A paper that has been accepted for publication in a journal has recently been found to be unsuitable for publication. The authors have been highlighted in other journals for disseminating misinformation regarding the treatment of COVID-19. An expression of concern has been issued on another article, similar to the one we are close to publishing, in another publisher's journal. We looked into the paper in our system further and have decided the peer review comments are insufficient. We would like to rescind the decision on this manuscript and send it to peer review for further comments. We do not want to add to the literature casuing misinformation regarding the tratment of COVID-19.
During the production process of the manuscript, the author has emailed the production editor saying people will try to pay us to shut down the paper before it is published. This has added an extra layer of concern to how the situation is handled.
We think we should take the following steps:
Contact the author to explain the manuscript requires further peer review.
Send the manuscript for two further sets of peer review comments.
Provide the author with the new set of peer review comments.
Question for COPE Council
- Is this the right way to handle this case?
Advice on this case is from a small number of COPE Council Members. Most cases on the COPE website are presented to the COPE Forum where advice is offered by a wider group of COPE Members and COPE Council Members. Advice on individual cases is not formal COPE guidance.
Journal editors have complete authority over what is published in their journal, at any stage, so even if a paper has received an acceptance decision, if concerns have been raised the editors may choose to have additional peer review, delay publication, or even overturn the decision to accept. That said, given the additional publicity in this case, the journal may want to minimise the risks of bad publicity or claims of censorship, and place the manuscript on hold rather than rescinding the acceptance at this stage.
Any decision to have the paper re-reviewed should be based on whether there are deemed to have been flaws in the original peer review process or in the paper itself. For example, if the peer reviewers missed fact checks, exaggeration bias, or overconfident reporting of findings or the literature, the technical editor can ask for revisions directly rather than sending it out again. Conversely, re-review would be recommended there were concerns about the original peer-review process, such as undeclared conflicts of interest or recognition that reviewers were author-nominated reviewers and returned too fast and favourable reports. Additional review would also be necessary if the editor realises that the dataset or research ethics are questionable because of a retraction elsewhere involving the same study, or there has been duplicate or salami publishing. If the authors did not declare that their other article was under investigation this would be a failure to disclose information that ‘would have unduly affected interpretations of the work or recommendations by editors and peer reviewers,’ which is a factor noted in the COPE Retraction Guidelines.
Basing a decision on what has happened at another journal generally seems unwise, especially if the current editors lack all of the facts behind that decision, and do not know how relevant that case is to their own paper. Also, a journal should generally avoid lengthening their review process by ‘holding’ a paper while issues at another journal are resolved. Similarly, it is difficult to judge the merits of any comments written in the public space about the authors or their study. However, given the background issues of potential COVID-19 disinformation and the claims about external interference or censorship, in this case it is recommended that the editors take great care to make sure that there are no processual issues with the way that the paper has been handled and that peer review has been robust. This also protects the journal from allegations of censorship if the article is not eventually published.
The way that the review process should be handled will also depend on whether the article has been made available anywhere. If the article has been accepted but not yet posted anywhere it could be held for further evaluation (perhaps from Editorial Board members rather than by further external review), and the decision to accept could be changed to a rejection if the paper is found to be faulty. If it is already online early and available to readers, then a retraction would be required.
Even before getting to this stage, however, the journal should consider its own review process. Since the manuscript was accepted based upon what are now considered insufficient reviewer comments, what will be done differently to make the next round more robust? If this can all happen quickly, might the editor even delay notifying the author of the additional reviews until someone has confirmed the reasons for concern?
It would be worth reviewing the journal’s policy on accepted manuscripts to determine whether it suggests anything about further actions that may be required. Will the authors be obliged to take the article down from anywhere else they may have deposited the accepted version, for example? Can they legitimately say the article has been accepted by the journal or not while the re-review is going on?
Finally, the journal should consider giving the authors the choice to withdraw their article rather than go through a re-review.