FDA Blues

NEITHER THE recent Senate testimony of David J. Graham, an unhappy Food and Drug Administration employee, nor the controversy over Merck Inc.'s decision to withdraw its painkiller Vioxx from the market have provided conclusive evidence that the FDA's coziness with the drug industry has made the agency, in Mr. Graham's words, "incapable of protecting America against another Vioxx." But they do provide evidence that intellectual fashions in the field of drug regulation have changed. In 1992, Congress passed a law designed to speed up the rate of new drug approval, because too many ill people were having to wait too long for new treatments. Now, the FDA's higher speed of drug approval is sometimes cited as evidence that the agency is too friendly to the drug industry. Similarly, when large numbers of drugs were withdrawn from the market in the past, the agency was accused of having to do so because of sloppy approval procedures. But when fewer drugs are withdrawn, as has recently been the case, the agency is said to be too reluctant to hurt the companies whose drugs it has approved, despite the fact that the lower rate of withdrawal also reflects a smaller number of new drugs on the market.

The current discussion has also demonstrated that the tools the FDA uses to regulate drugs once they are on the market are too blunt. Soon after it had been approved, for example, the agency suspected that the use of Vioxx might cause cardiovascular problems. But officials had to negotiate for a year before they were able to persuade Merck to add that risk to the drug's labels. The negotiation was time-consuming because, other than withdrawing a drug from the market -- which the agency is rightly reluctant to do unless serious risks have been proven (and the risks of Vioxx were not) -- the FDA doesn't have any way to force drug companies to change their labeling. Yet drug labels can not only warn patients of possible side effects, they can also alert physicians to categories of patients who should not be taking particular drugs. In a world in which there are more kinds of drugs than ever and in which some risks will be considered worth enduring for some categories of patients but not for others, labels play an important role and must be reliable.

The same is true of post-approval testing of drugs already on the market. The FDA can recommend that a company undertake a clinical trial, but the agency cannot demand that a company do so, nor can it order independent or comparative testing. For that to be possible, the agency would need greater resources, as well as greater enforcement powers, both of which would have to come from Congress. In recent years, the number of drugs that Americans take has risen rapidly, yet the agency's drug safety funding and capacity have remained flat. When the new Congress takes its place in January, members should look harder at the tools the FDA uses in regulating drugs and at whether they should be improved.