Conflict-of-Interest.net

Two recent stories on COI in the context of US government health research advisory committees caught my attention.

1) FDA to Broaden Disclosure on Advisers’ Conflicts of Interest
- In a move to redress some of the major concerns about COI on the part of researchers who serve on Food and Drug Administration (FDA) advisory committees (e.g., researchers with pharma or medical device industry ties), while accepting that some COI may be inevitable (e.g., because there are only a few expert researchers and they all have financial interests) the FDA is proposing new guidelines dealing with the public declaration of financial interests.

2) Senator pushes CDC to disclose experts’ conflicts
- in a similar vein, attention is being brought to bear on the Centers for Disease Control (CDC) and their handling of COI on the part of expert researchers giving advice to the CDC committees.

In both of these cases, the issue is about ensuring that the topmost government health agencies — which seek advice from the academic research community in setting policy or making funding decisions — do so in a manner that recognise the interests of the parties involved, and put in place mechanisms to manage any conflicts.

For any readers in the Montreal are, my COI team and I will be presenting a number of papers (in French) at a workshop — Recherches au pluriel et conflits d’intérêts — organised in the context of ACFAS, on May 12. The abstracts are listed below. (P.S. here’s a radio interview I did with French language Radio Canada)

Bryn WILLIAMS-JONES, Université de Montréal
Défis contemporains dans la gestion des conflits d’intérêts dans la recherche universitaire
Le problème des conflits d’intérêts (CI) dans la recherche universitaire a fait l’objet de beaucoup d’attention ces dernières années et pour cause. Les CI peuvent menacer l’objectivité et la crédibilité du processus et des résultats de la recherche et conduire à une perte de confiance de la part public à l’égard de la recherche scientifique. Les CI les plus évidents se retrouvent dans les situations où les chercheurs ont des intérêts financiers, notamment dans le contexte de la recherche biomédicale, provenant du financement privé de la recherche. Mais il existe des CI autres que ceux reliés aux intérêts financiers et également dans la recherche en sciences humaines et sociales! De plus, les CI ne devraient pas être considérés comme étant toujours intrinsèquement contraire à l’éthique. Parfois, le fonctionnement institutionnel engendre des CI et les rend même inévitables. Ce qui importe sur le plan éthique est que les individus et les institutions possèdent les outils nécessaires pour traiter de façon appropriée les CI lorsqu’ils surviennent. Cette présentation donnera un aperçu de ma vision des principaux enjeux reliés aux CI dans la recherche universitaire, notamment ceux qui persistent et qui, de façon générale, sont moins bien examinés dans la littérature.

Ghislaine MATHIEU, Université de Montréal
Un évaluation des politiques de conflits d’intérêts dans les universités canadiennes qui abritent des facultés de médecine
Afin de préserver la confiance du public, tant les chercheurs que les universités elles-mêmes doivent s’acquitter de leurs obligations professionnelles et rendre compte de leurs actes. Le classement des 149 écoles de médecine des universités et des collèges américains, que l’American Medical Student Association (AMSA) publie à chaque année, apparaît comme un modèle des exigences attendues des institutions de recherche en santé en matière de CI. Une analyse de ce classement soulève toutefois des questions. D’abord, comment les universités canadiennes qui abritent des facultés de médecine se positionneraient suivant ce type de classement? Est-ce que ce type d’évaluation mériteraient d’être amélioré, voire justifieraient d’améliorer la qualité des politiques de CI des universités canadiennes? Notre présentation veut répondre a ces questions, en démontrant que, comme les institutions américaines, presque toutes les politiques de CI des universités canadiennes ont tendance à se concentrer presque exclusivement sur les CI de type financier en ignorant d’autres CI, tels que la loyauté, le prestige ou l’avancement scientifique. Ce n’est donc pas surprenant que le classement de l’AMSA a pour faiblesse de limiter l’évaluation des institutions aux mesures mises en place pour garantir la divulgation d’un CI exclusivement financier et en relation avec les entreprises pharmaceutiques.

