Class of 2008 Abstract / Objectives: To determine the cost-effectiveness of adding a thiazolidinedione (TZD) versus insulin glargine (glargine) as a triple regimen for treatment of Type 2 diabetes mellitus for patients not controlled with metformin and a sulfonylurea. Methods: A decision analytic model was developed to compare the clinical outcomes and costs of triple therapy with either a TZD or glargine. Published literature was used to determine treatment efficacy and the frequency of clinically important adverse effects. Cost data were obtained from the 2007 Physician Fee Reference and North Carolina Industrial Commission website. The decision tree was built using TreeAge software. Clinical outcome measures included HgA1c (A1C) control, hypoglycemia frequency, and the development of edema associated with the use of these medications. A Monte Carlo probabilistic sensitivity analysis was conducted to determine the mean and 95% CIs for both treatment efficacy and costs. Results: There was no statistically significant difference in the efficacy of adding either a TZD or glargine in achieving a goal A1C ≤ 7%. However, glargine triple therapy was estimated to be significantly less costly than TZD triple therapy ($3,161/yr; 95% CI $3,116 to $3,356 versus $3,769/yr; 95% CI $3,667 to $3,902, respectively). Conclusions: Most patients requiring triple therapy for the management of T2DM should receive glargine rather than a TZD due to the significantly lower cost producing similar clinical efficacy.