Data fabrication

In November 2014, the first author of a decade old paper in our journal and a 15-year-old paper from another journal informed us that he faked the data in two figure panels in the paper in our journal and one figure panel in the paper in the other journal. The main gist of the manipulation was loading unequal amounts or delayed loading of gel lanes.

Self-admission of data falsification is a serious charge that is difficult to disprove, and we felt a challenge to identify evidence to counter or support this type of allegation. As general guidelines, we felt there were three types of evidence that could help resolve the standoff:
(1) compelling original raw data with evidence for or against unequal or delayed gel loading;
(2) verified replication already existing within the published literature; and
(3) as a last resort, a replication study performed by a wholly independent laboratory.

In December 2014, we asked the first author to contact the corresponding author of both papers and the institute, but he refused. We informed the first author that we would inform the corresponding author of the papers and this might result in violating his confidentiality. In January 2015, we informed the corresponding author that we had received self-admission of fraud from the first author and asked the corresponding author to retrieve original raw data for the figures in question and provide them to us. We also urged the corresponding author to engage the institute ethics committee and get in touch with the first author in gaining a deeper understanding of the challenges.

In February 2015, we spoke with the corresponding author by telephone. The corresponding author did not believe that the first author had faked the data. We discussed potential ways to counter a compelling self-admission and agreed that the corresponding author would provide us with the raw data by the end of March 2015 and would inform his institute.

In March 2015, we were contacted by his institute. On request, we provided the contact information of the first author to the institute’s investigation committee after obtaining permission from the first author. The corresponding author told us that he has located all of the raw data. In contrast with what we agreed by telephone, he told us that he would not be sending us the raw data directly but would pass them to the committee. The first author provided both us and the committee with data that he said was contemporaneously produced and showed a different result from what was published, that he said was without manipulation with gel loading and showed no experimental effect.

In April 2015, we asked the committee to share their investigation results with us and also asked the corresponding author to provide the copy of the raw data to us. But the corresponding author and committee refused to share any useful information with us. The committee told us by email that they have an obligation to protect the corresponding author’s reputation.

In May 2015, we spoke with the research integrity officials of the institute by telephone and they agreed to share more information with us. In early June 2015, we received a summary but not the full version of the committee report, which cites that no further action is warranted because the evidence they gathered can neither support nor refute the first author’s self-admission. We were not provided with any of the original data. The summary report included information such as promotion schedules of both the first and corresponding authors, but these seemed irrelevant to us. We felt it essential that we have access to the full scientific information on which the committee based its recommendation.

In late June 2015, the institute shared with us the full report of their investigation. We were able to understand from this that their conclusion that no further action is warranted was based on the fact that there was no recorded falsification in the laboratory notebook. We feel this reason is not sufficient to counter self-admission of fraud as someone who intentionally fakes data would not likely record it in their laboratory notebook. We therefore were unsatisfied with recommendation to take no further action.

In July 2015, we interviewed the first author via Skype and asked him to describe again how he generated the data and how he intentionally manipulated the data to fake the results. What he described over Skype was consistent with what he has described to us in previous correspondences.

In July 2015, we also spoke with an institute senior official and explained again that only contemporaneous data collected by the first author, contemporaneous data collected by other members of the laboratory, or direct replication of the data by an independent laboratory reported in the published literature would be necessary to counter the first author’s self-admission of fraud. Since none of these avenues turned up evidence to counter the self-admission, we suggested that the experiments in question could be independently repeated by a third party or the paper will need to be retracted.

In August 2015, the corresponding author agreed to proceed to have the data in question independently repeated by a third party. We are now instructing the corresponding author to reach out to a laboratory to start repeating the experiments. While he agreed in principle, the corresponding author is dragging his feet and we are uncomfortable sitting on a serious allegation and eager to move forward with a resolution in a timely and responsible manner.

Question(s) for the COPE Forum
• What are COPE's recommended procedures for handling self-admissions of fraud?
• What is the journal’s responsibility when one author self proclaims fraud and another author says no fraud occurred?
• What is the responsibility of the journal if the journal thinks an institutional investigation was not evidence based.

In 2014 we received a communication from the Research Integrity Officer of an academic institution informing us that a paper, published in our journal in 2013, included falsified or fabricated data. We were informed that, following an investigation, they had determined that scientific misconduct had occurred.

Within a few days we received a communication from one of the authors of the paper (who is no longer at the institution) reiterating this assertion and providing some further explanation; that a former student had fabricated data and that it affected the paper (but providing no specifics).

Over the next week or so, other journals by the same publisher received similar notifications from the same author. Initially, we were presented with no information regarding who the perpetrator was or the specifics of the affected data. We were therefore unable to determine how severely affected the validity of the overall paper was and whether a retraction or correction was necessary.

