What happens before a retraction? A behind-the-scenes look from COPE

Charon Pierson

Ever wonder how editors figure out whether a paper should be corrected, retracted, or left as-is? For a window into that crucial decision-making process, the Committee on Publication Ethics (COPE) publishes a number of anonymized cases per year, in which they weigh in on a dilemma faced by a journal editor. The organization has weighed in on more than 500 such situations since 1997. We spoke with Charon Pierson, Editor-in-Chief of the Journal of the American Association of Nurse Practitioners and the Secretary of the Trustee Board and Council at COPE to find out more information about these cases – including the one that affected her most.

Retraction Watch: How does one of the COPE cases get opened?

Charon Pierson: COPE will open a case if it’s brought forward by the editor of record for a journal that’s among one of our more than 10,000 members. Often it’s a situation where a reader has contacted the editor about a potential issue in a paper, and the situation isn’t addressed in any of our guidelines or flowcharts. To be clear – cases cannot be opened by a reader who has questions about an individual paper; they must be initiated by a member journal.

These are real editors dealing with real problems, which they bring forward to COPE when the journal is trying to make its decision about how to handle them. The cases are, in a way, a behind-the-scenes look at how journals make these decisions. Typically, they are discussed during one of our Forums, held four times per year.

RW: What have you learned about publishing in the process of dealing with these cases?

CP: My main takeaway, after five years of dealing with cases, is just how tough an editor’s job is. When a reader comes forward about a paper, it’s the editor’s responsibility to handle the complaint – whatever it is – in an open and transparent manner. But they are really disadvantaged in their ability to pursue a case – often they don’t have the authority or resources to properly investigate an allegation. And the situations are often complex, and editors face restrictions depending on where the authors are from. For instance, if a biomedical study was conducted with U.S. federal funding, editors aren’t allowed to have any input until the Office of Research Integrity finishes its investigation.

I know how hard it is to be an editor fielding complaints from readers first-hand, and I’ve had to retract articles from my journal for plagiarism after a reader approached me about the problem. The first time this happened, my first thought was: ‘Why would any author plagiarize?’ My next thought was: ‘What should I do about it?’ These questions led me to COPE more than 10 years ago. As an editor, you’re constantly focused on logistics – getting reviewers, reviewing the page count, and corresponding with authors. There are so many details you have to focus on that adding an investigation to the mix can feel overwhelming. And remember, the majority of editors do this job part-time in addition to an academic, clinical or research career.

RW: So the cases are one of the ways COPE tries to help editors navigate the complicated process of dealing with reader complaints?

CP: Yes, and as we as an organization continued to weigh in on cases, they served as our template for creating our other resources for editors, such as our guidelines, code of conduct, and flow charts. These now exist to help editors solve problems so they don’t have to submit cases – what do you do if someone says a reviewer has a conflict of interest, for instance? Or a reader says an article is a translation of another paper the authors wrote in another language?

We hope these resources help make editors’ jobs easier, and when they’re not enough, that the cases will help them handle complaints as transparently as possible.

RW: Given your desire for transparency, why anonymize the cases?

CP: Given that we are weighing in on an ongoing situation, before the journal has reached its final decision, we don’t want readers or other editors to have any pre-formed opinions about a specific situation. Additionally, some cases involve problems during the submission and review of manuscripts, so confidentiality is essential.

RW: Is there any case that stands out in your mind more than others?

CP: I’d say definitely case number #12-12 – meaning, it was the 12th case we handled in 2012. In this case, a journal had discovered that an author had created fake email accounts for reviewers, and asked us how to handle the other papers published by the same author, in which the author did not admit to fabricating reviews. We suggested:

The Forum advised re-reviewing the remaining published papers to which author A has not admitted influencing the peer review process. If the journal wishes to stand by these papers, then it is essential that all of the papers are re-reviewed. In the meantime, an expression of concern should be issued for all of these papers. One suggestion was to inform the author of the course of action that the journal is going to undertake and see if he wishes to retract all of these papers.

The Forum noted that the journal should take some responsibility for failure of their peer review system. Good practice is always to check the names, addresses and email contacts of reviewers, and especially those that are recommended by authors. Editors should never use only the preferred reviewer.

This case was a major eye-opener for us, as it was the first time we’d ever heard about reviewers setting up dummy email accounts to review their own papers. I think people were jaw-dropped at this; it didn’t even occur to me that it could happen. It represented a turning point for a lot of people about how much journal publishing has changed in the digital era.

RW: Did anything about this case change practice at COPE?

CP: Yes – soon afterwards, we established the COPE ethical guidelines for peer reviewers, which included the line: “recognize that impersonation of another individual during the review process is considered serious misconduct.” Of course, this wouldn’t have much of an impact on authors posing as fake reviewers, but it reflects our attention to peer review, which had now moved to the forefront of the minds of journal editors. I think after learning about all the many cases of fake reviews, journals became more cautious about their internal processes to check reviewers’ identities.

