Misuse of Received Manuscripts by Peer Reviewers: A Cross-sectional Survey
Darren Taichman,1 Jill Jackson,1 Deborah Cotton,1 Cynthia Mulrow,1 Jaya Rao,1 Mary Beth Schaeffer,1 Catharine Stack,1 Sankey Williams,1 Christine Laine1
How often peer reviewers use information from unpublished manuscripts in a manner inconsistent with the goals of peer review is not known. Annals of Internal Medicine recently experienced an egregious occurrence involving the plagiarism of an entire research study by a peer reviewer who had evaluated the manuscript for Annals of Internal Medicine. The reviewer subsequently published the study in another journal as his own work. We therefore aimed to assess peer reviewers’ views and practices including their self-reported use of any information in manuscripts they reviewed.
All recipients of Annals manuscripts sent for external review in 2015 and 2016 were invited to complete an anonymous online survey between December 8, 2016, and January 17, 2017. Two reminder emails were sent.
A total of 1431 of 3275 invited reviewers (44%) returned the survey; 1068 of 1398 respondents (76%) reported working in an academic setting and 1249 of 1388 (90%) reported being involved in research. Nearly half indicated having reviewed and published more than 50 manuscripts and having mentored others in peer review. Reasons reported for agreeing to review included keeping up to date in a research field (957/1417 [68%]), a sense of obligation to peer review (1316/1417 [93%]), to plan one’s own work (425/1417 [30%]), and to know what competitors are doing (190/1417 [13%]). One hundred sixty-nine of 1417 (12%) had agreed to review manuscripts from authors with whom they had conflicts of interest; of these, 61 (36%) did so without informing the journal’s editor. One hundred fifty-three of 1413 (11%) showed manuscripts to colleagues without seeking permission. Twenty-six of 1414 (2%; 95% CI, 1%-3%) indicated having used the information in a reviewed manuscript for personal or academic benefit prior to the paper’s publication. Such reported use included using what was learned to alter one’s own research plans, speeding up journal submission of one’s own work related to the subject of the manuscript being reviewed, and copying some part of the reviewed manuscript for one’s own work.
Trust that reviewers will treat manuscripts received for peer review as confidential communications is an essential tenet of peer review. Although self reported and of uncertain generalizability, these results suggest that breaches of this trust do occur. Larger studies involving multiple journals should be considered to assess the generalizability of these results and to inform targeted educational initiatives aimed at promoting the highest ethical standards among peer reviewers.
1Annals of Internal Medicine, American College of Physicians, Philadelphia, PA, USA, firstname.lastname@example.org
Conflict of Interest Disclosures:
Christine Laine is a member of the Peer Review Congress Advisory Board but was not involved in the review or decision for this abstract. Jaya Rao reports that she has stock holdings/options in Eli Lilly and Pfizer. Catharine Stack reports that she has stock holdings in Pfizer and Johnson & Johnson.