Background: Menstrual, reproductive and hormonal factors have been related to ovarian cancer risk, but further quantification of their role in various populations is required.
Patients and methods: Cases were 1031 women, below age 79, with incident, histologically confirmed epithelial ovarian cancer, and controls 2411 women, admitted between 1992 and 1999 to a network of hospitals in 4 Italian areas for acute, non-neoplastic, diseases. Odds ratios (OR) were obtained using multiple logistic regression.
Results: Multiparity was associated with a significant reduction in risk of ovarian cancer (OR = 0.6 for 3, and 0.5 for > or = 4 births). No consistent association was observed with time since first or last birth, nor with spontaneous or induced abortions. Late age at menarche (OR = 0.8), and early menopause (OR = 0.6) were inversely related to risk, as did long-term oral contraceptive use (OR = 0.5, for > or = 5 years). Hormone replacement therapy in menopause was associated with a non-significantly elevated risk (OR = 1.4). The pattern of risk was similar for women with and for those without family history of breast or ovarian cancer.
Conclusions: This uniquely large study confirms and further quantities the relation between hormonal and reproductive factors and ovarian cancer. The pattern of risk observed cannot be totally explained by a role of ovulation in ovarian carcinogenesis.