The use of exogenous sex hormones in relation to the risk of rheumatoid arthritis (RA) was examined in a cohort of married nurses 30-55 years of age followed since 1976 in the Nurses' Health Study. Baseline information on the use of oral contraceptives, replacement estrogens, and other potential risk factors was obtained in 1976 and updated every 2 years. During 8 years of followup, 217 incident cases of polyarthritis were ascertained (115 RA and 102 undifferentiated polyarthritis). When compared with women who had never used oral contraceptives, the age-adjusted relative risk was 1.0 (95% confidence interval [CI] 0.7-1.3) for past users; however, too few women were currently using oral contraceptives for a reliable estimate of its effect. Among postmenopausal women, 123 cases of RA were reported. Compared with postmenopausal women who never used replacement estrogens, current users had an age-adjusted relative risk of 1.3 (95% CI 0.9-2.0), past users had an age-adjusted relative risk of 0.7 (95% CI 0.5-1.2), and ever users had a relative risk of 1.0 (95% CI 0.7-1.4). These data do not show a protective effect of past use of oral contraceptives or replacement estrogens for RA; however, a modest protective effect of current oral contraceptive use cannot be excluded.