Aims: To evaluate patient outcomes for a novel integrated primary/specialist model of community care for complex Type 2 diabetes mellitus management compared with outcomes for usual care at a tertiary hospital for diabetes outpatients. Methods: This was a prospective open controlled trial performed in a primary and tertiary care setting in Australia. A total of 330 patients with Type 2 diabetes aged >18 years were allocated to an intervention (n=185) or usual care group (n=145). The intervention arm was a community-based model of care led by a general practitioner with advanced skills and an endocrinologist partnership. Usual care was provided via the hospital diabetes outpatient department. The primary end point was HbA1c concentration at 12 months. Secondary end points included serum lipids and blood pressure. Results: The mean change in HbA1c concentration in the intervention group was −9 mmol/mol (−0.8%) at 12 months and in the usual care group it was −2 mmol/mol (−0.2%) (95% CI −5,1). The percentage of patients in the intervention group achieving the HbA1c target of ≤53 mmol/mol (7%) increased from 21 to 42% (P<0.001); for the usual care group there was a 1% increase to 39% of patients attaining this target (P=0.99). Patients in the intervention group experienced significant improvements in blood pressure and total cholesterol compared with those in the usual care group. The percentage of patients achieving clinical targets was greater in the intervention group for the combined target of HbA1c concentration, blood pressure and LDL cholesterol. Conclusions: A community-based, integrated model of complex diabetes care, delivered by general practitioners with advanced skills, produced clinical and process benefits compared with a tertiary diabetes outpatient clinic.