BACKGROUND: Previous studies convincingly showed an increase in risk of breast cancer associated with current or recent use of oral contraceptives from the 1960s to 1980s. The relation of contemporary oral contraceptive formulations to breast cancer risk is less clear. METHODS: We assessed lifetime oral contraceptive use and the specific formulations used among 116,608 female nurses ages 25 to 42 years at enrollment in 1989, and subsequently updated this information every 2 years. We related this information to risk of breast cancer up to June 1, 2001. RESULTS: During 1,246,967 person-years of follow-up, 1,344 cases of invasive breast cancer were diagnosed. Past use of any oral contraceptive was not related to breast cancer risk [multivariate relative risk (RR), 1.12; 95% confidence interval 0.95-1.33]. Current use of any oral contraceptive was related to a marginally significant higher risk (multivariate RR, 1.33; 95% CI, 1.03-1.73). One specific formulation substantially accounted for the excess risk: the RR for current use of triphasic preparations with levonorgestrel as the progestin was 3.05 (95% CI, 2.00-4.66; P