About these ads

Retraction Watch

Tracking retractions as a window into the scientific process

After retracted study’s cited, editors ask, “time to add scientific integrity to the downside of print on paper?”

with 8 comments

JneuranesAs we — and others — have written, retracted articles don’t necessarily creep off to some little island somewhere never to be heard from again. After all, the electronic versions of about a third of retracted papers aren’t marked as retracted. Sometimes, like Napoleon, those papers return from exile to wreak havoc: They get cited as if they had never been retracted.

To wit: The Journal of Neurosurgical Anesthesiology has published a letter to the editor regarding a 2012 article by a group of Italian researchers. The topic of the paper in question was “Perioperative pregabalin for postoperative pain control and quality of life after major spinal surgery.”

That happens to have been an area of interest for one Scott Reuben, a Massachusetts anesthesiologist and pain specialist turned federal inmate who was, for a brief while anyway, the unofficial record holder for most retractions by a single author. Before he was eclipsed by another anesthesiologist, Joachim Boldt. Before he was eclipsed by a third anesthesiologist, Yoshitaka Fujii. (In fact, we can’t really say Reuben was the record holder because, although he has more than 20 retractions to his name, we started this blog in the wake of his misconduct and so have a rather limited time horizon.)

Sure enough, the Italian group cited one of Reuben’s retracted papers on the treatment of pain after spine surgery, a 2006 article titled “The analgesic efficacy of celecoxib, pregabalin, and their combination for spinal fusion surgery,” which appeared in Anesthesia & Analgesia.

That prompted the following letter from David Polaner, an anesthesiologist at the University of Colorado:

I was very interested to read the paper by Gianesello et al on the use of pregabalin to attenuate postoperative pain and improve the quality of life after spine surgery. I am surprised and concerned, however, that one of the citations in the manuscript (and one that is commented on at some length in the discussion) was a paper by Reuben and colleagues that has been retracted. The increased visibility of fraudulent, falsified, or unethically conducted studies appearing in the medical literature in recent years has been sufficiently discussed and needs no further comment; however, we as authors, investigators, and editors need to be vigilant that we do not taint our own properly conducted research by citing the work done by others that has been discredited.

In response, the editors of the JNA — James Cottrell and John Hartung – wrote of a key weakness in the defenses against these invidious returns: Paper journals on bookshelves and in library stacks that gather dust but don’t update.

We agree completely with Dr. Polaner’s concern, and we wish to raise an additional concern: does the accelerating rate of retractions mean that readers (authors, investigators, editors, reviewers, and regular readers alike) should stop accessing medical journals by pulling off printed versions from the shelves? Unlike PubMed versions, print versions are not accompanied by retraction notices or reference to follow-up commentary. In addition to environmental concerns, is it time to add scientific integrity to the downside of print on paper?

Perhaps. But while we’re not arguing for killing more trees, it’s not clear how the Italian group accessed the retracted Reuben article. If it was indeed a paper copy, a watermarked electronic version — which this one was — wouldn’t have done much good. But if it was a PDF downloaded before the study was retracted, just to sit in someone’s personal repository, that wouldn’t have helped, either, whether or not a dead tree version was buried in the stacks.

At the end of the day, it’s not the presence of an outdated version that leads to problematic citations, it’s whether or not authors check the most recent electronic version — a task with which CrossMark, as we’ve suggested elsewhere, can help.

About these ads

Written by amarcus41

December 13, 2012 at 8:30 am

8 Responses

Subscribe to comments with RSS.

  1. Every paper copy of a journal should come with a little pack marked “Retraction Use Only”. Then when a paper is retracted the publisher emails librarians and tells them to break the seal, in which they will find a red pen.

    Maybe.

    Neuroskeptic (@Neuro_Skeptic)

    December 13, 2012 at 8:39 am

  2. The problem of persistence of error in the literature is a real problem. Researchers download copies of articles and save them to their personal librarians. There is little incentive to go back and check whether the article was corrected or retracted at a later date. There are hundreds of copies of retracted articles freely available on non-publisher websites; unfortunately, many of them are available in PubMed Central. The PDF version of these articles contain no indication that they were retracted. See:

    Davis PM. 2012. The persistence of error: a study of retracted articles on the Internet and in personal libraries. Journal of the Medical Library Association 100: http://dx.doi.org/10.3163/1536-5050.100.3.008

    Phil Davis

    December 13, 2012 at 8:49 am

  3. I believe you meant to say “wreak” havoc, not “wreck”.

    vhedwig

    December 13, 2012 at 9:24 am

    • Fixed — thanks.

      ivanoransky

      December 13, 2012 at 9:33 am

  4. I don’t know, it is not my field, but the real problem here seems to be that the authors (and reviewers and editors) were not up on the literature. They should have know it was retracted if they are claiming to be working in the field. The literature is full of misinformation, retracted or not, and a healthy dose of scepticism and obsession about what is really known never hurt anyone, even a medical scientist.

    Dave

    December 16, 2012 at 6:36 pm

    • Exactly. Additionally, a good editorial system will automatically check the references of a submitted manuscript. PubMed supplies retraction information in XML format, so this would be a trivial thing to cross-check. So I assume that the Journal of Neurosurgical Anesthesiology simply does not have a good editiorial system.

      Bernd

      December 19, 2012 at 6:26 am

  5. Surely the onus should be on the authors of any article to verify their sources and discount anything retracted. Editors should assist this process by insisting that those submitting articles must certify they have scrutinised all references and none have been retracted.

    Dr Mark R Nel

    January 2, 2013 at 9:02 am

    • A good idea. Maybe “none has been retracted as of the date of manuscript submission.”

      Now, where should researchers check? I tell graduate students to check the URL of every on-line source, to be sure the item is still available and the URL is still correct, on the day before they submit the final version of a thesis/dissertation for copyrighting and archiving. But several commenters on RW have reported that journals do not always flag the retractions on the journal’s web site. Some retracted papers have “RETRACTED” splayed across the page and others do not. Where should a person look to find a complete list?

      JudyH

      January 2, 2013 at 10:37 am


We welcome comments. Please read our comments policy at http://retractionwatch.wordpress.com/the-retraction-watch-faq/ and leave your comment below.

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 33,721 other followers

%d bloggers like this: