Author A published a paper in Journal X, which presented evidence of failure by another research group to declare a serious conflict of interest in a paper that had been published some years before in Journal Y. This conflict of interest centred around the undeclared involvement of a third party with a vested interest. Evidence for this was presented in the form of correspondence from the third party stating explicitly that they had developed the work and written a draft of the paper. The paper reached conclusions which had significant favourable policy implications for the third party.
A year later, the research group submitted a response to Journal X, in which they strenuously denied Author A’s conclusions. They accused Author A of libel for suggesting that this undeclared involvement had taken place. They stated “We conceived of this paper, collected the data on which it was based, performed the statistical analyses, wrote the manuscript, and had the final say as to what went into it.”
Most importantly, the research group stated explicitly that they had had no contact with the third party other than to obtain some publicly available, although expensive, information. This remark was made in the context of their emphatic denial that the third party had had any role in the conception, data analysis and preparation of the manuscript.
The editor of Journal X asked Author A, along with three other reviewers, to review the research group’s reply. Each reviewer raised significant criticisms, which prompted the editor to request a radical resubmission from the research group.
Author A’s review presented authentic documentary evidence in the form of correspondence between the third party and the research group. This referred explicitly to research being undertaken on the same data used in the published paper. This contradictory evidence thus revealed thor the the research group had made a serious misleading statement in their original reply to Journal X.
The research group duly resubmitted their revised response. Critically, it did not contain the statement they had originally made, but reiterated that the third party had had no role in their published research.
Author A was shown their revised response and invited to write a reply for simultaneous publication. He was advised that he could not refer to the statement in his reply because this no longer appeared in the revised manuscript. Author A’s response however, did reference the contradictory evidence, in the process of reiterating that the research group had failed to establish that the third party had not been involved in their original paper.
Journal X then accepted both submissions. But the research group withdrew their reply, compelling the withdrawal of Author A’s reply, because it relied on the publication of the research group’s response.
The editors felt the research group had submitted a misleading statement, which they imagined would never be revealed as such, and intended that it would constitute a defence for why they had not declared the involvement of a significant vested interest in their original paper to Journal Y. Had the editor not chosen to select Author A as a reviewer, this intended deception might never have never been revealed.
The review process allowed the authors to sanitise their reply for eventual publication. The editor, Author A, and the research group thus remain the only people who know about this. By withdrawing the paper, the research group may have calculated that they would thus avoid the embarrassment of having the contradictory evidence exposed.
When authors attempt to mislead readers and are found to have done so during the review process, do editors have any right or duty to publicise such attempts in their journals?
When authors withdraw a paper, do editors have any right to override that withdrawal?
Should the editors’ duty go beyond taking the matter up with the authors’ institutions?
Notwithstanding his agreement that the review process is confidential, should Author A have any right or duty to publicise his experience in uncovering this attempt at deception?
It was intriguing that the authors denied the allegation, when they seem to have lied and tried to cover up what they had done.
It was not clear if the lead author was a university employee.
Editors do have a right to publicise attempts to mislead; it is also in the public’s interest.
The editor should publish and be damned, but after having sought legal advice.
The journal should write a very strong letter to the author's institution.
Author A should have the right to publicise his experience of uncovering deception.
A group of authors, based in private practice, submitted three manuscripts to Journal A and one to Journal B. All the manuscripts described the application and effectiveness of a spinal manipulation technique.
The first manuscript in Journal A was a case series of 21 patients. After publication, a member of the journal’s editorial board pointed out several flaws in the study design, including the fact that the authors did not say whether consent had been obtained from the patients. The editorial board member was invited to post a comment detailing his concerns; he has not done so yet.
The second manuscript in Journal A was a case report describing the application of the technique on a patient with a rare spinal disease. The manuscript was rejected on the grounds of poor quality. A reviewer also said that the authors should not be able to make a claim about the effectiveness of this technique when it was only being applied to one patient, and when one of the authors was the director of the institute that “invented” and promotes the technique. This was not picked up during peer review, partly because the author was less than clear about his competing interests.
The third manuscript in Journal A was another case report describing the technique being tested on one patient. The manuscript was rejected before peer review for this very reason.
