Niacin was one of the treatments compared in the Coronary Drug Project, a placebo-controlled, multicenter trial of lipid-lowering drugs in the secondary prevention of coronary heart disease. A total of 1119 men, aged 30-64 at entry, were randomized to niacin and 2789 to placebo by the end of recruitment in March 1969. Although side-effects interfered with adherence to the niacin regimen, it was the most effective agent in achieving cholesterol-lowering (10% overall); other agents in the trial were clofibrate, dextrothyroxine, and conjugated equine estrogens. At the scheduled conclusion of the trial in February 1975, the niacin-treated group exhibited a statistically significantly lower incidence of definite, non-fatal myocardial infarction (MI) than the placebo group. There was a trend toward improvement in the life-table mortality curve, but this was not statistically significant. In 1981 an extended follow-up was carried out concerning vital status for the 6008 men who were still alive at the end of treatment and active follow-up in the trial in 1975 (827 in the niacin group and 2008 in placebo groups). Vital status was determined for 99.1% of these men after a mean of 9 years from conclusion of the trial. In the group previously randomized to niacin, there were 69 (11%) fewer deaths than were expected on the basis of mortality in the placebo group. This difference was significant (z = -3.52; P = 0.0004). The data also suggested that patients with a higher baseline cholesterol experienced greater benefit from niacin therapy, as did those with the best response to the drug.(ABSTRACT TRUNCATED AT 250 WORDS)