Editors should consider retracting a publication if:
They have clear evidence that the findings are unreliable, either as a result of major error (eg, miscalculation or experimental error), or as a result of fabrication (eg, of data) or falsification (eg, image manipulation)
It constitutes plagiarism
The findings have previously been published elsewhere without proper attribution to previous sources or disclosure to the editor, permission to republish, or justification (ie, cases of redundant publication)
Author Elizabeth Wager, Virginia Barbour, Steven Yentis, Sabine Kleinert on behalf of COPE Council Version 12009 Version 2 2019 How to cite this
COPE Council. COPE Guidelines: Retraction Guidelines. November 2019. https://doi.org/10.24318/cope.2019.1.4
Attribution — You must attribute the work in the manner specified by the author or licensor (but not in any way that suggests that they endorse you or your use of the work).
Non-commercial — You may not use this work for commercial purposes. No Derivative Works —
You may not alter, transform, or build upon this work. We ask that you give full accreditation to COPE with a link to our website: publicationethics.org
Full page history
17 May 2020
Text replaced with summary text from 2019 guidelines.
An allegation of data fraud was not satisfactorily resolved by correspondence with the authors. We then went to the lead institution and asked for an investigation. Within 10 days we had a report clearing the authors, but interestingly using some of the exact same phrases the authors used in their responses to us. We felt that the report was too superficial and approached the other institution involved.
A new investigation was started; this investigation took many months. The report said there was insufficient justification to take the matter to formal assessment and the institution was not minded to investigate further.
We remain concerned. The second investigation asked the authors to provide some data for re-analysis: that came out close enough for the committee to be reassured. However, provision of the same data for re-analysis is likely to produce the same result. In our previous experiences, only checking of at least some of the case report forms would uncover fraudulent data. We are therefore not reassured that the paper is sound and we have no direct evidence (CRFs) that the data are genuine.
Questions for the COPE Forum • When two institutional reviews have failed to investigate thoroughly enough to reassure editors, what further investigations might be warranted? And by whom?
• Is there sufficient doubt remaining for an expression of concern? Or should we accept the results of the investigations even if we consider them inadequate?
The Forum suggested that if the original data cannot be produced, it would be reasonable to retract the paper. No access to the primary data are grounds for retraction. The Forum asked if it would be possible for the journal to have access to the anonymised data?
The Forum agreed that an expression of concern would be appropriate if the journal believes that the institution has not done due diligence. It is not within the journal’s remit to carry out an investigation. The COPE retraction guidelines state that editors are welcome to issue an expression of concern if the investigation by an institution has not been fair or conclusive.
A suggestion was to contact an organisation that oversees these types of investigations at the institution. Is there a national body than could be contacted?
The role of the editor is to safeguard the literature and prevent readers from being misled, so an expression of concern is entirely warranted in this case.
A reader contacted the journal to raise concerns about a paper containing a potentially manipulated figure. The editor-in-chief agreed with the assessment that the figure had been manipulated and attempted to contact the corresponding author, without response. Following further contact with the co-authors and institution, it was established that the corresponding author had retired after publication of the paper, and no current contact details could be found.
No co-authors were able to confirm how the figure was constructed, but explained that it was an old image that was made by or for the corresponding author, and that the location of the raw image or original data was not known due to the corresponding author’s laboratory being dismantled on retirement. The figure is also present in a previous publication from 2007. The figure manipulation does not appear to affect the scientific results or conclusions of the paper.
Question for the COPE Forum • Given that the corresponding author cannot be contacted to confirm if the nature of any figure manipulation merits retraction, would it be appropriate to publish an expression of concern which will remain in place if no additional information is forthcoming?
The Forum advised that as there is no true confirmation of the evidence that the figure was manipulated, a retraction does not seem to be warranted without further investigation. A suggestion was to contact the journal of the previous publication, as there may be an issue of duplicate publication or copyright issues related to the figure. The editors can then discuss together how to deal with this issue. The other journal may have the original data. If the copyright resides with the first journal, then the editor may have to have a different type of notice on the paper.
The Forum agreed that the most reasonable solution would be to publish an expression of concern, explaining the issues. An expression of concern provides an opportunity for further information to be made available at a later date, and then any further action, if necessary, may be taken. A different view expressed was to remove the figure and publish a correction—if other researchers have re-done the work on the subject it may be possible to replace the figure with a citation.
Following advice from the COPE Forum, the editor contacted the journal of the previous publication. The journal is proceeding with an expression of concern to explain the issues. The editor considers the case closed.
