Disclosing Our (or Our Institution’s) Potential Biases. Or Not.

You may also like...

2 Responses

  1. Kathleen,

    Great post, as usual. Two points:

    First, I applaud your focus on institutional COIs. My group has done some work reviewing COI policies, and even where they do exist, they almost universally address COIs faced by individual faculty, fellows, investigators, etc. This is misguided for a number of reasons, not least of which is that, as you suggest, many of the most significant conflicts occur at the administrative/institutional level. The fracas at VCU in the last year is a perfect example of this; there was no question that had an individual facultyperson submitted a proposal for the arrangement, it would never ever have been permitted.

    VCU itself has a long relationship with the tobacco industry, for better or worse, and it is difficult to see how that relationship did not color the decision of whether to accept money with all sorts of restrictions on the use of the data that have long since been proscribed at most universities.

    Second, and related, while I applaud the efforts to increase disclosure, it is a source of frustration with me that so much of the discourse on COIs focuses on disclosure, as we have excellent evidence suggesting that disclosure itself may do little to mediate the behavior of partiality that we hope to avoid in context of COIs. There is some social psych evidence that it may make matters worse inasmuch as it signals to the audience that deep enmeshment with commercial interests is unproblematic so long as everyone knows about it.

    This is not to say that disclosure isn’t worth it; of course it is, but I admit to irritation that even while we have good evidence that disclosure is merely a ceiling, a start, and that the only surefire way of preventing the problems some (including myself) see WRT to COIs is to preclude the relationships that give rise to the conflicts.

    Of course, given that our systems of health care delivery and finance, as well as research, depends on the existence of those relationships, the problem becomes a difficult one indeed once one moves past the idea that disclosure is the answer (I am not suggesting you believe this, only expressing my frustration that this is typically regarded as “the” solution rather than simply a small beginning).

  2. Please forgive the grammatical errors in the above comment. Must learn to make better use of “preview” function . . .