- Research on synergistic effects of patient targeted interventions combined with physician-targeted interventions has been limited. To compare a combined physician-patient intervention to physician feedback alone on a composite outcome of glycemic, lipid and blood pressure control. In this cluster study 417 patients with adult-type 2 diabetes from four primary care clinics were randomized to receive either a physician-only intervention or a combined physician-plus-patient intervention. Physicians in all clinics received diabetes-related quality performance feedback during staff meetings. Patients at combined-intervention clinics also received a letter encouraging them to remind their doctors to address essential aspects of diabetes care at the next visit. At 1 year follow-up, outcome measurements included hemoglobin A1c, low density lipoprotein-cholesterol and systolic blood pressure: namely, the proportion of patients with HbA1c 9%, LDL <130 mg/dl and SBP <140 mmHg both as separate outcomes and combined. After adjusting for patient characteristics and baseline measures, follow-up levels of HbA1c (7.5% vs. 7.8%, P = 0.09), LDL (104.7 vs. 110.7 mg/dl, P < 0.05) and SBP (135.6 vs. 139.9, P = 0.10) were marginally better for combined-intervention patients compared to physician-only intervention patients. Significantly more patients in the combined-intervention (38.8%) than physician-only intervention (24.2%) met all three targets (HbA1c (9%, LDL (130 mg/dl and SBP <140 mmHg) as a single combined outcome (adjusted odds ratio 2.4, P < .01). Compared to physician-feedback alone, a dual intervention combining a patient letter with physician feedback produced modest improvements in glycemic, lipid and blood pressure control individually, but substantial improvement in a combined measure of these three outcomes together. Using composite outcomes may detect meaningful improvements in the management of complex chronic disease.