Camille ASSÉMAT, Université de Montréal
Conflits d’intérêts et éthique de publication : revue systématique des lignes directrices des journaux médicaux
Dans le domaine de la recherche scientifique, il y a des conflits d’intérêts (CI) qui peuvent biaiser le jugement impartial et désintéressé des professionnels de même que le processus de recherche. Notamment, les relations qu’entretiennent les cliniciens et les chercheurs avec l’industrie font l’objet de préoccupations majeures en particulier au niveau de la qualité et de l’objectivité des publications. De nombreuses revues à fort impact ont progressivement demandé aux auteurs de soumettre, en plus de leurs résultats de recherche, une déclaration explicite de CI. En 2007, le Comité international des rédacteurs de revues médicales (CIRRM) a émit des directives davantage détaillées quant aux CI liés aux engagements de l’auteur, au financement des projets de recherche, ainsi qu’aux engagements des examinateurs et des rédacteurs de journaux. Il n’est pas clair si les principes éthiques émis par le CIRRM sont rendus suffisamment compréhensibles et accessibles aux chercheurs, ni combien de revues médicales internationales ont effectivement mis en place des politiques de CI. Inspirée par le modèle du CIRRM, nous présentons une analyse éthique comparative des politiques en matière de CI des vingt plus grands journaux biomédicaux permettant d’évaluer le niveau de conformité de ces politiques aux standards attendus.

Christiane ROCHON, Université de Montréal
Les conflits d’intérêts et les conflits de rôles professionnels
Le concept de conflit d’intérêt (CI) est parfois restreint aux conflits d’ordre financiers ou alors il est abordé de façon très large en englobant plusieurs autres concepts, dont le conflit de rôle. Quelque soit le type de conflits, la crainte est qu’ils influencent le jugement des professionnels en leur faisant perdre leur objectivité. Pourtant, les études en psychologie on démontré que de façon générale les jugements sont rarement objectifs et qu’ils sont facilement motivés par des intérêts personnels. Quelle que soit l’ampleur ou la nature de l’intérêt personnel le jugement s’en trouve biaisé et souvent de façon inconsciente. Devant un tel constat, il y a lieu de réfléchir sur les mécanismes qui sont généralement mis en place pour contrer les conflits d’intérêts et se demander s’il est possible, voire même utile, de chercher à tous les éliminer. Les conflits d’intérêts ne sont pas tous de même nature ni de même niveau. Il en ressort donc qu’une seule politique ou une seule solution ne peut à elle seule couvrir tous les conflits. Certains doivent simplement être gérés. La présentation traitera de la diversité des conflits d’intérêts et des différentes questions qu’ils suscitent.

One of my current research interests – actually an ongoing project – is the study of institutional policies dealing with academic integrity, and in particular, policies on COI. So it was with great interest that I read the following stories about the progress being made in some US universities & medical schools: Many medical schools update ethics policies.

The University of Minnesota is not alone in tackling conflict-of-interest issues. Allan Coukell, director of the Pew Prescription Project, which tracks such policies in medicine, has said that in the past two years, one-third of university medical schools have either approved new policies or are in the process of doing so.

As in this case (and others – Conflict-of-Interest Policies: A Detailed Look) attention is still to a large extent on the COI posed by relations between professors/clinicians and the pharmaceutical industry. While this is an important issue, and probably a legitimate focus, institutional COI policies should not only deal with the issues arising from interaction with the pharma industry. They should also attend to the broad range of interests at stake in academia — including those in the context of research, teaching, consulting, etc. — and with the range of public and private sector actors (government, NGOs, companies) that interact with universities.

The potential for COI related to the use of academic letterhead may seem pretty bizarre, at first glance. Yet as this story about a professor at the University of Maryland (UM professor reprimanded for apparent conflict of interest) helps demonstrate, the inappropriate use of letterhead is not banal.

When I write a letter, say recommending a graduate student for a scholarship, I am doing so in my capacity as a university professor — writing references for my students is part of my academic responsibility — and so using official university letterhead is appropriate. Letterhead is an visual means of formally demonstrating my professional affiliation…and it brings with it the imprimatur (the formal stamp of approval) of the university. That means I should probably be careful how I use this formal affiliation.

