Availability of Comparative Efficacy Data at the Time of Drug Approval in the United States

CONTEXT: Comparative effectiveness is taking on an increasingly important role in US health care, yet little is known about the availability of comparative efficacy data for drugs at the time of their approval in the United States. OBJECTIVE: To quantify the availability of comparative efficacy data for new molecular entities (NMEs) approved in the United States. DATA SOURCES: Approval packages publicly available through the online database of drug products approved by the US Food and Drug Administration (FDA). STUDY SELECTION: Identification of efficacy studies that supported approval of each NME approved by FDA between 2000 and 2010. DATA EXTRACTION: We determined whether eligible studies were head-to-head active controlled trials and whether the results of such studies were available in the approval packages. We recorded the approved indication, whether the NME was an orphan product, whether the NME had undergone priority review, and whether the control group was a specific active comparator or standard care. RESULTS: Of 197 NMEs identified that met eligibility criteria, 100 (51% 95% confidence interval CI, 44%-58%) met criteria for having comparative efficacy data available at the time of market authorization. After excluding NMEs designated as orphan products (n = 37) and those approved for indications for which no alternative treatments existed (n = 17), this proportion increased to 70% (95% CI, 62%-77%). The proportions of NMEs with available comparative efficacy data varied widely by therapeutic area, from 33% (95% CI, 9%-67%) for hormones and contraceptives to 89% (95% CI, 56%-99%) for diabetes medications. CONCLUSION: Publicly available FDA approval packages contain comparative efficacy data for about half of NMEs recently approved in the United States and for more than two-thirds of NMEs for which alternative treatment options exist. We did not investigate the extent to which available comparative efficacy information is useful for clinical guidance.