A journal published a paper that is now under investigation by the host institution for misconduct. All authors signed that they agreed authorship and took responsibility for the content of the paper. After the investigations started, an author asked to be removed from authorship.
Questions for the COPE Forum • What should the journal do in this situation? • Should the journal permit the author to withdraw, or does agreement to authorship have irrevocable responsibilities?
The Forum agreed that the best course of action is to postpone any decision until after the investigation. In the meantime, the journal could consider publishing an expression of concern stating that an investigation on the paper is being conducted but avoiding stating that there is an authorship dispute. The journal should await the outcome of the investigation before making any changes to the paper.
The Forum suggested this could be thought of in terms of an authorship dispute and so the journal should handle it as it would for changes in authorship. Hence the journal may need to go back to the institution, as for an authorship dispute.
Are all the authors from the same institution? The author may have a legitimate reason for wanting to be removed if he is from a different institution. A suggestion for the editor was to ask the author why he wishes to be removed from the article.
The presenter of the case confirmed that the author signed the agreement in good faith and that the signature was genuine. Hence the author signed and consented to publication. According to the International Committee of Medical Journal Editors (ICMJE) guidelines (4th criteria), (http://www.icmje.org/recommendations/browse/roles-and-responsibilities/defining-the-role-of-authors-and-contributors.html) all authors have responsibility for the data and agree to help in any investigation: “Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.”
Most of the Forum agreed with a robust “no” to the request and with contacting the institution.
If the editor decides not to allow removal of the author following the investigation, he could give the dissenting author the option of publishing a comment on the published paper.
The journal found the advice from the Forum very useful and intends to follow the advice. The journal considers their concerns resolved.
A journal appointed a new editor-in-chief to their journal. He had previously been on the editorial board of the journal for 10 years and the editorial registrar for 5 years. During the handover period, it came to the journal’s attention that he was due to appear in front of a tribunal for research fraud. By agreement with the journal, he stepped down until the outcome of the tribunal, and the editor-in-chief of another journal took over as acting editor-in-chief in the interim.
The outcome of the tribunal was that many of the charges against the editor were upheld, so he has stepped down permanently. The charges that the tribunal found him guilty of (which did not relate to any papers published in the journal) included fabricating data, accessing electronic patient records without permission, breaching patient confidentiality, submitting a paper knowing that his coauthors had not approved it as a final version, forging his coauthors’ signatures on copyright forms, and referencing a particular fictitious individual in the acknowledgments (apparently as a private joke).
During his time on the editorial board, he published numerous articles in the journal, including two original research articles, nine reviews, three editorials/commentaries and one case study. As far as the journal is aware, there are no substantive issues with any of these papers, which underwent the usual review procedures, but several reference the fictitious person in the acknowledgements.
During his time as editorial registrar, he had input in editorial decisions, as a reviewer and associate editor, and by assisting the then editor-in-chief with decisions on the two categories of papers in our journal which undergo internal review rather than full double-blind peer review. During the weeks when the editor-in-chief role was being handed over, before he stepped down, he made the final decision on several manuscripts as editor-in-chief.
Questions for the COPE Forum
• To what extent should the journal consider the editor’s previous publications in the journal as 'suspect'? The journal will publish a correction relating to the fictitious acknowledgments; should they publish any additional note of editorial concern against his papers? • To what extent should the journal revisit editorial decisions he has previously been involved with during his 10 years' association with the journal? • The acting editor-in-chief has not been appointed editor-in-chief according to the stringent procedures recommended by COPE/ICMJE. Is there anything the journal can do to mitigate this situation during the process of appointing a new editor, which may take several months?
The Forum noted that thankfully this is a relatively rare event. The journal needs to handle this well to avoid reputational damage. Hence the journal needs to think in terms of a public explanation or statement of what has happened
The Forum agreed there needs to be a thorough investigation. Previous papers written by the editor need to be handled on a case by case basis—hence all of the papers should be looked at if practically possible, in particular if any have medical/patient implications. The journal will need to carry out due diligence for these papers. While the acknowledgement issue is relatively minor, it is very unprofessional behaviour. The core issue is whether there is any likelihood of problems with the papers that were written by the editor. This is not dissimilar to institutional handling of research misconduct—is the misconduct a one-off or a systemic problem?
Regarding his input in editorial decisions, as a reviewer and associate editor, and in assisting the then editor-in-chief, the advice was to do a spot check of, say, 10% of the decisions, so the journal is reassured there are no problems with the process or the outcome of any of the decisions. In particular, decisions he took himself or where he went against a decision of the reviewers, for example, should be looked at carefully.
