Contextual effect of different components of social capital on health in a suburban city of the greater Tokyo area: A multilevel analysis

Abstract

Social capital (SC) can be broken down into a number of aspects and dimensions, but few studies have differentiated between the effects of different components of SC on health. This study examined the relationship between contextual SC and health (self-rated health, and co-occurrence of lifestyle risk factors such as smoking, drinking, overweight/underweight and physical inactivity) among the general population in a Japanese suburban area. The specific research question was to explore which components of contextual SC had what effects on health. In 2009, we randomly selected 4123 residents, aged 20 years and over, from 72 districts in the city of Kashiwa, Chiba prefecture (a typical suburban city of Tokyo) to participate in a cross-sectional survey using mailed questionnaires. We used four indicators of SC: cognitive/horizontal (trust in neighbors), cognitive/vertical (institutional trust in the national social security system), structural/horizontal (participation in groups with egalitarian relationships) and structural/vertical (participation in groups with hierarchical relationships). District-level SC was calculated by aggregating the individual responses of each SC indicator within each district. The response rate was 42.1% (1716 questionnaires), 43.7% of the respondents were male, and the mean age was 54.8 ± 16.4 (ranging from 20 to 97). A multilevel analysis showed that higher district-level institutional mistrust was associated with self-rated poor health, but higher district-level mistrust in neighbors was inversely associated with it, after adjusting for individual-level covariates. There was no contextual effect of any SC components on co-occurrence of risk factors. Our findings showed that institutional trust has a beneficial effect on self-rated health, but trust among neighbors might negatively affect the health of the residents in a Japanese suburban city. These unique findings could suggest the advantage of breaking down SC to examine more specific relationships between SC and health, and the importance of accumulating the evidence in specific cohorts to develop customized health promotion strategies.