Peer reviewers can meet journals’ criteria for authorshipBMJ 2013; 346 doi: http://dx.doi.org/10.1136/bmj.f166 (Published 16 January 2013) Cite this as: BMJ 2013;346:f166
- 1University Hospital of Cologne, Cologne, Germany
- 2Center for Laboratory Medicine, University Hospital Münster, Westphalian Wilhelms-University of Münster, Germany
- 3Institute for Biomedical Ethics, University of Basel, Basel, Switzerland
- Correspondence to: DM Shaw
The question of exactly who should receive credit and bear responsibility for scientific publications has been discussed repeatedly in recent years. The majority of journals in the medical arena adhere to the authorship standards that have been developed by the International Committee of Medical Journal Editors (ICMJE). The ICMJE’s Uniform Requirements for Manuscripts Submitted to Biomedical Journals aim to combat ghost and guest authorship and to allow appropriate recognition and accountability for what was done and how it is published. However, consideration of the ICMJE’s Ethical Considerations in the Conduct and Reporting of Research leads to the following questions. Should substantial contributions by peer reviewers be openly declared? Should some reviewers perhaps even be credited as authors?
Admittedly, reviewers’ contributions arrive after a study has been planned, conducted, analysed, and written up. However, diligent reviewers examine how research was designed and data collected, analysed, and interpreted (meeting ICJME criterion 1).1 In addition, reviewers suggest modifications to both design and text that often substantially improve the paper (criterion 2).1 The reviewer’s recommendation to accept the paper for publication could qualify as final approval of the paper (which is criterion 3).1 In principle, therefore, some contributions by reviewers may justify credit for authorship as recommended by the ICMJE. This is similar to the suggestion that ethics committees sometimes act as authors or should at least be acknowledged with regard to their substantial contributions to published work.2 It is recognised that peer review provides added value,3 4 and it may be unethical not to acknowledge reviewers’ contributions if all three criteria are met.
An example helps to illustrate this point. Let’s imagine that Susan and Louise conduct a study, write a paper together, and submit it to a journal for publication. They receive a “revise and resubmit” decision from the journal along with substantial, helpful comments from one reviewer. They implement all the comments, which include suggestions for some additional experiments and analyses, and resubmit the paper. At this point, the reviewer has already met the ICMJE’s first and second criteria for authorship, having influenced the design of the study and the format and content of the resulting paper. If the reviewer is now happy with Susan’s and Louise’s additions and revisions, she or he will communicate this to the editor of the journal, who is almost certain to accept the paper for publication. In so doing, the reviewer fulfils the third ICMJE criterion for authorship. (Note that if Susan conceived the study and Louise carried it out, the reviewer has actually been involved in more of the criteria than either of the authors—but then, of course, neither Susan nor Louise would qualify as an author under the ICMJE criteria.5)
The case of Susan and Louise highlights a paradox of authorship that lies at the heart of the current peer review system. If they had shown their paper to Justine, a casual acquaintance at another university, and she had suggested substantial changes and additional experiments and analyses that they implemented, wrote up, and confirmed with her before submission, she would clearly qualify as an author under the ICMJE criteria, and would understandably be aggrieved if she ended up as a ghost author. Yet under the current system, if Justine had been approached as a reviewer and made the very same suggestions, she would get no credit at all. This is illogical and unethical.
We sometimes notice the statement “thanks to anonymous reviewers for helpful comments” in the acknowledgements section of published papers. No journals seem to specify how to acknowledge advice from reviewers who contributed substantially to the final paper, and a few even discourage such acknowledgements. Although the notion of qualification as formal authors might go too far, it seems appropriate for reviewers to be included as contributors and have their precise contributions described. (And of course, it might be awkward for the original authors if the contributions of a reviewer clearly merited full author credit.) Moreover, full disclosure of reviewer contributions could be relevant with regard to responsibility and accountability, as reviewers’ comments can (whether intentionally or deliberately) bias a paper.6 In such cases, readers would have a chance to use reviewers’ specified contributions to judge how those peers may have shaped, and sometimes distorted, material and interpretations.
Given these potential distortions, it may also be necessary for reviewers to declare to readers—and not to editors alone, as is frequent practice in numerous journals today—any potential conflicts of interest, whether theoretical or financial. After all, if authors must declare conflicts of interest, and some reviewers qualify as authors, it would actually be odd not to demand such a declaration. These suggestions could also have the beneficial side effect of ensuring reviewers do their jobs well.
The case of Susan, Louise, and Justine highlights the absurdity and inappropriateness of the current system. Accurate interpretation of evidence in medicine requires accurate evidence regarding the role of reviewers; at the very least, it should be recognised that some reviewers are currently contributing more to research than some authors. Greater transparency demands that reviewers receive due credit and shoulder due responsibility for their work.
Cite this as: BMJ 2013;346:f166
TE and ME had the original idea for the article and wrote the first draft. DS substantially expanded the article, and all three authors approved the final version.
Competing interests: All authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.
Provenance and peer review: Not commissioned; not externally peer reviewed.