How Should Journals Handle the Conflict of Interest of Their Editors?Who Watches the “Watchers”?

Conflict of interest (COI) policies at most academic medical centers require that investigators disclose research-related financial interests to their institutions so the institution can review the potential risks of the COI. To manage COI, academic medical centers typically direct investigators to disclose their financial interests to research participants and to journal editors. If an institution follows the guidelines of the Association of American Medical Colleges,¹ except for situations in which a researcher has a significant financial interest in research involving humans, investigators generally are permitted to retain their financial interests while performing related research as long as there is transparency about their financial interests via disclosure to the consumers (eg, research participants, co-investigators, and individuals who receive reports of the research, including those listening to a presentation of the research or reading an article about the research).

Is this “buyer beware” approach for the consumer sufficient transparency regarding an investigators’ COI? Even though each research institution establishes its own policies regarding when to impose more stringent management than disclosure, the reliance on disclosure presumes that risks of COI-based bias are mitigated by the regulations and norms that apply to scientific research. For example, institutional animal care and use committees as well as institutional review boards typically must approve protocols and oversee the conduct of the study. These requirements should guarantee that protocols, including those involving an investigator’s COI, are vetted by experts, and that there is some oversight of study conduct. In scientific reporting of clinical research (ie, a randomized clinical trial, observational studies, and meta-analyses) that follows reporting recommendations (ie, the standards recommended by the EQUATOR Network²), there generally is clarity with regard to the role of any investigator with a COI and the role of any industry sponsors. Disclosure of a COI in a publication invites readers to consider whether the study design was objective, data were appropriately analyzed, and interpretations and conclusions accurately reflect the data.³

In the case of systematic reviews without meta-analysis, narrative reviews, or opinions (eg, editorials, invited commentaries, or other viewpoints), disclosure may not be judged adequate to manage situations in which an author has a related financial interest or involvement by a financially interested company. These submissions are susceptible to greater potential bias than are reports of research involving an investigator COI or industry sponsorship. Thus, in JAMA as well as in the family of journals composing The JAMA Network, although Original Investigations might involve authors with a related financial COI that is managed through disclosure to the reader, most opinion pieces do not involve authors who have a COI related to their opinions. As stated in the AMA Manual of Style, chapter 5.5.1, “editors of some journals prefer that authors of some types of articles, such as editorials, commentaries, and reviews, not have financial interests in the subject matter….”³ All opinion articles are reviewed by the editor in chief or senior editors designated by the editor in chief to ensure that an article is not biased.

The process of disclosure to journals requires that editors act as “watchers” over authors because editors filter submitted manuscripts on behalf of the readers and are expected to safeguard the integrity of the journal. They do that in part by publishing authors’ COI disclosures. Readers also trust that editors will not have a COI that might influence their decisions about which articles to accept or reject. This trust extends to judging when to reject a research or nonresearch manuscript that appears biased by an author’s financial COI such that, although the conflict is disclosed, the article is deemed unacceptable for publication.

But who watches the watchers? Because such “external” protections ordinarily are not in place for the editorial process (ie, an editor of the editor), journal editors should be held to a high standard. Editors exercise substantial control over the biomedical literature, and the literature, in turn, can have enormous influence in health care. Some peer-reviewed medical journals are considered the most authoritative sources of scientific knowledge by physicians, researchers, promotions committees, regulatory agencies, federal and private grant-making bodies, courts, the media, and the public. Editors can shape the literature by determining which original research, opinion pieces, reviews, and other educational materials are published. They also may choose which articles are given most prominence in the journal, on its website, and in other platforms. Editors also may select publications to be highlighted or promoted to the media, although the media ultimately decides what to feature in their own publications.

Consequently, the potential for biased decision making associated with an editors’ COI is of particular concern. For example, can an editor who holds patents on an imaging device objectively judge whether to publish a study of a novel technology that directly competes with her invention? What if an editor was a collaborator on a study submitted to his journal? Institutional COI also may affect decision making. Can an editor who holds a senior faculty position at a university fairly decide between 2 equally strong manuscripts when one is by authors from her institution and the other is from another university? Because the stakes potentially are so great, COI policies of journals should promote the highest possible level of objectivity in the editorial process.

The “Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals” by the International Committee of Medical Journal Editors (ICMJE) states that editors should not have a COI that could lead to bias or the appearance of bias in decision making: “Editors who make final decisions about manuscripts should recuse themselves from editorial decisions if they have conflicts of interest or relationships that pose potential conflicts related to articles under consideration.”⁴JAMA and The JAMA Network’s journal policy indicates that “final decisions regarding manuscript publication are made by an editor who does not have any relevant conflicts of interest.”⁵ This includes a financial and nonfinancial COI. These journals have a formal recusal process in place to help manage a potential COI of an editor.^4,5 One example of an explicit procedure for addressing an editor’s COI is offered by JAMA Ophthalmology, for which one of us (N.M.B.) is editor in chief. JAMA Ophthalmology clearly delineates conditions that create an unacceptable COI for an editor and outlines a detailed procedure for recusal from editorial decision making (Box). The ICMJE recommendations, JAMA’s policy, and JAMA Ophthalmology’s procedures prioritize elimination of bias or an appearance of bias in the review of manuscripts over other considerations, such as the availability of relevant expertise.

Since the JAMA Ophthalmology procedures were implemented when the current editor in chief and deputy editors began their terms in July 2013, 442 submissions (8%) of approximately 5335 submissions necessitated recusals by the editor in chief to a deputy editor. There has been only 1 instance in which the JAMA Ophthalmology editor in chief and the deputy editors needed to recuse themselves from making decisions with respect to an Original Investigation submission. In that instance, the JAMA editor in chief oversaw an unconflicted JAMA Ophthalmology editorial board member with expertise in the area of the submission, who advised the JAMA editor in chief regarding editorial decisions on the manuscript.

Although minimization of conflicts is desirable, a “purist” approach, whereby an editor has no potential COI and recusal policies are therefore unnecessary, may lower the quality of the scientific literature. In particular, editors who are engaged in academic medicine and participate in the ever-evolving process of design, implementation, and reporting of research also have up-to-date knowledge and broad expertise in their fields and bring important perspectives to the role of editing journals. Recognizing that editors in chief still might be involved in Original Investigations, both JAMA and JAMA Ophthalmology indicate within the Article Information section, as explained in the AMA Manual of Style,³ when a decision-making editor was recused (eg, Dr Bressler, Editor in Chief of JAMA Ophthalmology, had no role in the editorial review or decision to publish this article). In our roles as an editor (N.M.B.) or associate dean for policy coordination (J.D.G.), which both involve review of an individual or an institutional COI, we are unaware of instances in which these rigorous recusal policies have resulted in prospective editors being unwilling or unable to serve as an editor.

Setting high standards for an editor’s COI and “watching the watchers” with clear recusal procedures should help maintain the trust of readers, authors, and the public. Although sources of bias will never be completely eliminated, journals can navigate COI to successfully disseminate advances in medicine and science.

Corresponding Author: Neil M. Bressler, MD, Office Maumenee 752, Johns Hopkins Hospital, 600 N Wolfe St, Baltimore, MD 21287-9277 (nbressler@jhmi.edu).

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Ms Gottlieb reports being associate dean for policy coordination at the Johns Hopkins University School of Medicine, which oversees review of outside interests, including financial conflict of interest. Dr Bressler reports receiving grants from Bayer, Novartis, Roche/Genentech, and Samsung to his institution; and previously serving as chair of the Johns Hopkins University School of Medicine’s committee on outside interests, which reviews financial conflict of interest. He was not involved in the editorial evaluation or decision to accept this article for publication.