We report the results of an analysis of information from a prospective cohort study of women examining the association between cholecystectomy and use of supplemental estrogens (never, current, and past use). There were 55 cases of cholecystectomy in those who had never used estrogen and 105 cases in those who had ever used estrogen. After adjustment for age, the relative risk of cholecystectomy in those who had ever used estrogen was 2.1 (95% confidence interval, 1.5-3.0). In women classified as current users based on information available in 1977, the relative risk of gallbladder disease was 2.7 (95% confidence interval, 1.8-4.0) and in past users as of this date, it was 1.6 (95% confidence interval, 1.1-2.5). When cases known to have reinitiated estrogen use after 1977 are removed from the numerator for past users and added to the numerator for current users, the relative estimate for past use decreased to 1.1 (95% confidence interval, 0.7-1.8), and the risk estimate for current use increased to 3.9 (95% confidence interval, 2.6-5.9). Although an increase in the risk of gallbladder disease in women using supplemental estrogens has been consistently observed in previous studies, the risk after cessation of use has received little attention but could be important. Our data suggest the possibility that the risk of gallbladder disease in estrogen users persists after use of the drug ceases. An increase in the risk of cholecystectomy in estrogen users that persists after drug use ends is plausible if estrogen-induced increases in the lithogenicity of bile cause gallstones to form during drug use and if these gallstones fail to dissolve even after bile lithogenicity returns to normal upon cessation of use. The effect of estrogen use on the gallbladder should be considered in weighing the net risk-benefit ratio of these drugs.