AIMS: To determine the prevalence and associates of depression in Aboriginal and Anglo-Celt (AC) Australians with type 2 diabetes. METHODS: Community-based patients were screened using the Patient Health Questionnaire (PHQ-9) as part of detailed assessment. The prevalence of any current depression, major depression and antidepressant use by racial group was compared after adjustment for age, sex, educational attainment and marital status. Multiple logistic regression was used to determine associates of current depression. RESULTS: The 107 Aboriginal participants were younger (mean±SD 54.3±11.8 vs. 67.2±10.6 years), less often male (34.6% vs. 50.9%) and married (39.3% vs. 61.7%), and more likely to smoke (44.6% vs. 8.1%) than the 793 AC subjects (P≤0.002). Fifty-two Aboriginal (48.5%) and 772 AC participants (97.4%) completed the PHQ-9; these Aboriginals had similar socio-demographic, anthropometric and diabetes-related characteristics to those without PHQ-9 data. A quarter of the Aboriginals had current depression vs 10.6% of ACs (P=0.16), 15.4% vs. 4.1% had major depression (P=0.029), and 68.8% vs. 29.7% had untreated depression (P=0.032). Compared with non-depressed participants, patients with current depression were younger and more likely to smoke, to be overweight/obese and to have worse glycaemic control (P≤0.024). Significant independent associates of current depression were educational attainment (inversely), smoking status, body mass index and fasting plasma glucose in the AC group and alcohol use in the Aboriginal group. CONCLUSIONS: Although prevalence of depression was not significantly increased in the Aboriginal patients, it was more likely to be major and untreated. Depression complicating type 2 diabetes is associated with adverse cardiovascular risk.