Share |





The Hidden Costs of Conflicts of Interest Regulation

Richard A. Epstein
Medical Progress Today
October 5, 2006

The problem of conflict of interest is pervasive in all cooperative human endeavors. No amount of moralizing can deny that individuals, especially in high–stakes market environments, face conflicts of interest that can turn their heads from the proper roles. That simple and inescapable truth, however, is only the prelude to the more critical question, which is what steps should be taken to either eliminate or control the conflicts that do emerge in medical research, as everywhere else in life.

The initial burst of indignation often leads to the wrong policy prescription: ban conflicts whenever and wherever they arise. But that course of action, if taken literally, bans not only dangerous conflicts but also beneficial cooperation. Any sensible evaluation of the problem requires that we set off the loss against the benefit. But all too often I fear that the policies in question are designed to ward off adverse publicity of scandal, which never takes into account all the successful interactions that take place in the shadow of the conflicts that are managed right now. To my mind, the proper approach is always one of watchful management, but not one of ingrained suspicion that seeks to root out all conflicts at, it seems, almost any price.

I am sorry to report that it seems that I am much on the minority on this critical question. An influential article that Troyen Brennan and his colleagues published in JAMA in January, 2006 rejects any involvement of pharmaceutical houses in the internal affairs of academic medicine on the grounds that these necessarily undermine the undivided loyalty that hospitals and their physicians owe their patients.

In order to forestall these baleful influences the article proposed wide ranging new restrictions on drug companies' marketing and educational efforts towards physicians. Since then, several prominent academic medical centers—Yale, Penn, and effective October 1, 2006, Stanford—have implemented new regulations that prohibit, among other things, physicians from receiving any small trinkets from drug companies or free samples for their patients. In addition, the new regulations prevent drug companies from offering any support, such as free meals, to run faculty seminars. Finally, there is a strong move to keep any physicians with industry ties from sitting on committees that make decisions on which drugs the hospital should use in their patient services.

All of these policies are directed at legitimate concerns that are duly taken into account, and collateral costs, that are assiduously ignored. Thus one objection to the use of free samples, which could be given to poor patients, is that they constitute "a powerful inducement for physicians and patients to rely on medications that are expensive but not more effective." The phrasing of this objection treats the worst case outcome as if it were the only possible outcome. There is no probability assessment as to frequency or severity. Nor is there any effort to envision other scenarios, as when the high–priced drug is in fact more effective than the existing treatment in ways that more than justify the cost differential.

Think, now, of the consequences of the current ban. It will reduce the rate at which this product is introduced into the market. It makes drugs more expensive, which in turn denies the benefits of new treatments to some less wealthy individuals. The slower rate of diffusion in the market place will therefore reduce both the utilization of, and the rate of return on, innovative new drugs, which in turn will reduce the resources committed to their development in the first place. Medical administrators who look only to one margin, and ignore all the others are bound to make serious mistakes. Yet it is highly unlikely that patients will be able to coordinate their opposition to proposals that will, in a substantial set of cases, work to their detriment. Fear of scandal is not an appropriate basis for medical policy.

Similar criticisms can be lodged against other initiatives that generate unfortunate unintended consequences in their effort to root out all conflicts of interest. The new proposals make it impossible for drug companies to sponsor research seminars at universities. Yet there is no showing of which I am aware that indicates that they have sought to pressure the participants in these studies to skew their selection of speakers or topics. And the financial shortfall that arises has to be made up from somewhere else, at costs to other parts of the medical curriculum. Where? It is never said.

Last but not least, I think that it is most unwise to adopt any policy that keeps physicians with drug interest ties off internal hospital purchasing and utilization committees. To be sure, it would be foolish to constitute such committees consisting solely of this type of physician. And it would be equally foolish to allow them to serve without disclosure of their outside connections, or to make decisions in which their outside companies have a direct financial interest.

Once again, however, a fuller analysis reveals costs both ways. Many physicians with drug company contacts are the ablest within their profession and the loss of their good advice and knowledge counts as a real cost. In addition, one has to worry about the risks of stacking committees in the other direction, with physicians who are so wedded to the cost savings of generic drugs that they miss the therapeutic advances that new products can provide. The unfortunate bias in favor of the status quo could in the long run prove harmful to pharmaceutical innovation. There is an instructive parallel here to the current corporate fetish to insist that every board have its large contingent of independent directors to avoid conflict of interest, even if those people lack the commitment and knowledge to help the company flourish. In both contexts a mix of expertise from all backgrounds may well prove ideal, and no blanket prohibition can hope to approach that outcome.

All in all, each and every one of these supposed antidotes to the conflict of interest problem has real costs, which their proponents blithely ignore. A far better strategy in my view is to impose harsh sanctions on any abuse of trust that does arise out of the course of this relationship, which should deter any untoward efforts to obtain undue influence. With so much at stake, and so little hard evidence of real abuse under the current conflict of interest regimes, I think that it is little short of incredible that American academic institutions will take it on themselves to block a set of interactions from which both sides can benefit. One always has to monitor conflicts for abuse, but in this instance, the right war cry is not "too little, too late." It is, regrettably, "too much, too soon."

Richard A. Epstein is the James Parker Hall Distinguished Service Professor of Law at the University of Chicago, and the Peter and Kirsten Bedford Senior Fellow at the Hoover Institution. He has also consulted over the years for both Pfizer and the Pharmaceutical Research and Manufacturers of America (PhRMA).

home   spotlight   commentary   research   events   news   about   contact   links   archives
Copyright Manhattan Institute for Policy Research
52 Vanderbilt Avenue
New York, NY 10017
(212) 599-7000