Industry funding of the FDA: effects of PDUFA on approval times and withdrawal rates

The time for FDA review represents a significant component of the overall time of drug development.In 1992, the US Congress, in an attempt to reduce the time and cost of drug development, enacted the first in a series of Prescription Drug User Fee Acts (PDUFAs). PDUFA I authorized the FDA to collect fees from sponsors submitting a new drug application (NDA) or biologics license application (BLA), and enabled the FDA to hire additional review staff to facilitate more rapid review.According to the PDUFA legislation, in exchange for collection of user fees, the FDA is legally obliged to “review and act on” NDA/BLA submissions, but not necessarily approve them more rapidly.Congress renewed the 1992 PDUFA legislation under the Food and Drug Modernization Act of 1997 (PDUFA II) and again under the Bioterrorism Preparedness and Response Act of 2002 (PDUFA III).Detractors of PDUFA suggest that payment of user fees by the industry undermines the regulatory abilities of the FDA and has created a federal agency that is beholden to the very companies it is mandated to regulate. Proponents of PDUFA maintain that the FDA's decisions on drug applications are independent of the source of the revenue, because user fees from the sponsoring organization are not paid directly to any individual reviewer or division within the agency.In an attempt to assess the impact of PDUFAs, in this article, we review data on drug approvals and drug approval times over the past 25 years. We also consider drug withdrawals during this period, which are often used as a proxy to assess the FDA's safety record.We document that implementation of PDUFA led to substantial incremental reductions in approval times beyond what would have been observed without these acts (6–7% annual declines during PDUFA I and ∼3–4% during PDUFA II).A preliminary examination of the trends in the number of new molecular entity (NME) withdrawals also indicates that the proportion of approvals ultimately leading to safety withdrawals prior to PDUFA and during PDUFA I and II were not statistically different (ranging between 2% and 3%, depending on the method of analysis).