The joint effect of use of combination-type oral contraceptives and other exposure factors on risk of endometrial cancer was examined in data from a multicenter case-control study conducted in 5 areas of the United States. Cases were 405 women with histologically confirmed invasive epithelial endometrial cancer first treated at one of 7 participating hospitals. A total of 297 population-based controls of similar age, race, and geographic area were selected as a comparison group. Information on exposure factors was derived from in-person interviews. Combination-type oral contraceptive (COC) use was associated with a significant reduction in risk of endometrial cancer, with an adjusted odds ratio (OR) of 0.4 (95% confidence interval 0.3 to 0.7) for ever compared to never use. Long-term (> or = 10 years) users experienced a markedly lower risk (OR = 0.2). Women who discontinued COC use > or = 20 years earlier remained at reduced risk (OR = 0.7) compared with non-users. The negative association with COC use was apparent regardless of the presence or level of several other risk factors for endometrial cancer, including age, menopausal status, parity, obesity, ever-use of menopausal estrogens, smoking history, or history of infertility. The magnitude of the negative association observed in COC users, however, was considerably diminished in women with no full-term births and in women who subsequently used replacement estrogens for 3 or more years. These results provide new evidence that the protective effect of COC use lasts for 20 or more years after use is discontinued, and highlight several sub-groups of users in whom the level of protection is attenuated by the presence of other risk factors for this disease.