Association of body mass index, physical activity, and reproductive histories with breast cancer: a case-control study in Gifu, Japan.

To further clarify risk factors for breast cancer in Japanese women, a self-administered questionnaire was completed by 157 cases with histologically confirmed breast cancer from 1989 to 1993 and by 369 age and residential area matched controls in Gifu, Japan. Conditional logistic regression model was used to assess the relations. Multivariate analyses showed that breast cancer risk decreased with body mass index for premenopausal women (RR = 0.45; 95 % CI = 0.22-0.92 for BMI > or = 23 vs. < 21 (kg/m2)), but the risk increased with body mass index for postmenopausal women (RR = 1.98; 95 % CI = 0.86-4.55 for BMI > or = 24 vs. < 21.5 (kg/m2)). The risk increased with a small number of births in pre- and post-menopausal women (1.83; 1.11-2.99 and 6.06; 2.40-15.3 for 1-2 births and nulliparity, respectively, vs. > or = 3 births). Ex- or current smoking increased the risk of breast cancer (2.31; 1.19-4.49). Reduced risk of premenopausal breast cancer was associated with high energy expenditure in physical activity during teenage, although the trend was not statistically significant.