Drug Safety: What Studies Don't Report
Years of extensive testing are required by the Food & Drug Administration before a drug is approved. Yet several drugs, including Rezulin, for diabetes, Propulsid, for reflux, and the anti-angina drug Posicor have been pulled from pharmacy shelves in the past two years because of adverse reactions that showed up after approval. How could this happen?
A new study sponsored by the U.S. Agency for Healthcare Research & Quality suggests one reason: Large, randomized, controlled clinical trials--the gold standard for evaluating new drugs--may regularly underreport drug-safety problems. The study, in the Journal of the American Medical Assn.'s (JAMA) Jan. 24 issue, examined safety reporting in 192 drug trials, each involving a minimum of 100 patients. The trials ranged across seven medical areas. Say the researchers: "We found no instances where the safety reporting can be deemed satisfactory."
The study found that the severity of medical side effects was adequately reported in just 39% of drug-trial reports and drug toxicity in just 29%. Specific reasons why patients had to be withdrawn from a trial were given only 46% of the time. Also, none of this information was easy to find. The authors report that, on average, medical-journal articles on drug trials devote only a third of a page to such information.
A second JAMA article reported that Phase 2 clinical trials, which are meant to ferret out safety problems before a drug is tested on a larger group, may be adding to the problem. These studies, say the authors, are usually too small and their patients too carefully selected to produce the very data on potential adverse outcomes they are meant to uncover.