The epidemiological associations of gallstone disease were evaluated in a general population sample of 29,584 individuals (15,910 men and 13,674 women; age range, 30-39 years) belonging to 14 cohorts examined between December 1984 and April 1987. Subjects were screened for the presence of gallstones by gallbladder ultrasonography, completed a questionnaire, and underwent a physical examination and blood chemistry tests. Participants were considered to have gallstone disease if they had already had cholecystectomy or gallstones. Statistical associations were established by univariate analysis of the age-standardized data and by stepwise multiple logistic regression. Increasing age and body mass index and a maternal family history of gallstone disease were the most consistent associations (both at univariate and multivariate analysis and in both sexes) found in this study. Personal history of dieting was associated with gallstone disease in men, and at univariate analysis, in women. Decreasing serum total cholesterol levels and increasing serum triglycerides were associated with gallstone disease in both sexes in the multivariate analysis. In women, associations were also found with a number of pregnancies and paternal family history of gallstone disease. A slight but negative association with contraceptive pill use was identified only at multivariate analysis. Associations (investigated at univariate analysis) were also found with diabetes, cirrhosis, angina or myocardial infarction, and peptic ulcer. There was no association with smoking habits and use of aspirin or antirheumatic drugs.