Intended for healthcare professionals

Analysis

Poor visibility of retracted articles: a problem that should no longer be ignored

BMJ 2023; 381 doi: https://doi.org/10.1136/bmj-2022-072929 (Published 20 June 2023) Cite this as: BMJ 2023;381:e072929
  1. Christophe Boudry, associate professor1 2,
  2. Katherine Howard, digital research analyst3,
  3. Frederic Mouriaux, professor45
  1. 1Normandie Univ, UNICAEN, Média Normandie, Caen, France
  2. 2URFIST, Ecole Nationale des Chartes, PSL Research University, Paris, France
  3. 3University of South Australia, Adelaide, Australia
  4. 4INSERM UMR_S_1242, Faculty of Medicine, Rennes University, Department of Ophthalmology, CHU Rennes, Rennes, France
  5. 5CUO-Recherche, Centre de Recherche du CHU de Québec – Université Laval, Axe Médecine Régénératrice, Hôpital du Saint-Sacrement, Québec, Canada
  1. Correspondence to: C Boudry christophe.boudry{at}chartes.psl.eu

Christophe Boudry and colleagues call for better identification of retracted articles on publishers’ websites and academic databases to avoid propagating scientific error

Article retraction—the withdrawal of an article published in an academic journal1—is “a mechanism for correcting the literature and alerting readers to articles that contain such seriously flawed or erroneous content or data that their findings and conclusions cannot be relied upon.”2 Although retractions are still rare, with around five retractions per 10 000 articles published,345 rates have increased (box 1).378 Retraction rates were high in the early, acute phase of the covid-19 pandemic14—up to four times higher than during outbreaks of other infectious diseases such as HIV, H1N1 influenza, or Ebola.15 However, the rate is now estimated to be “consistent with the expected overall rate of retraction.”16

Box 1

Rise in retractions of biological and medical science research articles

  • The retraction rate increased from 0.38/10 000 publications in 1985 to 2.03/10 000 publications in 2000, and 5.95/10 000 publications in 2014

  • Most retractions occur within 1 year of publication, with lower retraction rates over time since publication

  • On average, articles are retracted 3.8 years after their publication4

Retraction rates vary over time with changes in fraud and error6 as well as efforts by the scientific community to detect and report them.378 Pressure to “publish or perish” in order to secure research funding may contribute to an atmosphere in which some people could be tempted to selectively report results, or worse, commit outright fraud, both of which may lead to article retraction.9

Accelerated research publication, as in the acute phase of the covid-19 crisis,101112 may be associated with less rigorous peer review,13 further increasing the risk of retraction.

RETURN TO TEXT

Theoretically, citations should stop as soon as articles are retracted. In fact, retractions …

View Full Text

Log in

Log in through your institution

Subscribe

* For online subscription