We developed and evaluated a comprehensive pharmacist-led, primary care-based diabetes disease management program for patients with Type 2 diabetes and poor glucose control at our academic general internal medicine practice. The primary goal of this program was to improve glucose control, as measured by hemoglobin A1c (HbA1c). Clinic-based pharmacists offered support to patients with diabetes through direct teaching about diabetes, frequent phone follow-up, medication algorithms, and use of a database that tracked patient outcomes and actively identified opportunities to improve care. From September 1999, to May 2000, 159 subjects were enrolled, and complete follow-up data were available for 138 (87%) patients. Baseline HbA1c averaged 10.8%, and after an average of 6 months of intervention, the mean reduction in HbA1c was 1.9 percentage points (95% confidence interval, 1.5-2.3). In predictive regression modeling, baseline HbA1c and new onset diabetes were associated with significant improvements in HbA1c. Age, race, gender, educational level, and provider status were not significant predictors of improvement. In conclusion, a pharmacist-based diabetes care program integrated into primary care practice significantly reduced HbA1c among patients with diabetes and poor glucose control.