Sunday, December 6, 2009

RE: Appointment of Dr. Bernard Prigent to the Canadian Institutes of Health Research (CIHR) Governing Council (GC)

(See previous entry) The basic assumption in Steven Lewis' argument is that because Dr. Pringent works for a pharmaceutical company and may have an obligation to protecting shareholder interests, he will not be able to play an unbiased or in anyway valuable role on the Governing Council (GC).

By using such language as, "Seemingly countless systematic transgressions of pharma against scientific integrity and honest marketing", Lewis appears to imply that no representative of “evil-pharma” deserves a seat on the Council.

While making no apologies for pharma's sales practices, in light of the recent "climategate" scandal and the countless examples of systematic transgressions of researchers against scientific integrity, one could easily make a similar case for not having any scientists on the GC. I am not sure that this would be in line with what Lewis is trying to achieve.

Furthermore, as pointed out by David Allison, Professor of Biostatistics at University of Birmingham, Alabama, in a letter to Science [1] and in an article in the International Journal of Obesity [2], it appears that scientists are not in any way "immune" to taking a biased view of scientific data. Indeed, one may argue that industry driven bias may be far less perfidious (because the cards and intentions are on the table for everyone to read) than the many biases that are far more difficult to recognize, like the ideological (anti-industry) "white-hat” bias, self-aggrandization bias, or simply the all-too-prevalent “love-for-their-favorite-hypothesis” bias.

In fact, if we accept Lewis' argument against having representative of "evil-pharma" at the table, we may as well exclude all industries (certainly no food industry, no energy companies, no banking, no private health care, no information technology, no medical devices, in fact, no representation from any sector that is primarily driven by profit and could potentially benefit from strategic decisions made by the GC).

In contrast to what Lewis implies, I would readily assume that everyone on the GC likely represents a constituency or interest group and brings a bias to the table even if their views are only coloured by their personal perspectives, ideologies, views, or beliefs. Indeed, I have no doubt that were I to be offered a seat at the GC, in my role as Scientific Director of the Canadian Obesity Network, I would do my best to promote more investment in obesity research and vehemently oppose any attempt of other GC members to cut funding for obesity because of their personal beliefs that obesity is simply a "self-inflicted" matter of choice and does not deserve public funding to address (incidentally, it turns out that individual choice has surprisingly little to do with the obesity epidemic after all [3]).

Nor does the argument that a seat on the GC would provide an unfair advantage to Prigent, and thus Pfizer, over its competitors fly. The same could easily be said for any researcher who sits on the GC. Indeed, I have sat on numerous scientific review panels that have provided me with (unfair?) insight into the projects of my (scientific) "competitors" and I have often had the opportunity to steer committees towards funding projects that I thought and felt were more important and away from others that I believed (for whatever reasons) were perhaps less deserving of funding. The notion that scientists on committees are not in someway representing their own views, areas of interest, and ideologies is naive and unfounded. Nor do I expect representatives from research institutes, foundations, patient-advocacy groups or any other constituency to be fair and unbiased towards all possible decisions of the GC. In fact, it is this very diversity of opinions and interests that makes for a strong and effective Council.

I believe that having an industry representative at the table provides an important voice in the discussion, a chance to bring in arguments and views that may be important to consider and insights that may not be immediately obvious to GC members with little insight or experience in the private sector. The "holier-than-thou" argument presented by Lewis is simply discriminatory and unfair in singling out one member on arguments that may in principle apply to all members of the GC
Finally, simply in light of a democratic process one could even make the argument that the pharma industry, as a major tax payer and player in the health sector, very much has a right to sit at this table - after all a significant chunk of public funding that the GC will decide over comes from the taxes that this very industry pays into the government coffers - no taxation without representation?

Rather than demonizing or declaring outrage about the appointment of representatives of certain constituencies to the GC, the simplest way to deal with any actual or perceived conflicts of interest is to have these openly declared (my guess is that it is likely far easier for Pringent to identify and declare his obvious areas of conflict than for most other members on the Council, who I am sure all have interests and biases of their own). It is easy enough (and hopefully common practice during GC sessions) to expect Pringent (and others) to excuse themselves from voting or even from being in the room when matters are discussed where such a conflict occurs.

Excluding anyone with a putative personal, professional, ideological or political conflict of interest from the GC will most likely result in not having a council at all.

Arya M. Sharma, MD/PhD, FRCPC
Professor and Chair of Obesity Research and Management
University of Alberta

1. Allison DB. The antidote to bias in research. Science. 2009;326:522-3

2. Cope MB, Allison DB. White hat bias: examples of its presence in obesity research and a call for renewed commitment to faithfulness in research reporting. Int J Obes. 2009. [Epub ahead of print]

3. http://www.drsharma.ca/obesity-lifestyle-choice-or-lifestyle-chance.html

No comments: