Authorship dispute and possible unreported protocol amendment

Case number: 
16-14

Case text (Anonymised)

Our journal accepted a randomised controlled trial for publication which has not yet been published online. In the submitted paper, the randomised controlled trial is described as commencing in 2004 with completion in 2011. We have received an email and telephone call from an individual not listed as an author or reviewer of the paper with the following alleged disputes:
• He was an investigator on the trial between 2004 and 2008.
• He analysed the data from this study in 2008 which were published as an MD thesis in 2010 with the institution at which the study was carried out with the trial protocol possibly changed after that time. The results were submitted to a journal but the submission was rejected and was not submitted elsewhere.
• He is not listed as an author of the manuscript although his contribution to the study has been acknowledged in it.
• The data collected up to 2008 were unblinded for the purpose of then data analysis before the full data up to 2011 were fully analysed.
• There is a dispute as to whether some subjects at the time of the assessment of the primary outcome measure should be included in the data analysis.

Steps taken by the journal so far:
1) We have written to the corresponding author of the manuscript to inform him about the receipt of this complaint and the duty to investigate.
2) The publication process for this manuscript has now been suspended pending the outcome of this investigation.
3) We have requested from the complainant the following documents: original trial protocol in 2004 in order to compare it with the registered version; MD thesis published in 2010 in order to compare its results with that reported in the currently submitted manuscript; draft manuscript of the results of the study submitted to the other journal which was submitted in 2010 but rejected for publication for comparison of its results with that reported in the currently submitted manuscript.
Next steps
4) We will seek confidential input of peer reviewers to provide their feedback on alleged discrepancies, when additional information is received.
5) If we find that the above discrepancies appear to be valid, we will contact the author of the current submission and ask for clarification of the connection between the two works and a response to any discrepancies found.
6) In light of the above, we will revisit the decision on manuscript acceptance.
7) If we conclude that any confirmed concerns are serious and unfixable through revision of the manuscript, we reserve the right to reject the manuscript.
8) The publisher and its publishing partner will be notified so that its legal department has prior notification of the case.

Question(s) for the COPE Forum
• Is the journal taking the right steps?
• Is there any further action required?
• If the current authors amend the manuscript and authorship appropriately in light of any discrepancies found, we intend to accept the paper. Is this ok?
• In the above situation what is our obligation with respect to informing the authors’ institutions?

Advice: 

The Forum was updated that the journal has received the documents requested in step (3) above. The journal did not involve the peer reviewers—the editors reviewed the materials. On reviewing the documents, the editors found discrepancies and have contacted the authors for an explanation.

The Forum cautioned about the editors getting involved in what is essentially an authorship dispute. There may be a limit to what the journal can do. The recommendation was to refer to the institution, in a neutral way.

The Forum discussed the journal policy on trial registration. As the trial began in 2004, technically the authors are not required to prospectively register it. Some journals ask for the full trial protocol (not the one on clinical trials .gov), as it is usually more detailed. The journal may want to consider doing this to avoid a similar situation arising in the future.

One other suggestion was to ask for documentation from the ethics committee that approved the study. The ethics committee should also have evidence of any protocol changes. If there are any discrepancies, these can also be brought to the attention of the institution.
 

Resolution: 
On-going
Year: