We recently provided data from a prospective cohort study of postmenopausal women which suggested that a first livebirth at age 30 or older (cf before age 20) was associated with a twofold increased risk of breast cancer in women without a family history, but a 5.8-fold higher risk in women with a positive family history. To address the question of whether these observations reflect difficulty becoming pregnant or maintaining a pregnancy, we performed additional analyses in which the outcome of each pregnancy was considered. During five years of follow-up, 620 incident cases of breast cancer were identified in the 37,105 women at risk. There was little evidence for an increased risk associated with a history of spontaneous abortion (relative risk [RR] = 1.1; 95 percent confidence interval [CI] = 0.9-1.4), nor was the risk higher among women who reported two or more spontaneous abortions in consecutive pregnancies (RR = 1.0, CI = 0.7-1.4). Although women who reported that they had tried unsuccessfully to become pregnant had only slightly and nonsignificantly elevated risks of breast cancer (RR = 1.1, CI = 0.9-1.3), a more pronounced and statistically significant association was noted in women with a positive family history (RR = 2.1, CI = 1.4-3.2). There was a strong inverse association between failure to become pregnant and parity (P < 0.0001); nearly 50 percent of the nulliparous married women reported having tried and failed to become pregnant, whereas the frequency was only 6.8 percent among married women with five or more livebirths.(ABSTRACT TRUNCATED AT 250 WORDS)