Previous epidemiological studies suggesting an association between oral contraceptive use and ulcerative colitis incidence have been weak and conflicting. To measure a possible association, 46 incident cases of ulcerative colitis patients, women aged 18-50 years, were compared with peer-nominated age-matched and sex-matched controls. There were no differences between case and control patients in demographic characteristics. There was no association between oral contraceptive use and ulcerative colitis (current use: odds ratio 0.70, 95% confidence interval 0.27-1.83; former use: odds ratio 1.14, confidence interval 0.41-3.15; current or former use: odds ratio 0.86, confidence interval 0.40-1.85). Stratifying by disease location (pancolitis or left-sided disease) also failed to identify an association. Controlling for possible confounding effects of cigarette smoking did not alter the lack of association between oral contraceptive use and ulcerative colitis. Similarly, testing for interaction failed to demonstrate any effect modification. Analyzing for duration of current oral contraceptive use or time interval since last use failed to demonstrate a "dose-response" effect. The study was of sufficient size to detect statistical significance for oral contraceptive use for odds ratios of 2.8 and higher. In this matched case-control study of incident cases and community controls, there was no association between oral contraceptive use and ulcerative colitis incidence. To date, there is no evidence suggesting that women predisposed to the development of ulcerative colitis should be advised to avoid oral contraceptive use.