David Resnik, a National Institute of Environmental Health Sciences (NIEHS) bioethicist who has published on retractions and corrections, is apparently hoping to do so again. In a request for information that went out from the NIH last week, Resnik seeks:
1. Information about the journal’s written policies (if any) concerning retracting or correcting published papers.
2. Information concerning disputes involving authors or editors concerning retractions or corrections in papers published in the journal in the past five years. Comments can also briefly describe the issues at stake.
3. Any other comments concerning issues related to retracting or correcting papers.
“Responses are invited from editors of scientific journals,” he writes, noting:
The comments collected will be analyzed for the purpose of publishing an article on issues related to retractions and corrections. All responses will be kept confidential and only general information will be published. Despite this, proprietary, classified or confidential information should not be included in your response.
The deadline is May 3. Find out more here.
Like Retraction Watch? Consider supporting our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, and sign up on our homepage for an email every time there’s a new post.
I would not imagine journal editors are the best people to ask for such information, given their rather huge conflict-of-interest in these matters. Perhaps if the RFI were extended to those of us outside the role of editor who nevertheless have significant experience in this arena, a more balanced set of information for this study may become available? Of course, good luck verifying this with the journals, some of whom will simply deny there’s a problem or refuse to respond to your contact asking for verification. Here are some of my experiences, for which full details can be made available on request…
1) Currently I have a complaint in to COPE regarding a matter of alleged misconduct, wherein the case was sent to a sub-editor at the journal. Response of the journal was “no fault found”. Sub-editor trained in the lab’ of a trainee of the senior author. Journal is currently incommunicado, even for purposes of verifying receipt of new allegations about the paper.
2) Different journal, alleged manipulation (gel splicing). Journal acknowledged that splicing occurred and this does not follow “best practices”, but since paper was published before they had a policy on this, the journal will not apply the policy retroactively to issue a correction. Journal considers the matter closed. Further evidence provided, showing some items on left and right side of splice lines are duplicated, but now journal is now incommunicado, even for purposes of verifying receipt.
3) Rejected a paper I was sent to review. It showed up a few days later revised. I rejected it again, and was told by email from editor that final decision was reject. A few weeks later paper was in press, with an accept date 5 days after editor’s email to me. Editor refused to enter into conversation regarding what happened between telling me it was rejected, and then accepting it.
4) Journal replaced supplementary information in a paper which was documented to be problematic. No correction notice or any other indication that the problem was ever there. Journal refuses to answer emails regarding this matter. PubMed also silent on lack of notification.
5) Failure of a journal and its entire editorial board, and the leadership of the scientific society that governs it, to respond to a multitude of contact requests regarding alleged misconduct by one of their former associate editors. COPE CC’ed on emails, refuses to act.
6) Paper was corrected, but the correction itself contained questionable data. Paper was subsequently retracted with a totally opaque notice (“withdrawn by authors”) despite widely documented data problems. Completely at odds with journal policy on such matters.
7) Found problem data in paper for review, made allegations, journal investigated and found no problem, considers matter closed, paper published anyway. Problem data still there in the paper. Journal refuses to release any details on exactly what the investigation entailed. Editor claims to have seen original data, but will not release it so it can be verified by the person who originally made the complaint (me).
8) Flying in face of COPE (and other) guidelines on simultaneous submission, 2 papers submitted from the same lab’ a week apart contain a 6 panel figure reproduced in its entirety including the legend. Both journals published “corrections” merely pointing out the similarity.
9) One of our papers was held up in review for 12 weeks (at a journal claiming 11 days average turnaround time). One reviewer referred to a paper that was not even in press at the time we submitted. Our paper was rejected on 2nd round, largely due to comments from same reviewer. Then a month later the author of the referenced paper published data which would have been impossible to publish if our paper had been out first. We suspected this person was the reviewer, (they were delaying and citing their own work) and have since confirmed this via a 3rd party.
10) Paper came out in big journal making bold claim, but with numerous errors. I wrote letter to editor with data, but journal refused to publish it – “we don’t publish things intended primarily to refute our already published papers”. I went back and assembled a who’s who of the field (10 authors), got more data from knockout animals refuting claims. Journal took 1.5 years to publish our letter, cutting it by a page at the last minute, and allowing authors the last word in a response letter that we were not allowed to see during editing. Our findings have since been validated by other labs.
