utilisation. 36 The effectiveness of the referral system and patient-centred care team in reducing oral diseases among pregnant women has been emphasised in various studies. 37 38 Participants in a study emphasised that effective communication and interaction among healthcare team members, particularly dentists, are crucial for performing oral care practices. Therefore, dental professionals must develop and maintain strong interpersonal communication skills to establish trust and enhance patient satisfaction with their care previous studies have also confirmed this finding. 39 40 The participants suggested that Telegram and WhatsApp are promising platforms for promoting oral health among pregnant women. A study conducted in 2020 also highlights the potential of smartphones and social networks as an initiative to promote prenatal oral healthcare. 41 The participants in the study emphasised the importance of family, government and cultural support to maintain oral health and improve self-care behaviours. In line with our findings, according to a 2019 study, women whose medical providers promoted oral healthcare were nearly wo times as likely to have a dental visit during pregnancy. 40 Therefore, health centre education can influence dental visits during pregnancy. Educating oral health in families, encouraging dental visits, integrating oral health into general health and paying specific attention to prenatal oral health are some recommendations that can implemented to improve oral healthcare for pregnant women in Iran. --- Limitations There were some limitations to this study. The results of the study may not be generalisable to pregnant women who do not visit public health centres. Another constraint was that the pregnant woman's socioeconomic status might have affected her oral health. Maximum diversity sampling was used to recruit participants from all health centres in different areas of the city. Also, Group discussions could have contributed to a deeper understanding of the concepts but were not carried out due to restrictions caused by the COVID-19 epidemic. --- Implication for future research This study may contribute to future research on needs assessment to create a new specific scale of oral care in pregnancy and improve training programmes. However, more cross-sectional studies are needed to evaluate all perspectives and establish causal relationships between variables. --- CONCLUSIONS Considering the increasing importance of oral and dental health in pregnancy and the existence of health guidelines in this field, there is a fundamental need for a deeper understanding of the experiences and determinants of oral and dental care. By providing an image of the factors affecting the oral care of pregnant women at different individual, physiological, social-cultural and organisational levels, this study emphasises the importance of understanding the unique needs and experiences of pregnant women in research and healthcare settings. --- Contributors All authors participated and approved the study design. MGholian-aval contebuted to designing the study. MAbdollahi collected the data. MAbdollahi,HT,MM and AN_K, analysed the data. The final report and article, were written by MAbdollahi, MM, HT and MGholian-aval and all authors read and approved the final manuscript. MGholian-aval was a guarantor who has full responsibility for the work and the coduct of the study, had access to the data andcontrolled the decision to publish. Competing interests None declared. Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research. --- Patient consent for publication Not applicable. Ethics approval This study involves human participants and was approved by the Ethics Committee of Mashhad University of Medical Sciences with ethics code IR.MUMS.REC.1399.323. Participants gave informed consent to participate in the study before taking part. Provenance and peer review Not commissioned; externally peer-reviewed. --- Data availability statement

Objectives This study aimed to gain a comprehensive understanding of the determinants that influence oral healthcare behaviours among Iranian pregnant women. Study design Qualitative study. Setting Comprehensive health service centres. Participants A sample consisting of all people involved in the process of oral care during pregnancy, 18 pregnant women, 7 midwives/healthcare workers, 3 supervisors of prenatal care services and 3 dentists) were purposefully sampled in terms of demographic characteristics. Methods The qualitative content analysis study conducted 31 semistructured individual interviews in 2022, utilising MAXQDA V.10.The participants identified 3 main categories and 11 subcategories: individual and physiological determinants (care needs, perceived importance, motivation, oral health literacy and inherent pregnancy limitations), organisational determinants (costs, access to equipment and services, review of service delivery process and professional behaviour) and social-cultural determinants (educational services and support from family and friends). Conclusions The results can be used in interventions to improve oral healthcare for pregnant women. This study highlights the importance of addressing individual, organisational and social-cultural determinants to improve oral healthcare during pregnancy.

INTRODUCTION The research project upon which this paper is based is concerned to understand whether, and if so, how, science-based SMEs engage with CoPs in order to acquire new knowledge and whether any innovations are generated through such engagement. The project addresses a number of specific questions relating to CoPs in science-based SMEs: 1. Are CoPs to be found such organizations? 2. If so, how and why do they emerge and/or are they constructed? 3. How is knowledge acquired? 4. What innovations, if any, are generated thereby? The study extends our understanding of CoPs in science-based SMEs through the new empirical material the authors have gathered and the development of a framework for constructing CoPs in science-based SMEs. --- COMMUNITIES OF PRACTICE In knowledge management terms, a community of practice is a group of people informally bound together by shared expertise and a passion for a joint enterprise . The initial theoretical formulation of CoPs focused on the way in which individuals learn, and argued that learning is socially situated, that is, it arises from and within the context, activity, and culture in which it is developed and used. CoP theory has been adapted by knowledge management theorists and used to highlight its value in relation to increasing firms' absorptive capacity, i.e. the ability to recognize the value of new, external information, assimilate it, and apply it to commercial advantage , improve learning and innovation . The emergent and self-selecting membership features of CoPs can contribute to strategy and innovation, help start new lines of business, solve problems on the spot, transfer best practice, develop professional skills and help the recruitment and retention of talent . Through the sharing of practice participants learn together, focusing on problems directly related to their work . This has some affinity with Orr's ethnographic study of photocopy repairmen, although Orr used the term 'occupational communities' rather than CoPs. Lave and Wenger's and Wenger's earlier CoP research failed to acknowledge that a lot of research and technological development work is now done between organizations-these early formulations being rather amorphous, group-centric and inward looking . CoP theory has often ignored the inter-organizational networks and relationships that are required in complex, technology-rich, innovation-driven organizations, where research and development is organized and conducted at distributed locations . Moingeon et al. argue for the support and development of inter-organizational communities of practice , noting that such CoPs can bring together professionals belonging to different organizations and that 'for the organization, IOCoPs indirectly represent a powerful monitoring and innovation force, making both knowledge production and distribution easier' . Similarly, Allee comments that if CoPs emerge in the social space between project teams and knowledge networks they could be beneficial in supporting learning communities that extend both inside and outside organizational boundaries. For science-based SMEs wishing to innovate, there are two clear imperatives: support the development and circulation of knowledge within and pursue alignments across CoPs. Innovation occurs both within and across the boundaries of business firms, universities, research laboratories, suppliers and buyers . 