Contextual effects of social fragmentation and material deprivation on risk of myocardial infarction - Results from the Stockholm Heart Epidemiology Program (SHEEP)

Abstract

Background. Socioeconomic deprivation has been suggested as a contextual feature of importance for cardiovascular disease and mortality, whereas the effect of social fragmentation has largely been studied in relation to suicide. In this study we examine the contextual effects of social fragmentation and material deprivation on the incidence of myocardial infarction (MI). Methods. A population-based case-control study (SHEEP). The study base included all Swedish citizens aged 45-70 living in the Stockholm metropolitan area. Cases (n = 1631) were all first events of MI during 1992-1994. Exposure information on individual risk factors was obtained from a questionnaire. Areas (n = 862) were classified according to the Townsend index, measuring material deprivation, and the Congdon index, measuring social fragmentation. Results. We found increased incidence of MI in both materially deprived and socially fragmented contexts that were not due to confounding from individual social risk factors being more prevalent among subjects in deprived settings. The adjusted relative risk of MI was 2.0 (95% CI: 1.3, 3.1) for women living in the top quartile of materially deprived areas. For men, the adjusted relative risk (RR) was 1.6 (95% CI: 1.2, 2.1). Women living in the top quartile of socially fragmented areas had an RR of MI of 1.6 (95% CI: 1.0, 2.5) after adjustment, while the corresponding figure for men was 1.4 (95% CI: 1.0, 1.8). Conclusion. Our findings support the notion that the social context in which people live has an impact on the risk of coronary heart disease. We could not determine which of the contextual aspects under study made the most substantial contribution. Mutual adjustment of the two indices suggests that material deprivation is the dominating factor, especially for women. However, the indices were highly correlated (r = 0.87), and it cannot be ruled out that they partly measure the same underlying phenomenon. © International Epidemiological Association 2004; all rights reserved.