We set out to detect a transient increase in risk of breast cancer following childbirth, the existence of which has been postulated, but for which empirical evidence is contradictory. Breast cancers and births occurring among the cohort of Swedish women born after 1939 were linked, yielding 3,439 cases and 25,140 age-matched controls with at least two children. Within three years of their last childbirth, women had an estimated rate of breast cancer of 1.21 (95 percent confidence interval [CI] = 1.02-1.44) times that of women whose last birth was 10 or more years earlier, after adjustment for parity and age at first birth. Further analyses suggested that this effect reflected, in part, a small transient increase in breast cancer risk that lasts for about three years following completed pregnancy. The effect of age at first birth on breast cancer risk appears to be confounded by time since last birth; the parity-adjusted rate ratio for having a first birth at age 35 years or more compared with under 20 years is reduced from 1.72 (CI = 1.14-2.58) to 1.36 (CI = 0.88-2.09) on additional adjustment for time since last birth. A transient increase in breast cancer risk after childbirth appears thus appears to account for part of the effect of age at first birth on breast cancer risk.