• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services


  • Print
  • Share
  • E-mail

Has The Rate of Serious Adverse Events Increased?

Table of Contents: Managing the Risks from Medical Product Use: Creating a Risk Management Framework

Previous Section :Part 2: Is the FDA maintaining the Quality of its Premarketing Reviews



To answer this question, we first looked at a 1990 report by the U.S. General Accounting Office (GAO) on adverse events first reported after a medical product goes on the market.1 GAO's 1990 report evaluated serious adverse events discovered during the postmarketing period for drugs approved between 1976 and 1985. The report defined previously unanticipated serious adverse events as events resulting in (1) the withdrawal of a product, (2) the addition of label warnings, (3) the making of a significant label change, and (4) issuing a Dear Healthcare Professional letter.2

Applying the same methods used in the GAO report, we examined the rate of previously unanticipated serious adverse events occurring after approval under the faster PDUFA-era reviews. We then compared the GAO's pre-PDUFA datasets to our own post-PDUFA datasets. To evaluate only those drugs where the review was primarily conducted under PDUFA, products approved in 1992 or 1993 were not included. And, to allow the effects of a product's postapproval market rollout to be considered, products approved in 1998 or 1999 were not included. (See Appendix A.)


The rate of withdrawals has decreased

The results of our comparison of the data showed that there has been no increase in the rate of drug withdrawals in the United States since PDUFA was enacted.3 As the graph at the end of this section shows, the Nation has experienced a 1- to 3.5-percent rate of postmarketing withdrawals for new products during the last several decades.4 In most cases, withdrawals occur during the first or second year following approval. But there have been cases where drugs were withdrawn 3, 4, and up to 5 years after approval. Of the five drugs withdrawn for safety reasons after 1992, two were approved before PDUFA was implemented.5 As a result, some additional drugs approved under PDUFA could be withdrawn in the future. Nonetheless, because the rate of withdrawals since 1992 shows a downward trend, even if a proportionate number of late-appearing problems were to result in withdrawals, an increase in the overall rate as compared to the pre-PDUFA era will most likely not occur.

Safety based NME Withdrawals


The rate of serious adverse events has decreased

The Task Force has found that available evidence does not support the charge that unanticipated serious adverse events are occurring at a higher rate since the implementation of PDUFA. We found that under PDUFA, there has been a lower rate of serious adverse events identified during the postapproval phase (30.3 percent of products) than during the 1976 to 1985 baseline years (51.5 percent of products). The table, below, shows the new molecular entities with significant postapproval label changes resulting from reports of serious adverse events, 1976 to 1985 compared to 1994 to 1997.




Table - New Molecular Entities with Significant Postapproval Label Changes, 1976 to 1985 compared to 1994 to 1997




Labeling Changes Associated with Significant Postapproval Risk





Percent with Significant Label Changes




1976 - 1985





1994 - 1997





Among drugs approved following implementation of PDUFA, the highest annual rate of postmarketing serious adverse events was still well below that in GAO's baseline data. As in the GAO audit, our comparison was limited to new drugs. New biological therapeutic and vaccine products had five significant postapproval events for 29 products approved from 1994 through 1997 (17 percent); new medical devices subject to premarketing approval were not evaluated.

Although the 30-percent proportion is better than that previously found, it still raises the question of why these serious risks are not discovered before marketing. There are several reasons for this. For example, some kinds of serious side effects, such as those resulting from drug overdoses, cannot be studied ethically in humans and can only be learned about from overdoses of drugs that are on the market. In addition, in some cases, the Agency approves drugs intended to treat serious and life-threatening diseases with less information than usual, knowing that more will be learned during the postmarketing period.6 Finally, as discussed later in this Part, it is impossible to detect or predict before medical product approval every possible drug interaction, unusual clinical situation, or rare side effect that could lead to harm once a product is on the market. Nevertheless, the Agency's goal is to minimize the number of serious adverse events that occur after a medical product is approved.

Next Section : How well is the Agency's Quality Control System Working?