Author! Author!Author! Author!

To the Editor: The study by Dr Bates and colleagues¹ shows that the disclosure of authors’ contributions has improved but not solved the problem of honorary authorship. The practice of including undeserved authors is not easy to eradicate because it provides some benefits and, remarkably, does so without detrimental consequences to the real authors. Bibliometric analysis may transfer the impact factor of a journal to its articles, and ultimately to the articles’ authors. This seems reasonable for single-author articles, but is difficult to justify for multiauthored articles. An article published in a journal with an impact factor of 15 could provide 15 points for its authorship if there is a sole author, but 150 points if there are 10 authors. This unfair system encourages authorship inflation and, at the same time, may put honest authors at a disadvantage in academic competition.

A more appropriate system to address this problem would be to transfer the score of the impact factor to authorship, but to divide it among all authors when there are more than 1. Several studies have shown a good correlation between authors’ position in the byline and their relative contributions.^{2 - 5} Based on this, I propose a scale where the individual reward of authorship would represent a percentage of the value of the article, according to authors’ position. The respective percentages for authors in the first, second, last, and middle (those who are not first, second, or last) positions would be as follows: 1 author, 100/0/0/0; 2 authors, 75/25/0/0; 3 authors, 60/20/20/0; 4 authors, 60/15/15/10; 5 authors, 55/15/15/15; 6 authors, 55/13/12/20; and 7 or more authors, 50/10/10/30. Middle authors would have to divide their corresponding percentage among all of them. This scheme could be applied regularly, unless the authors of an article propose a different distribution. Should journals publish the value of authorship for each author in a footnote of every article, bibliometric assessments would be much easier, and the idea that multiauthorship decreases each individual’s reward would be evident. Actual authors would still be able to add other names on the byline, but they would probably feel less prone to do so because it would be done at their own expense.