A Logistic Regression Analysis of Risk Factors In Relation to Women Breast Cancer

The data were collected in 8 hospitals in Guangzhou, Shaoguan and Zhaoqing of Guangdong Province. Altogether 125 matched sets were investigated in retrospect. There were one case matched two controls in every set.The data were analyzed by using multiple conditional Logistic regression model.The results suggested that habit of drinking-tea OR=0.3913 was regsrded as a protective factor and psychic trauma OR=3.4338, history of mastitis OR=3.4099, passive smoking OR=2.0796, average of oil intake OR=3.1773, abortion OR=1.3544 and family history of cancer OR=5.6797 were regarded as risk factors.