Background: Ethylene oxide (ETO) is a sterilant gas considered to be a human carcinogen, due primarily to excess hematopoietic cancer in exposed cohorts. ETO causes mammary tumors in mice, and has been associated with breast cancer incidence in one small epidemiologic study. Methods: We have studied breast cancer incidence in a cohort of 7576 women employed for at least one year and exposed for an average 10.7 years while working in commercial sterilization facilities. Breast cancer incidence (n = 319) was ascertained via interview, death certificates, cancer registries, and medical records. Interviews were obtained for 68% of the cohort. Results: The standardized incidence ratio (SIR) for incident breast cancer in the whole cohort using external referent rates (SEER) was 0.87 (0.77–0.97). The SIR for those in the top quintile of cumulative exposure, with a 15 year lag, was 1.27 (0.94–1.69), with a positive trend of increasing SIR with increasing exposure (p = 0.002). SIRs are under-estimated because breast cancer incidence in the whole cohort was under-ascertained, due to incomplete response and lack of complete coverage by state cancer registries. In internal nested case–control analyses of those with interviews (complete cancer ascertainment), controlling for reproductive risk factors, a positive exposure–response was found with the log of cumulative exposure with a 15-year lag (p = 0.0005). The odds ratio by quintile of cumulative exposure were 1.00 (0 exposure due to 15 year lag), 1.06, 0.99, 1.24, 1.42, and 1.87. Conclusions: Our data suggest that ETO is associated with breast cancer, but a causal interpretation is weakened due to some inconsistencies in exposure–response trends and possible biases due to non-response and incomplete cancer ascertainment.