Multivariate prediction of coronary heart disease in the Western Collaborative Group Study compared to the findings of the Framingham study

Abstract

The Western Collaborative Group Study (WCGS) is a prospective epidemiological study of 3,154 initially well men, aged 39059 years at intake in 1960-61, who were employed in ten participating companies in California. Clinical coronary heart disease (CHD) occurred in 257 men during a follow-up period of eight and one-half years. Coronary heart disease risk is predicted using the additive multiple logistic model with the risk factors: age, cholesterol, systolic blood pressure, hematocrit, ECG status, smoking at intake, and relative body weight. The predicted individual CHS risk levels, using the logistic results derived from the WCGS data, are highly correlated with predicted risk levels using a Framington study (FS) equation for these same risk factors with 12-year follow-up. The observed number of CHS events in the WCGS is not significantly different from the expected number of events derived from the FS logistic equation, after correction of length of follow-up. Multiple logistic analysis of the direct association between CHD incidence and behavior pattern gives an approximate relative risk of 1.9 (P = 0.0006) and 2.1 (P = 0.0015) for Type A compared to Type B men aged 39-49 and 50-59 years, respectively. It is estimated that removal of the excess risk associated with Type A behavior would correspond to a 31% (standard error = 6.6%) reduction of coronary heart disease incidence in the Western Collaborative Group Study population.