Relationship between socioeconomic status and ischaemic heart disease in cohort and case-control studies: 1960-1993

Open Access
Publisher Website
Google Scholar
Abstract
Much of the information on the relationship between socioeconomic status (SES) and ischaemic heart disease (IHD) comes from ecological, cross-sectional and longitudinal mortality studies. Such studies may present methodological problems which affect cohort studies and case-control studies to a lesser degree. A systematic review was conducted into the relationship between SES and IHD, as reported by cohort and case-control studies in which SES had been measured by education or occupation. A bibliographic search, covering the period 1960–1993, was carried out using Index Medicus, MEDLINE, Sociological Abstracts, Social Scisearch and the references cited in papers identified in such databases. Finally, 34 studies were selected, with data being extracted separately by two reviewers working independently. Where differences of opinion arose, these were resolved by discussion between the two. Multiple linear regression analysis was used to identify determinants of inter-study heterogeneity. Risk of IHD was inversely related to educational level in the period 1982–1993. There was a lower risk of IHD among manual workers versus non-manual workers until approximately 1970; thereafter the relationship became inverted. The odds ratio for HID in manual workers rose progressively over the period 1960–1993, something that could not be explained by study design, study size, study country or control for confounding factors. Education- and occupation- related differences in risk of IHD affected all SES levels. These findings agree with those yielded by ecological, cross-sectional and longitudinal mortality studies, although they cannot be extrapolated to women, nonwhite ethnic groups, or developing countries. The inversion of the relationship between SES and IHD, and the increase in socioeconomic differences in LHD in recent years suggest that the problem is potentially controllable and that the necessary policies should be implemented as a matter of urgency.

This publication has 0 references indexed in Scilit: