Friday, November 13, 2009

COI: My Reply to Stell

In the last post I turned the microphone over to my esteemed colleague, philosopher and bioethicist Lance Stell. He apparently accepted the definition of 'conflict of interest' attributed to him in the editorial by Michael Weber that was the subject of the post immediately preceding. He then proceeded to offer a further defense of that definition of COI, and cited as his inspiration for that definition the work of a Yale law professor, John Langbein.

I have now spent a pleasant couple of hours reading the 63-page article by Langbein that Prof. Stell refers us to. My conclusion is that it makes excellent sense, is grounded in what appears to be a thoughtful historical analysis, and I have no problem at all saying that he makes a good case for the conclusions that he argues for. (Whether he is actually right or not hinges on matters of trust law about which I am wholly ignorant, so I must withhold judgment. All I can address is the internal logic of his arguments.)

It also seems clear to me that Prof. Langbein's article has nothing at all to do with the debate over physicians and COI in relation to Pharma--or, if it does, that it supports conclusions opposite to those drawn by Stell and ACRE.

What Prof. Langbein sets out to show is that a principle of the law of trusts, the sole interest principle, is outmoded and should be replaced. According to this principle, if you are a trustee, you are assumed to be disloyal to the person you are the fiduciary for, if you enter into any deals that benefit you in any way--even if the result for the trust is a net benefit from the deal.

Prof. L. notes that the law has already been modified to grant him a good deal of what he wants--because the law now recognizes a bunch of exceptions to the sole-interest principle. That is, the law recognizes now that there are many ways in which it might be in the mutual interests of both trustee and fiduciary to carry out certain business arrangements. All Prof. L asks for is that these exceptions be made the new default principle.

Thus far it is not at all clear why Prof. Stell should imagine that this article is at all informative about COI in medicine/Pharma. The key line he quotes is, "The very term 'conflict' is an epithet that prejudices our understanding..." So I read Prof. L's article with the question in mind--does what he say here undermine or support the view of COI that I , as a card-carrying "pharmascold," propose is best? My conclusion is that Prof. L. says a lot more in support of my preferred view of COI.

Prof. L. wishes to allow transactions in which a COI exists but in which the interests of all parties are nevertheless best served by allowing what the sole-interest rule would prohibit. This is completely in keeping with most COI policies in academic medical centers. An example commonly discussed is the physician who owns a business interest in some new discovery. The ordinary rules would preclude that doc from running a research trial to test that discovery. But the discovery may be such that the only person on earth with sufficient knowledge of the new discovery, to be able competently to carry out the necessary research, is the doc who discovered it in the first place. The policies then commonly state that the doc may serve as the principal investigator for the study, but that he must accept additional oversight to be sure that he is not abusing his privilege as PI. So far nothing here to disagree with Prof. L's take on the matter. Prof. L at several places in his article calls for further regulation to prevent possible abuses.

One point that Prof. L. makes, that would at first glance seem to recommend his views to Prof. Stell and to ACRE, is the ubiquity of COI. You cannot get rid of COI because it's everywhere you look, he says. But then see where he takes this point. First, he distinguishes unavoidable and avoidable COI and agrees that it may generally be sound policy to prohibit the avoidable COI even if you end up having to tolerate the unavoidable. We pharmascolds, of course, argue that most of the COI we object to is avoidable. Docs can buy their own dinners and pay for their own CME. Academic docs who want to promote new discoveries can communicate freely with industry, but need not become paid consultants and speakers on behalf of industry--or, if they do want to do this, they can work part-time as salaried industry employees, and only part-time at the academic medical center. Famous academic "thought leaders" can actually write their own articles--or, if they want hired ghostwriters to do it for them, admit that and make the real writer the first author.

Further, while demanding changes in the sole-interest principle, Prof. L. nevertheless acknowledges the core reasons why the principle exists. For example (p. 953), he criticizes the principle, "What the sole interest rule does in such a case is to identify some conceivable but conjectural evil and then conclusively presume that this farfetched plot actually transpired, by refusing to let the putative evildoer prove that no such thing happened." Sounds pretty uncompromising. But he follows quickly with, "I do not mean to say that the trust tradition is wrong in being alert to the possibility of abuse [in the sorts of cases being referred to]." Now, this is precisely where we pharmascolds and ACRE/Stell seem to part company. Because they think that COI is sometimes applied in too sweeping a fashion, or because they think COI is ubiquitous and unavoidable, they do conclude that it is all a lot of hogwash, and that Pharma money and gifts and inducements create no ethical problem of any sort whatsoever, except in the fevered imagination of people like me. But if they conclude that, I would say, they cannot turn to Langbein for support.

Another point at which Prof. L. seems to offer much more support for my own position than for Stell and ACRE is where we call upon heightened standards of professionalism in medicine to get docs to refuse to accept Pharma freebies and bribes/handouts. Prof. L. includes in his historic survey (947-48) that one of the changes that has occurred since the sole interest rule made sense, is that the business of being a trustee has become much more professionalized. He mentions as his key example enhanced recordkeeping, driven by the "process values" that charaterize "good trust administration." With these improved records, if the trustee actually cheated the trust, you an always go back into the records and find out when and where this happened; and knowing this provides a strong incentive to the trustee not to abuse his power. Gosh--sounds to me as if Prof. L. is invoking professional standards as a defense against abuse and misconduct.

Final matter--I take Prof. L. to be making a claim that could be characterized as an account of successive pendulum swings. There was a time in history, he says, when the sole interest rule made good sense. Then a bunch of things changed, but the rule remained petrified. Nonetheless the pendulum has swung back toward the middle because more and more exceptions to the rule have been recognized. Prof. L. now calls for the final pendulum correction by replacing the rule with a more up-to-date principle.

Again we see why this analogy is superficially attractive to ACRE adherents. But let's ask--over the past several decades, which way has the Pharma pendulum actually swung? ACRE claims that at least in the last few years, the pendulum has swung wildly in the direction of discouraging financial exchanges between docs and Pharma, even when these exchanges might make patients better off (as they believe nearly always occurs). We pharmascolds on the other hand believe that the past 30-40 years have seen a huge pendulum swing in the direction of normalizing and accepting all sorts of blatant abuses and unprofessional behavior on the part of docs, and the reforms of the last few years have only just begun to edge the pendulum a few notches back toward the center-- though we still have an incredibly long way to go. I leave it to the reader to fill in your preferred account of which "pendulum" narrative seems more plausible to you.

So for all these reasons I believe that a fair-minded reading of Prof. Langbein's paper, far from undermining my own preferred definition of COI, would actually support it. As to what Prof. Stell said in the post that I reprinted from his e-mail to me, the only point to which I would call attention is his remark, "Some commentators quote GB Shaw as holy scripture [presumably when he said that it makes no policy sense to pay a surgeon or cutting off your leg unless you want surgeons to cut off a lot of legs]. And if so, buy into his socialist biases, whether self-consciously or not." C'mon, Lance--you cannot mean that in order to decide that fee-for-service payments to docs constitute a form of COI that incentivizes overtreatment, you have to be a socialist? Whatever those ACRE people were smoking at their July meeting, you should never have inhaled.

Langbein JH. Questioning the trust law duty of loyalty: sole interest or best interest? Yale Law Journal 114:929-990, 2005.

No comments: