A matter of context : Social inequalities in incidence of myocardial infarction

Abstract: Coronary heart disease is the leading cause of death and makes the separate largest contribution to social inequalities in the burden of disease in Sweden. The purpose of this dissertation is to study the potential influences of local social environment on incidence of myocardial infarction (MI), thereby contributing to understanding of the social aetiology of MI. Specific aims are to define social dimensions of the neighbourhood context, and to analyze their possible contextual effects on the incidence of MI. Further, to explore mechanisms underlying such effects, with an emphasis on the interplay between social context and individual socioeconomic position, and to evaluate the contribution of established biomedical risk factors and health behaviours. Finally, to evaluate whether economic residential segregation contributes to social inequalities in incidence of MI. The analyses in the studies are based on the Stockholm Heart Epidemiology Program (SHEEP), a population-based case-control study of first events of myocardial infarction, which comprised all Swedish citizens aged 45-70 in Stockholm County, 1992-1994. The studies indicate that structural material and economic conditions in neighbourhoods are relevant to individual MI risk. We find graded effects of concentration of affluence, level of economic resources, and material deprivation. The effects of social fragmentation and contextual dimensions, as defined by factor analyses, were weaker and less graded. Socioeconomic homogeneity only had an effect in combination with resource level within neighbourhoods. Our results show that women living in economically disadvantaged contexts have an incidence of MI twice as high as those living in economically advantaged contexts; for men the incidence is 50% higher. These figures have been adjusted for family socioeconomic position, disposable income, own education, position on the labour market and cohabitation, all of which influence both where a person lives and the risk of coronary heart disease. Further,we explored mechanisms that might underlie the relations. First, we considered whether individuals in different social strata are differentially vulnerable to their neighbourhood context. The results suggest that women with a low disposable income are more vulnerable to the effect of a lowincome context. Further, we find an indication of increased vulnerability to contextual effects among lower strata of both non-manual employees and manual workers. Second, we evaluated the contribution of established biomedical risk factors and health behaviours as mediators. We find that smoking and insulin resistance make notable contributions to the excess risk of MI in economically disadvantaged areas, after making prior adjustment for individual social characteristics. Finally, our results indicate that structural economic conditions in neighbourhoods do matter with regard to adult social class inequalities. The effect is not restricted to lower social strata, but since individuals in lower social strata more often live in disadvantaged contexts, and also seem to be more vulnerable to the effects of these contexts, we conclude that economic segregation creates neighbourhood contexts that contribute to social class inequality in incidence of MI.