To investigate the effect of lactation on the risk of ovarian cancer for Chinese women, a case-control study was conducted in Hangzhou, capital of Zhejiang province, China. Cases were 275 patients with histologically confirmed epithelial ovarian cancer. Controls were 623 women without neoplasm. All participants were parous women who had given at least one live birth and had been residents of Zhejiang province for at least 10 years. Information was collected using a validated and reliable questionnaire on total months of lactation, number of children breastfed, and average duration of lactation per child. Multivariate logistic regression models were used to assess the association between ovarian cancer risk and lactation variables, accounting for age, locality, full-term pregnancy, oral contraceptive use and family history of the cancer. The adjusted odds ratios were 0.51 (95% confidence interval (CI) 0.3-0.9) and 0.44 (95% CI 0.2-0.9) respectively for women with over 12 months of lactation and at least three children breastfed, compared with those with 4 months or less lactation and one child breastfed. The corresponding dose-response relationships were also significant (P<0.05). Therefore, prolonged lactation could reduce the risk of ovarian cancer foe Chinese women. (C) 2004 Lippincott Williams Wilkins.