The question of whether FDA should shoulder the burden of comparative effectiveness research crops up from time to time, but there is more faith in the appropriateness of that notion than I believe is warranted. I only bring this up because Steve Grossman of the Alliance for a Stronger FDA (Washington) raised the issue recently at the Gooznews blog. So let's take a quick look at this issue and examine the question of not just whether FDA is sourced to do this, but whether it should even consider such a task.

As for resources, this is a no-brainer. While the folks at the Centers for Medicare & Medicaid Services might disagree, the prevailing view seems to be that FDA suffers from the most serious mismatch in terms of resources and tasks, a view that is far from a novelty. This alone ought to quash any further such discussion, but apparently is not.

Another point to be made here is that comparative effectiveness is a very different animal than safety and efficacy, and I just don't see that FDA as it is currently constituted is prepared to deal with CE considerations. After all, the point of CE is not to discern which treatment is superior (or non-inferior), but for which patients is which treatment superior. My impression is that the base of data for making such evaluations is not particularly well built out as yet. In the case of the highly iterative therapeutic medical device, the data are obsolete by the time you have them anyway.

This question of data obsolescence does not apply to all devices, that's true. It's also true that comparing medical management to a device type may be illuminating, but a different coating on a drug-eluting stent or a different algorithm in an implantable defibrillator can quickly render a comparison irrelevant.

I'd also hasten to point out that those who believe FDA should handle CE are doing a nice job of ignoring the fact that the taxpayer is already financing two entities whose charge is just this, namely the Agency for Healthcare Research and Quality and the Federal Coordinating Council for Comparative Effectiveness Research. Really, is there any need for FDA to get involved at all?

And finally, one must ask how would FDA obtain the data needed to conduct CE research. The short answer? Clinical trials financed by industry. No longer would it suffice to demonstrate safety and efficacy compared to the current standard of care. Now you have to enroll even more. If you are a big fan of device flight, you'll love this idea. What better way to drive more device makers out of the U.S.?

So I'd like to make a request of everyone inside and around the Beltway. Please, enough of this business of having FDA conduct CE research. It's a lousy idea any way you slice it and even if the agency could do it, it'd be redundant. May we now finally lay this idea to rest?