Institutional Conflicts of Interest Policies: Some Assembly Required

The National Pharmaceutical Conflict of Interest Survey Factory (they're not really called that, it just seems that way), a.k.a. Eric Campbell and buddies, has published their latest study, this time on how many medical schools have adopted policies on institutional conflicts of interest (COI). Our national conflict-of-interest/professionalism guru, David Rothman of Columbia (Center on Medicine as a Profession), authored the accompanying editorial, making the points that far too few of the responding schools had institutional COI policies and predicting that government entities would probably step in to regulate the mess.

As I try to explain in HOOKED, institutional COI is a difficult problem to wrap one's head around. Individual conflict of interest (faculty member Dr. X owns stock in Pfizer and is PI on a big research study funded by Pfizer) is by contrast much easier to grab hold of. So for that reason alone it would not be surprising that more med schools would have individual COI policies compared to institutional COI. Nor is it surprising that many schools construe institutional COI to be a variant of individual COI--that is, do your president, trustees, and other highly placed officials individually own stock in companies, etc.?

When a problem is difficult to get a handle on, it helps to enumerate some concrete examples. Unfortunately, Ehringhaus et al. (that's the Campbell gang) did not, so far as I could notice, include any examples in their study report, so we are left to guess what sorts of instances they had in mind when they designed their survey questions. Rothman gives one key example, the 1999 Jesse Gelsinger case at Penn (subject dies in gene therapy experiment; university owns stock in the biotech company that would eventually market the therapy if it had worked). In HOOKED, I gave two other examples that are worth considering:

• The University of Toronto's reversal of its hiring of psychiatrist David Healy to run a research institute, after Healy gave a lecture viewed as unfriendly toward Prozac, when the University was hoping to land a major grant from Prozac's manufacturer, Eli Lilly
• UC-Berkeley's decision to virtually sell its plant biology department to Novartis, in exchange for first dibs on any patentable research produced by that department over a period of years

In asking whether a med school has an effective institutional COI policy, we'd have to ask what the policy would say, for example, about each one of these three arrangements (which hardly exhaust the possibilities but are at least a place to start). It seems clear that Ehringhaus et al's survey addressed cases like Gelsinger and Toronto; it's not as clear to me that the Berkeley case would be covered by one of their categories. (I'd assume that the money to run the department was classed as a grant and not as a "gift.")

Two points need to be made in summary:

• It is not clear to me what one can do about institutional COI when the forces working on the average medical school are aligned as they are today. Basically a med school dean is under tremendous pressure to bring in commercial sources of research support. NIH dollars are drying up as fast as subprime mortgages are imploding; and the going buzzword (I'll pay a reward to anyone who can explain to me what it means in plain English) is "translational research," which means that research results must be moved much more quickly "from bench to bedside," which seems to be a directive to med schools to jump into bed with industry even faster and deeper (I think that is what "translational research" means in plain English). Exactly what is a paper policy supposed to do in the face of that perfect COI storm?
• In HOOKED I cite the conclusion of some Canadian thinkers on this matter who no doubt have carefully pondered the Toronto example. They offer the conclusion that there is simply no such thing as an institutional COI policy so long as there is no body outside of and above the university to whom concerns can be referred or appealed. Exactly who at Berkeley was supposed to have decided that selling the plant biology department to a private company was an institutional COI, given that everyone from the President to the Dean on down was obviously salivating at the dollar signs? I have yet to see an institutional COI policy in the US that addresses the incredible difficulty of an institution policing itself in such matters, and that suggests how such an extra-institutional appeals board should be created. (For med schools it would presumably have to be a creature of the Association of American Medical Colleges (AAMC). For universities in general--who and where?) Without some enforcement mechanism with teeth, that resides outside of any individual university or med school, I will continue to regard any institutional COI policy as nice window dressing (a somewhat harsh overstatement but only slightly). Indirectly that backs up Rothman's perediction that the government is probably coming, whether we like it or not.

Ehringhaus SH, Weissman JS, Sears JL, Goold SD, Feibelmann S, Campbell EG. Responses of medical schools to institutional conflicts of interest. JAMA 299:665-71, Feb. 13, 2008.

Rothman DJ. Academic medical centers and financial conflicts of interest. JAMA 299:695-97, Feb. 13, 2008.