Does social trust at school affect students’ smoking and drinking behavior in Japan?

This study examined the individual and contextual effects of cognitive social capital at school on cigarette smoking and alcohol drinking in Japanese high school students. Self-administered questionnaires were distributed to 3248 students in grades 10–12 (aged 15–18 years) at 29 public high schools across Okinawa, Japan in 2008. The individual-level cognitive social capital studied was generalized trust. Using multi-level logistic regression models, the effects of individual- and contextual-level cognitive social capital on smoking and drinking were analyzed. Contextual-level cognitive social capital was measured on the basis of aggregated individual responses to the trust question at school level. After adjustment for the covariates, individual-level trust was negatively associated with smoking and drinking among boys and girls. Similarly, after adjustment for the covariates, school-level trust showed an inverse association with smoking for girls. A similar but not statistically significant association was observed for boys. On the other hand, school-level trust was not associated with drinking among boys or girls. After adjustment for individual-level trust and the covariates, these findings were in the same direction, but the school-level trust for girls no longer had a significant contextual effect on smoking. The findings suggest evidence of the individual effect of cognitive social capital on adolescents’ smoking and drinking, and that the contextual effect of social capital on smoking was inconclusive. In addition, no association of contextual-level social capital with drinking was observed.

Research highlights
► This study examined the individual and contextual effects of cognitive social capital at school on cigarette smoking and alcohol drinking in Japanese high school students.
► The individual-level social trust was negatively associated with smoking and drinking among boys and girls.
► This study suggests a contextual effect of social trust at Japanese school level on levels of smoking, but the association is inconclusive because it did not reach statistical significance.
► There was also no association between contextual-level social trust and alcohol drinking.
► This study expands the research on social capital and health in neighborhoods to Japanese schools, and provides important implications for school health promotion strategies.