'Open innovation' in particular involves searching outside the firm's boundaries for mutually beneficial relationships , and focuses upon how firms manage their network of internal and external relationships . Open innovation has increasingly emphasized the role of communities in creating, shaping and disseminating innovations . Autio et al. argue that firms involved in collaborative innovation should purposely foster the development of CoPs. There are clear links between open innovation and absorptive capacity, particularly with reference to the sourcing and exchange of externally developed knowledge . CoPs may provide an introductory vehicle for SME knowledge management , a means to boost technological learning, and a means to commercially exploit new innovations . Notwithstanding the above, the authors and others have found that, for science-based SMEs at least, the construction of CoPs can be problematic. It is unlikely, for example, that SMEs can spare the resources required for constructing CoPs . These resources can include time for CoP participation, training for CoP leaders, and IT tools to facilitate participation in CoPs . Given that they have access to fewer resources than larger organizations, Tödtling suggests, however, that they might compensate precisely by engaging in CoPs to support innovation and knowledge management . Probst and Borzillo advocate the use of 'governance committees' to assess the activities of CoPs and management sponsors to guide them, but such practices might be seen by SMEs to divert resources away from core business activities . In the next section the authors briefly outline the research methodology and case study organizations. --- METHODOLOGY The research project used the critical case sampling technique to identify and select cases relevant to the research questions . This helped to generate the four sciencebased SMEs, access to which was negotiated and agreed between key gatekeepers in these organizations and one of the authors. Semi-structured interviews were conducted with 21 individuals employed in technical and commercial roles . The interviews lasted for between 40 and 80 minutes and were audio-recorded with the permission of the interviewee, and subsequently transcribed verbatim. The names of the organizations and interviewees have been anonymized for confidentiality reasons. Template analysis was used to thematically analyze the resultant data. This deploys hierarchical coding, beginning with broad a priori themes, moving to sequentially narrower, more defined themes as the analysis progresses. This method was chosen because it is in sympathy with the adopted social constructionist epistemology, and was used, inter alia, to explore relationships, the meanings attributed to actions, and the various perspectives and orientations of the interviewees. Nvivo qualitative data analysis software was used to code the data and create the final thematic template. Nvivo was helpful in organizing and storing the data as well as generating a hierarchical structure. It also supported an iterative process, allowing for codes to be added, deleted or changed more readily than a 'cut and glue' paper-based approach; it did not, however, replace the iterative process of generating codes, which is where Template Analysis came in. This, argues King , is a flexible, iterative methodology that facilitates a systematic yet reflexive approach to data analysis. Appendix 1 lists the themes which emerged during coding. Whilst the authors' sample is not claimed to be representative of the wider population of SME science-based firms, it does present new empirical material on CoPs, knowledge acquisition and innovation in such organizations, an area where there has been only a limited amount of research to date. The six organizations are now briefly outlined. Company A is a contract chemical processing company providing bespoke outsourcing solutions to the process industry, such as agrochemicals, biocides and herbicides. It is the largest and longest established of the six firms. Company B manufactures a range of in-vitro diagnostic products for use in the detection, prevention, and monitoring of medical conditions related to haemostasis and platelet function. Company C is a university spin-out, manufacturing semi-conductor materials in the cadmium telluride family, which have a variety of applications in medical imaging, security screening, industrial inspection and space exploration. Company D specialises in the design and manufacture of gas sensors and analyzers and provides solutions to gas monitoring in a variety of environments. See Table 1. --- Table 1 about here The following section presents and discusses the findings. --- FINDINGS AND DISCUSSION The findings are presented and discussed below in relation to the four research questions. In Company B there was evidence of apprentice-based, which emerged to support individual learning and internal knowledge-sharing related to individuals' domain of scientific knowledge. Junior scientists moved from 'novice' to 'master' and, through their participation as CoP members, in the process gaining the firm-specific scientific knowledge that they needed to participate in innovation activities: [Name omitted], our R&D Manager, he, and our other less experienced scientists, through [the MD's] mentoring really, is getting in touch with this and that group of people and the knowledge is transferring to [R&D Manager] so he is getting more confident in passing this on to colleagues. . Intra-organizational CoPs were leveraged for a variety of purposes including individual learning, internal knowledge-sharing, and for supporting noncanonical practices. Scientists and engineers pooled their expertise in problem-solving activities. These CoPs emerged through such mechanisms as 'corridor meetings', and shared many similarities with Orr's study of a community of photocopy repairmen. As one participant commented: 'people work in teams generally, but often the most productive solutions happen by accident, like when we chat at lunch about formulations... we discuss possibilities'. In Company D, intra-organizational CoPs were constructed through the use of a boundary object, the 'Imagineering Wall', which was used to encourage informal collaboration and stimulate innovation. This is not dissimilar to the CoP cultivation approach adopted by the organizations in Probst and Borzillo's study. The 'Imagineering Wall' was a large tridivided whiteboard mounted on an office wall, where employees posted ideas for new products or improvements to existing products. Employees were encouraged to post ideas on the wall, no matter how bizarre they might appear, in order to encourage sharing and The company did not keep records of each employee's informal 'wall activities' until these became designated as formal projects. This meant that an employee could move from one project to another without informing anyone and, therefore, he/she might be engaged in multiple CoPs at any one time. Inter-organizational CoPs emerged between SMEs and two distinct sets of customers: commercial organizations, focused on developing new products and processes, and R&D organizations represented by universities and other research organizations seeking to commercialize their inventions. These CoPs were particularly evident in Companies B, D and E. One interviewee commented on the importance of regular contact for building and maintaining strong relationships with customers through informal contact: 'We very much have to work with these customers and build relationships, and that can only be achieved through regular contact' . Recurring interaction with customers produced a sense of shared understanding. By working together towards a mutual goal and engaging in a joint enterprise, inter-organizational CoPs emerged, where the participants sought to develop innovative solutions in partnership with customers: We work together with the customer to provide solutions in a way we both understand. Someone may call, a chemist or engineer for example, or someone else... call to discuss progress, and that's significant in terms of maintaining good relations and meeting their demands for a solution. . The leveraging of personal relationships helped bring people together, including customers, competitors and suppliers. In Company B, the MD had built an extensive inter-organizational CoP consisting of diverse members, including hospital doctors, nurses, GPs, practice managers, hospital technicians and academics. It appeared that an important ingredient in the success of such inter-organizational CoPs was the strength of the social presence of a particular individual or individuals, in this case the MD of Company B. Social presence