We’ve since added an “Expectations Post Review” section, which reminds reviewers to continue to keep details of a manuscript and their reviews confidential. Now with so many preprint websites like PeerJ and Academic Karma, there is a greater expectation that reviewers will post their reviews. And with sites like PubPeer, more opportunity for them to do so.

RW: Any other cases stand out in your mind?

CP: There’s case #15-03, in which an author appealed a journal’s decision to reject his manuscript because he had experimented on himself – depriving himself of a nutrient and recording his symptoms. Here, the forum had to decide 1) is it ethical for scientists to experiment on themselves, and 2) are data from one person worthy of publication? Here, it ended in a judgment call, where the Forum concluded an author can perform self-experimentation, but it’s up to journals to decide if they want to publish the findings.

RW: Thanks so much for your time, Charon. For more of a behind-the-scenes look at how journals decide what to do about potentially problematic papers, you can peruse the 539 cases COPE has adjudicated since 1997.

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our new daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.

8 thoughts on “What happens before a retraction? A behind-the-scenes look from COPE”

  1. “To be clear – cases cannot be opened by a reader who has questions about an individual paper; they must be initiated by a member journal.”

    Therein lies the greatest problem.

  2. I am already waiting 241 days on receiving the correspondence of COPE with publisher Taylor & Francis about a paper in a TF journal with deep concerns about almost all of the raw research data which are presented in this paper. See for backgrounds
    https://pubpeer.com/publications/CBDA623DED06FB48B659B631BA69E7 (the Pubpeer entry about the paper in question).
    See https://pubpeer.com/publications/7DA806A8062EF9474F1A53717B9D1D#fb36200 for some details about my contacts with COPE (which includes an e-mail from COPE, dated 26 July 2015, in which it is stated that I will receive this correspondence).

    1. Thanks for your comment. These are 2 separate processes – the cases are about ethical issues that arise from handling and publishing manuscripts; a complaint about the actions of an editor should go through the proper channels at the journal/publisher. Every journal should have a mechanism for handling complaints from readers about editorial actions . Often that process begins with contacting the journal publisher.

  3. I do not understand the above statement: “. . . editors face restrictions depending on where the authors are from. For instance, if a biomedical study was conducted with U.S. federal funding, editors aren’t allowed to have any input until the Office of Research Integrity finishes its investigation.”

    Editors in the U.S. have frequently communicated with the federal Office of Research Integrity (ORI) as well as with research institutions, sending concerns raised by readers, reviewers, or journal staff that the editor is unable to resolve, as possible research misconduct. ORI staff scientists have also provided assistance to journals with image analysis techniques in their journal cases.

    ORI conducts aggressive oversight reviews of investigation reports and procedures done at research institutions (not its own investigations). Of course, editors are acting as complainants, witnesses or resource persons in research misconduct cases handled by institutions or ORI; editors are not part of the investigation and misconduct decision-making team. However, institutions and ORI will inform editors of the outcome of investigations when recommendations are made to correct or retract a figure or paper.

    As former ORI Associate Director for Investigative Oversight (2000-2006) and since then a consultant on research misconduct, I was invited to talk about ORI and editor interactions and cooperation at the Council for Science Editors’ Publication Ethics course in 2012, 2014, and 2016 (coming on May 15th) with COPE members.

    1. For clarification, I meant that if editors receive a complaint, and it’s the editor’s responsibility to pursue an investigation in an open and transparent manner, this may not be possible if ORI is involved. If we refer the case to ORI, then we cannot respond directly to the complainant, and some editors have been harassed for not doing anything to resolve the complaint. Yes, once ORI has completed its investigation, then the editor is free to retract the paper or do whatever is necessary to make the recommended corrections, but meanwhile, during the investigation, it appears that nothing is happening. ORI has also requested that editors not contact the authors before contacting ORI when there is a suspicion of serious misconduct so that the authors do not have the opportunity to destroy evidence. Editors and ORI need to have good communication, but that communication cannot be completely transparent to journal readers until an investigation is complete.

      1. I guess the problem is the idea that “it’s the editor’s responsibility to pursue an investigation in an open and transparent manner.” In 17 years at ORI, and I believe after that (2006), ORI has always been willing to discuss cases with the editor and provided the editor assistance (as I mentioned earlier), but of course that is done in confidential communications with the editor, which ORI asks not to be made public, since the ORI regulations require confidential treatment of individuals involved in research misconduct investigations.

        But the editor needs to do his or her job, including communicating with the complainant and the respondent (or the respondent’s research integrity officer or other senior official). An increasing number of editors in the past decade have been publishing Editorial Expressions of Concern about questioned papers, when the authors do not provide original data or sufficient answers to questions raised by editors (as from complainants, PubPeer, etc.) Then the research community has been notified of questions raised, which hopefully are resolved by institutional investigation and/or ORI oversight of them (of course, often months or years later for complex and contested cases).

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.