Studying mechanisms underlying social inequality in postmenopausal breast cancer is important in order to develop prevention strategies. Standard methods for investigating indirect effects, by comparing crude models to adjusted, are often biased. We applied a new method enabling the decomposition of the effect of educational level on breast cancer incidence into indirect effects through reproductive patterns (parity and age at first birth), body mass index and health behavior (alcohol consumption, physical inactivity, and hormone therapy use). The study was based on a pooled cohort of 6 studies from the Copenhagen area including 33,562 women (1,733 breast cancer cases) aged 50-70 years at baseline. The crude absolute rate of breast cancer was 399 cases per 100,000 person-years. A high educational level compared to low was associated with 74 (95% CI 22-125) extra breast cancer cases per 100,000 person-years at risk. Of these, 26% (95% CI 14%-69%) could be attributed to alcohol consumption. Similar effects were observed for age at first birth (32%; 95% CI 10%-257%), parity (19%; 95%CI 10%-45%), and hormone therapy use (10%; 95% CI 6%-18%). Educational level modified the effect of physical activity on breast cancer. In conclusion, this analysis suggests that a substantial number of the excess postmenopausal breast cancer events among women with a high educational level compared to a low can be attributed to differences in alcohol consumption, use of hormone therapy, and reproductive patterns. Women of high educational level may be more vulnerable to physical inactivity compared to women of low educational level.