Objective: The purpose of this analysis was to determine if high-risk cardiovascular patients with concomitant diabetes and a pre-treatment A1c measurement greater than 9% experienced better outcomes with insulin containing anti-diabetic medication regimens than similar patients not taking insulin. Methods: High-risk cardiovascular patients with concomitant diabetes and a pre-treatment A1c measurement of greater than 9% from a large western United States integrated health-care system were evaluated for an A1c measurement at least three months prior to treatment initiation and a follow-up measurement at least three months following initiation. In the case of patients taking insulin, the three month follow-up period started with the initiation of insulin. Change in A1c resulting from medication treatment was evaluated using a two-step endogenous treatment regression model, with insulin as the endogenous treatment variable. Exogenous independent variables included hypertension diagnosis, hyperlipidemia diagnosis, age, gender, and distance from treatment goal (A1c less than 7%) at baseline. Standard errors for beta coefficients were computed using HCCM3. Results: Of 11,181 diabetic patients, 707 (314 on insulin) patients met the inclusion criteria. Although both insulin and non-insulin containing treatment regimens reduced A1c from baseline to follow-up, patients on insulin were associated with a greater reduction in A1c than patients not on insulin. The difference in change was 2.6% and was statistically significant (p=0.0029). Conclusion: Use of insulin in an anti-diabetic regimen in high-risk cardiovascular patients with concomitant diabetes and a pre-treatment A1c greater than 9% resulted in a significantly greater reduction in A1c compared to patients not taking insulin.