Our initial response was to request further information from the institution and the author. Initially, we were informed by both parties that, as a result of Federal privacy laws, they were unable to divulge any details pertaining to the investigation, aside from what they had already told us. In the meantime, we decided to publish an expressions of concern on all four papers affected by our publisher with identical notices detailing what we knew for certain and stating that we would seek further details from the institution.

Sometime later we heard back from the institution providing further specific information (ie, outlining the fabricated data) for three of the four papers. Of these three papers, two are now in the process of being retracted, while an academic editor has been consulted to advise on whether the third should be retracted or corrected, based on the additional scientific information now available.

However, in regard to the fourth paper, published in our journal, we were told by the institution that no further information was available. The author who contacted us has not provided any specific information either. Therefore, we find ourselves unsure of how to proceed next, as we still do not know to what extent the conclusions in the paper are valid.

Question(s) for the COPE Forum
• Should we proceed with a retraction but simply state that we cannot provide further information (something we feel is unsatisfactory for our authors)?
• Should we instead leave the expression of concern online but update it to say that we will not be able to provide any further information?
• Does the Forum have any other suggestions?

In June 2014 we received a manuscript by four authors from a well known research institution. They described a randomized trial comparing a variation in a procedure with standard care. In total, 200 patients were randomized, 100 to each arm. As measured by an interview, patients undergoing the new procedure were statistically significantly more content than those in the control arm. This manuscript was submitted 116 days after the same group of authors had sent us a first manuscript on the same topic.

The first manuscript, however, described an observational study: 50 patients had chosen the new procedure, 50 underwent conventional treatment. The patients rated the new procedure higher (statistically significantly). In their discussion, the authors mentioned as a limitation the non-randomized status of their study and called for a randomized comparison. At the time we rejected the manuscript because we were not convinced by the non-randomized design of the study. The senior author appealed our decision saying that it was very difficult and almost unethical to carry out a randomized trial. We did not change the decision but I granted the author that we would evaluate and possibly review a manuscript on a randomized study.

In the cover letter of the second manuscript, dated June 2014, the authors referred to this discussion and stated that 100 patients had been randomized to each group. [As an aside, in an online source detailing procedures carried out in the department of the authors, the procedure in question is said to have been performed more than 1200 times a year. As a consequence, it is conceivable that the authors have randomized 100 patients to each study arm during a period of 3–4 months. In his appeal to the rejection of the first manuscript, the senior author mentioned that the ethics committee had already expressed approval. And yet, common experience with randomized trials indicates that the present study would be an extremely fast trial regarding screening, consent, inclusion, examination (4 days after the procedure), and analysis.]

Here is the problem: the results are identical in manuscripts 1 and 2. In numerical form the results are only presented in tables (not in the main text and not in the abstract). In all three tables, the values are identical in both manuscripts. All three tables were submitted as one file, leaving open the possibility that the author mixed up files. The figure (a horizontal, stacked bar chart) is slightly different but the numbers indicating the results, however, are identical. This figure was submitted as a different file.

The main text of the second manuscript is identical to the first one except for minor updates in relation  to the numbers of subjects and study design. All four photographs illustrating the procedure in both manuscripts are identical. The reference list is identical.

I can think of only two ways to make sense of this submission: sloppiness or fraud. Under the sloppiness  assumption, the authors would have submitted a text referring to their randomised control trial and tables, referring to an earlier observational study. This is conceivable mostly because it is hard to imagine that authors believe they can get away with submitting the same data in two manuscript describing two completely different trials and separated by only 4 months. On the other hand, the tables and figures differ in layout and several details from those submitted with manuscript 1. If sloppiness is not the reason, it must be fraud, and we can only reject the paper.

I feel we should be frank with the authors about our decision to reject the paper. Confronted with this decision, however, the authors have no incentive to cooperate with us and to send us, for example, original data. Rather, they may blame the mess to an unfortunate confusion.

Question(s) for the COPE Forum
(1) We have decided to reject the paper but how does the Forum think we should now proceed?

Journal A has accepted a meta-analysis for publication. As is standard practice for many articles accepted in this journal, a key expert (Professor X) in the relevant field was invited to submit a commentary on the paper. Professor X expressed concerns to the journal that “we believe that some of the papers included in the review could be either fabricated or at best are heavily plagiarised”. The papers included in the meta-analysis are all primary studies published in peer-reviewed journals.

Journal A requested some evidence for the concerns raised by Professor X.

Professor X has already tried to investigate the potential research misconduct of the primary studies. He sent a comparison of five studies, three of which were included in the meta-analysis accepted by journal A. Professor X claims strong evidence of plagiarism, and questions whether the trials took place at all. He also notes that he has previously written to the authors of the trials but says that few have responded. Those that did respond, he believes, have failed to provide reassuring responses.