The journal needs to reassure potential authors that they have processes in place to be confident in what they have published in the past—otherwise the journal risks serious reputational damage.
The Forum suggested that this may be too difficult and indeed inappropriate to handle only internally and the journal might consider engaging an external, independent group of people to deal with the issue on behalf of the journal.
The journal needs to develop a process to appoint a new editor-in-chief that ensures that this situation does not happen in the future. Again, it is essential to have external people in the selection process.
An intentional satire of a randomised controlled trial was published in a journal. In addition to multiple overt clues that the article was fake in the text, the article ended with a clear and direct statement in the acknowledgments that it was satire.
Investigators conducting a systematic review on the topic inadvertently included the satire article in their review as a legitimate manuscript, including generating a table based on some of the ‘data’ from the satirical article. This systematic review was eventually published in another journal. The authors of the satirical article saw the published systematic review and immediately contacted the editor of the journal in which it appeared to explain the situation. The editor of the other journal blamed the authors of the satirical article for the situation and demanded that they apologise to the authors of the systematic review and retract the original satirical article. The editor’s argument was that there is no room for ‘nonsense’ in scholarly publishing, and that such articles take publication space away from real scientific articles that could be published in their place.
The authors of the satirical article responded that there has always been a place for humour in scholarly publishing, and several established medical journals regularly publish satire. They commented that the authors of the systematic review failed to thoroughly read the satirical article and did not fulfil their scholarly responsibility in performing the review.
Question(s) for the COPE Forum • Does the publication of satire in a scholarly journal usurp space that should be reserved for legitimate investigations? • Is the journal that published the satirical article at fault when authors performing a systematic review do not thoroughly read and vet the articles they cite? • Is it reasonable for the other journal editor to request the retraction of the satirical article?
The Forum noted that it is up to individual editors or publishers to decide what they publish, and if publishing these types of articles is a valuable use of their page budget. Editors should not be told by other editors or journals what they can and cannot include in their journal. Hence it is not reasonable for the other journal editor to request retraction of the satirical article. There are no grounds for retraction.
The Forum agreed that there should not be editorial censorship but journals and publishers have an obligation to tag satirical articles clearly. They need to be appropriately and responsibly flagged up as such. A view expressed was that in this era of “fake news”, editors have an increased responsibility to ensure that the scientific record is not corrupted and co-opted, and that satire does not end up having unintended consequences on public discourse, including development of public policy. It was suggested that the metadata should also be tagged so that a machine can quickly understand that this is satire. This is especially relevant in terms of text mining ecosystems so that anyone designing a study would have a very easy means of filtering out articles that have been tagged as satire.
From a legal standpoint, journals need to meet a reasonable standard of not being misleading. If the article is clearly marked, with clear headings, and no suggestion this is proper research, then the reader has a responsibility to read things carefully.
The authors of the systematic review are at fault for not carrying out their methodology correctly and should have read the paper properly. The journal that published the systematic review needs to take steps to correct the systematic review.
The journal did not retract the article and agreed with the Forum that the onus was on the researchers to read the paper, which clearly indicated that it was satire.
The journal will take the Forum’s other recommendations into consideration on future articles of this type (eg, ensuring metadata indicate that it is satire in addition to noting in the article type and within the article itself).
A reader, Dr A, wrote to the editors explaining a number of concerns she had with some of the figures in a paper published in the journal. The editors sought the advice of an associate editor with more expertise in the subspecialty of the paper. The associate editor concurred with Dr A’s opinion of the paper and the authors were invited to respond. After some back and forth correspondence, the authors agreed with the editors that an erratum should be published containing the revised figures.
Out of courtesy, the erratum was sent to Dr A, who replied stating that she did not feel the erratum to be adequate and voicing more concerns about the modified figures. After further lengthy back and forth discussion with the authors and Dr A, the editors decided that the erratum should first be published and that Dr A should write a formal letter for publication in the journal expressing her concerns about the paper, with the authors then being given the right of reply to this letter.
Dr A duly wrote a letter but the nature of the concerns she raised has led the editors to conclude that this approach might never resolve the matter, and that the issue should best be handled by correspondence directly between Dr A and authors. The editors have therefore decided that the matter should be formally closed in public by publishing the erratum, and that any subsequent discussion should be handled privately between Dr A and the authors. The erratum has not yet been published.
