Comparison of four measures of cholesterol for predicting men and women who will develop coronary heart disease within 8 to 10 years. Cohort study. 1898 men who received placebo (the placebo group of the Lipid Research Clinics [LRC] Coronary Primary Prevention Trial [CPPT]), 1025 men and 1442 women who participated in the 1970-1971 Framingham Heart Study biennial examination, and 1911 men and 1767 women without coronary heart disease who were from the LRC Population Prevalence Study. Total cholesterol, low-density lipoprotein (LDL) cholesterol, ratio of total cholesterol to high-density lipoprotein (HDL) cholesterol, and the ratio of LDL to HDL. Outcomes were coronary heart disease in the CPPT and Framingham studies and death from coronary heart disease in the Prevalence Study. Independent information in the total cholesterol/HDL ratio added risk-discriminating ability to total cholesterol and LDL cholesterol measures (P or = 9). The total cholesterol/HDL ratio is a superior measure of risk for coronary heart disease compared with either total cholesterol or LDL cholesterol levels. Current practice guidelines could be more efficient if risk stratification was based on this ratio rather than primarily on the LDL cholesterol level.