A case-control study of breast cancer (BC) was conducted from 1994 to 1996 in Chongqing, People’s Republic of China, in order to explore the etiological role of passive smoking (PS, so-called second hand smoking) as well as other early life factors (weight, height, socioeconomic status and history of suffering from a disease resulting in hospitalization). These factors were reviewed both in childhood (age less than 10 years) and in the teenage years (youth: 10 to 16 years). One hundred and eighty six cases of newly diagnosed and histologically confirmed BC, aged 24 to 55 years, were individually matched by day (within six months) and age (within 2 years) at diagnosis as well as marital status to 186 controls selected from outpatients not suffering from cancer. All subjects, cases and controls were never-smokers. A standardized questionnaire was used for interview in a face-to-face situation. After adjustment for a wide range of covariates using multiple logistic regression analysis, PS was found to be a statistically significant risk factor for BC. This was found for exposure to PS in childhood [odds ratio (OR) and 95% confidence interval (95%CI): 1.24 (1.07-1.43)], in youth [1.15 (0.90-1.47)] and in adulthood for either exposure at home [4.07 (2.21-7.50)] or at work [1.27 (1.04-1.55)]. For exposure to PS in childhood, a significant dose-response effect was evident (test for trend, p<0.05) with the number of smokers in the home, as well as the perceived level of exposure to PS, and in adulthood with the number of smokers at work. Our study also found an increased risk of BC in those with a past history of suffering from a disease requiring hospitalization [2.41 (1.21-4.81)]. Subjects with a low body weight in childhood and with a poor economic status in youth were associated with increased risk of BC in their adulthood [1.54 (1.09-2.18) and 1.03 (1.00-1.06)]. Being overweight as an adult, however, was associated with a high risk of BC [1.76 (1.02-3.04)]. Age at menarche [0.83 (0.72-0.95)] was associated negatively with risk of BC. A history of benign breast disease [2.05 (1.01-4.16)] or a history of life stress [2.32 (1.54-3.48)] were both associated with increased risk of BC. Our results therefore indicate a small but definite effect associated with PS, the credibility of which is enhanced by a dose-response relationship to BC risk. The other early life factors, such as age at menarche, history of suffering from a disease requiring hospitalization, history of benign breast disease, being overweight as an adult and life stress are similar to those consistently found in other countries. The associations involving low body weight, low socioeconomic status in early life and subsequent high risk of BC require further study.