has been found to be associated with successful community building , and has some resonance with Coakes and Smith's notion of 'innovation champions' who transcend organizational boundaries and support collaboration with other communities and organizations. Company B had been particularly successful at building external connections and, as the R&D scientist acknowledged: 'Well I suppose it's 'who you know' -especially who [the MD] knows as he's worked in the industry a long time'. The MD in this instance exhibited a particularly high level of saliency in both his networking and relationshipbuilding. An example of CoPs emerging between SMEs and their suppliers can be taken from the interviews in Company A, which held occasional events for their suppliers in order to encourage informal knowledge sharing: … we invite major suppliers here … and we will give them, if you like, seminars on different sort of process techniques, which sort of can help them go and find more businesses… and has a reciprocal effect in that they put us in touch with businesses who want help… sort of knowledge [sharing] . There were a number of references in the authors' case study interviews to mutual engagement, shared repertoires, and joint enterprises between these SMEs and their customers and suppliers. In the next section the authors explore how and why these CoPs were constructed. --- How and why are CoPs constructed in science-based SMEs? CoPs in science-based SMEs in this study were found to be an essentially emergent phenomenon. There was also some evidence of attempts to cultivate CoPs. The use of boundary objects stimulated the cultivation of multiple intra-organizational CoPs. The 'Imagineering Wall', for example, brought together scientists and engineers who, through the building of shared repertoires, mutual engagement and joint enterprise, formed a variety of CoPs focused on both incremental and radical innovation . There were implementation problems, however. For example, some employees felt that they could not always use the 10% time allocated to become involved in projects that interested them. Others highlighted the fact that, once they had posted their initial ideas on the wall, they struggled to generate new ideas: It's a good idea… when it was first developed because everyone had been sitting for a few years saying "why didn't we do this", "why didn't we do that"… and when the wall was put there everyone just jumped on it… and everything is discussed and bashed out… and then, after the novelty is worn off everybody's sort of got rid of their ideas and it is now sort of sat… we don't really have enough external stimulation to keep it going… . The wall initially acted as a boundary object, encouraging practice-based creativity, but in order to maintain momentum and help generate new innovative 'big ideas', the organization perhaps needed to move beyond capturing the skills, expertise and creativity of its employees and seek to create or join external, inter-organizational, CoPs. Other efforts to cultivate CoPs included building and maintaining trust-based relationships and credibility with customersclosely linked to developing reciprocal relationships. Building trust was essential in order to facilitate effective knowledge sharing and to help SMEs to better understand their customers' needs -as Field has observed, 'relationships matter'. Through mutual engagement, sharing expertise internally with colleagues and externally with customers and suppliers, SMEs were able to build effective trust-based inter-organizational CoPs, thus generating social capital, developing their capabilities and improving their ability to problem solve on behalf of their customers. This finding is in contradistinction to Hamburg and Marin , who argue that SMEs are unreceptive to knowledge sharing due to a lack of trust. Trust was found to be an essential ingredient for building social capital in inter-organizational CoPs, and the SMEs employed a variety of trust-building activities to engage customers and suppliers. For example, Company A held free seminars to which they invited their customers and suppliers, whereas Company B provided free specialist advice to a university. These companies used such events to build trust and stimulate informal networking. The interorganizational CoPs that developed in the aftermath of such events frequently led to more formal partnerships being established between firms, often based on a mutual need for expertise not possessed internally. These organizations, by offering on-going advice and support, and thus sharing their expertise, were able to encourage informal interaction between members of different cognate organizations, which sometimes formed the basis of interorganizational CoPs. The establishment of trust between members of the different firms was linked to building credibility with customers. Some larger SMEs, notably Companies A and E, encountered similar difficulties when engaging in trust-building activities, and on occasion adopted a pragmatic approach in order to source external expertise. For example, for Company C, which was engaged in radical innovation in developing new products, trustbuilding could be a secondary consideration after identifying someone with the necessary expertise, particularly if the person was not within their existing external network. However, overall, the study found that SMEs were able to build external trust-based collaborative relationships and, CoPs sometimes developed through regular mutual engagement with customers. This helped them to better understand their customers' needs and provide appropriate solutions. Reciprocity was an important part of the trust-building process and was often achieved through offering free advice or additional services. Companies A, B and D all offered their services free to both commercial and R&D customers. Company A provided free services to commercial customers in order to maintain regular, informal contact, build collaborative relationships and generate future business. For non-commercial customers, e.g. universities, these reciprocal arrangements often led to informal knowledge exchanges and the emergence of inter-organizational CoPs. As the MD of Company B commented: … in the past [N] university's got a grant to do this research, then we say 'well, we'll do this', but then that's how we got the contacts, so 'cause we've done this, or that for them, if we want something to be externally validated, we can say, you, 'Do you mind doing this?' . The relationship here was cultivated by the MD of the company. Entrepreneurs such as these, who displayed high levels of social presence, were able to build strong, long-lasting, trustbased relationships. The ability to 'be out there' making connections was an essential factor in building successful inter-organizational CoPs, based on trust and reciprocity. Thus, the knowledge exchanges that occur in CoPs are an essential part of the innovation process in science-based SMEs; in the next section the authors discuss how this knowledge is acquired. --- How is knowledge acquired in science-based CoPs? As has already been seen, internal and external knowledge acquisition is a key part of the innovation process for science-based SMEs. Individual learning was often informal and both apprentice-based and intra-organizational CoPs emerged in response to the need to support the sharing and dissemination of knowledge. Apprentice-based CoPs emerged to support members' needs to expand their existing scientific knowledge and they thus often supported informal 'on the job' training and development. Although individual learning was recognized as being important for innovation, the view of managers was that it had to be aligned with the firm's commercial goals. This could inhibit the emergence or cultivation of CoPs by restricting the knowledge domain: We might have to do bits of research, but the learning for learning's sake… or because of what's interesting, isn't really what we do. We are a commercial company and we need to make products, we need to make money. That's what we're in it for. We're not in to widen everyone's understandings… (New Technology Manager, --- Company C). The individual learning that occurred in the apprentice-based CoPs in Company B was more aligned to notions of incremental innovation, where there is a high level of reliance on building a shared repertoire that facilitates the passing on of existing scientific knowledge internally between CoP members. Conversely, for the SMEs involved in radical or disruptive innovation, such as Company E, the emergence and/or cultivation of apprentice-based CoPs was inhibited by the absence of a shared repertoire, which in turn was related to