Neighborhood sociodemographic structure, neighborhood cohesion, and coronary heart disease mortality: A multilevel path analysis of a large Swedish cohort

Various studies have documented an effect of the neighborhood socioeconomic environment on coronary heart disease (CHD), but few have investigated its underlying mechanisms. Besides effects of the physical environment, neighborhood social interactions may contribute to CHD risk. Following a theoretical model inspired from the American sociological literature, 1) we examined whether neighborhood residential instability (i.e., population turnover) was a structural determinant of neighborhood cohesion; 2) we assessed neighborhood cohesion effects on CHD mortality; and 3) we used a path analysis approach to estimate the extent to which neighborhood cohesion mediated the effect of residential instability on CHD mortality. We applied the ecometric approach to the Health Survey in Scania, Sweden (n = 5,545), to measure neighborhood cohesion, and used the LOMAS database (n = 71,258) to assess cohesion effects on CHD mortality. The average length of residence of inhabitants in their neighborhood was a major structural determinant of the level of cohesion in the neighborhood. After controlling for individual health and socioeconomic variables, both neighborhood poverty and residential instability were associated with an increased CHD mortality. In addition to these neighborhood structural effects, CHD mortality increased with decreasing neighborhood cohesion, with a stronger association among residents settled for a certain time in their neighborhood. A path analysis model indicated that a significant share of the effect of residential instability on CHD was mediated by neighborhood cohesion. Besides possible effects of the physical environment, the neighborhood sociodemographic structure may shape social interactions among neighbors, which may, in turn, affect CHD risk.