To examine the relation between alcohol intake and the incidence of coronary heart disease among Japanese, the authors analyzed data from a prospective study of 8,476 Japanese male employees, who were 40-59 years old at baseline (between 1975 and 1984) and worked for 13 urban companies in Osaka, Japan. These men were followed until the end of 1993, on average, an 8.8-year follow-up. Eighty-three coronary heart disease events (54 myocardial infarction, 32 angina pectoris) occurred during the employment period under study. Compared with the risk of coronary heart disease for never drinkers, the age-adjusted relative risk for those with an increased ethanol intake was lower, but the risk did not appear to be reduced further with the intake of > or = 69 g of ethanol per day. The multivariate relative risk adjusted for age, serum total cholesterol, cigarette smoking, body mass index, left ventricular hypertrophy, and a history of diabetes mellitus was 0.83 (95% confidence interval (CI) 0.24-2.86) in exdrinkers, 0.69 (95% CI 0.37-1.29) in drinkers of 1-22 g/day of ethanol, 0.55 (95% CI 0.29-1.05) in drinkers of 23-45 g/day, 0.41 (95% CI 0.19-0.88) in drinkers of 46-68 g/day, and 0.59 (95% CI 0.23-1.51) in drinkers of > or = 69 g/day. The inverse association with alcohol intake was similar between myocardial infarction and angina pectoris. Alcohol intake seemed to prevent the premature incidence of coronary heart disease among urban Japanese middle-aged men.