« Wendell Berry on farming, Jesus, tradition, and corporate tyranny | Main | Butt slogans for undergrad girls »

drug trial registries: pro and proer

I'm usually a "big picture" kind of guy, but when it comes to the recent controversy over whether drug companies should be required to post all of their clinical trials to a public registry, it's worth dwelling on the details.

First, the major proposals to create registries would require varying amounts of information to be made public. The still-in-the-oven proposal from the International Committee of Medical Journal Editors (scheduled for release later in the year) would only require drug firms to publicize the existence of all clinical trials in a public registry. While this would be an improvement over current practices, doctors and patients would still not have access to the data. Instead, the registry would be useful only if it encouraged a healthy skepticism: "I see this study in JAMA reporting the favorable effects of a new drug, but I see from my own perusal of the public registry that only one study has been published out of the five that were started..."

There are two reasons why this proposal is too limp: (1) not every doc is going to have the time or the inclination to check every positive, published study against the public registry--especially when she's running late because of that burrito lunch with the drug rep. (2) The skepticism might sometimes be unfair, because a study could have been stopped for reasons that do not reflect poorly on the drug--all clinical trials at an institution could have been suspended because their IRB was asleep at the wheel, for example.

For these reasons, the public registry proposal from the AMA is better. It would require that the outcomes of every clinical trial be posted to the registry. For this sensible idea to work, though, the FDA would have to stop going along with the absurd notion that the study data was proprietary. The need for physicians and patients to evaluate the safety and effectiveness of drugs aggressively marketed to the public should not be held hostage to some misguided notion of intellectual property more suited to the manufacture of Pez dispensers (how do they do that, anyway??)

Some critics make the argument that any registry would increase the costs of research, in part because drug firms would be "afraid to make a mistake." But this fear is much greater when a firm knows that its competitors can hide their mistakes. The public understands that some drugs won't work; that's why we do studies in the first place. Besides, arguments about cost lose their luster when you realize that drug firms spend about as much on marketing as on research.

As an aside, even when and if a public registry is implemented, don't assume that the pharmaceutical companies will comply with any required reporting... (pdf file, small. Go ahead! Open it!)