Comments and Discussion (on Fees, gifts, and steak dinners from drug companies to doctors; conflicts of interest, but doctors still don't get it!)

Last updated 14 March 2009


From: Steven Bagley

4 October 2008


I looked at your essay about the problems of doctors and drug companies. I agree there's a problem, but mostly disagree with what you wrote. (I disagree with most of what everyone's written on this topic.) Here's why.


The drug companies are clearly out to influence prescribing behavior, and I think the balance of the evidence suggests that they are able to do so. In some cases, they are increasing their own profits. In other cases, they are increasing the tendency of physicians to prescribe anything -- maybe not even their own drug -- in place of nonpharmacologic treatments. I don't know if there are any studies to show that this has produced great harm, but common sense suggests that overprescription risks extra side effects, and probably wastes money, if there are effective and cheaper alternatives.


But so what? There's ample evidence from cognitive and social psychology that all humans are subject to a variety of influences and biases that function outside of conscious awareness. In a very small number of cases, we (humans) have set up formal systems to reduce potential bias. One example you cite is for journalists. Another is that most media organization set their newsgathering departments separate from their marketing departments. But significant biases and incentives remain. Essentially all "professionals" (tax advisers, lawyers, doctors in private practice) have an incentive to keep clients in the consulting relationship as long as possible (assuming their are being paid piecemeal, not flat-rate). I think particularly for psychiatrists in private practice who provide long-term psychotherapy that the incentive to keep the gravy train functioning would be a strong one. Also, there's good evidence that medical costs are influenced by supply factors -- the number of physicians and hospital beds. I'm sure that there are many more surgeries than are needed (whatever that means) because surgeons are paid by the operation. I'm not saying that any of this is conscious, willed, or intended, only that we know that incentives affect behavior, even when people are intentionally trying (if they are) to ignore or even counteract them.


The proponents of individual choice ignore the realities of human motivation, and the proponents of restrictions and putatively ethical behavior seem sanctimonious (and also ignorant of those realities). Nearly all of the arguments on both sides miss this larger picture. Nowhere have I seen any of this laid out in a rational framework that would allow people to compare the effects of various kinds of distorting effects. Should we ban drug dinners? Should we ban all drug advertising? Should we ban all advertising? One could make good claims for all of those, but given the current liberties granted corporations under the guise of protection of commercial speech, it would be difficult to implement in the US, at least. I'm not even sure that corporations are the root of all the evil, as those who work for government aren't free from bias, just free from bias to maximize shareholder value. I don't see how you could establish any means to make the current payments to physicians be illegal. What if I'm a doctor and my spouse works for a drug company -- isn't that likely to produce bias, too?


So: yes, there are problems of bias. But I don't see anybody confronting the problems in an intelligent manner that both respects the realities of how humans process information, and the decisions our society has collectively made about rights and responsibilities.


=================================================


Reply from: David Sirkin

7 October 2008


The point about prescribing a drug vs. no drug is very apt and something I hadn't thought of.  I disagree with your more-or-less dismissal of the concept of professional ethics as applied to receiving gifts from parties who stand to gain financially by biasing your work for your clients/patients.  By lumping the receiving of such gifts with other types of conflict of interest and with exposure to advertising, you are making an argument similar to Treisman's.  I think there is something different about receiving such gifts that makes it wise and correct for all professions to have the concept that it is unethical.  That it is difficult to eliminate all types of conflict of interest is not for me an argument that we should not censure or even outlaw this particular type.  And there is a clear difference between seeing an advertisement from a drug company and receiving a dinner or other gift from one; a prescriber is in no way beholden, and can not be seen as being beholden, to a company from merely seeing or reading its advertising ephemera. Accepting a gift on the other hand suggests that the prescriber is being bought; it smells of corruption. Saying no to a gift demonstrates integrity, even if it does not guarantee that the prescriber is immune to all biasing influences (for example the articles he reads in the free magazines/"journals" whose content may be much more controlled/strongly influenced by the drug companies than most readers suspect or realize...).

=================================================



From: Steven Bagley

8 October 2008


I don't completely dismiss ethical prohibitions. My point is that all the discussions either reject that there's a problem (which seems ignorant), or focus on ethics and restrictions, without looking to see if (1) those would work, (2) the effects they would counteract are dwarfed by larger effects. I think having a spouse work for a drug company would have more motivating power than a drug dinner or a gift. However, I can't prove it, and it would nice for some more principled data collection and analysis to figure this all out. I think that the ethical prohibitions can also be used as a rhetorical device ("see, we have ethics") to defuse other important work, the net effect perhaps being the opposite of that intended. I didn't read Treisman's article so if my points were similar it's a coincidence. I'm not trying to lump conflicts or biases together to water them down but so that they can be compared and the larger ones controlled for. I'm not convinced that ethical prohibitions will do that. I do think they would advance the case of removing appearances of conflict of interest, which might be worthwhile, but it's hard to be sure.


=================================================


Reply from: David Sirkin

4 December 2008; updated 14 March 2009


I am basically a pragmatist, but I think that some ethical principles are worth respecting even if their practical effect size cannot be proven. Every doctor I have talked to who goes to these drug dinners assures me that they do not let the free meal influence them in their prescribing. Let's leave aside the fact that the drug companies have access to prescribing data, and that therefore we can presume from the fact that they continue to offer the dinners that the dinners are paying off. If a judge accepted a dinner at a fancy restaurant from a party in a case on which he had to pass judgement, and he assured everyone that the free dinner would not influence him, would we say it was ok? The ethical principle that one should not accept gifts from a party who has an interest in a decision that one is supposed to make impartially is a principle the adherence to which, across all professions, makes society run more fairly and smoothly and efficiently, both because irrational decisions are reduced, and because malignant suspicions are also reduced. As prescribing physicians, we are indeed like judges, or government purchasing agents, with power and responsibility. We have some power over our patients to decide if they will receive a medication or not, and if so, which one. We have a responsibility to our patients to make decisions regarding treatment that are rational and in their best interest (and cost-effectiveness might be in their best interest sometimes too, or, if not, then in the interest of another payor: a private insurer or the government, to whom we also have some responsibility).