Full and accurate disclosure of conflicts by researchers does not ensure the prevention of publication of articles that are misleading or otherwise biased.
In the book On the Take: How Medicine's Complicity With Big Business Can Endanger Your Health, a former editor of the New England Journal of Medicine, Jerome Kassirer, MD, calls for "increased scrutiny by medical editors of all financial conflicts of authors with full disclosure not only of the company relationship but also the specific relevancy of the conflicts to the subject matter (specific drugs and devices)."
The position of the Accreditation Council for Continuing Medical Education (ACCME) is even more conservative. ACCME believes that managing or resolving conflicts of interest related to educational presentations must involve the content of the CME activity.
For example, when an individual has been the principal investigator on a project funded by a pharmaceutical company or other commercial interest, the CME provider has been advised to limit the individual's presentation to the data and results of the research.
Change the content
Someone else should be assigned to address the broader implications and recommendations for clinical care. Alternatively, the focus of the proposed activity can be changed so that the content is not about the products or services of the commercial interest that are the basis of the conflict of interest.
The federal government is considering legislation to compel all pharmaceutical companies to provide critical information in a publicly accessible Internet database when a trial is begun and full results when a trial is completed — currently companies are doing this voluntarily — with stiff penalties for noncompliance.
Shield medical students
And of course residency training directors and other academic faculty members are attempting to shield medical students and residents from the influence of pharmaceutical companies by restricting or preventing interactions with salespeople, reminiscent of the NIH ban against collaboration with outside companies. Draconian actions like these are tantamount to throwing the baby out with the bath water.
Alan J. Gelenberg, MD, editor of the Journal of Clinical Psychiatry, the most widely read psychiatric journal in the world, put it this way: "Only a naïve observer would argue that bias is impossible or never occurs. When we take funds, either directly as individuals or through our academic departments, when we develop friendships with industry executives, as occurs routinely, we are subject to influence. In today's paradigm, either one is faced with a panel of experts with acknowledged interests and arrangements or one settles for a panel with less expertise."
Perhaps we should go back to a fundamental precept taught in ethics courses, namely, that the goal is to recognize and manage the conflict rather than mandate almighty solutions.
Just as important, individuals who read reports of clinical trial results, especially trainees, must become their own gatekeepers. They must learn to evaluate statistical gaffes, shoddy study designs, and unwarranted conclusions for themselves. At a minimum, this entails an awareness of:
- any deviations from the treatment protocol or any features of the treatment protocol that bias the results in a particular fashion,
- conclusions that go beyond what the data warrants,
- compelling reasons to think that the trial results extend beyond the particular patient population in the trial,
- meta-analyses combining divergent patient populations; and
- study authors who are strong proponents of one side of a medical controversy.
Managing research conflicts requires vigilance on the part of researchers and readers. Professional organizations, academic institutions, journal editors, and the government cannot be surrogates for individuals conducting clinical trials and interpreting scientific results, no matter who sponsors the research.