The manuscript in Journal B was rejected after peer review. The lack of ethical approval for the study concerned the reviewer, who suggested that the results might have been fabricated because the manuscript involved the treatment of 50 patients, and in the past he had peer reviewed two manuscripts by the same authors presenting exactly the same results from the treatment of only 15 patients.
When the author was questioned about ethics approval, he said that he had previously gained consent from a specialist research society for his original study of15 patients. But when this manuscript was rejected, he retrospectively looked at some more patients to make the number up to 50. He said that he did not think retrospective case studies had to be approved by a review board. The webpage for the specialist research society cannot be located.
Do retrospective case studies require ethical approval?
Can we ask the authors for proof of ethics approval for their previous study of 15 patients?
Could it be applied to the later retrospective study or do researchers normally have to re-apply for approval?
Should we further investigate the suggestion that the author may have fabricated the results?
What do we do if we can’t actually locate the people from whom they allegedly obtained ethics approval?
A new intervention would require ethics committee approval.
It seems very odd not to have included the other 35 patients.
A real retrospective study does not require ethical approval
From whom do authors in private practice obtain ethics approval?
With no institution involved is it very difficult to take action.
Get a statistician to look over the data.
Find out who certifies the authors and send a letter to their oversight group.
Contact the relevant research misconduct agency.
Write to the authors, asking for details of ethics committee approval and the address of the board.
An editorial board member of a journal submitted an unsolicited review article on a drug. The editor said the journal would consider the article, but suspected that the article had been commissioned or even written by a drugs company. S/he stipulated that the author must provide a financial disclosure statement before the article could be accepted. The journal published the review article, which had been refereed by two independent reviewers. The author disclosed in his competing interests that he had been a paid consultant for the company that markets the drug. Several months after publication, an agent for the drug company ordered reprints of the article. The agent requested the wording: “This literature review was supported by [X]” be included on the cover sheet of each reprint. The agent was advised that this statement could not be added because the author had not disclosed it. The agent insisted, so the journal contacted the author. The author asked: “Does the final article have these words or something that states the article was in part supported by [X]?” A copy of the agent’s wording and the competing interest statement from the published article were sent to the author, who replied that he was fine with it as long as the publisher was. The author was then asked to explain the extent of the drug company’s involvement in writing the review article. The author replied that the competing interest statement in the article was accurate; the review had been written independently of any pharmaceutical company, and that the requested statement from the agent was inappropriate. The author was contacted again to point out the contradiction in his two replies. At the same time the agent was asked to question the drug company as to whether it had paid the author to write the review, and to confirm the extent to which the drug company had been involved in preparation of the manuscript. The agent did not reply; neither did the drug company. Eventually, the agent cancelled the reprint order. The author finally replied to confirm that he had been confused by the original request, thinking that clarification of whether he was a paid consultant to the drug company was required. He said that when it became apparent in a follow-up email that the drug company wanted the extra statement added, he realised it was inappropriate. The author assured the editors that the drug company would write a letter of explanation soon. The letter has yet to arrive.
- This case raises serious concerns. The connection was not made clear and this is a full conflict of interest. - The paper should be retracted. - The author should be asked explicitly if s/he had been paid by drug companies to write this review.
A journal published an animal study on the use of drug X for the treatment of clinical condition A. The authors did not declare any competing interests. A few months after publication, a journalist contacted the editors to say that the corresponding author had several patents on drug X, was listed as an inventor of the drug, and that the public charity of which he is the director recently announced that they were seeking approval for clinical trials of drug X in condition B. He also said the corresponding author co-owned a commercial company with whom the charity does business. The authors were asked to clarify any competing interests, and were directed to the journal’s policy, which is posted online as part of the instructions for authors. The corresponding author replied saying that the commercial company was a subsidiary of the charity and had no ownership rights to the drug. He did not specifically say whether there were any competing interests to declare. Instead, he wanted to know how this matter had come to light. During a telephone conversation he confirmed that the charity had applied for a patent for the use of the drug for the treatment of clinical condition A, but it was yet to be approved. He also said that no company had been licensed to develop/manufacture the drug for either condition. However, he also said that if the drug was ever licensed, the charity might choose to pass on some of the royalties to the inventors. It was suggested that the following competing interest statement should be published: “Authors 1, 2 and 3 are employees of the not for profit institute, which has applied for a patent for the use of drug X in the treatment of clinical condition A. The institute is a public charity that currently holds patents for the use of drug X in other clinical conditions, with Author 1 listed as an inventor.” - Should the authors have included a competing interest declaration on the manuscript? - If so, should it be any different to what was suggested? - In general, what should be done about studies on drugs that are potentially lucrative? Should the authors declare if they “might” make money? - In this case, what does “not-for-profit” really mean? - And what does all this mean for researchers at universities that make money through spin-off companies?
