By Sidney Wolfe, M.D.
Amid the furor involving pain drugs, it may be of interest to ponder the theoretical problem of the Food and Drug Administration prematurely deciding to ban a drug whose benefits outweigh its risks. But we must confront the reality that during the past 10 years, there is overwhelming evidence that the FDA repeatedly delayed withdrawing many drugs long after clear, unequivocal evidence of risks that outweighed any demonstrable benefits.
The often-failed FDA strategy of “risk management,” a dangerous interim alternative to taking drugs off the market, has ensured that thousands of people were injured or killed after certain drugs should have been pulled.
A recent example is the diabetes drug Rezulin, first marketed in 1997. By that December, it was taken off the market in the United Kingdom because of liver damage, many cases having occurred in the United States.
In July 1998, Public Citizen petitioned the FDA to ban the drug, which, although lowering blood sugar by a different mechanism than earlier diabetes drugs, had no evidence of improved mortality or morbidity. By July 1998, there were 560 reported cases of liver damage, including 26 deaths.
After failed efforts at “risk management,” Rezulin was withdrawn from the market in January 2000, by which time there were hundreds of additional cases of liver damage and 63 deaths. Other belatedly banned drugs include the painkiller Duract, the blood-pressure drug Posicor and the cholesterol-lowering drug Baycol.
Decisions to approve or remove drugs must be based on adequate evidence of both benefits and risks. For Vioxx, Celebrex and Bextra — the three COX-2 inhibitor pain/arthritis drugs — at approval there was no evidence that they were more effective than older drugs.
Only Vioxx, now withdrawn because of a significant increase in heart attacks, proved to be less dangerous to the gastrointestinal tract than older drugs. Celebrex and Bextra did not. Although at approval there was no evidence of increased cardiac risk from Vioxx or Celebrex, less than a year later a 2001 study on Vioxx found it was five times more likely than naproxen (Aleve) to cause heart attacks, and a Celebrex study prompted FDA concerns about cardiac risks for that drug.
Removing Celebrex from the market will be a major step forward for public health. For naproxen, the evidence of its harm is preliminary at best, and it is unlikely to pose cardiac risks as high as those documented for Vioxx or Celebrex. There is no basis for removing it from the market.
Unique risks without unique benefits should always be the algorithm for removing drugs.
Sidney Wolfe is the director of Public Citizen’s Health Research Group and co-author of the new edition of Worst Pills, Best Pills. Public Citizen’s Web site, www.worstpills.org, lists what it says are safer alternatives to 181 drugs.
