With major city residence as the reference category, the relative rate (95% confidence limits) of mastectomy was 1.03 (0.99, 1.07) for women from inner regional areas and 1.05 (1.01, 1.10) for those from more remote areas. Low annual surgeon case load (≤10) was predictive of mastectomy, with a relative rate of 1.08 (1.03, 1.14) when compared with higher case loads. Tumour size was also predictive, with a relative rate of 1.05 (1.01, 1.10) for large cancers (40+ mm) compared with smaller cancers (<30 mm). These associations were confirmed in multiple logistic regression analysis. Conclusions: Results confirm previous studies showing higher mastectomy rates for residents of more remote areas, those treated by surgeons with low case loads, and those with large cancers. Reasons require further study, including possible effects of surgeon and woman's choice and access to radiotherapy services.