OBJECTIVES: Meta-Analyses of oral hypoglycemic agents (OHA) revealed that Rosiglitazone (Rosi) increased the risk of Myocardial Infarction (MI) and Heart Failure (HF), and Pioglitazone (Pio) increased the risk of HF and decreased the risk of MI. The objective of this research is to compare OHA utilization patterns, before and after these publications. METHODS: CareFirst BlueCross BlueShield’s claims were analyzed for patients continuously enrolled from January 2005 through December 2007 who started on mono-Rosi or mono-Pio. The “pre-publication” period (November 1, 2006 – March 31, 2007) was compared to the “post-publication” period (July 1, 2007 – December 1, 2007) using a difference-in-difference approach. Multinomial logistic regression (MLR) explored discontinuation; continuation with monotherapy or adding another drug; and switching after adjusting for gender, age, type of insurance, history of MI or HF and risk factors for MI and HF. RESULTS: The number of monotherapy Rosi users decreased from the pre (N=368, 5.94%) to post (N=170, 2.87%) period, while monotherapy Pio use was stable across the two periods. The proportion who switched from Rosi to non-Rosi drugs changed from 2.17% in pre-period to 5.88% in post period. Adjusted relative risk was 2.66 (95 % CI 1.0569, 6.7689). Pio to non-pio drugs switching was 1.48% in pre-period and 1.16% in post-period (relative risk not significant). Therefore, the impact of the new studies on Rosi users to switch to non-Rosi drug, relative to Pio users before and after the publication was 3.6189 (90%CI 1.051, 12.457). CONCLUSIONS: Consistent with prior research, the utilization of Rosi declined by more than half in the post-period. Additionally, Rosi users were three times more likely to switch to a non-Rosi drug in the post period, relative to Pio users. Therefore, our results show that the publications about safety risks had differential impact between the two drugs within the therapeutic class.