We reviewed epidemiological data on oral contraceptive (OC) use and colorectal, liver, lung and other nonfemale neoplasms. The data for colorectal cancer are suggestive of a favourable effect of OC, in the absence, however, of any duration-risk relation. Current, but not past, OC use is associated with excess risk of benign liver tumours, and a modest excess risk of liver cancer. The association with liver cancer was smaller for recent, low-dose OC. There was no evidence of an association between OC use and lung, digestive tract neoplasms other than colorectum, cutaneous malignant melanoma, thyroid cancer and any of the other neoplasms investigated.