a lack of a shared domain of knowledge in a particular field of scientific research where there was a high level of novelty or uniqueness associated with the innovation. As one interviewee observed: 'We're not doing projects that are digging ditches or building walls'. Both the emergent and cultivated CoPs which focused on specific domains of scientific knowledge were seen to support knowledge sharing between various professional groups, such as scientists and engineers. Members of these CoPs participated in a range of knowledge-sharing activities, including what one interviewee called 'the obvious stuff', as well as things they 'didn't yet understand'. The impetus for this was often a wish or need to support members in problem-solving activities conducted on behalf of customers. This is illustrated by an interviewee's reference to informal 'corridor meetings' with colleagues in Company C, where internal knowledge exchanges often occurred. CoP members pooled their expertise, participating in problem solving activities, a kind of incremental innovation, on behalf of customers: Internally, there is a network, which is probably the four people who I talked about, you know, the IT guy, the detector guy, myself and Ben, and that's an informal network, you know, but we're… we tend to do corridor meetings, you know: 'how's that?' … our Venn diagrams of influence cross over so regularly.. . It is important to note, however, that the construction of CoPs could be inhibited. In Company C, for example, the Chief Technical Officer commented that for senior scientists occupying strategic roles in SMEs, it was often difficult to access intra-organizational CoPs due to a lack of 'band width' inhibiting their capacity to build CoPs relationships and leverage the social capital necessary for innovation. CoP participation became more difficult because they could not maintain regular contact with their network of scientific colleagues. Additionally, conflict frequently emerged between scientists and engineers, the former viewing innovation as a 'journey' whilst the latter saw it more in terms of a 'destination'. This sometimes restricted the emergence of intra-organizational CoPs. In general, scientists were considered to be less commercially aware than engineers, who were more focused on developing an end product: An engineer wants to get somewhere. In fact an engineer's happy if something never happens again, you know I think that's the thing that… a scientist likes to do a thing over and over and over, you know, 'we'll try that', 'I'll do that experiment'. An engineer's target, I would say, is to make it to have nothing to do in a day. . In SMEs focused on radical innovation, i.e. Company E, this conflict inhibited knowledgesharing activities generated through the emergence or cultivation of CoPs, thereby restricting their ability to turn innovations into commercial products. The literature suggests that in networks people might fear losing their expert status, whereas organizations might fear disclosure of their competitive knowledge, thus questioning whether organizations are willing to initiate innovation projects with new partners . SMEs in particular might refrain from participating in such innovation networks because of the perceived increased risk to their competitiveness . However, Meeus and Oerlemans observe that the limited resource base usually available to small firms does not negatively impact on innovativeness as they are able to develop the adaptive behaviour conducive to innovation. The present study found that collaboration was an essential part of the commercialization of innovation for all the SMEs for a variety reasons, including sourcing external scientific expertise from other firms and universities, seeking commercial expertise to help them exploit their innovations, as a route to market through licensing agreements, and as a source of funding. At the same time, customers often sought collaboration with SMEs to acquire external knowledge. Companies A, B and G in particular used this approach to bind customers to them, in effect cultivating an inter-organizational CoP by building a shared repertoire through regular, informal contact. There was much evidence of the internal pooling of expertise and exchanges of tacit knowledge through collaborative activities, often resulting in or from the emergence of CoPs. SMEs acted as 'key experts', solving problems on behalf of their customers, and there was evidence of enhanced absorptive capacity based on SMEs' collaborative activities with customers. The capability to access internal and external pools of knowledge suggests that although these SMEs may have no formal strategy for knowledge acquisition, they did possess some absorptive capacity based upon the collaborative relationships that develop through emergent CoPs : Right at the outset they'll [customers] come to us and say, 'Look, we've got a particular problem. Do you think you can solve it?' And, coming back then to the pool of knowledge and experience we have within the business, then we can apply ourselves and we can usually come up with solutions, or recommend alternative solutions… but, you know, we can either develop... bring in outside expertise or pass perhaps potential enquiries on to other people that we know or source solutions from people in our network. . The SMEs which were collaborating predominantly with R&D customers tended to value collaborative trust-based relationships more than those mainly working in partnership with commercial organizations. For the former, collaboration with universities was an important source of external scientific knowledge, as well as providing access to research and development funding and specialist scientific equipment and facilities. The latter tended not to consider themselves innovative; although they engaged in collaboration with customers, they were generally suspicious of, or failed to see the value in, collaboration with other external organizations. This suggests that those SMEs who see themselves as engaged in innovative activities tend to be more aware of the importance of collaboration and are also more likely to actively seek collaborative relationships, engaging in the CoPs that emerge through shared repertoire and joint enterprise, focused on their specific scientific domain of knowledge. In summary, the findings illustrate how science-based SMEs acquire internal and external knowledge through the construction of, and participation in, CoPs. CoPs support individual knowledge acquisition and improve absorptive capacity, both of which are necessary for successful innovation. In the next section the authors discuss the impact this participation has on their innovation activities. --- What innovation is generated from collaborative activities in CoPs? The findings of this research suggest that in relation to radical innovation, the construction of CoPs is inhibited by a lack of shared repertoire, especially for those science-based SMEs at the forefront of their respective scientific field. The literature indicates that participation in intra-firm knowledge sharing activities is often difficult because it is dependent on CoP members' willingness to voluntarily share their knowledge and expertise . The findings support the view that CoPs are more likely to support incremental innovation in science-based SMEs. For SMEs working with R&D organizations, innovation was often described in terms of novelty or uniqueness, i.e. aligned with ideas of radical or disruptive innovation and verified by 'new to the world' products and processes. The SMEs working with commercial customers considered innovation more in terms of 'problem solving' activities, i.e. aligned with views of incremental innovation, evidenced by collaborative activities as problemsolvers for their customers. The SMEs involved in producing novel innovations sometimes found it difficult to engage with external organizations because the uniqueness of the knowledge base associated with their innovation resulted in the lack of a shared repertoire. In effect, the SMEs were frequently the 'knowledge leaders' in their particular field. Although novelty provided capability advantages, some interviewees considered it a barrier to collaboration: 'We are the only people in the world doing this… innovation is quite difficult 'cause it can only come from ourselves' . This finding suggests that for SMEs generating novel innovations, their status as 'knowledge leaders' restricts their ability to engage in inter organizational CoPs because of a lack of shared repertoire. However, some interviewees in Company E, generating novel innovations, were aware that further incremental innovation, often generated