THE PURPOSE of this paper is to review the recent medical literature on social class and coronary heart disease.? Specifically, the aims are (1) to summarize the basic findings of these studies; (2) to examine the indicators of social class that have been employed; (3) to consider the importance of the social context from which the indicators of social class were drawn; and (4) to assess the implications and the value of the various indicators of social class that were used. This paper does not consider the significance of the selection of specific indicators of the dependent variable, coronary heart disease. Nevertheless, it is worth noting that the criteria employed certainly could influence the relationship between the independent and dependent variables. In essence, one finds that two classes of data have been employed as indicators of coronary heart disease. The first type is based on physical examinations performed on living persons. The second type is based on a physician’s statement on a death certificate that death was due to coronary heart disease. Eight of the sixteen articles reviewed employ death certificate data at some point; the other eight do not. This paper does not undertake a systematic evaluation of the over-all methodology of each of the studies. The problem of methodology shades off into the previous one (i.e. specifying the dependent variable). For instance, the study at Du Pont by PELL and D’ALZONO considers only first events of coronary heart disease in order to eliminate the possible bias introduced by a prior coronary occurrence. [l] However, none of the studies using death certificates employ such refinements. Oftheeight studies which conducted physical examinations, only the Du Pont, Framingham [2] and Chicago utility [3] studies explicitly state they focus on first events. Every study reviewed found some relationship between class and heart disease, although the strength of the relationship varies markedly. In fact, one paper, by STAMLER and his associates, upon examining their own and others’ data, concludes that
Read full abstract