Background: Medications are increasingly being approved with limited, short-term evidence regarding safety. Regulatory safety concerns may emerge for these drugs but later may be reversed if additional evidence suggests no warning is indicated.
Objective: To describe trends over time in the initiation of rosiglitazone and pioglitazone-both in the thiazolidinedione (TZD) class-and medications from the dipeptidyl peptidase-4 (DPP-4) inhibitor class before and after the FDA removed a black box warning and restricted access program for rosiglitazone regarding an increased risk of myocardial infarction.
Methods: This retrospective study evaluated initiation of TZDs and DPP-4 inhibitors using 2001-2015 administrative claims data from a U.S. commercially insured population. Patients were aged 18-64 years and were new users of either a TZD or DPP-4 inhibitor. Among all patients who were new users of either a TZD or a DPP-4 inhibitor during each quarter-year (Q), the percentage of patients who initiated rosiglitazone, pioglitazone, and DPP-4 inhibitors were calculated.
Results: There were 630,977 patients eligible for the study. During 2007, rosiglitazone initiators decreased from 39.1% to 8.0% in 2007 Q4 when the black box warning was implemented. During 2010, rosiglitazone initiators decreased from 7.6% to 1.0%, as safety evidence accumulated and the restricted access program requirement was announced. Rosiglitazone initiation remained below 1.0%, even after regulatory restrictions were removed in November 2013. Pioglitazone initiation decreased from 46.4% in 2010 Q1 to 14.8% in 2011 Q4 and remained relatively constant between 14.5% and 17.8% after regulatory restrictions for rosiglitazone were removed. After DPP-4 inhibitors first became available in 2006 Q3, initiation of this medication class increased rapidly, stayed relatively constant between 42.8% and 45.5% in 2009, and then quickly rose and remained above 80% from 2012 through 2015.
Conclusions: This case study provides some evidence that adding and later reversing drug safety warnings-particularly those with restricted access requirements-may affect the uptake of the targeted product into the population when multiple clinically relevant treatment alternatives are available (such as type 2 diabetes). Once a treatment falls out of favor, removal of safety warnings and/or restricted access programs may not lead to increased use.
Disclosures: This project was not directly supported by any funding. Hickson was supported by the National Heart, Lung, and Blood Institute through a National Research Service Award (NRSA) training grant (4T32HL007055-41) as a postdoctoral research fellow with the Cardiovascular Disease Epidemiology Program at The University of North Carolina at Chapel Hill (UNC-CH). Cole was supported by a NRSA Predoctoral Traineeship from the Agency for Healthcare Research and Quality sponsored by The Cecil G. Sheps Center for Health Services Research, UNC-CH (grant no. T32-HS000032) and a predoctoral fellowship from the American Foundation for Pharmaceutical Education. Unrelated to this project, Cole was a part-time employee of Truven Health Analytics/IBM Watson Health. Dusetzina has nothing to disclose.