Hall wonders if Vioxx was overpromoted compared to what the alternatives are. While fingers are being pointed at pharma and FDA, he says, "There's a basic question about how we reimburse for drugs. Your insurance company reimburses you for prescription drugs—Nexium, Vioxx, what have you—while many over-the-counter drugs have to be paid for out of your own pocket. The patient really has no choice after a while." Another thing: "If you have a $3 billion to $4 billion drug and you're spending a quarter of a billion dollars on advertising, you've got to believe that it's being used beyond what the strict indications warrant," says Hall.

Will the Vioxx issue spur tighter regulation of DTC marketing? Says Morris: "The Democrats are interested in putting some limits on DTC. It does provide them with ammunition."

Expect the Unexpected The Vioxx recall underscores the need to do better postmarketing surveillance of drugs, says Sandy Schwartz, MD, professor of medicine, health care management, and economics at the University of Pennsylvania School of Medicine. "We all recognize that initial clinical trials are not designed to pick up unanticipated side effects," he says. "Unanticipated side effects are just that—unanticipated. But statistically we know they're going to happen. And we don't have an adequate system in place to detect them."

Schwartz says, "We're looking at the risk-benefit of drugs—something the public doesn't understand and the lay press doesn't do a good job with. We're not talking about things that are good or bad. With all these drugs, it's a question of how good, in which patients, under what conditions. Every drug, if enough people take it in pharmacologic doses, is going to have some adverse events. The real question is whether the benefits of the drug are sufficient to significantly outweigh the risk. To the degree a drug is used in the highest-risk patients, then even its chance of doubling the risk of a cardiovascular event might be acceptable if the benefits are sufficiently high. However, to the degree it is used in people for whom the expected benefit is minimal—people who don't really need it—then even a very small risk is unacceptable. FDA—the staff and advisory committees—does an excellent job dealing with that. But it's not always appreciated by the general public. The public thinks a drug's either good or bad."

But, he adds, "There's a tendency to use some of these new drugs rather indiscriminately instead of targeting them to the people who would benefit the most. I don't think there's any question that overprescribing has been going on in this class of drugs. Drugs should be used where there's a significant benefit. We have to recognize there are always going to be unanticipated and unmeasured adverse events, what we call a Type 2 error. There's a chance you'll miss something because you're not looking for it. In this case, we were lucky it was found."