Question(s) for the COPE Forum • Have the editors handled the situation reasonably? • Could the editors have handled it differently? • How might the journal prepare for dealing with similar situations in the future?
The Forum questioned the decision by the journal to invite Dr A to write a formal letter to the editor. It may have been more useful for the editor to make it clear to Dr A that any letter to the editor would go through the normal peer review process.
If concerns are raised by a reader, the usual approach would be to contact the authors regarding these concerns and to determine whether there are errors, and hence an erratum is needed, or if there is just a difference of opinion. If the latter is the case, the editor could suggest that Dr A submit a comment on the paper, which would go through the normal peer review process, with the authors given the chance to submit a reply. The comment and reply can then be published in the journal. In this way, the full discussion on the article is in the journal, giving clarity to the reader and enabling them to draw their own conclusions.
Some journals have a stated policy that allows no more than one letter to the editor from one specific reader on any one particular article, to avoid ongoing dialogue.
The journal took the following actions: 1. They proceeded with the course of action previously described, writing to Dr A and the authors to let them know that the journal would first publish the erratum and then one exchange of correspondence between Dr A and the authors. The editor made it clear that this concluded the matter as far as the journal was concerned and that any subsequent exchange of correspondence should take place privately between Dr A and the authors. 2. The journal added the following statement to their editorial policy on items of correspondence: “Only one letter may be submitted by any single author or group of authors on any one published paper”.
A paper was accepted in 2012 but there was a lengthy disagreement between the four authors regarding the order of authorship. The authors were advised that the paper would not be published unless all authors could sign a written agreement on the order of authorship and copyright form.
An agreement was received in 2015 that specified the order of authorship and named one of the authors as “the final corresponding author to see the paper through the rest of the process for the paper’s publication”. At the end of the agreement it was stated, “Please address any correspondence to all authors."
Subsequently, the corresponding author attempted to make ‘minor’ changes and another author, author B, rescinded his acceptance of the agreement. The corresponding author later agreed not to make changes at that time and author B stated the terms of the agreement could stand.
During the production process, the proofs were sent to the corresponding author. Changes were made during the proofing stage which author B has subsequently disputed. The corresponding author stated that all authors (including author B) were given multiple opportunities to provide specific changes and comments on the changes that other co-authors suggested.
The paper was published on early view later in the year. In 2016, author B requested retraction of the paper immediately, alleging that the agreement was voided by the changes made during proofing. The paper is still on early view and has not been included in a print issue.
The journal has corresponded with all four authors and advised them that they need to agree on the final version of the article or the journal will be forced to retract the paper because of irreconcilable differences among the authors. The correspondence has not produced any agreement so date and the authors have individually raised the prospect of litigation.
Question(s) for the COPE Forum • On balance, would the Forum agree that an ethical choice in this difficult situation would be a decision to retract with an option to resubmit with a new author byline? • What other options or advice would the Forum suggest?
The Forum advised referring this to the institution, and asking the institution to verify who should be authors on the paper and what the authorship order should be. It is not up to the editor or journal to investigate this issue. As three of the authors are based at one institution, it would be reasonable to ask the institution to mediate in this situation.
There may be copyright issues, if the dissenting author no longer agrees with the content of the article, as all authors have joint copyright. Again, the institution needs to resolve this—this cannot be decided by the editor.
One view was to tell the authors that unless they sort out their differences their manuscript will be retracted. However, another view was that if there are no scientific concerns with the paper, it would be difficult to make a case for retraction. If the journal is confident that the data are valid, then there are no grounds for retraction, but an Expression of Concern could be published stating the concerns of all of the authors. Another suggestion was to issue a correction notice stating which authors support and which authors do not support the current version of the paper, thereby avoiding having to retract the paper in what is essentially an authorship dispute.
Does the journal ask for contributorship statements from the authors? This may clarify the issues around authorship.
Authorship order is a common problem and the issue of who should be listed in what order differs by discipline. There can also be cultural differences as well as different practices in different countries. Hence it can be difficult for authors to navigate.
The journal has reached agreement in principle with the authors on a revised author order and statement regarding the research, but to date, only one author has returned the sign off sheet. The journal is hopeful to receive the rest shortly.
An article was published in July. In October, a corrigendum was published to correct large sections of unattributed text. Two weeks later the journal and publisher received a complaint from a reader who accused the author of the published article of using text from an unpublished collaborative manuscript on which the published author was participating. This participation on the collaborative work was initiated and ongoing during the time that the manuscript was being prepared for publication at the journal. The unpublished collaborative work has not yet been published.