- In the interests of transparency the authors should have declared a conflict of interest. - The holder of a patent should declare that they hold such a patent even if any profits made go elsewhere, because they stand to derive an intangible benefit from being associated with that patent. - There is nothing wrong with having a conflict of interest, but it must be declared. This allows readers to decide for themselves the relevance of the conflict of interest to the paper’s conclusions. - It is good practice for journals to ensure that they have a clear policy on conflict of interest. - Some journals send the paper back to the original reviewers and ask them to comment on whether the conflict of interest would have changed their opinion on the paper. - Retraction of a paper for an undeclared conflict of interest would only be considered in very serious cases where this seemed to undermine the validity of the data. - The publication of a correction about the undeclared conflict of interest is usually the most appropriate course of action. - The reference to the not-for-profit institution was a red herring as there was a potential for the authors to still receive both financial and non-financial benefits from the product. - The editor should publish a correction stating the nature of the undeclared conflict of interest. - The journal should also consider whether they need to state their conflict of interest policy more clearly.
A new correction was drafted, which excluded the authors’ affiliation to a non-profit organisation. The editors contacted the corresponding author to let him know of the change, why it had been done, and to ask for his approval of the reworded correction. The new correction read as follows: “In our recent article, we failed to declare the following competing interests: A, B, and C are employees of X Institute, which has applied for a patent for the use of drug N in the treatment of haemorrhagic fever. The Institute currently holds patents for the use of N in other clinical conditions, with D listed as an inventor.” The corresponding author said that he would not agree with a declaration that did not mention that his institution is a public charity. He also stated that he felt the guidelines for declaring competing interests posted in the instructions for authors were not explicit enough. The editors responded that the “non-profit” status of the institution was not relevant, potential competing interests might still exist, and that it should be declared. The author refused to consent to a declaration that did not mention “non-profit” and wanted to contact COPE to discuss the matter in more detail. The author also threatened legal action if the editors proceeded to publish the correction without his consent. He suggested a new draft for the text of the correction. In our recent article , we did not declare the patent rights related to N technology. Since they may potentially be viewed as non-financial competing interests, despite the fact our employer is a non-profit public charity, we are now providing the following supplemental information: A, B and C are employees of X Institute, which has applied for a patent for the use of N in the treatment of haemorrhagic fever. X Institute currently holds patents for the use of N in other clinical conditions, with D listed as an inventor.” To clarify the journal’s position, the editors decided to post the following comment on the correction, which the author also agreed to: “[The publisher] thanks the authors for clarifying their competing interests and wishes to make clear its view that even employees of non profit public charities may have competing interests (financial or otherwise) and that it is always best to err on the side of declaring these. However, at the time of publication of the original article, our competing interests policy was not sufficiently explicit on this point. We will, therefore, be updating it in due course.”
Several years after a case series was published, a journalist with serious allegations of research misconduct contacted the editor. These allegations were that: - Ethics approval had not been obtained, contrary to a statement in the paper; and that the reported study was completed under the cover of ethics approval granted to a different study - Contrary to a statement in the paper that the participants had been consecutively referred, they were, in fact, invited to participate. - The children reported were also part of another project funded by legal channels. - The results of the study were passed to lawyers before publication. - An important financial conflict of interest by the corresponding author, which would have influenced the judgments made of the paper at the time, was not declared to the editors. - The journal launched an investigation with the authors' knowledge, and in conjunction with the institution, where the research was conducted. This concluded that the first two allegations were unfounded. But the undisclosed conflict of interest allegation was upheld. Ten of the 13 co-authors subsequently issued a partial retraction and an investigation by the national regulatory body began.