through collaborative activities, was required in order to successfully commercialize their inventions. A lack of shared repertoire restricted their ability to engage in informal networks, or in the informal collaborative relationships that often result from CoP participation. Nevertheless, there was collaborative relationship-building and the emergence of shared repertoires through mutual engagement with a range of external organizations, including customers, suppliers and competitors. Tension was noted between scientists and engineers regarding their views of innovation. At one extreme, there were scientists who said they sought to discover entirely new processes and knowledge. At the other extreme, there were engineers who used their knowledge to build useful devices. In general, scientists viewed innovation as a 'journey' and appeared to be less commercially aware than engineers, who were more focused on developing an end product. In those SMEs focused on radical innovation, the level of conflict between scientists and engineers was often more pronounced because of the absence of a shared repertoire. This was related partly to a lack of common existing knowledge, but also because of the uniqueness or 'novelty' associated with radical innovation. This aspect sometimes inhibited the internal knowledge-sharing activities associated with the emergence or cultivation of intra-organizational CoPs. On the other hand, these same SMEs, in particular Company C, generally found it easier to collaborate with R&D customers, such as universities, and there was evidence of both resource sharing and knowledge sharing through inter-organizational CoPs. We still collaborate with the university… in fact we built a system… one of the four systems is at the university, you know, so we have a fairly strong collaboration with The interviewees expressed a range of views about innovation per se. Those who saw their firm's innovation activity as being towards the radical end of the spectrum generally considered their firm's products and processes to be unique. Others saw the innovation activity in more incremental terms, e.g. offering 'better' solutions by making a product simpler in order that customers find new uses for it. Uniqueness was considered a source of competitive advantage to the SMEs, who were often operating in niche markets. However, interviewees also recognized the importance of engaging in incremental innovation in order to develop new products based on their initial science-based innovation, or when they were commercializing their innovations. Although novelty or uniqueness was considered a capability advantage, the lack of shared repertoire restricted the community-building activities associated with inter-organizational CoPs. Interviewees in Company A did not consider their organization to be innovative because it did not manufacture its own products; those in Company D expressed a similar view because they created new products by integrating existing technology rather than through their own R&D activity. However, the Imagineering Wall had been successful in generating a number of small incremental innovations to existing products; the CoPs that emerged from the associated activity had produced 6 new versions of products that were on the market, and a further 4 projects were in the early stages of development . Given these organization's incremental approaches to innovation and their problem-solving activities on behalf of or in partnership with customers, one can see many similarities with the sort of 'user-led' innovation communities discussed by Heiskanen et al., and Von Hippel and Finkelstein , which, the authors would argue, are in essence a form of interorganizational CoP. Problem solving requires in-depth collaboration and, as these examples show, SMEs share knowledge with their customers in order to develop complex innovative solutions. They use a variety of methods to engage with customers, including hosting events to showcase their expertise and visiting customers to offer advice and training related to their products. Thus customers were sometimes reliant on the sharing of scientific expertise with the SMEs, and this required individuals from the organizations to work closely together, engaging in problem solving activities and generating the mutual engagement and building of shared repertoires necessary for inter-organizational CoPs to emerge. --- CONCLUSIONS --- Findings The empirical findings of this research project confirm the existence of CoPs in sciencebased SMEs and show how they emerge or are cultivated for a variety of purposes, including knowledge acquisition and innovation. This confirms the widely held view in the extant literature that CoPs exist in most organizations . Three types of CoPs were identified: apprentice-based CoPs that support individual learning; intra-organizational CoPs that facilitate internal knowledge sharing ; and inter-organizational CoPs that emerge between SMEs and external organizations , and were leveraged to support a range of collaborative innovative activities. However, the SMEs did not always recognize either the presence of CoPs or the benefits they might gain from explicitly engaging in their construction. The findings also identified instances where, although the conditions for constructing CoPs were present , CoPs did not emerge. This suggests that science-based SMEs could do more to develop and support the cultivation of CoPs. --- Limitations Although template analysis is a useful tool for analyzing qualitative data, issues arose in relation to its use in this research project. Deciding on the extent of the initial template and the number of codes was a necessarily subjective process, potentially leading to accusations of researcher bias. In order to mitigate such bias, the data was coded, discussed and reviewed at various stages by the authors. Although this provided some consistency in terms of applying the template method, it did not allow for the inclusion of multiple perspectives. Involving a wider range of reviewers, or even involving the interviewees themselves, might have generated some insights when developing the template. Sampling bias: these sciencebased SMEs represented a heterogeneous group of firms from a range of scientific disciplines. However, they had science at the core of their business and, therefore, could be argued to represent a variety of typical, although not necessarily representative, cases that were useful in generating data and making comparisons and contrasts. Whilst the authors cannot make any claim for the representativeness of the sample in relation to the wider population of science-based firms, they do maintain that this new empirical material and theorizing on CoPs and innovation in science-based SMEs makes a contribution to a research domain which has been dominated, especially in recent years, by a focus upon large organizations and there has been an over-reliance upon the voice of senior managers . In the next section the authors present a contextualized framework for the construction of CoPs in science-based SMEs. --- Implications The contextualized framework for constructing CoPs (Figure 1 SMEs should also encourage individuals who display high levels of 'social presence' to act as brokers and boundary spanners, drawing external knowledge and expertise into the organization, and thus enhancing their firm's absorptive capacity. SMEs which pursue these sorts of activities and take these sorts of initiatives can thereby encourage and support CoP cultivation, and thus build trust and reciprocity, leading to enhanced social capital. Participation in intra-organizational CoPs increases firms' absorptive capacity; participation in inter-organizational CoPs stimulates open innovation . This helps SMEs to build new networks and facilitates more effective knowledge transfer, thus helping to generate more focused innovation in science-based SMEs. --- Suggestions for future research The findings provide a number of insights and pointers for future areas of research. Firstly, whilst it was not possible, given the time constraints of the current research, to conduct a longitudinal study, this would be helpful for examining in more detail, inter alia, the types of CoP activities taking place and how they change over time, along with the membership of the CoP. Secondly, future research would benefit from a larger sample size to establish a better understanding of CoPs in science-based SMEs. Thirdly, this research focused on the views and experiences of managers, scientists and engineers-it would be valuable to gather the views and experiences of a wider range of people, both internal and external to the focus organization.

