By David Tuller, DrPH
Last week, Brian Hughes and I sent a letter to Occupational Medicine, which recently published yet another of Professor Trudie Chalderâ€™s awful papers. Among other problems, Professor Chalder and her four co-authors completely misstated their own findings in the text of the paper. We called for retraction of the paper.
In the past, I have preferred to send such letters directly to a journalâ€™s editor or editorsâ€”as a way of creating more attention and perhaps triggering some action. My interest has been more in prompting editors to address major issues than in having my letters officially published. In this case, Occupational Medicineâ€™s masthead does not include e-mails for specific editors, so we sent our letter to the journal through the formal submissions process. We also posted it on a pre-print server.
On Monday, we received an e-mail from an assistant editor at Occupational Medicine. They relayed a request from the editor-in-chief that we trim our letter from 1326 words to the strict maximum of 500 words and re-submit it. The editor-in-chief also suggested that we send the full, unedited version of our letter to the paper’s corresponding author–i.e. Professor Chalder.
This response did not seem commensurate with the gravity of the concerns Professor Hughes and I raised. This paperâ€™s mangled percentages were submitted by an experienced team of investigators from Kingâ€™s College London, including three professors. The article passed peer-review and was published by the journal, which is sponsored by the Society of Occupational Medicine. Shouldnâ€™t an editor-in-chief be aghast that this happened on their watch?
Instead of promising to investigate the matter, the journal offered us the chance to publish a bare-bones letter. The response we received provided no indication that further steps would be pursued. The journal also placed on us the burden of informing the authors that they submitted a manuscript with disqualifying flaws. But delivering the bad news to Professor Chalder and suggesting a course of remedial statistics is the journalâ€™s responsibility, not ours.
Professor Hughes and I therefore have sent the following response to Occupational Medicine.
Dear [name redacted]â€”
Thank you for your message. We understand the Editor-in-Chief’s request, but we wish to decline that suggestion.
Our concern is not with having a ‘Letter to the Editor’ published in Occupational Medicine. Rather, our intent is to formally notify the Editor-in-Chief of significant technical problems with the paper by Stevelink et al. that was recently published by the journal.
As outlined in our letter, the paper contains basic technical statistical mistakes, inaccurately stated results, and other major lapses. Our main goal in writing was to urge the Editor-in-Chief to investigate the issue and ultimately retract the paper, not to engage in an exchange of opinions with the original authors about why they submitted a problematic manuscript.
We believe it is the journal’s role, not ours, to pursue these issues with the authors (and, possibly, with the peer reviewers who recommended the paper for publication). It is for the journal to determine how this paper came to be published, and to inform readers of what is being done to address the matter.
To repeat, we are not concerned with whether or not our letter is published in the journal. This is a formal correspondence with the Editor-in-Chief concerning our call for retraction of the paper by Stevelink et al. We therefore ask that you forward the letter again to the Editor-in-Chief for his urgent attention.
We look forward to seeing prompt action on the concerns we have raised.
National University of Ireland, Galway
University of California, Berkeley