Reproductive factors are known to be aetiologically important in breast cancer, but less is known regarding their effect on breast cancer prognosis. We have investigated the prognostic effect of age at first birth and total parity using data from the Danish Breast Cancer Cooperative Group that, since 1977, has collected population-based information on tumour characteristics, treatment regimes and follow-up status on Danish women with breast cancer. Details of pregnancy history were added from the Danish Civil Registration System and the National Birth Registry. Included in the study were 10,703 women with primary breast cancer. After adjusting for age and stage of disease (tumour size, axillary nodal status and histological grading), the number of full-term pregnancies was found without prognostic value. However, women with primary childbirth between 20 and 29 years experienced a significantly reduced risk of death compared with women with primary childbirth below the age of 20 years [20-24 years: relative risk (RR) = 0.88, 95% confidence interval (CI) 0.78-0.99; 25-29 years: RR = 0.80, 95% CI 0.70-0.91]. Further adjustment for oestrogen receptor status did not influence these results. The effect was not modified by age at diagnosis, tumour size or nodal status. In conclusion, low age at first childbirth, but not parity, was associated with a poor prognosis of breast cancer. We speculate whether women who develop breast cancer despite an early first full-term pregnancy might represent a selected group with a more malignant disease.