Crawling Toward Better Health Info

Let's get wonky.

In a move that's long overdue, the Food and Drug Administration and the National Institutes of Health are teaming up to create an electronic database for all clinical research data about regulated medical products, "from the earliest research phase to postmarket surveillance," the online newsletter FDA Webview (subscription required) reports. The agency solicited input from private vendors on how to create such a system earlier this week.

This effort bears close scrutiny by researchers and consumer groups, even though it is still years away. The FDA has bungled previous efforts at setting up user-friendly databases, and anybody that spends any time on its website knows that the agency is hopelessly incompetent at developing consumer-friendly tools for imparting information. Moreover, there's likely to be huge fights over who will have access to the data, with industry fighting to keep its proprietary information away from prying eyes.

Such restrictions would make this database much less useful to academic and independent researchers, who need a central repository for all clinical trial outcomes data about drugs, biologics, devices and medical procedures to be comprehensive. That's why NIH's participation is crucial: it already runs ClinicalTrials.Gov, which serves as a central repository for listing the existence and protocols for clinical trials. The next step logical step is creating a central database for listing the results.

Once that's up and running, it should be easier and quicker for researchers to conduct meta-analyses on problematic therapies. In a meta-analyses, a researcher pools the results of numerous trials to show effects that may not have been apparent to any of the researchers who conducted the individual trials in the mix.

In addition, the FDA and NIH could expand the database to include medical outcomes data. Indeed, large insurers and Medicare should be consulted on how to design such a system so that their soon-to-be computerized medical records (yes, some day the U.S. will catch up with the rest of the industrialized world and digitize individual medical records) can be easily incorporated. The agencies will have to take steps to insure that individuals' privacy is properly protected, of course.

How could this outcomes data be used? David Graham, the safety expert turned whistleblower at FDA, used Kaiser Permanente data to prove that Vioxx was causing a significant increase in heart attacks among its users. His job would have been a lot easier if he had access to a centralized repository that incorporated a much wider range of outcomes data.

A central repository of outcomes data would also be a valuable resource for researchers who want to conduct post-treatment comparative effectiveness studies, which is key to holding down skyrocketing medical costs. Sen. Hillary Rodham Clinton, ever the wonkette, has championed comparative effectiveness research on the campaign trail.

Posted by gooznews at March 30, 2007 07:26 AM