A paper was submitted to a medical journal, reporting the beneficial effects of a treatment with an expensive biological preparation. The author list included one employee of the company that produces and sells the preparation. Specific employees of the company were also thanked for medical input, epidemiological advice, programming support and copy editing; several authors declared having received speaker fees from the company for lectures related to the product. Nevertheless, the authors stated that they were responsible for all of the content and editorial decisions.
After editorial assessment, revisions were requested. The revised version of the manuscript included (per the journal policy) a copy of the revised text with changes from the original tracked. The author of all changes was identified by the word processor tracking as someone whose name appeared neither in the author list nor in the acknowledgements. The company’s website lists this individual as a ‘scientific communications manager’.
The editor felt that this created a transparency issue and contacted the authors. Their response was that the individual involved had replaced the company employee previously thanked for copy editing and “was extremely helpful in assembling the comments and suggestions from all of the co-authors after the data re-analysis, and assisted in preparing the revised version of the paper for submission”. Not thanking him in the acknowledgements was an oversight which the authors are willing to correct. They argue that this input did not fulfil the criteria for authorship.
The editor thinks that the described contribution goes much beyond copy editing.
Question(s) for the COPE Forum • Is “assembling the comments and suggestions from all of the co-authors after the data re-analysis, and assisting in preparing the revised version of the paper for submission” simply copy editing or does it justify authorship? • The editor believes this is quite significant intellectual contribution. Does the Forum agree?
COPE has guidelines on ‘What constitutes authorship’, which the editor may find useful to consult (http://publicationethics.org/files/u7141/Authorship_DiscussionDocument_0_0.pdf). It is very difficult for an editor to make a definitive decision on who qualifies for authorship. The Forum advised that the editor cannot make a decision on this—it is the responsibility of the authors and their institution. The advice was to contact the institution and ask them to resolve the issue.
The editor might want to consider contacting this person directly, rather than liaising through the corresponding author, and asking this person directly about his/her contribution. Self declarations of involvement are often the best evidence that can be obtained. The editor could point to the ICMJE guidelines for authorship and ask him/her if they believe they fulfil these criteria for authorship.
From the information available, it would appear that this person was akin to a medical writer whose job it is to help authors put their paper together and they probably had no intellectual input into the paper. If that is the case, then an acknowledgement would seem to be the most appropriate option, perhaps with more detail about who paid them to do the work. But it is not the editor’s role to decide who or who is not an author—the institution needs to make that decision.
The matter was resolved by including the company employee, who organized the revisions, in the acknowledgements section of the paper. The editor received email confirmation from the company employee that he agreed to be mentioned in the acknowledgements and not be listed as a co-author. The paper was then published.
Our journal uses an internally transparent process where throughout the editor or peer review process, authors, editors and reviewers are all aware of the identities of who is involved. Reviewers are also told—when initially solicited to do a peer review—that they will be named on the final article manuscript as a reviewer. Prior to publication, the pre-print version of a text is sent to reviewers for their approval to be named (or not) as a reviewer on the article. We do not currently publish the content of the peer reviews.
We recently had concerns raised by one reviewer who disagreed with the content of the manuscript and its suitability for publication; the second reviewer was enthusiastic about the manuscript, and the editors decided to publish the text. The first reviewer accused the editors of behaving in a non-transparent manner and even of being unethical, because: (1) we did not publish the content of the critical peer review and (2) we did not have a disclaimer on the text stating that reviewers were not responsible for the content of the published manuscript (which we had assumed was obvious).
We have thus begun the process of adding the following disclaimer to all our peer reviewed texts (and backdating to all those previously published): “Reviewer evaluations are given serious consideration by the editors and authors in the preparation of manuscripts for publication. Nonetheless, being named as a reviewer does not necessarily denote approval of a manuscript by the reviewer; the editors of the journal take full responsibility for final acceptance and publication of an article”.
Question(s) for the COPE Forum • What are the benefits of going to fully transparent review, with publication of the content of peer reviews? • We are aware of the risks (eg, reviewers feeling inhibited from making critical comments for fear of reprisal). Do the benefits outweigh the risks?
The Forum recommended consulting the previous Forum discussion on “Who owns peer reviews” which was discussed on 9 September 2015 and covers some of the issues raised here. A summary of the discussion can be downloaded from the COPE website (http://publicationethics.org/resources/discussion-documents).
The Forum participants provided examples of different peer review models at their journal or publisher. Some journals have a fully transparent open peer review system and publish the reviewer reports alongside the article, with signed comments by the reviewers. These journals believe the benefits outweigh the risks. Other journals do not publish reviews because of the potential confusion or redundancy of publishing criticisms that no longer relate to the final published paper (the concerns having been addressed during revision). However, yet other journals post all versions of the manuscripts, with the revisions, along with the correspondence so people can follow the full history of the paper. This can prompt comments from readers, sometimes negative comments, but they feel this can be refreshing, and authors and readers value it. With this model, it can sometimes be harder to find reviewers, but it does not seem to have any impact on the decisions they recommend. Some publishers only publish those reviews which have approved acceptance of the paper—those who have recommended rejection do not have their names published on the paper to avoid conflict.
The Forum agreed that the journal has done the hard work in establishing an open peer review model with the reviewers, and getting the reviewer names on the published article. The next step is making the reviewer reports public. Some believe the benefits of openness outweigh the drawbacks.
We have implemented our proposed declaration on all peer-reviewed manuscripts, but after discussions we have decided not to proceed at this time with fully open peer-review. We are still exploring ideas.