Retraction Watch

Tracking retractions as a window into the scientific process

Non-retraction notice: Editors explain why two similar papers aren’t redundant

with 12 comments

abdominal radiologyEditors have published a notice to let readers know why they’re not retracting a couple of papers.

One paper examined whether the results of CT scans could be used to stage patients with uterine carcinoma; the other considered whether CT scans could be used to predict overall survival in uterine carcinoma. Both papers — by researchers at the Memorial Sloan Kettering Cancer Center — used data from the same 193 women. After they appeared in in different journals, the editors considered whether they were redundant — a quality that can spell retraction for a paper.

The editors explain why they decided the papers were unique in a brief commentary — a non-retraction notice, if you will — published in a third journal, Abdominal Radiology:

The two articles were examined following the Committee on Publication Ethics (COPE, www.​publicationethic​s.​org <http://​www.​publicationethic​s.​org>) guidelines with regard to suspected redundant publication. C. Daniel Johnson, M.D., Editor in Chief of Abdominal Imaging and Dr. Charles M. Balch, M.D., Editor in Chief of Annals of Surgical Oncology recognize the overlap in the patient cohort analyzed. Yet, both Journal Editors felt that different rationales and criteria were used in the two publications that lead to different, though complimentary, conclusions on different aspects of the dataset studied. It was decided to publish this notice to state this conclusion.

The articles that are staying in the scientific record are:

Yulia Lakhman and Evis Sala are the first and last authors on both papers, respectively. All three journals are published by Springer.

COPE’s flowchart on what to do if you suspect a publication is redundant is here.

We’ve reached out to the editors to ask who brought the overlap to their attention, and why they decided to publish the commentary in another journal.

Update, May 10, 3:00 pm:
We’ve written about one of the co-authors of these papers before — Hedvig Hricak, chair of the Department of Radiology at Memorial Sloan Kettering. She had a paper pulled last year for containing “similar text and illustrations” to those in other articles. (It was unclear if those articles were by different researchers.)

An anonymous commenter pointed out that Hricak also has an addendum, published last year for “MR imaging and MR spectroscopic imaging in the pre-treatment evaluation of prostate cancer,” in the British Journal of Radiology:

We would like to raise the awareness of readers of BJR to the following.

Some material published in this review article was discussed previously in the review article listed under reference 37: Pretreatment evaluation of prostate cancer: Role of MR imaging and 1H MR spectroscopy. RadioGraphics 2004; 24: S167–S180.

The Editors of BJR were not aware of the earlier publication and were informed of the overlap only after the submission, peer review and publication of Br J Radiol 2005; 78: S103–S111.

That paper has been cited 76 times. Hricak is an author on the RadioGraphics paper that it borrows from. Blogger Leonid Schneider has more background on the addendum, as well as a comparison of the text between the two papers.

The commenter also pointed out that last author Evis Sala has a retraction that’s a few years old. “Magnetic resonance imaging of uterine abnormalities” was pulled from The Obstetrician & Gynaecologist in 2012.

Here’s the retraction notice:

The following article from The Obstetrician & Gynaecologist, 14:1, pp. 1–8, January 2012, ‘Magnetic resonance imaging of uterine abnormalities’, by Penelope Moyle, Lorenzo Mannelli, Mahmood Shafi and Evis Sala, published online on 26 January 2012 in Wiley Online Library (www.wileyonlinelibrary.com), has been retracted by agreement with the journal Editor-in-Chief, Jason Waugh; Blackwell Publishing Ltd; and the authors. As the result of feedback from experts in the field of disorders of sex development, it has been determined that the paper contains some factual errors and out-of-date terminology and could be misleading to the readership. All associated continuing professional development (CPD) questions have also been withdrawn.

The journal is not indexed in Thomson Reuters Web of Science.

Hat tip: Rolf Degen 

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our new daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.

Comments
  • Brandon May 10, 2016 at 9:34 am

    “…different, though complimentary, conclusions…”

    Should be complementary, unless the papers were saying nice things about each other.