- The failure to declare conflicts of interests is common. - Journals do have editorial responsibility as to how papers are presented. In this instance the journal’s efforts had been overshadowed by the authors’ press conference. - The paper reported preliminary findings, which had been overlooked by the media. - The journal’s retraction had been appropriate. - No action required at present, but the options should be reviewed once the national regulatory body’s investigation have completed.
Some time after a single authored research article was published a journal received a letter pointing out that the same article had been rejected by another journal because of unresolved authorship and acknowledgement issues. At that time the paper had 12 authors. The correspondent said that the single author had a patent application related to the topic of the paper. This was declared as a competing interest in the published paper. The correspondent claims that the single author insisted that all the other researchers sign an agreement transferring all intellectual property rights to him before they could be included as authors on the second paper. The correspondent added that the author had already been investigated by his university for ethical issues related to his research and was also being investigated by a national body. The editor contacted the correspondent asking his permission to put the allegations to the single author and requesting information on the investigations. The author replied that all the other 11 authors had agreed to be “withdrawn” from the paper before it was submitted to the second journal as they had not actually produced the article. However, it was difficult to understand how substantially the same paper had such a different authorship list. Either the initial submission listed authors who did not deserve the credit or the second publication did not list those who did deserve it and should have been accountable for the data.
- If the author is currently under investigation, the national body may welcome further information from the editor. - The first journal should pursue these issues even after rejection as it was likely that they would resurface again elsewhere with another attempt at publication. - The editor should contact the institution and the investigational committee with the allegations. The editor should inform the author of his intentions.
A paper submitted to an international medical journal was reviewed externally and the authors were subsequently invited to submit a revised version. The initial submission included authors from two different research institutions and one author from a corporate sponsor. The initial submission was accompanied by an appropriate description of the individual authors’ contributions, a negative conflict of interest statement, and an appropriate acknowledgement section. When the revised manuscript was submitted, the covering letter stated that to comply with the requested revisions and with the International Committee of Medical Journal Editors’ definition of authorship, the list of authors had been amended, but that this amendment could be renegotiated if the editorial team considered that necessary. The amendment involved the removal of all the authors from the second of the two research institutions, leaving the authors (a husband and wife) from the first research institution and the funder. The second research institution was now only mentioned in the acknowledgement section. The conflict of interest declaration had also been changed in the revision, stating that the remaining authors had filed a patent application for the technique described in the manuscript. The paper was subsequently re-reviewed internally and externally and a further revision of the manuscript was invited and submitted accordingly. The final version of the paper was accepted for publication. At proof stage, the senior author from the second research institution contacted the journal to enquire about the progress of the manuscript. During the course of this discussion it became clear that neither he nor his colleagues were aware that they were no longer authors, nor that the paper had been accepted for publication. On their instigation an investigation was initiated by the appropriate authority at the first research institution, and subsequently by the federal government, because the second research institution had received government funding for the project. On discovering the authorship dispute, the journal cancelled the planned publication and informed the corresponding author that the authorship dispute would have to be resolved before publication could be considered. The remaining authors at the time of acceptance initially refused to cooperate with the investigation and formally withdrew the manuscript. They also requested that the journal should not communicate with the authors who had been removed and should not provide a copy of the revised manuscript to any external party. The journal cooperated with the investigations and released information on the paper despite this request.
- The journal had been correct to cooperate with the institutional investigations despite the authors’ objections. This was an example of the public interest overriding the duty of confidentiality between editor and author. - Usually institutions have significant rights over work done internally. - Some journals write to “removed” authors to double check that they have agreed to this. - Major changes in authorship, including priority of authorship, should prompt journals to question authors further.
A published study reviewed the use of particular devices for performing a clinical manoeuvre. One of the authors worked for a consultancy, but declared that he had no conflict of interest. Subsequently, the journal received a letter pointing out that the consultancy had been set up explicitly to persuade governments and their regulatory organisations of the virtues of new drugs and technologies. This editors felt that this was a conflict of interest that should have been declared, and asked the author if he agreed. The journal’s advice is that if there is any doubt authors should declare an interest as there are rarely any problems taking this approach. The author explained that he had moved jobs after carrying out the research and writing the paper, but before the peer review and publication process had occurred. The author had cited his current address as being the place of contact. The journal is considering publishing a notice of undeclared interests, but should it do more?
