Study: New Drugs Have Side EffectsMay 1, 2002 | AP - New research estimating that 20 percent of recently approved prescription drugs have serious and even life-threatening side effects suggests doctors should prescribe older medication whenever possible.
"It's like playing Russian roulette when a doctor prescribes a newly approved drug that doesn't have a big breakthrough," said Dr. Sidney Wolfe of Public Citizen Health Research Group, a co-author of the study.
The findings, published in Wednesday's Journal of the American Medical Association should prompt the Food and Drug Administration (news - web sites) to consider raising its threshold for approving new drugs when safe and effective alternatives exist, the researchers said. They said pressure from pharmaceutical companies and doctors' failure to closely read warning labels are partly to blame.
An accompanying editorial by two FDA experts said the analysis overstates the problem.
Safety studies that are conducted before a drug wins approval typically involve a few thousand patients and may not detect all side effects, especially relatively rare ones, Drs. Robert Temple and Martin Himmel said.
"Frequent post-marketing label changes are therefore inevitable and should be anticipated," they wrote.
Temple also noted that some medications cause side effects in only certain groups of patients, such as pregnant women, which does not mean a drug is dangerous for everyone.
The researchers, led by Dr. Karen Lasser of Cambridge Hospital and Harvard Medical School (news - web sites), analyzed 548 drugs approved from 1975 through 1999. Of these, 56, or more than 10 percent, were later given a serious-side-effect warning or taken off the market for safety reasons. The number climbed to about 20 percent when the researchers took into account drugs that were approved toward the end of the period studied.
"When a drug that comes on the market has a 1-in-5 chance that it's going to have to be banned or get a black-box warning is pretty worrisome," Wolfe said.
Most troublesome new drugs do not represent any advance in treatment and are at best no better than older, safer drugs already on the market, he said.
The study analyzed what are known as "black-box" warnings listed in the Physicians Desk Reference, a compendium of drugs and labeling information published annually. Black-box warnings highlight the most serious side effects.
Sixteen drugs studied were withdrawn from the market, about half of them within two years after winning FDA approval.
They include the diabetes drug Rezulin, which was approved in 1997 but has been linked to dozens of cases of fatal liver damage. Lasser said doctors continued to prescribe it in an unsafe manner even after it was given a black-box warning, and it was ultimately withdrawn from the market in 2000.
Two allergy drugs, Seldane and Hismanal, were linked with potentially fatal heart problems in certain patients but were not removed from the market until several years after receiving black-box warnings.
The FDA is correct in saying doctors don't pay enough attention to warning labels, but that is "all the more reason to do the right thing on the front end," Wolfe said. "The remedy should be don't put the drug on the market unless it's a breakthrough drug."
The FDA has said that while its drug review process has gotten shorter in recent years, the procedure is still adequate. But the agency has expressed concern over doctors not reading drug warning labels closely.
Temple said that while doctors are getting better at reporting side effects to the FDA and drug companies, the agency is seeking further improvements, including a proposal to include a drug's approval date on packaging inserts.
"I don't think anybody believes that we're absolutely at the best we can do, but it's better," Temple said.
The study "adds to the body of evidence that we really have to enhance our system of risk management and evaluation of drug safety and effectiveness," said Christopher Milne, assistant director of Tufts Center for the Study of Drug Development, who was not involved in the research.
But it also risks "going overboard" and making all patients on new drugs think they should stop taking them, which could be harmful, Milne said.