The reader requested retraction of the published article, with the possibility of a republication only when all collaborators of the unpublished work were in agreement with the article content.
The publisher and journal initiated the procedure outlined in the COPE flowchart 'Suspected plagiarism in a published manuscript'. The editor-in-chief has requested feedback from the published author on the reason for the large overlap in text.
Question(s) for the COPE Forum
• Should the publisher and journal publish an 'expression of concern' while continuing with the investigation? • Is this the correct approach in this situation?
The editor told the Forum that she had checked the articles and the degree of overlap of text was nearly 20%. The Forum warned that percentages can be misleading, and the editor needs to look carefully at where the overlap occurred (in the introduction and methods may be fine, but it can be more problematic in the results or conclusion sections).
The institution(s) may need to be involved in this case.
The Forum would advise against an expression of concern on the article as these are generally used for ongoing unresolved cases. In this case, there is nothing proven or finalised —it may ‘end’ in a correction or a retraction. Hence it may be premature to publish an expression of concern and the editor should wait for a response from the authors. There is also the issue of the negative connotations of an expression of concern and/or stigma for the author, which may be unwarranted
A suggestion was for a less permanent ‘Editor’s note’ on the article for now, written in neutral terms.
The journal’s review of the guidelines on text recycling led to the conclusion that the scientific content was not disputed, and in fact the article adds to the body of knowledge. Also, the text recycling was not in the discussion or conclusions but rather in the methodology. The journal decided not to publish an expression of concern or retract the publication. The editor considers the case closed.
A group of unspecified members of an organisation have written an expression of concern (letter via email) to the editors wherein they request that an article previously published in the journal be retracted since they believe it is biased and inaccurate about regulation details within the organisation. They are further requesting that their letter be published in the journal.
The editors of the journal are unsure how best to proceed with this request. They believe the article should not be retracted (no unethical misconduct was committed by the author as far as they are aware, and the article did go through peer review with two reviewers having minor corrections and recommending the editors publish the article, which they accepted), as the reasons stated in the letter for requesting the retraction are not grounds for doing so. The reasons given are:
1) The absence of important information on the organisation's accreditation and professional regulations 2) The calling into question of the organisation's ethical standards and practices 3) The general misinformation about the organisation's business model 4) The overall poor research methodology and writing standards Note: not all members of the organisation were aware that the letter was sent (one of the editors is a member of this organisation), and the author of the article and certain members of this organisation have a history of disagreeing on previous articles that the author has published in other journals on the same topic (furthermore, the author once had a submission accepted at this organisation’s conference and after a disagreement they uninvited him).
The editors have shared the expression of concern letter with the author and the publisher. The author has responded to all the points in their letter with counter arguments. The editors have also crafted a response letter to this organisation (standing by their decision). The publisher does not believe the article should be retracted, but instead that the editors should respond to the complaint with an explanation about when retractions are appropriate and also the nature of how the paper was accepted/published.
However, the editors are considering whether they should publish the letter with the responses (if it would benefit the community, which is a very close knit group). The author has also requested that his response be published in the journal.
Question(s) for the COPE Forum • Is it wise/is there justification to publish the letter? (it could be beneficial if it is conducted in a scholarly manner but also could be unproductive if major disagreements sit on both sides historically.) • If COPE recommends publishing on this topic, how does it suggest the editors do so (what format/are there procedures that should be followed)? • If COPE suggests responding to the organisation via email is best, are there any other steps the editors should take to resolve this disagreement? • Has a situation like this one occurred before/been handled successfully?
The Forum agreed that there does not seem to be a case for retraction, but the editor may wish to consult the COPE Guidelines for retracting articles (http://publicationethics.org/files/retraction%20guidelines_0.pdf). The advice was to have the conversation played out in the journal and allow the exchange to be published. It would be reasonable, if time consuming, to do the following: have the letter peer-reviewed by a qualified reviewer who can evaluate the accuracy of the claims in the article; have a peer-review of the response of the author; and if both get through the review process (with revisions if necessary) then publish them both.
This is an editorial judgement call. COPE would not have a “publication ethics” position here, besides that contained within its regular guidance for publishing (ie, to manage transparency, permissions, conflicts of interest, etc). Publishing letters or debate papers would make the process transparent. If the journal decides to publish the letters, the Forum recommended ensuring that the letter from the organisation is signed by an individual or individuals. The journal should not publish an anonymous letter.
The Forum noted that in cases such as these, it is very important to choose reviewers carefully. Selection of objective reviewers is especially important when organisations are involved.