To determine whether atypical antipsychotics, when compared to typical antipsychotics, increase the risk of breast cancer. We conducted a retrospective cohort study using a nested case-control analysis within the United Kingdom General Practice Research Database population. We identified all female patients prescribed at least one antipsychotic (either typical or atypical), between 1 January 1988 and 31 December 2007, with follow-up until 31 December 2010. All incident cases of breast cancer were identified and matched up to 10 controls. Adjusted rate ratios (RR) of breast cancer associated with ever use of atypical antipsychotics was compared to ever use of typical antipsychotics. The cohort included 106,362 patients prescribed antipsychotics during the study period. During a mean follow-up of 5.3 years, 1237 patients were diagnosed with breast cancer (overall rate: 2.7 per 1000/year). Compared to patients who only used typical antipsychotics, exclusive users of atypical antipsychotics were not an increased risk of breast cancer (RR: 0.81, 95% CI: 0.63, 1.05). These results remained consistent after considering specific atypical antipsychotics known to significantly increase prolactin levels such as risperidone (RR: 0.86, 95% CI: 0.60, 1.25). Furthermore, no dose-response was observed in terms of cumulative duration of use and cumulative dose in olanzapine equivalents. The results of this study should provide reassurance that compared to typical antipsychotics, atypical antipsychotics do not increase the risk of breast cancer.