Potential Similarity of Serum Cholesterol Multivariate Coefficients in the Prediction of Coronary Heart Disease Across Populations: A Review From the Seven Countries Study

This study aimed to review the unique contribution of the Seven Countries Study (SCS) of cardiovascular diseases to the possible non-heterogeneity in multivariate coefficients of serum cholesterol in predicting coronary heart disease (CHD) mortality across different populations. This study reviewed five published analyses from the SCS, which together encompassed 16 cohorts of middle-aged men from eight nations across seven countries in the USA, northern and southern Europe, and Japan. In total, these analyses included 12,763 participants and follow-up periods ranging from 25 to 60 years after the baseline examination. Serum cholesterol was measured using uniform, standardized procedures that differed from those reported in the literature. Marked differences in mortality rates were observed, with higher rates in the USA and northern Europe and lower rates in the other regions. A systematic comparison of serum cholesterol coefficients did not reveal significant heterogeneity across cohort combinations or follow-up durations of 25, 40, 50, and 60 years for CHD mortality. In all cases, coefficients were adjusted for three additional risk factors: age, cigarette smoking, and systolic blood pressure. Variations in CHD mortality rates across populations were explained by differences in serum cholesterol levels. In contrast, the magnitudes of serum cholesterol coefficients were relatively similar across groups, although not necessarily homogeneous. These findings support the idea that predictive models of CHD mortality developed for a specific population can also be applied to other populations, since the expectation, at least for serum cholesterol, is to obtain similar relative risk estimates, not absolute risk, which aligns with the limited evidence that is available in the current literature.