The dangers (?) of public funding of drug trials

Ezra Klein relays Jim Manzi's worry that public funding of drug trials

exposes you to the inverse problems of the current system. Namely, "bureaucrats and politicians tend to have enormous career risk from an unsafe drug introduction, but almost none from a rejected drug that would have been effective had it been introduced...[it] would likely result in fewer new drugs being brought to market."

There's a bit to this. But it misses something important: The biggest problem with the present system may not be that deeply unsafe drugs are approved but that too many drugs that carry modest safety issues (and most drugs carry some safety issues) but little if any benefit are approved because their benefits are overstated. The risk-benefit ratio gets misrepresented, in other words, not so much because risks are understated (though that is done too) as because the benefits are oversold.

There are exceptions, of course, like Vioxx, in which an effective drug (it reduced pain) is more dangerous than revealed. But many of the most problemmatic drugs -- most notably psychiatric drugs, which get taken longer and by many, many people -- won approval and entered the market and became popular and then became safety problems, through sheer excess of unmonitored and ill-advised use, only because their benefits were exaggerated enough to win FDA approval and then get marketed as the Next Great Thing.

if the problem is overstated benefit rather than understated risk, then a move to public financing is not as apt to create an "inverse problem" in which risk-averse regulators Just Say No. TIghtening the risk filter, for instance, would probably have kept out only the most egregriously dangerous of the recent drugs-of-scandal, which we would hope is a good thing. A more balanced, skeptical approach to weighing risk against benefit, meanwhile, can only be a good thing. And I'm not sure how reasonable it is to fear that overcautious regulators would "likely result in fewer drugs being brought to market." Much of the drug problem the last few years is that too many drugs were brought to market -- drugs that improved little if at all on previous drugs (or even placebo) and that served mainly to increase health-care costs and, through the sheer hope of the New Good Thing, the number of people taking drugs with questionable efficacy.

Public funding of drug trials seems to me a promising idea, especially if, as Lisa Bero recommends, it emphasizes head-to-head drug competition (see bottom of post) rather than the rather empty requirement that a drug merely beat placebo rather than existing drugs.

More like this

Virginia Postrel says It's the worst regulatory news I've heard in a long time--and it predates the new administration by a half year. Sidney Wolfe, who seemingly never met a new drug or device he thought should be legal, has been named to four-year term on the FDA's Drug Safety and Risk Management…
Ronald Bailey at Reason has an article about the costs of the FDA black box warning on antidepressants: Excessive caution is risky, too. Back in 1992, Congress, worried about the slow rate of approvals, passed legislation imposing FDA user fees on pharmaceutical companies. Flush with these new…
You probably don't know what PDUFA is (pronounced pah-doofah) but it's about to expire. Which is good. If they let expire. "They" are our friends in congress. PDUFA stands for the Prescription Drug User-Fee Act and it is an integral part of the FDA drug safety program. You know. The one that…
The first two posts in this series are href="">here and href="">here.   The final editorial in the NEJM's three-part series…

As someone with experience trying to develop new drugs, I'm having a very hard time making sense of these arguments.

First, we already have a group of politicians and beaurocrats with high career risk from unsafe drug introduction but almost none from rejection of safe drugs: the FDA.

Second, what does public financing of clinical trials have to do with the approval process? Paying for the trials is completely independent of approving the drugs. (Witness our current system, where pharma pays but FDA approves.)

Third, I don't quite see how overstating benefits is primarily a problem during drug approval. After all, FDA has access to all the clinical data. They can and do determine for themselves what the benefits of a potential new drug really are. Sure, the drug sponsors try their best to convince FDA that the benefits are high and the risks low. But if anyone is in a position to see past overstated claims, it's FDA, precisely because they get to see all the data.

IMO, overstated benefits (and understated risks) are mainly a problem when pharma markets their drugs to MDs and consumers. We only get to see the data summaries in the package inserts and in whatever papers pharma chooses to publish. But that's essentially all post-approval.