Social integration and immigrant mental health: the role of social capital and neighbourhood ethnic composition

Sammanfattning: Background: Nearly 20% of Sweden’s population are immigrants and research has identified inequalities in mental health between certain immigrant groups and the native Swedish population. The role of social integration factors such as social capital, defined as: “features of social organization such as networks, norms, and social trust that facilitate coordination and cooperation for mutual benefit”, has been minimally explored in understanding differences in immigrant health. The influence of neighourhood context, as a prominent arena for immigrant social integration, has been increasingly studied, given that European immigration is commonly characterized by neighbourhood economic and ethnic segregation. Previous research has established that living in an economically deprived neighbourhood can negatively impact mental health. However, other research indicates that living among coethnics or other immigrants in general may offer some protection even in deprived settings. Thus, neighbourhood ethnic composition, measured in terms of overall ethnic density, diversity, and own-group ethnic density, have become areas of interest for researchers. Few studies have researched this phenomenon in Sweden. Aim: The aim of this thesis is to contribute to the overall body of knowledge on understanding the relative importance of various aspects of social capital and neighbourhood ethnic composition, as individual and contextual social integration factors, for the mental health of immigrants in Sweden. Materials and methods: All studies were cross-sectional, deriving data from the Stockholm County Council’s Stockholm Public Health Cohort (SPHC) sampled from Stockholm’s general population (n ~ 50,000; Articles I, II, & III) and the Red Cross University’s Resiliency, Mental Health, and Social Integration of Refugees (RMSR) cohort of refugees from Syria (n = 464; Article IV). The specific objectives and methods were to: 1) Use logistic regression to identify inequalities in psychological distress between Swedish-born individuals and various immigrant groups, defined based on their reason for immigration and duration of residence in Sweden. The Baron & Kenny method was also utilized to investigate if social capital explains these inequalities (Article I); 2) Use multi-level Poisson regression to investigate the association between neighborhood ethnic composition, defined as overall ethnic density and own-group ethnic density, on psychological distress and whether aspects of individual and contextual social capital explain this relationship (Articles II & III); 3) Conduct multi-group Structural Equation Modelling to investigate the association of social participation on depressive symptoms within a cohort of newly-arrived refugees resettled from Syria, and whether it is moderated by social network type or mediated by social support (Article IV). Results: Article I found that psychological distress does vary for immigrants based on reason for immigration and duration of residence in Sweden. All immigrant men, regardless of reason for immigration or duration of residence, had greater psychological distress than their Swedish-born counterparts whereas only refugee women living in Sweden 10 years or more had greater psychological distress than Swedish-born women. Social capital did explain differences between migrants and their Swedish-born counterparts, with social support showing the strongest effect. Preliminary results from the cohort of refugees from Syria in Article IV shed light on the potential role of network type in accessing social support for newly-arrived migrants, finding that participating within bonding networks amplifies social support, resulting in a twofold decrease in depressive symptoms compared to those participating primarily within bridging networks. Regarding neighbourhood ethnic composition, Article II found that the association between overall ethnic density and psychological distress differed for individuals depending upon ethnic background. There was no apparent association for respondents with foreign-background (ie. Swedish-born with two foreign-born parents). However, for both foreign-born and respondents with Swedish-background, the detrimental effect of overall ethnic density on psychological distress was mainly explained by socioeconomic factors, with social capital having a small effect. Preliminary results of Article III found that among eight of Sweden’s largest immigrant groups, neighbourhood own-group ethnic density was associated with psychological distress for only immigrants from former Yugoslavia after accounting for demographic and socioeconomic factors. This could indicate either a lack of accessible culturally-meaningful resources and/or the continuation of ethnic tensions after resettling in Sweden given that many came as refugees during the Yugoslav wars in the early 1990s. Social capital did not appear to affect the relationship between own-group density and psychological distress. Conclusions: The findings of this thesis contribute to the scientific evidence that social capital and neighbourhood ethnic composition, as individual and contextual social integration factors, can influence immigrant mental health. Regarding social capital in particular, this thesis further confirms that while weaker, more diverse social ties can prevent mental ill-health, the strong social ties characterising social support and co-ethnic networks offer even more protection. Furthermore, the detrimental relationship between neighbourhood ethnic composition and mental health was explained by socioeconomic factors, indicating that economic hardship, both at the individual and contextual level, might overpower any influence of social ties. Taken together, these results indicate that policy initiatives should seek to minimize the economic and social exclusion of immigrants.