Purpose -Recent research into communities of practice (CoPs) has focused on large organizations, suggesting they can be constructed for the purposes of knowledge acquisition and innovation. The study provides evidence that for small science-based firms CoPs are more likely to emerge unplanned to support incremental innovation in the form of problem solving activities. Design/methodology/approach -Thematic template analysis was used to analyze 25 indepth interviews conducted with a range of employees in science-based SMEs. Findings -Analysis of the interview data from a case study of seven science-based SMEs provides evidence that both intra and inter-organizational CoPs were leveraged for a variety of purposes, including the acquisition of knowledge and improving firm's ability to generate innovative solutions. Research limitations/implications -Whilst there cannot be any claim of 'representativeness' of the sample in relation to the wider population of science-based firms, the authors do maintain that the new empirical material has made a contribution to a research area which has been dominated, especially in recent years, by a focus upon large organizations that have adopted a managerialist orientation. Originality/value -The challenges of understanding the role of CoPs in science-based SMEs and the factors that influence their success, or otherwise, have not been sufficiently investigated and have received little attention in the extant literature. Hence, this paper makes an original contribution to this body of knowledge.

Background The World Health Organization estimates that more than 4 million health workers are needed to fill the health workforce gap globally. This includes 2.4 million physicians, nurses and midwives. Fifty-seven countries are defined as having a critical shortage of health staff; of these, 36 are in sub-Saharan Africa. Of the total world's health work force of 59.2 million, Africa has only 3% of the world's health workers in spite of having 25% of the global burden of disease [1,2]. The shortage of health staff cripples the health delivery system. It is also a threat to provision of essential, life-saving interventions such as childhood immunizations, provision of safe water, safe pregnancy and childbirth services for mothers as well as access to treatment for AIDS, tuberculosis and malaria. Health workers are critical to the global preparedness for and response to threats posed by emerging and epidemic-prone diseases. Different interventions have been tried to address these shortages [3][4][5]. Retaining health staff in rural areas has proven extremely difficult as young professionals increasingly prefer urban postings and health systems do not reward rural service [6, 7. 8]. For example, Wilson andcolleagues, 2009, Dovlo, 2003 and Kruk and colleagues, 2010, found that rural exposure, poor working conditions, low job satisfaction, political and ethnic problems, and sometimes, civil strife and poor security in most underserved areas, predispose new graduates to select cities [9][10][11]. Qualitative research has also shown the importance of health care providers' personal characteristics and value systems, such as religious beliefs and sociopolitical convictions, to their motivation towards rural practice. Emigration of skilled professionals to high-income countries is another barrier to adequate staffing of health facilities. A study in Ghana in 2006 on trainee physicians and nurses revealed that the majority had considered emigrating. More physicians than nurses considered emigration. These findings imply that achieving improvements in the health status of people living in low-income countries, and particularly, in rural areas, will be extremely difficult and the attainment of the United Nations Millennium Development Goals 4, 5, and 6 by 2015 [12,13], in Ghana is unlikely. While previous research has looked at incentives and working conditions to promote uptake of rural posts, few studies have focused on motivation crowding and its effect on willingness to accept postings to rural area. Motivation crowding [14] is the conflict between external factors , such as monetary incentives or punishments, and the underlying desire or willingness to work in areas needed most. Students may have a mix of extrinsic and intrinsic motivations for studying medicine. Extrinsic factors may either undermine or strengthen intrinsic motivation, led by the belief that medicine has the imperative to help others, as enshrined in the Hippocratic Oath [15][16][17]. Current monetary incentives, which favour urban practice, may crowd-out the intrinsic desire to give back to society by working in underserved areas [18,19]. This could have debilitating effects on health worker retention in rural areas [20][21][22]. To tackle the maldistribution of human resources for health , understanding the factors that crowd-out the intrinsic motivation of medical students and their willingness to accept postings to rural underserved area is integral. This paper analyzes the effect of extrinsic versus intrinsic motivational factors on stated willingness to accept postings to rural underserved areas in Ghana. --- Methods --- Study site Ghana is located in the west coast of Africa with an estimated population of 23 million [23]. It is mainly an agrarian economy with estimated per capita income of USD 1500. More than two-thirds of the economy is rural, with the cities of Kumasi and Accra having the highest population densities due to brisk economic activities, and relatively strong economic and cultural infrastructure [24]. These attributes make the two main cities destinations for rural-urban migration and a highly heterogeneous socio-cultural context. There are four medical schools in Ghana: the University of Ghana , Kwame Nkrumah University of Science and Technology , University for Development Studies , and University of Cape Coast . In Ghana, medical education consists of three years of basic science/para-clinical studies, three years of clinical training at a teaching hospital, and a two-year rotating housemanship. The study was conducted with two public universities in Ghana: University of Ghana in Accra and Kwame Nkrumah University of Science and Technology in Kumasi. These universities were selected because all the fourth year medical students in the public universities had their clinical training at either UG or KNUST at the time of the study. All 310 fourth year medical students in the country were invited to participate in the study; no sampling was conducted. Fourth-year medical students were selected because they had completed the BSc. Human Biology and had also been exposed to field work, but had not yet made their final decisions about rural or urban practice. --- Data collection Data collection was preceded by discussions with the heads of medical training institutions, who informed the content of the questionnaire and provided access to the student population. The data collection instruments were developed after seven focus group discussions of 6-8 participants in each group facilitated by trained social scientists were held with third and fifth year medical students at UG and KNUST. The themes for the focus group discussion were motivation, willingness to work in deprived areas, experience in the field, and the influence of background characteristics on willingness to work in deprived areas. The survey instrument, which included semi-structured questions as well as a discrete choice experiment, were then pretested and finalized for the study. The questionnaires were administered electronically with Sawtooth Software SSI Web CAPI [25], in computer laboratories in UG and KNUST. --- Ethical considerations The study received ethics approval from the Ghana Health Service Ethical Review Committee; the UG Medical School; the KNUST Committee on Human Research, Publications, and Ethics; and the University of Michigan Institutional Review Board. All respondents voluntarily participated after the intent and the design of the study had been explained to them and signing informed consent forms. --- Statistical Analysis The study used STATA v10.1 for statistical analyses [26]. The main outcome of interest was the willingness to work in a deprived area after graduation. Students were asked to rate how likely they were to work in a deprived area on a scale from 1-4, where 1 represented "I will definitely not work in a deprived area;" 2 "I am unlikely to work in a deprived area;" 3 "I am likely to work in a deprived area;" and 4 "I will definitely work in a deprived area." We collapsed this response set to a dichotomous willing versus unwilling to practice in a deprived area. We defined deprived area as "a rural area that is distant from the big cities with few social amenities such as schools, roads, pipe-borne water, etc."-as per Ministry of Health definition. Predictors of interest included motivation , demographic characteristics, and rural exposure. Students were asked to identify the top three factors that motivated them to study medicine from a list of twelve factors identified as important by the focus group discussions. Five intrinsic motivations included: desire to help others, desire to give back to their home community or country, interest in medicine as a subject matter, inspiration by a role model, and loss of a loved one. Seven extrinsic motivation factors included: income of physicians, job security and lifestyle, social status/prestige, proposed by parents, opportunities to travel and work internationally, ability to use new cutting edge technologies, and research opportunities. Respondents were coded as having "strong" intrinsic or extrinsic motivation if two or more of their motivational factors were intrinsic or extrinsic. Thus, "strong" intrinsic and extrinsic motivation groups were mutually exclusive. Demographic factors included: sex, age, ethnicity , partnership status , and parental parental professional and educational status . The Akan peoples are the largest ethnic group, representing approximately half of Ghanaians; we grouped all other smaller ethnic groups together as "non-Akans" for this analysis. High parental professional and educational status was defined as having a mother and/or father who is a University-trained professional and low PPES was defined as neither mother nor father is a University-trained professional. Rural exposure factors included: birth location , location of pre-medical studies , having ever lived in rural area , exposure to rural service in medical training . Bivariate associations and 95% confidence intervals were estimated using logistic regression. In model 1, the influence of strong intrinsic/extrinsic motivation on willingness to accept postings to rural area were assessed. Socio-demographic factors were added to the regression in Model 2, and rural exposure factors were further added in Model 3. --- Results --- Socio-demographic characteristics Of the 310 eligible medical students, 307 participated in the survey . Of these, five survey files were corrupted by viruses or lost due to computer malfunction; thus the analysis was conducted with 302 total records. The socio-demographic characteristics of respondents are presented in Table 1. Of the 302 respondents recruited for the study, the majority were male , with a mean age of 22.9 . Most respondents were born in or around urban areas and had never lived in rural underserved area . In terms of socioeconomic status, 173 students came from high PPES families. About half of the respondents were exposed to rural service. --- Intrinsic and extrinsic motivation and likelihood of working in underserved area The intensities of current motivational factors are presented in Table 2. Overall, 158 of students stated that they were likely to or definitely would work in an underserved area. More than 6 in 10 respondents had strong intrinsic motivation. A higher proportion of respondents who had strong intrinsic motivation indicated willingness to work in a rural area, compared to those with weak intrinsic motivation . These results were reversed for those with strong extrinsic motivation . --- Regression analysis of motivations and the willingness to accept postings to rural underserved area after graduation We present the multivariate logistic regression results for the strength of intrinsic motivation and willingness to work in a rural underserved area after graduation in Table 3. In the unadjusted model there was a significant association between strong intrinsic motivation and willingness to work in rural underserved area . In Model 2, with demographic factors, female gender and high PPES were associated with reduced willingness to practice in a deprived area and , respectively, while age was associated with greater willingness to practice in a rural area . Rural exposure factors were not significant when added to the model with intrinsic motivation and demographic factors . In the adjusted models 2 and 3, motivation was no longer a significant predictor of willingness to practice in a deprived area. In Table 4 the influence of strong extrinsic motivation on willingness of students to work in rural underserved area is presented. In Model 1, having a strong extrinsic motivation reduced the odds of being willing to accept a job in an underserved area to . In the model adjusting for demographics, Model 2, female gender and high PPES were associated with reduced willingness to practice in underserved areas respectively, while age was associated with greater willingness to practice in a rural area . Rural exposure factors in model 3 did not influence the outcome of willingness to work in rural underserved area. In the adjusted models 2 and 3, motivation was no longer a significant predictor of willingness to practice in a deprived area. --- Discussion We found that twice as many students reported high intrinsic motivation compared to high extrinsic motivation to study and practice medicine. This may reflect the underlying altruistic motivation for many students entering a profession focused on serving others [12,13]. There may also be an element of social desirability bias in the students' responses as intrinsic motivation may be thought to be more socially acceptable than extrinsic motivation. Nonetheless, we found that high extrinsic motivation was associated with low self-reported likelihood of rural practice and that the converse was true for high intrinsic motivation [2,3]. Interestingly, this association lost statistical significance at the 95% confidence level in models with demographic and rural exposure confounders, whereas socioeconomic status retained a highly influential role, as discussed below. In this study, rural origin did not influence students' willingness to practice in rural areas after controlling for intrinsic/extrinsic motivation and demographic characteristics. This is in contrast with studies which have found rural origin to be an important motivator for rural practice [8,17,23]. Our findings highlight the heterogeneity of trends in motivation dynamics for rural practice and the importance of locally-relevant data for decision making. High PPES, measured using parental education and profession, was consistently associated with lack of willingness to work in rural areas. This is concerning as nearly 6 in 10 medical students in this cohort were from high PPES backgrounds-which is typical for Ghanaian medical schools [7]. These findings suggest that admission policies that favour well-to do applicants may be reducing the pool of students willing to consider rural practice. Female gender was also strongly associated with reduced interest in rural practice for women even after controlling for extrinsic/ intrinsic motivation and rural exposure variables. This is consistent with similar studies among health staff which revealed that women are less likely to accept positions in remote areas due to varying family reasons; they would like to live where their husbands jobs are, have difficulties convincing their husbands to follow them to rural areas and want their children to have better education in the urban areas [10,11,22]. The studies further explained that female doctors rarely live in the same village as their assigned post and have higher overall absentee rates in rural practice [19,20]. With increasing representation of female healthcare professionals in many places in sub-Saharan Africa, [10,11], it is likely that the supply of health staff to rural underserved areas will remain a major setback if professional motivations are designed to attract more female students to rural practice. Although our study showed a lower proportion of female medical students in Ghana compared to other areas, they are likely to become a more important cadre in the coming years [11,19]. More research is urgently needed to determine how female healthcare professionals' motivations towards rural practice can be better engaged by policy-makers. The limitations of this study include the possibility of social desirability bias in responses on motivation and likelihood of rural practice, as noted above. Despite the fact that study participants were assured of anonymity, confidentiality, in responding to the questions, some social desirability bias is likely. For this reason, we selected a measure of high intrinsic and extrinsic motivation for use in the regression models. Research comparing students stated intentions with their actual career choices during internship is urgently needed as few studies on matched follow-ups are available. In addition, most the students participating in the study were young and had not yet tasted the rigors of working in a rural area, which may affect their job preferences. Thus the findings of this study may not be applicable to practicing physicians. Finally, these findings are only generalizable to students in the current medical education system. The findings may be different if selection criteria for medical school admission change. The major strengths of this study are its high response rate of 99% and that its ability to capture an entire population of young medical students who are one of the targets for addressing the rural-urban health staff recruitment imbalance. Surveying practicing physicians would have missed out those who had migrated. This study has several implications. First, the majority of students profess high intrinsic motivation for rural service. More research is needed to determine the potency of this motivation source in real-life decision making and how to best engage it via HRH policy. It is possible that emphasizing the community service aspect of medical practice and elevating the status of rural primary care in under-graduate and post-graduate training may help narrow the gap between motivation and eventual career choice in favour of rural areas. In addition, well-supervised and supported rural placements in which students experience the rewards of rural practice may help to persuade students who are largely unfamiliar with rural life. However, the success of these rural rotations is likely to depend heavily on having adequate local infrastructure and mentorship [13,23]. Second, admission criteria may need to be reconsidered in light of the strong relationship between high PPES and lack of interest in rural practice. For example, medical school admission slots might be reserved for qualified students from poorer families. These students may not need to come from rural areas as we found that none of the rural exposure factors were significant after controlling for motivation and demographics. Third, our results suggest that programmes to promote and support rural practice after graduation may have some success. Our focus groups and discrete choice experiment suggested that students may be willing to commit to short-term placements of 2 years or less in rural areas [9]. The Ministry of Health may want to consider the possibility of short contracts that rotate physicians in and out of difficult to staff rural areas. --- Conclusions The majority of students still claim high intrinsic motivation and therefore it is important to appeal and build on this in medical school curricula and in designing rural postings. However, extrinsic motivation and, perhaps most importantly, PPES, will likely continue to be an important factor in deciding on job postings. Future research should explore how motivations could be directly supported, how motivations are formed, the influence of contextual factors on motivation among medical students, and motivation crowding among practicing health professionals in rural underserved areas. Qualitative work may be especially informative in this effort. Our research also suggests that increasing efforts to recruit medical students from low PPES backgrounds may be the most effective current pathway to increasing the yield of physicians willing to practice in underserved areas. --- Authors' contributions PA-B, SRK, JCJ, EN, MG, KA, MD, JK, RCS, and MEK jointly conceived the study and designed the survey. PA-B, JCJ, MG, EN, and KA carried out data collection, under supervision by MEK. All authors were involved in the interpretation of study findings. PA-B, wrote the first draft of the manuscript. All authors reviewed and critically revised the manuscript for important intellectual content and agreed to submit the manuscript for publication. --- Competing interests The authors declare that they have no competing interests.

Background: Retaining health workers in rural areas is challenging for a number of reasons, ranging from personal preferences to difficult work conditions and low remuneration. This paper assesses the influence of intrinsic and extrinsic motivation on willingness to accept postings to deprived areas among medical students in Ghana. Methods: A computer-based survey involving 302 fourth year medical students was conducted from May-August 2009. Logistic regression was used to assess the association between students' willingness to accept rural postings and their professional motivations, rural exposure and family parental professional and educational status (PPES). Results: Over 85% of students were born in urban areas and 57% came from affluent backgrounds. Nearly twothirds of students reported strong intrinsic motivation to study medicine. After controlling for demographic characteristics and rural exposure, motivational factors did not influence willingness to practice in rural areas. High family PPES was consistently associated with lower willingness to work in rural areas. Conclusions: Although most Ghanaian medical students are motivated to study medicine by the desire to help others, this does not translate into willingness to work in rural areas. Efforts should be made to build on intrinsic motivation during medical training and in designing rural postings, as well as favour lower PPES students for admission.