Elsewhere in this issue of the journal, David Resnik argues that published disclosures of financial conflicts of interest (COIs) are often incomplete because the disclosed funding sources may not be tracked back to their actual origins, usually in industry.1 It is not difficult to obfuscate the true origin of financial support. You can hide it, sanitize it by moving money through an accredited medical education company, or even launder the money through a newly created front organization whose credibility may be enhanced by a pretentious name and membership of prominent academics. Over the past decade, numerous major medical journals have been stung by the revelation that they had failed to elicit or report overt financial COIs. Reporters and other whistleblowers have raised the ire of the public about failures to disclose; both the Senate and House of Representatives have held hearings; and a Sunshine bill would set up a searchable Web site of all physicians who have received payments from industry worth $100 or more.2 If the bill passes, we probably will have more disclosures than we know what to do with.
Even when we have perfect information about sources of financial support, what does such disclosure tell us? When we listen to a lecture by an expert who has acknowledged that she is a paid speaker for a given company, how do we interpret her recommendation about the use of that company’s drugs (or, for that matter, the products of its competitors)? When we examine a clinical practice guideline compiled by a cadre of experts, most of whom are paid company speakers or consultants, how do we know whether their recommendations are unbiased? Should we apply such guidelines uncritically to the care of our patients? When such guidelines are being used as a quality indicator and form the basis for payment decisions, can we be sure that our national resources are being properly used? The point is that a financial COI raises a suspicion of bias, but it does not prove bias, and because of this ambiguity, the recipient of information from financially conflicted individuals has to be a mind reader in trying to assess motives. Is the information biased, and if so, in which direction, and what are the consequences? The problem with financial COIs is that you simply don’t know what to believe.
I am not suggesting that we eliminate disclosure but that we appreciate its limitations. Disclosure is a necessary aspect of dealing with financial conflicts, but it does not sufficiently address the issue of COI.3 Although the spotlight has been on the failure to disclose (or to adequately disclose) financial relationships with industry, the problem with COIs is not the lack of disclosure but the existence of the conflict itself. In fact, assuming that “sunlight is the best disinfectant”4 and making disclosure the standard fare may exacerbate the problem. First, some physicians are not embarrassed to have their financial arrangements aired widely; some even consider it a badge of honor to be paid handsomely by industry. And for some, the more “badges” the better. Second, disclosure allows them to maintain or even strengthen their conflicts and say or write whatever they wish. Economist James Surowiecki summed it up as follows: “It has become a truism on Wall Street that conflicts of interest are unavoidable. In fact, most of them only seem so, because avoiding them makes it harder to get rich. That’s why full disclosure is so popular: it requires no substantive change.”5
So, if the conflicts are the problem, not the lack of disclosure, what is the remedy? The answer is to eliminate financial conflicts whenever possible. Here we should remember that financial conflicts are optional: When approached by industry representatives, practicing physicians and physician–investigators can either accept or reject offers to be paid speakers, consultants, or the like. With a sufficient cadre of nonconflicted individuals, medical journal editors could readily find experts to write their review articles, FDA authorities could convene panels free of conflicted individuals, and professional societies would find it easier to appoint physicians to guideline committees without concern that money might influence the clinical recommendations.
Nevertheless, some conflicts are inevitable if academics are going to pursue industry-funded research that has the potential to benefit patients. Although some have recommended developing an independent federal agency to perform comparative studies of drugs,6 it is quite likely that the current approach of industry funding will prevail for some time. How should we think about COIs among such clinical trial researchers, even those who refuse to take honoraria or consultation fees? Are researchers who run or participate in industry-supported drug studies subject to bias? There is ample evidence that some industry- supported research that yields unfavorable results has never been published. Moreover, industry support of research does promote a tendency to produce favorable results for the sponsoring company. Investigators who regularly publish studies that do not support their funding company’s marketing aims are not likely to be supported for long. Unfortunately, many techniques exist to enhance the likelihood of a positive research outcome, including those involving the design, interpretation, and reporting of the work.7
How do we guard against bias in industry-supported clinical trials? First, by having nonconflicted expert panels assess all aspects of a study independently.8 Failing that, individual physicians must do the hard work of study analysis themselves. They must ask whether treatment was randomized and whether the investigators were properly blinded, whether study groups were similar at the start, whether all subjects were accounted for, whether groups were treated equally (except, of course, for the study drug), whether follow-up time was sufficient, and whether real end points were assessed. It is unfortunate that there is a need to guard against bias, but optimal patient care and appropriate use of resources demand that we do so.
A vocal minority considers efforts to reduce financial COIs as overkill. They ridicule the widespread concern over such conflicts, describe conflict-riddled problems as “occasional lapses,” and contemptuously dismiss as self-interested those who have raised these issues.9 Despite this push-back, academic medical centers all across the land are developing or revising their existing COI policies, and some professional societies are following suit. Some have restricted free food and gifts; some have dropped industry support of continuing medical education; some have proscribed participation in speakers’ bureaus. All are making the rules more stringent. Yet some organizations and some individuals tenaciously hold to the discredited notion that maintaining the conflicts is okay as long as disclosure disinfects and sanitizes. They may be taking the easy way out, but they are fooling only themselves.
1 Resnik DB. Disclosing hidden sources of funding. Acad Med. 2009;84:1226–1228.
3 Kassirer JP. Medicine’s obsession with disclosure of financial conflicts: Fixing the wrong problem. In: Snyder PJ, Mayes LD, Spencer D, eds. Science and the Media: Delgado’s Brave Bulls and the Ethics of Scientific Disclosure. San Diego, Calif: Academic Press; 2008.
4 Brandeis LD. What publicity can do. Harper’s Weekly. December 1932;20:1913.
5 Surowiecki J. The talking cure. The New Yorker. December 9, 2002:54.
6 Angell M. The Truth About The Drug Companies and What to Do About It. New York, NY: Random House; 2004.
7 Psaty BM. Conflict of interest, disclosure, and trial reports. JAMA. 2009;301:1477–1479.
8 Kassirer JP. On the Take: How Medicine’s Complicity With Big Business Can Endanger Your Health. New York, NY: Oxford University Press; 2004.
9 Shaywitz DA, Stossel TP. It’s time to fight the “PharmaScolds. Wall Street Journal. April 8, 2009:A11.