So what if I use letterhead when I am consulting for a 3rd party, such as a corporation, or a labour union as in the UM case? Well, it depends on 1) what my role or expertise is (professor?) that is being contracted, and 2) the norms of my institution about doing consulting alongside my academic responsibilities. In the UM case, Mr. Fred Feinstein, an adjunct professor at the School of Public Policy, wrote a letter in the context of a labour dispute, and provided an expert judgment based on his capacity as a lawyer (he had also been a paid consultant for one of the parties involved in the dispute).

“Mr. Feinstein violated university procedures by improperly using university letterhead in the course of his outside work,” Donald F. Kettl, dean of the School of Public Policy, said in a statement. “This activity was wholly unrelated to his work at the University of Maryland, which has no involvement or stake in this outside matter. He should not have written the material on university letterhead nor invoked his title as a university employee. In addition, he should have disclosed the payment he received from one of the parties in the issue on which he commented.”

The conflict arises because Mr Feinstein implied, through the use of institutional letterhead, that he was commenting in his capacity as a professor, and thus also in a sense involving the institution. Now this may seem like a bit of a stretch. Surely academic liberty allows professors to make all sorts of comments, sign petitions, be active in political debates, etc., and use the public credibility given to them by being professors? Maybe, but I think it depends on the context, whether the role (as professor, as academic) is reasonably related to the activity in question, and how the institution is being involved (directly or indirectly).

With the privilege of academic liberty that is provided by my institution, also comes the responsibility to be professional in my behaviour. I can be critical of my own and other institutions, but I should also use good judgment, and think carefully about where my role as academic (professor, public thinker) ends and my personal space begins…

Here’s a story that caught my attention in the last few days, because its not about one of the “usual suspects” of drug companies, the need for policy in med schools or journals, etc. While financial interests are at the heart of the case, the mechanisms and harms are somewhat different and so worth thinking about…

Conflict of Interest and Textbooks

The issue here is about profs using their own textbooks in classes they teach. The COI is a financial one, because in choosing one’s own textbook, the prof receives royalties on the sale of the book. In the context of large Intro classes with hundreds of students, the royalties (e.g., 3-5% per text, on a $100 book, for a few hundred students) could be in the order of thousands of dollars. The concern then is that this not insubstantial interest could bias the judgment and choice of text for the class. There’s also the indirect interest of having students read the prof’s own textbook (good for the ego!), and the fact that this may stymie freedom on the part of students to critique or critically engage with the text, undermining another important element of the university learning environment. So the challenge, like with many COI, is twofold. 1) Determine the extent and magnitude of harm, and 2) Decide what to do about the situation.

1. Potential Harm: bias in choice of text (maybe not the best), loss of critical judgment on the part of the class. Clearly this will vary based on factors such as the size of class (a large Intro class vs a small advanced undergrad or grad class), and also maybe the subject – we can imagine that in some areas there simply aren’t many texts to choose from, or the contrary, that there’s so much choice that one text is as good as another. In both cases, this factors may reduce our perception that there’s a real bias at play.
2. What to do? Eliminating such COI by banning a prof’s right to choose their own teaching material seems like an extreme measure, and also an undue restriction on academic freedom. After all, if I’m hired to teach a course, surely there should at least be the presumption of confidence in my ability to choose appropriate material. And an extreme measure such as prohibition also undermines the possible positive impact of disclosure of the COI to the class, debate about the issue (a great learning opportunity!), and the development of innovative solutions to separate the interest such as deciding (together with the class) that royalties from sales of the text will go to fund research, for extra resources or opportunities for the class (a trip, a party, etc.), or to a charity.

This story brings to our attention the fact that COI can occur outside the bioscience lab, in the classroom, and thus should also be a concern for scholars in the humanities and social sciences. It also leads us to think about appropriate responses to COI, and in particular, question the scope and magnitude of harms and possible appropriate (and innovative) mechanisms to deal with the conflict.

Thanks to Chris MacDonald of the BusinessEthicsBlog for discussions on this subject and thinking through possible responses…

P.S. Apr 15, 2010: See this story about some universities prohibiting profs from collecting royalties – Colleges move to stop profs from cashing in on textbooks.