Example response from authors sent to Professor X include the following: “The work has been actually undertaken after proper clearance. And details of the same are available with the competent authority.” “We don't want to be get disturbed as I discussed with our main author.” “Excuse us..Bye”.

Journal A has now halted publication of the meta-analysis.

The editors of journal A are unsure how to proceed, as the potential research misconduct lies with research not submitted to the journal, but rather primary studies included as part of a meta-analysis submitted based on the “available data”.

Question(s) for the COPE Forum
(1) How do we establish whether or not the primary studies are fabricated?
(2) Is it journal A’s responsibility to pursue this investigation or should it be the responsibility of the journals in which the primary studies are published?
(3) How should journal A proceed with managing the meta-analysis accepted for publication?

A sixth year medical student, with expected year of graduation of 2013 (Mr X), submitted 29 original articles and 17 letters to the editor in the period February 2012 to October 2012 to our journal. This amounted to an average of five submissions per month. Mr X is an author and corresponding author in every article. Of these, he is the first author of eight original research articles and 12 letters. In the remaining one he is a co-author. The articles are on very diverse subjects.

This set us thinking that, apart from his clinical work and studies, how he had time to conduct research, analyse the results and write the articles.

The journal first wrote to Mr X for the necessary justification. He responded promptly, “I am one of the best researchers of my country and have multiple publications in every field of medicine and have won multiple prizes”. He provided a list of 72 publications to his credit. He also provided the name and email of the chief of the research committee of the university.

We wrote to the concerned parties asking them to endorse the submissions as being ethical and valid for the purpose of publication. The chief replied that Mr X was a member of the student research committee with some research background in medicine which led to multiple awards and publications. He confirmed the research background in a vague manner and there were no more comments or endorsements of the submitted articles.

We then wrote to the vice chancellor of the university asking for verification and endorsement of the articles according to the ICMJE guidelines. The director of research affairs was also approached, who asked for details of all the articles submitted. These were duly sent.

In the meantime, Mr X contacted us stating that his e-mail had been hacked and someone else had sent letters and articles with his name. This was incorrect, as all mails had the same e-mail address. We also sent an email to the Publication Commission in our country on 6 March 2013. There has been no response.

We face a dilemma. The articles are lying unprocessed. It is a mystery as to why the higher authorities are not taking any action or replying to our emails.

Question
What would the COPE Forum suggest we do?

A scientific paper was submitted in January 2011. After initial assessment by the journal’s editor-in chief, it was allocated to one of the co-editors. By chance, the co-editor had reviewed the manuscript for another journal only a few weeks before. The manuscript had been rejected by the previous journal for a number of methodological flaws.

The resubmitted manuscript contained significant changes to both methodology and results apparently correcting the flaws noted in the previous reviewers’ comments. Realistically, these changes could have only been possible if the study had been repeated but as only a few weeks had elapsed since the previous rejection, the editor suspected fabrication of results.

The editor contacted the editor-in-chief to highlight a possible case of misconduct. The manuscript was rejected primarily because it was not of a sufficient standard to merit publication and also because of concerns regarding the possible falsification of results.

In his letter to the author, the editor-in-chief asked for an explanation of the differences between the two manuscripts. The reply claimed that additional patients had been recruited. Of note, the demographic details (including age, height and weight) in the revised paper were identical to the previous submission, making this explanation unlikely.

In view of the unsatisfactory response from the lead author, a letter was sent to the dean of the faculty of medicine at the author’s institution but no reply has so far been received. Information regarding the academic department in this university has been difficult to obtain as their website is unhelpful so it is not clear if this letter was received by those in authority. Consequently, after several months and following discussion with the journal’s publisher, it was decided not to pursue this enquiry further.

Recently, another manuscript from the same department has been submitted. The co-author of the previous paper is listed among the five authors of this new submission. It is not clear, however, if this author was complicit in the previous case of alleged misconduct. The lead author of the previous paper is not included.

This paper has been reviewed and is again of a poor standard that does not warrant publication. There is, however, no reason to suspect misconduct with this current study.

The editor-in-chief has so far not responded to the authors regarding publication but has asked for the contact details of the head of the academic department from the lead author. An email reply included a contact name. A letter to the academic lead asking for clarification of the previous submission has been sent. A reply is awaited.

At present it is impossible to establish whether the previous misconduct is the result of a single rogue researcher or an institutional problem with research governance. The failure to receive an adequate explanation may simply be a problem with contacting somebody in authority who can investigate the conduct of research within this department. It is however, difficult to believe that a co-author of a paper where misconduct almost certainly occurred was not aware of such behaviour.