Alex Tabarrok's contribution highlights one of the main difficulties the health care industry and patients will have to confront over the coming decades as we move in the direction of more personalized treatments. Ordinarily, innovation tends to drive prices down. But if drug firms continue to face very high fixed costs associated with the FDA approval process--even for products that treat increasingly smaller patient populations--third party payers will naturally balk at paying prices high enough to compensate for R&D expenditures.
Alex suggests moving "away from pre-market gatekeeping and towards post-market surveillance and information provision." But critics argue that a market for drug safety and efficacy information, a la the Consumer Reports model Alex suggests, just isn't feasible. Fortunately, we don't have to wait to see whether such a market might or might not develop. It already has - in the form of guidance for physicians on off label uses.
Briefly, off label use occurs when a doctor prescribes a drug for a condition for which the FDA has not approved it. It's not only legal, but often constitutes the medically recognized standard of care. And because they're not FDA approved, off label uses provide some pretty good insights into how the market can and does provide usable information even in an industry characterized by substantial information asymmetries.
Doctors learn about off label uses in a variety of ways, but one of the most important is their inclusion in practice guidelines and treatment compendia published by physician professional associations, non-profit organizations, and even the federal government. An off label use doesn't end up in a treatment compendium just because a drug manufacturer says it's safe and effective; there has to be some pretty strong evidence to support the use. Nevertheless, a study by Howard Beales that looked at off-label drug uses eventually approved by the FDA found that these uses appeared in official treatment compendia an average of two and a half years before FDA approval. Thus, arguments that a Consumer Reports model for pharmaceuticals could never work have already been proven wrong. In fact, in some regards, it seems to work even better than the FDA.