Many authors are unclear as to what constitutes a conflict of interest, and a balance has to be struck between exactly what should or could be declared. - The journal does not need to take further action other than that suggested by the editor. - The editor should consider publishing an editorial in the journal to help clarify the issue for authors.
A submitted paper reported on the investigation and management of an outbreak of a disease in a work environment (Company A). The authors acknowledged the referring physician from the workplace—who had declined on legal advice to be listed as an author—and also declared that the lead author had provided medical advice for remuneration to Company A during legal proceedings related to the outbreak discussed in the article. When the article was submitted, the outbreak was the subject of legal proceedings between Company A, where the outbreak occurred, and the company supplying the alleged causative agent of the outbreak (Company B). The lead author had signed a confidentiality agreement with Company A in regards to his/her evidence for the legal proceedings, but not for any information already known to the public through no fault of the author. The author had also added a handwritten addendum, stating s/he accepted the agreement “to the extent that my academic freedom to report findings of scientific and public health importance is not compromised.” On peer review the science of the paper was judged to be sound. The journal’s legal advisor had some concerns about publication; legal proceedings were active; the workplace physician though involved scientifically was not listed as an author; and the paper discussed the outbreak from the perspective of Company A. While the article was cogent and objective about Company A, there was no information about Company B’s knowledge of the outbreak. If the case resolved in favour of Company B, then the article would need to reflect this. The editor wrote to the authors, relaying the legal concerns and informed them that journal, on the basis of legal advice, could not publish while litigation was ongoing. The journal suggested that it would consider a revised version of the manuscript after the case had been resolved. The authors submitted a revised version of the article. As part of the revisions, the authors had deleted all references to the names and locale of the companies. The legal proceedings had been concluded with an out of court settlement; the lead author had no involvement in this. The terms of the settlement are subject to a confidentiality clause and it is not known whether liability was admitted or not. Company A does not wish the paper to be published on the grounds that this would violate the confidentiality agreement between the two parties. On the basis of legal advice from his/her institution, the author states that s/he is not bound by an agreement to which s/he was not party; that the handwritten clause in his/her agreement with Company A allows for publication of the article; and that the details of the outbreak were public as they had been presented in abstract form as well as briefly described in a local language publication. The lead author feels that the journal’s reluctance to publish on the basis of legal concerns is flawed. As originally relayed to the author, it was stated that the journal could be seen as “taking sides” in an ongoing legal dispute—a view that the author feels is “ethically unacceptable. ” Company A is threatening legal action against the authors if details of the case are published, and Company B would also potentially have an action for defamation. What should be done?
_ Editors have to be mindful of legal advice against publishing a paper and could base their decision not to publish on strong advice. _ In such a situation where legal advice against publishing was so strong, the author would have to indemnify the publisher against any legal action that might be taken. _ If an editor has been advised that publishing something was unlawful, prima facie that was immoral and publishing the article could potentially be held to be in contempt of court. _ In such a case then the author would also need to provide indemnity for loss of reputation. _ If the author’s institution has stated that, having taken legal advice, they are happy to publish, then the work could be published on the institution’s website. _ Publication would only be a possibility if both companies agreed to it.
A letter was published that provides guidance on prescribing a particular drug in children. There are anxieties about the use of this drug in children, and sometime back a letter from essentially the same group on the same subject was published in the same journal. The electronic version of this original letter included a conflict of interest statement, but the paper edition did not. This was a mistake. Unfortunately neither the paper nor the electronic version of the new letter included the conflict of interest statement. It clearly should have done, not least because it seems that one of the authors of the current letter received funding from the manufacturers of the drug. The intention is to go ahead and gather conflict of interest statements and publish them in both the paper and the electronic versions of the journal, but the lead author of the second letter seems to be opposed to this move. The journal plans to override his objections. Does COPE agree with this? A further issue raised by the second letter is that the third party wrote to say that three of the authors of the letter do not support everything that is contained in it. Wouldn’t most people who read a piece that is signed by many authors believe that all authors support what is published unless it specifically states otherwise? What action should be taken on this issue?
The exclusion of the conflict of interest statement from the paper version was the fault of the editorial process. Statements were included with other published articles but not “letters”.