OBJECTIVES: To evaluate the effect of pharmacist interventions on diabetic patient outcomes compared to usual care without pharmacist clinical intervention. METHODS: The study was implemented in “Safety Net” clinics located in Los Angeles County serving poor, Hispanic patients. Pharmacist services included review of the patient's medical and medication history, medication evaluation and adjustments or deletions under an established protocol, ordering and reviewing routine laboratory tests [e.g., HbA1c levels, LDL, HDL, etc.], monitoring patient compliance, patient education and scheduling follow-up visits. Control patients were selected retrospectively to match the intervention cohort. The primary outcome was measured as the reduction of A1c and the percentage of patients who achieved the treatment goal, an A1c <7%. Secondary outcomes focused on the diabetes-related outcomes, a decreased in blood pressure, lipid levels and BMI. METHODS: Multiple linear regression and logistic regression models were used to evaluate the effect of pharmacist intervention on patients' A1c levels after controlling for gender, baseline age, baseline A1c values, ethnicity, smoking status, insurance, and the duration between the index date and last visit. RESULTS: A total of 484 patients were included in the analysis: 222 who received pharmacist intervention, and 262 receiving usual care. The change in A1c between baseline and last visit was -1.7 percentage points in the intervention group, -1.1 percentage points with usual care. The estimated impact of the pharmacists' interventions was -1.38 percentage points (p<0.0001) after adjusting for differences in patient characteristics using ordinary least squares regression techniques. Similarly, the pharmacist intervention increased the likelihood of achieving the treatment goal of an A1c < 7% by four-fold (OR=4.037, p<0.0001). CONCLUSIONS: Pharmacist interventions do have significant benefit for patients with diabetes. The results demonstrate the importance of the pharmacist in diabetes control and can be applied to similar clinics serving mostly uninsured Hispanic patients.