Conflicts of Interest Linkfest

Lots of good posts out there in the COI-o-sphere.  The indispensable Health Care Renewal has a number of interest posts.  Cetona links to an article entitled "Allied Against Reform: Pharmaceutical Industry-Academic Physician Relations in the United States, 1945-1970," forthcoming in the Bulletin for the History of Medicine.  The article is remarkable, particularly in how it shows the dialectic of history, how often our present concerns and debates -- this time over the nature of the relationship between the pharmaceutical industry and academic physicians -- are reflected and constructed in the past.

The point of the emphasis on dialectic is to eschew a linear conception of history, to understand that past events, conditions, and ideas strongly shape contemporary events, conditions and ideas. 

Roy Poses has an excellent post on CT scans and cancer, and therein he queries, "At the minimum, this case suggests that academic medicine's current method of managing conflicts of interest through disclosure is not working well."  When I write, speak, or teach on COIs, one of my major objectives is to clear the detritus of what I consider to be the relatively simplistic picture of the effects of COIs on human behavior that still dominates academic medicine and research.  This picture tends not to take account of some of the more sophisticated psychological, sociological, and anthropological assessments of the way humans tend to behave when faced with such COIs.

One of the ideas I try to challenge is the notion, still quite prevalent even among well-intentioned critics, that disclosure of COIs is a panacea.  This idea is not supported by the evidence.  One particularly good 2005 study (Manchanda & Honka) compared the effects of gift-receiving on prescribing practices among groups of prescribers who were aware of the likelihood that such gifts influenced practices and groups who were unaware.  Interestingly, the results were the same; even the ones who were fully aware of the likelihood that gift-giving influenced practices still displayed the same behavior of partiality.

Disclosure and transparency, IMO, is a floor, not a cureall.  It is the minimum we should demand, but there is little reasons to suspect that such transparency alone significantly mitigates the effects of COIs on human behavior.  For more on this, see Sheldon Krimsky's seminal work, particularly chapter 8.  Full disclosure is an important start, but that is all it is.  And if we're really serious about trying to deal with the tremendous problems posed by the extensive structural and institutional COIs, we must, IMO, get past the notion that simply disclosing the extent of these conflicts is sufficient to ameliorate the problem.  The roots of the problem are much, much deeper than that, as Tobbell's article suggests.

Turning to the invaluable Clinical Psychology and Psychiatry Blog, ClinPsyc has some of the most detailed and thorough analysis of the recent spate of stories over SSRIs and atypical antipsychotics.

On industry funding of CME, there's no one better than Daniel Carlat, who brings word of the topic being treated in JAMA and BMJ, and the Oregon Academy of Family Physicians's decision to proscribe industry funding for medical education.

However, BMJ, at least, may not necessarily be in the best position to evaluate the implications of such funding, as Audrey Blumsohn documents in his continuing series of posts evaluating the breakdown of industry advertising in the journal (and FWIW, I seriously doubt the results would be much different in any of the other major biomedical journals).

Finally, Howard Brody notes an excellent analogy to the problem of COIs, supplied by Peter Mansfield of Healthy Skepticism:

Peter Mansfield: I think the situation that we are in now is similar to in the 1840s, doctors didn't believe that we needed to wash our hands before we did surgery. We couldn't believe that we could be infected by something invisible that could cause harm to our patient. Bias that comes from promotion is like a bacteria in that it is invisible, and at the moment, if you suggest to a doctor that they could have become biased, many of us will take that as a personal insult, in the same way that doctors in the 1840s felt insulted by the suggestion that they could be carrying infections.

This great quote has it all--how a point of view can absolutely permeate the medical culture, and yet still be dead wrong; how physicians can unwittingly be a source of harm to their patients; and how the first natural, human reaction of physicians, when this is pointed out, is to be angry and insulted.

One more finally: for an alternative point of view on most of these matters, check out DrugWonks.