EFFECTS OF MENSTRUAL AND REPRODUCTIVE FACTORS ON THE RISK OF BREAST-CANCER - METAANALYSIS OF THE CASE-CONTROL STUDIES IN JAPAN

To elucidate the magnitude of the effect of menstrual and reproductive factors on breast cancer occurrence among Japanese women, we reviewed eight case-control studies previously conducted in Japan and used a q uantitative method (meta-analysis) to summarize the data. While indivi dual studies have different methods and populations, the estimated odd s ratios (ORs) in the studies were statistically homogeneous for all s tudy variables. It was confirmed that early age at menarche, late age at first birth, and premenopausal status are significantly associated with risk of breast cancer; an estimated combined OR of 0.68 (95% conf idence interval (CI) 0.59-0.77) was obtained for women with onset of m enstruation after age 16 compared to those before age 14. Nulliparous women had higher risk than women with first birth before age 25 (OR = 1.56 95%, CI: 1.27-191). The OR for women with first birth after age 3 5 was 2.26 (95% CI: 1.85-2.77) compared to women at first birth before age 25. Premenopausal women had a higher risk than women with menopau se before age 50 (OR = 2.21, 95% CI: 1.53-320). We also found a signif icant protective effect of high parity after controlling for age at fi rst birth and the other menstrual factors. The OR estimate for 3 or mo re births compared to nulliparity was 0.68 (95% CI: 0.54-0.86). The me ta-analysis provided quantitative estimates of breast cancer risk amon g Japanese women with improved precision.