News from selected sessions of the International Congress of Biomedical Peer Review for Thursday, September 18:
Medical journals dogged by honorary, uncredited authors
Blinding and masking may not improve peer review
Medical journals dogged by honorary, uncredited authors
Prague, Sept. 18 – "Ghost" authorship and honorary writing credits continue to haunt major US-based medical journals, according to a study presented today at the International Congress on Biomedical Peer Review and Global Communications.
At least 17 percent of more than 1,000 articles published during 1996 in six peer-reviewed US journals either listed an author who did not meet recognized authorship criteria or failed to credit a legitimate author, said JAMA Senior Editor Phil Fontanarosa, MD, one of the study’s authors. The report confirmed earlier findings on the prevalence of "ghost" medical writers and honorary writing credits in such journals.
In the study, questionnaires about the contributions of each listed author and the existence of uncredited authors were mailed to 1,180 corresponding authors of randomly selected articles from six American medical journals, including JAMA and the New England Journal of Medicine. Of the 789 responses analyzed, 121 articles listed authors who did not meet International Committee of Medical Journal Editors (ICMJE) authorship standards, while 87 responses indicated that an uncredited author had done significant work on an article.
"It could mean one of two things: A, either people were brutally honest, or B, author criteria are too strict," Dr. Fontanarosa said. The results seem to agree with a previous study that found 26 percent of writers credited in multi-author papers didn't meet authorship criteria, he added.
In related presentations today:
- A survey of 450 authors in the Dutch Journal of Medicine revealed that while the ICMJE standards are "reasonably known and and met by many authors … they are of no help for determining order of authorship." The study suggests that "clearer and more practical guidelines should be developed."
- A survey of American medical colleges found that two-thirds of the respondent institutions did not have and were planning to establish their own authorship criteria.
- A study at the British Medical Journal showed that a "substantial portion of the increase" in articles published in the BMJ over the past 10 years could be attributed to authors who were professors.
Detailed information on this session
Back to Peer Review Congress coverage
Blinding and masking may not improve peer review
Prague, Sept. 18 – Concealing the identities of medical journal authors from peer reviewers, and of the reviewers from each other, is often unreliable and does not significantly improve the quality of peer review, according to research presented today at the International Congress on Biomedical Peer Review and Global Communications.
In one study, pairs of reviewers at the Annals of Emergency Medicine, Annals of Internal Medicine, The Journal of the American Medical Association (JAMA), Obstetrics & Gynecology, and Ophthalmology were randomly assigned to perform a masked or unmasked reviews of the same paper. Editors, unaware of which version the reviewers saw, then scored the quality of their reviews on a 5-point scale.
The study did not detect a significant difference in quality between masked and unmasked reviews, though the researchers reported that poor masking success at most journals may have compromised the potential benefits of concealing reviewers’ identities. A related study at the British Medical Journal, using slightly different methods, came to the same conclusion about the quality of reviews when authorship information is withheld from the reviewers.
A companion study attempted to confirm differences in the success rates between 7 biomedical journals at keeping authors’ identities hidden from reviewers, and to explain those differences. The researchers studied three journals with a long-standing policy of masking author identity and four without such a policy. Reviewers of masked manuscripts were asked if and how they could identify the author.
Masking success was significantly higher at The Annals of Emergency Medicine (83 percent) than at all other journals. There was no significant difference in masking success between journals with a policy of masking (62 percent) and those without (61 percent).
The researchers concluded that masking success does not appear to be related to a journal's policy of masking, but could be affected by other characteristics of a journal or specialty. They speculated that reviewers with greater research and reviewing experience – but not necessarily greater age or academic rank – could decrease the success of masking.
Detailed information on this session
Back to Peer Review Congress coverage
