Inadequate reporting of a trial, despite earlier rejection from a different journal
Case text (Anonymised)
We have been contacted by a reviewer after he spotted a paper he had reviewed for us (journal 1) now published in a second journal (journal 2). Both journals are members of COPE. The reviewer had advised we reject the paper when it was sent to him to review in September 2008. This was based on his assessment of the paper and also the supplementary material he was sent by us: protocol, CONSORT statement and trial registration details. Seeing the paper now published in journal 2 (April 2010), he still has the same concerns about the conduct of the trial and validity of the data presented—for example, the study was not really randomised (subjects were allocated alternately to treatment or control and participants’ trial numbers were known to the outcome assessors)—and reporting of the outcomes was inconsistent with scores given in the raw data.
Following the original rejection from journal 1, the authors appealed against our rejection and we declined their appeal. The letters from us made it clear that we had concerns about the methods and reporting, and both external reviewers’ reports were available for the authors to see.
So in spite of making it clear that there were problems with their paper, the authors still went ahead and submitted to journal 2.
When contacting us, the reviewer wanted to know if he could contact the editor of journal 2, make it known that he had seen supplementary material when he had reviewed the paper and explain his concerns to the editor. We advised him to write to the editor of journal 2 in confidence, explain what had happened and say that we had reservations about breaching confidentiality in this way but thought that the benefits of doing so outweigh the risks.
The editor of journal 2 has now contacted us asking to see the supplementary material. Unfortunately, we no longer have the files available (we do not archive them that long) but we do have the reviews.
Questions for COPE
• Should we share with journal 2 the reviews we still have?
• In principle, could we have shared the supplementary files and original submission (although they are no longer available)?
The Forum’s advice was that the journal should not share reviewer comments with another journal without first asking the reviewer’s permission. However, it is quite acceptable for the reviewer him/herself to decide to send their comments to another journal if they think this is the right thing to do (since the review belongs to the reviewer). Most people felt that the material submitted by the author to the journal (ie, the protocol) should not be sent to another journal, as this is privileged/confidential information. It is up to the second journal to contact the author and ask for this material. If the author does not respond, then the journal that had actually published his work (journal 2) should contact the author’s institution.
The editor of journal 2 contacted the author of the paper who provided some, but not all, of the answers which he sought. So, after discussions with the reviewer and a member of the editorial board, they decided not to dig deeper with the author but instead publish the letters that had been sent to the journal about the paper.
The editor of journal 2 published six letters in total: two letters about the paper with replies from the author, and a further letter about the commentary accompanying the paper, together with a reply from the author of the commentary.