Plasma levels of high-density lipoprotein cholesterol (HDL-C) at entry and subsequent changes from these baseline levels were inversely predictive of coronary heart disease (CHD) end points in hypercholesterolemic men followed for 7 to 10 years in the Lipid Research Clinics Coronary Primary Prevention Trial, especially in the 1907 participants receiving cholestyramine. When the men in this cohort were compared, each 1 mg/dl increment in baseline HDL-C (mean 44.3 mg/dl) was associated with a 5.5% decrement in risk of “definite” CHD death or myocardial infarction (Z = − 5.4), and each 1 mg/dl increase from baseline HDL-C levels (mean increase = 1.6 mg/dl) during the trial was associated with a 4.4% risk reduction (Z = −2.2). In the 1899 participants receiving placebo, the corresponding risk decrements were 3.4% and 1.1%. Although baseline HDL-C level (mean = 44.4 mg/dl) remained a significant risk predictor (Z = −3.8) in the placebo cohort, increases in HDLC (mean increase 0.5 mg/dl) were not significantly predictive of CHD (Z = −0.6) unless “suspect” as well as “definite” end points were analyzed (Z = − 2.0). When the associations between HDL-C (baseline plus change) and incidence of definite CHD end points within each treatment cohort were compared, their difference approached nominal significance (Z = 1. 9). The results suggest a synergistic interaction, in which cholestyramine treatment reduced CHD risk most substantially in men maintaining the highest HDL-C levels.