RESUMEN
A biologically motivated breast cancer incidence rate function was fit to data from a published case-control study conducted in countries whose incidence is high (Wales and the United States), moderate (Brazil, Greece, and Yugoslavia), and low (Japan and Taiwan). The data include personal characteristics of 3,925 breast cancer cases and 11,327 controls interviewed in selected hospitals in 1964-1968. Parameters in the function specify the dependence of age-specific breast cancer incidence rates on age at menarche, age at menopause, occurrence and timing of full-term pregnancies, and body mass. Parameters were estimated separately for high-, moderate-, and low-risk countries. Examination of residuals provided little evidence of inadequacy of the fitted function in describing combined effects of the characteristics studied. The following patterns were seen in all three risk groups: 1) Incidence rates jump to a higher level after first childbirth, but then increase with age more slowly thereafter. 2) Rates increase with age more slowly after menopause than before. 3) Rates change quadratically with body mass index among all women, although the main trend varies: Rates decrease with body mass among premenopausal women in high-risk countries, but increase with body mass in all other groups of women. Similarities of parameter estimates across countries suggest that reproductive events and body fat exert similar effects on all women, regardless of breast cancer rates in their country of residence.