  • Anonymous May 10, 2016 at 9:49 am

    The Obstetrician & Gynaecologist, 14:1, pp. 1–8, January 2012, ‘Magnetic resonance imaging of uterine abnormalities’, by Penelope Moyle, Lorenzo Mannelli, Mahmood Shafi and Evis Sala, published online on 26 January 2012 in Wiley Online Library (www.wileyonlinelibrary.com), has been retracted by agreement with the journal Editor-in-Chief, Jason Waugh; Blackwell Publishing Ltd; and the authors.
    http://onlinelibrary.wiley.com/doi/10.1111/j.1744-4667.2011.00077.x/pdf

    Related to Hedvig Hricak, a co-author:
    Abdom Imaging. 2015 Aug;40(6):2065. doi: 10.1007/s00261-015-0414-z.
    Retraction Note to: New Horizons in Genitourinary Oncologic Imaging H. Hricak Department of Radiology, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, Room C-278, New York, NY 10021,
    Retraction to: Abdom Imaging DOI 10.1007/s00261-005-0385-6
    This review article has been retracted upon the request of the author as it contains similar text and illustrations to previously published articles.
    http://link.springer.com/article/10.1007%2Fs00261-015-0414-z

    Hricak H. MR imaging and MR spectroscopic imaging in the pre-treatment evaluation of prostate cancer. Br J Radiol 2005; 78: S103–S111.
    http://www.birpublications.org/doi/full/10.1259/bjr.20139007
    We would like to raise the awareness of readers of BJR to the following:
    Some material published in this review article was discussed previously in the review article listed under reference 37: Pretreatment evaluation of prostate cancer: Role of MR imaging and 1H MR spectroscopy. RadioGraphics 2004; 24: S167–S180. The Editors of BJR were not aware of the earlier publication and were informed of the overlap only after the submission, peer review and publication of Br J Radiol 2005; 78: S103–S111.

  • Miguel Roig May 10, 2016 at 9:49 am

    This case strikes me as a classic instance of Salami Publication. I assume that no violations of statistical assumptions (e.g., multiple comparisons, independence of samples) occurred. Still, that future readers will not be able to tell that these separate study outcomes are derived from the same patient sample is troublesome to me. Frankly, unless the editorial comment can be linked to each publication, the situation merits a correction.

  • Alan R Price May 10, 2016 at 11:03 am

    Right decision. I had a case where a professor published much the same data in two different journals – when accusations about this were raised by others, the editors were both contacted, and both confirmed that the focus of the papers was different for the two different professional audiences, so the two editors had no concern about copyright infringement for their journals – nor had any ethical questions about the professor’s writings

  • TL May 10, 2016 at 12:49 pm

    One should not “publish data”, but rather conclusions of studies that are based on that data. If two clearly distinct studies happen to be able to reasonably draw on the same data, that just seems like good research economy. It is not sensible to require that something like the results of the Framingham Heart Study or the Rotterdam Elderly Study should only be allowed to be published in one single article lest it become “duplicate publishing”.

  • KT May 11, 2016 at 9:07 pm

    The journal “Abdominal Imaging” is the predecessor of “Abdominal Radiology,” so this is not the third journal. Authors of the commentary may hesitate to publish it in “Annals of Surgical Oncology,” which has higher IF than “Abdominal Radiology.”

  • Alison McCook May 12, 2016 at 9:45 am

    Fixed, thanks!

  • JA May 15, 2016 at 1:03 pm

    Alan – The decision may be right with your professor but if you look into the details of two publications the decision may be different or wrong. It all depends upon the degree, extent and intent of duplication or Salami Slicing.

    TL – Good comment. Here is an example, Framingham Heart published Jan 2016 in AJC validation of Framingham Heart Study for Atrial Fibrillation. Great paper. Could you imagine if the authors had published this data twice; once with the c-statistic and again separately with the 95% CI. Clearly, this would be redundant and salami sliced.

    See Pubpeer for more specific information about the degree and extent for these two papers.

    https://pubpeer.com/publications/F7240576B7BDD6BC270A0BF16D4854

  • Alan R Price May 15, 2016 at 7:25 pm

    As I said, in my professor’s care, the decision was right and the editors for both journals agreed that the dual publication of the data was appropriate for their two different professional audiences

  • Anonymous May 15, 2016 at 10:29 pm

    There needs to be one standard across all journals and publishers, independent of editors’ positions, which can be biased. There is always the possibility that friend’s of a professor can protect duplicate, redundant publications. Having a standard for all papers (i.e., one copy in the literature) eliminates the bias, or the possibility of bias.

  • Miguel Roig May 16, 2016 at 7:24 am

    Anonymous, such a standard could be conveyed as the following commandment: Thou shalt describe thy work in a way such that readers shalt never need to question the provenance of the data used in thy analyses. 🙂

  • Post a comment

    Threaded commenting powered by interconnect/it code.