By DAN RODEN, M.D. • October 10, 2008
Dr. Dan Roden is professor of medicine and pharmacology, director of the Oates Institute for Experimental Therapeutics, and assistant vice chancellor for personalized medicine at Vanderbilt University School of Medicine.
The Food and Drug Administration recently announced a public Web site listing potential drug side effects it is investigating. This move does seem to be in the public interest; after all, shouldn’t we know about drug side effects that are concerning enough to warrant an FDA investigation?
On further reflection, this turns out to be a pretty complicated question, and raises important issues about how doctors and patients should evaluate new information about the risks and benefits of drugs we use.
Before a drug is approved for marketing, it is tested in several thousand patients, and common side effects are identified. However, some side effects are only discovered after a drug is marketed. Some of these are so rare they just didn’t happen before a drug was approved. Others don’t turn up until they are identified in large, carefully conducted trials: An increase in heart-attack risk with Vioxx is a recent example.
There is no organized system in the United States, or elsewhere, to systematically examine the safety of marketed drugs. For many years, the FDA has run a voluntary adverse-events reporting system (AERS) that allows health-care providers and patients to describe possible side effects.
While AERS may generate clues to new side effects, the reports are often incomplete and may raise more questions than they answer. Is the rare effect really caused by the drug? What happens if a few people report something common, like an ulcer or a heart attack? How many people would have to report something before the FDA should get interested?
More reports often help
What is clear is that some of the 400,000 AERS reports submitted each year may represent real side effects, and others do not. One argument says that until further studies understand whether a drug actually causes a side effect, making a suspicion public would be counter-productive; for example, patients benefiting from the drug might stop needlessly.
I see a potential benefit to publicizing AERS reports that the FDA is investigating. Suggesting tenuous relationships between drugs and unusual side effects may prompt other physicians or consumers to recognize that they, too, may have seen (or been) similar cases. More reports do not establish cause and effect, but have helped in the past.
I also see an opportunity for coupling this kind of reporting to increasing public awareness of the risks vs. the benefits of treatment with any drug. An aware public will understand that because the FDA is investigating a potential relationship does not establish that a relationship even exists.
There is no drug that is 100 percent effective and never causes side effects. What we don’t want are drugs where the risks are much greater than the potential benefits, or where we can’t pick out the people at high risk for serious side effects. The new reporting system may, therefore, be one small step toward identifying unsafe drugs.