To determine the quality of care provided to non-English-speaking patients with non-insulin-dependent (type 2) diabetes mellitus compared with English-speaking patients, we did a retrospective cohort study of 622 patients with type 2 diabetes, of whom 93 were non-English-speaking and 529 were English-speaking. They were patients at primary and specialty care clinics at a university and a county hospital, and the study was based on clinical and administrative database records with a 12-month follow-up. Professional interpreters were provided to all non-English-speaking patients. Patients were identified using interpreter services records, which reliably included all patients who did not speak English. After adjusting for demographic differences, significantly more non-English-speaking patients received care that met the American Diabetes Association guidelines of 2 or more glycohemoglobin tests per year (odds ratio, 1.9; 95% confidence interval, 1.2-3.0) and 2 or more clinic visits per year (odds ratio, 2.6; 95% confidence interval, 1.2-5.4). More non-English-speaking patients had 1 or more dietary consultations (odds ratio, 2.8; 95% confidence interval, 1.3-6.1). No other significant differences were found in routine laboratory test use or in the number of ophthalmologic examinations. Outcome variables also did not differ, including standardized glycohemoglobin and other laboratory results, complication rates, use of health services, and total charges. At these institutions, the quality of diabetes care for non-English-speaking patients appear to be as good as, if not better than, for English-speaking patients. Physicians may be achieving these results through more frequent visits and laboratory testing.