OBJECTIVES: We assessed the impact of the use of a generic substitute of rosiglitazone (an anti-diabetic drug) on patient outcomes and treatment costs. METHODS: From the Maccabi Healthcare Services computerized databases we identified three groups of diabetic patients: 1. Patients who purchased the branded medication Avandia® and switched entirely to its' generic substitute Rossini® (same number of packages and dosage; the intervention group; N=1,632), 2. Patients who switched to Rossini® partially (different number of packages and/or different dosage; the control group; N=440), and 3. Patients who did not switch to Rossini® at all (N=897). For each patient we retrospectively obtained and compared the following parameters 6-months prior to, and 6-months following the switch: HbA1c levels, number of medication packages purchased, number of physician visits, hospital and emergency room admissions, and pharmaceutical costs. RESULTS: In patients who switched entirely from Avandia® to Rossini® we found a 45% decrease in treatment costs (p<0.001), as compared with a 29% decrease in costs in the control group that continued treatment with Avandia® (p<0.001) and 37% decrease (p<0.001) in the control group that partially switched to Rossini® (in the control groups there was a decline in the number of packages that were purchased after the switch). In the intervention group, we observed a reduction of 0.1% in HbA1c levels [from 7.6±1.3% to 7.5±1.4% (p<0.001)] in comparison to a non-significant difference in the two other groups [from 7.9±1.5% to 7.9±1.6% (p=0.198) and from 7.8±1.4% to 7.7±1.7% (p=0.261), respectively]. We found a decrease in the number of physician visits, and no difference in the number of hospitalization days prior to and following the switch both in the intervention and the control groups. CONCLUSIONS: The switch to a generic substitute in the case of rosiglitazone led to a reduction in health care costs, without affecting patients' health.