11) One of our publications in a glamor journal took 2 years from first submission to acceptance, with addition of >20 figures of supplemental data. During this time, rivals (suspected reviewers) published data with “surprising similarity” in low impact journals, thereby scooping us. During final stages of our paper being reviewed, journal pointed out their “right to reject even at this late stage, since the findings may have lost priority during the lengthy review process”.
12) Reviewed a paper, found fabricated data, reported it to editor who rejected it. Paper showed up in press in another journal a few months later, with all the fabricated data panels removed, but addition of a new fabricated image elsewhere. The author list was identical between the two papers. If you got a paper rejected for fabrication, wouldn’t you remove person responsible from subsequent versions? The 2nd journal refused to investigate concerns about the fabricated panel unless I was willing to have my real name passed on as correspondent. Paper is still out there with problem data. Lead author has ad-hoc’ed on panels that review my grants, so I do not wish to push the matter further.
I could go on, but you get the picture. You’re unlikely to hear any of the above from the rather protected stance of journal editors, but this is the kind of s*** that regular scientists have to deal with every day. This is the kind of behavior we have come to expect from the “gatekeepers” of scientific knowledge, to whom we pay thousands of our grant dollars to every year.
So, good luck with your study Dr. Resnik. So long as you’re only soliciting one side of the story, I doubt it will provide much illumination to the dark underbelly of the scientific publishing world exemplified by the above.
Considerable bitterness transpires…
A simple suggestion: Why doesn’t Paul Brookes get a job at NIH so he can pursue these topics full time and without fear of losing his job? Clearly, that is where his real passion is.
In light of the comments on this thread, and in the interests of balance, Dr Resnik may like to consider putting out a call for a second, parallel study with responses from authors, and those with an interest in such issues, only?
Wonder why the main bioethics folks at NIH — ie, http://www.bioethics.nih.gov — are not the ones doing this and whether the info will be shared.
David Resnick is a long-time, highly respected investigator and author on issues of integrity in research.
see his CV including publications at http://www.niehs.nih.gov/businesscards/docs/resnik_david_cv_508.pdf
In my experience at NIH / PHS from 1987-2006, the NIH Bioethics Office in the NIH Clinical Center was
narrowly focused on patient-related issues of medical ethics and related research. Dr. Miriam Kelty,
Research Integrity Officer at the National Institute on Aging, hosted for much of that period a more
global seminar on all kinds of research ethics issues, including misconduct in basic and clinical research.
This is slightly tangential, but I thought readers would be interested in this paper by Resnik on Research Misconduct that I commissioned as part of a series http://www.plosmedicine.org/article/info%3Adoi%2F10.1371%2Fjournal.pmed.1001406 …
That paper is not tangential at all. I am actually extremely concerned about the quality of that paper you have pointed out.
A close rexamination will reveal the following almost contradictory comments:
“Most high-income countries have developed policies and initiatives to address research misconduct” (Summary)
“Many high-income countries (HICs) have developed policies and initiatives to address misconduct” (first paragraph)
“different countries have adopted a range of policy instruments and governance mechanisms to preserve and oversee research integrity” (first paragraph)
“Although most HICs have developed comprehensive national policies, it is important to realize that many have not” (first paragraph)
The study then lists information about 5 HIC cases: the US, Canada, UK, Denmark and EU (as represented by the ESF), thus 5+28 = 33.
There are currently 76 HICs (http://en.wikipedia.org/wiki/High_income_economy). I cannot understand a) the importance of this unquantified paper, or b) why it was commissioned when the study is so superficially incomplete. Would you be willing to provide access to the “peer” comments on this paper that has failed to look at the issue in any quantifiable depth? I assume that if you commissioned this paper that you would be the acting EIC, or similar? I would be quite happy to write dozens of such loosely unquantified papers if I could get the paper published in an IF = 15.253 (http://en.wikipedia.org/wiki/PLoS_Medicine)
Perhaps tangential to my comment is an interesting analysis about PLOS One:
http://scholarlykitchen.sspnet.org/2014/03/07/plos-one-output-falls-following-impact-factor-decline/