|Home | About | Journals | Submit | Contact Us | Français|
I thank all who contributed to the discussion of bidirectional COIs at medical journals, published in the September 2009 and January and February 2010 issues of Mayo Clinic Proceedings. I also express appreciation to the members of the journal's editorial board and the peer reviewers who worked with Dr Hirsch, me, and the authors of letters to critique, select, and offer suggestions to improve the writings eventually published in this journal. I can speak only as a recipient of this process; I am impressed and thankful for the helpful suggestions offered.
The numerous letters, memos, and telephone conversations directed to the journal's staff, editors, and authors on the issue of bidirectional COIs represented one of the greatest responses to journal content in recent memory. Clearly, this topic struck a nerve with many Proceedings' readers: typically physicians engaged in the daily care of patients.
The overwhelming majority of communications were supportive of the Proceedings addressing the issue of bidirectional COI and of the writings by Hirsch and Lanier that expanded the discussion. Perhaps the most common comment I heard was astonishment that many leading journals, editors, policy makers, and published authors had so exhaustively represented 1 side of journals dealing with the ills introduced by financial COI (particularly those involving drug companies) but had been so remiss in presenting counterarguments or in discussing the larger risk-benefit equation (including a discussion of tort litigation) as it relates to the discovery, invention, and introduction of drugs and medical devices. No correspondent tried to excuse industry-associated misdeeds; he or she simply wanted a broader discussion of the issues.
From the various comments shared with me, I inferred that many of those speaking had firsthand experience with smaller specialty and subspecialty journals, in which editors tend to be more intimately related to those they serve and are chosen for their demonstrated expertise as clinicians, investigators, and providers of CME information. Such editors tend to remain in good standing with their sponsoring organizations and constituents because they are approachable and maintain (through ongoing activities) the clinical, investigational, and educational ideals their constituents admire. In contrast, the largest, most influential medical journals often function as multimillion dollar corporations. Editors may be selected not predominantly on the basis of the credentials aforementioned but instead on other credentials important to the journal, including a history of administrative experience at that journal or similarly complex (typically academic) venues and a presumed ability to advance the many missions inherent to a major journal (including numerous high-visibility presentations such as speaking publicly on issues of national and international medical ethics, health care policy, etc). The journal-as-a-corporation scenario carries with it an increased risk of editors acquiring distance and insulation from the journal's constituency.
Editors' ideas that evolve in isolation can be reinforced by like-minded individuals recruited to be a part of the journal's leadership team. It is not difficult to envision that, in such an environment, journal leaders can easily assume that they have special insights into critical issues and assume a position of advocacy for a given position. The rules of deportment that are critical to the integrity of scientific publishing may be seen as less applicable to supporters of a given position than to its critics or vice-versa. Rules of disclosure,1-5 condemnation of repetitive publication,6 and other issues may be applied unevenly,7-9 despite public comments otherwise: ie, “…don't publish papers in more than one place…authors must toil along one publication at a time….“6 Unfortunately, when making such standards operational, editors sometimes find it easier to talk the talk1-6 than walk the walk.7,9 For example, the 5 editorials from the ICMJE on reporting standards, disclosure of COI, and related issues initially published on its Web site1-5 have now been repeatedly published in multiple indexed medical journals for a total of 60 publications, inflating the PubMed listings of leading ICMJE members by some 55 items (last confirmed, January 8, 2010). The fact that there was repetitive publication and the reputational COI it introduced has not been appropriately disclosed by the authors.
Such capricious application of scientific publishing's rules and traditions will contribute to the long-term deterioration of journals' and editors' credibility and influence, to the benefit of no one.
Mayo Clinic Proceedings has taken another approach, outlined in numerous editorials, commentaries, editor's notes, and other publications,9-11 that, simply stated, we will treat those with whom the journal's leadership agrees and those with whom we do not agree with equal respect, using the same rules of engagement. We will attempt to avoid a unilateral advocacy position, recognizing that many of the great failings of science in general have resulted from advocacy on the part of “controllers of science.” Furthermore, Mayo Clinic Proceedings leadership has proclaimed that its work is not complete until the public has had a chance to respond to sentinel publications in the journal.9-11 This was true for our reports on Gulf War illnesses (August 2000)12,13 and the response(s) that followed (November 2000),14,15 physician involvement in capital punishment (September 200716-18 and January 200819-28), COI in clinical practice (May and August 2007),29-31 and the current discussion of bidirectional COI at medical journals. In taking this stance, the journal is critical of publications that, through advocacy, tip the scales in favor of a narrow position and, through executive decisions, tilted peer review, and Draconian restrictions on the collective space allowed dissenting authors to respond, prevent “science” (to use the term loosely) to self-correct its miscalculations, misstatements, and other errors.
This is not to say that any journal's approach to scientific publishing, including that of Mayo Clinic Proceedings, is fool-proof. Indeed, despite the efforts of the best-intentioned authors and reviewers, mistakes and misstatements are printed. That occurred in Hirsch's commentary: Errors appeared in the reporting of the correct amount of remuneration Drs Egilman and Krumholz had received when contributing to legal actions against Merck and the nature and scope of Dr Egilman's legal activities. The journal became aware of these errors shortly after publication of the September 2009 issue and wanted to publish a correction immediately. However, responsible correction demands that the journal not only know that an error was printed but also have some insights into an appropriate correction. This took additional time. Once the details of the correction were verified, the journal immediately released to its Web site a December 2009 correction, clarification, and apology from Hirsch, and the next printed issue of the journal (January 2010) repeated the correction and apology.32 Hirsch further addresses this error in the current issue of the Proceedings. I join Dr Hirsch in apologizing to the authors for these errors and thank all involved parties for making the errors and corrections public.
All who cherish and benefit from biomedical journals should reaffirm that peer review and the traditions of medical publishing that have evolved over decades and centuries have served us well. Journals function best when they use fair, balanced, and attentive peer review and their messages speak in terms of hypotheses, theories, and probabilities. Shortcuts do not serve us well, particularly when we attempt to jump to some new “truth” based on a portion of the story. Albert Einstein had it correct: “No amount of experimentation can ever prove me right; a single experiment can prove me wrong.” Such a realization of the limitations of science, and by extrapolation, scientific publishing, should keep us humble yet diligent to learn more. When we close the door to evidence, debate, and proper decorum, we mock the core of science and set ourselves up for failure.
Credible scientific publication demands adherence to concepts of right and wrong, fairness, and equal applications of core principles to all parties. These are broad brushstroke ideas. Efforts by contemporary journal editors to exhaustively codify author-journal interactions by introducing endless lists of highly specific rules1-5, 33 (eg, that authors disclose their religious and political affiliations5,34) in my opinion merely move science and scientific publishing away from its commonsense ideals. Furthermore, these contemporary lists of rules introduce new loopholes that will allow misdirected authors and editors to ignore the spirit of the rules while adhering to the letter of the rules. In such instances, journal reviewers and readers are deceived.
The long-established, sustaining principles of scientific method, peer review, and medical publishing do not need to be abandoned or undergo major revision in the current era. Hopefully, through the examples of contemporary publications in Mayo Clinic Proceedings, readers can appreciate how adherence to these traditional rules can be used appropriately for the advancement of knowledge and the benefit of society. The collective exchanges on bidirectional COI at medical journals, and previous exchanges on other sensitive topics, are provided as examples.