One-fifth of Major Journal Articles Have Misattributed Authorship

Here's a sobering article from BMJ. The authors (all senior editors from JAMA) surveyed in 2009 hundreds of corresponding authors on pieces that had published in high-impact medical journals in 2008. (The journals were Annals of Internal Medicine, JAMA, Lancet, Nature Medicine, New England Journal of Medicine, and PLoS Medicine). The surveyed authors self-reported 21% rates of ghost authorship and/or honorary authorship! ("Honorary" authors are listed authors whose actual contribution to the paper should not merit authorship under the International Committee of Medical Journal Editors criteria. "Ghost" authors are those who meet the criteria and aren't listed, or whom the corresponding author mentioned as participating in writing the article, but who remained unnamed.) Overall incidence of honorary authorship was largely unchanged since a similar survey was conducted in 1996, though ghost authorship was down to 7.9% from 11.5%. Fully a quarter of all original research articles in these top-notch journals had honorary authors--as did 15% of reviews and 11.2% of editorials. 11.9% of research articles, 6% of reviews and 5.3% of editorials had ghost authors. Of course, these numbers may overlap, because a ghost-written article can be signed by an honorary author.

This looks terrible, especially since one might expect under-reporting of bad behavior in surveys of this kind. Of course, we needn't think that all of the honorary authorships were the extreme case of people signing onto something they didn't work on; most probably fell into the category of people who contributed a lot to the piece but not enough to merit full authorship--they were made authors, in other words, when they should merely have been mentioned in an acknowledgement footnote. Still: a fifth of all articles, and a quarter of original research articles, gave the wrong kind of credit to the wrong people, and 8% failed to acknowledge the right people. And people wonder why NIH is requiring Responsible Conduct of Research training....