Advertisement

Regarding “Analyzing Spin in Abstracts of Orthopaedic Randomized Controlled Trials With Statistically Insignificant Primary Endpoints”

      We read with much enthusiasm the paper by Arthur et al.
      • Arthur W.
      • Zaaza Z.
      • Checketts J.X.
      • et al.
      Analyzing spin in abstracts of orthopaedic randomized controlled trials with statistically insignificant primary endpoints.
      and the editorial commentary by R.G. Marx.
      • Marx R.G.
      Spinning “spin” in randomized trials in orthopaedic surgery.
      Arthur et al.
      • Arthur W.
      • Zaaza Z.
      • Checketts J.X.
      • et al.
      Analyzing spin in abstracts of orthopaedic randomized controlled trials with statistically insignificant primary endpoints.
      analyzed the prevalence of a subtle and common, albeit lesser, highlighted form of reporting bias: spin in the orthopaedic literature. Although the paper does not discuss in depth the grades or classification of spin or describe comprehensive methods of preventing spin as did Boutron et al.,
      • Boutron I.
      • Altman D.G.
      • Hopewell S.
      • Vera-Badillo F.
      • Tannock I.
      • Ravaud P.
      Impact of spin in the abstracts of articles reporting results of randomized controlled trials in the field of cancer: The SPIIN randomized controlled trial.
      who were the first to analyze spin in the orthopaedic literature, the paper presents commendable findings that caution readers, editors, and reviewers of the top orthopaedic journals.
      As avid readers of Arthroscopy and other top journals in orthopaedics, as analyzed by Arthur et al.,
      • Arthur W.
      • Zaaza Z.
      • Checketts J.X.
      • et al.
      Analyzing spin in abstracts of orthopaedic randomized controlled trials with statistically insignificant primary endpoints.
      we wish to opine on the points of the editorial comments that we believe will set up healthy discussions in the future in this less-explored area of reporting bias.
      • 1.
        Spin is a type of reporting bias. As the term “reporting bias” implies, appropriate measures should be taken at the time of reporting of the trial to prevent or avoid them. Most readers may have access only to the abstracts. By reading the methodology of abstracts, one can get only an overall idea about the methodology used in the trial, and cannot understand nuances hidden in the methodology. Thus, discovering spin is a difficult task unless one is enlightened enough on this aspect. Suggesting reading the methodology section more thoroughly as the best way around this reporting bias is like accepting spin as a part of the reporting process. We believe efforts should be taken to remove the bias rather than accepting them and burdening the reader. This can be done, for example, by training journal reviewers to identify spin and by restructuring abstracts as suggested by Boutron et al.
        • Boutron I.
        • Altman D.G.
        • Hopewell S.
        • Vera-Badillo F.
        • Tannock I.
        • Ravaud P.
        Impact of spin in the abstracts of articles reporting results of randomized controlled trials in the field of cancer: The SPIIN randomized controlled trial.
      • 2.
        In 2018-2019 alone, reputed journals have published many articles highlighting the presence of spin in their fields of medicine.
        • Cooper J.G.
        • Jeong S.J.
        • McGuire T.L.
        • et al.
        Fibronectin EDA forms the chronic fibrotic scar after contusive spinal cord injury.
        ,
        • Checketts J.X.
        • Riddle J.
        • Zaaza Z.
        • Boose M.A.
        • Whitener J.H.
        • Vassar M.B.
        An evaluation of spin in lower extremity joint trials.
        The analysts of spin such as Arthur et al.,
        • Arthur W.
        • Zaaza Z.
        • Checketts J.X.
        • et al.
        Analyzing spin in abstracts of orthopaedic randomized controlled trials with statistically insignificant primary endpoints.
        in addition to their vast experience in their respective fields, have taken preresearch training to specifically differentiate spin. If they are at fault for misinterpreting the findings as spin, we believe that a large group of readers who happen to be clinicians are much more vulnerable to spin. The responsibility falls on the shoulders of authors and journals to convey research findings most understandably for the reader.
      • 3.
        Randomized controlled trials are predesigned and registered to avoid reporting bias. On the completion of the study, authors should objectively report the outcomes irrespective of their significance. Neglecting a primary outcome because it is not significant and highlighting a secondary outcome because it has significant association introduces spin and misrepresents the intention of the trial in the first place.

      Supplementary Data

      References

        • Arthur W.
        • Zaaza Z.
        • Checketts J.X.
        • et al.
        Analyzing spin in abstracts of orthopaedic randomized controlled trials with statistically insignificant primary endpoints.
        Arthroscopy. 2020; 36: 1443-1450.e1
        • Marx R.G.
        Spinning “spin” in randomized trials in orthopaedic surgery.
        Arthroscopy. 2020; 36: 1451
        • Boutron I.
        • Altman D.G.
        • Hopewell S.
        • Vera-Badillo F.
        • Tannock I.
        • Ravaud P.
        Impact of spin in the abstracts of articles reporting results of randomized controlled trials in the field of cancer: The SPIIN randomized controlled trial.
        J Clin Oncol. 2014; 32: 4120-4126
        • Cooper J.G.
        • Jeong S.J.
        • McGuire T.L.
        • et al.
        Fibronectin EDA forms the chronic fibrotic scar after contusive spinal cord injury.
        Neurobiol Dis. 2018; 116: 60-68
        • Checketts J.X.
        • Riddle J.
        • Zaaza Z.
        • Boose M.A.
        • Whitener J.H.
        • Vassar M.B.
        An evaluation of spin in lower extremity joint trials.
        J Arthroplasty. 2019; 34: 1008-1012