What is an author?

Erlend Hem About the author
Artikkel

This simple and banal question is possibly the most controversial one among medical researchers.

I often speak with scholarship holders and young researchers. Often, the conversation turns to authorship, and many have stories to tell. The most common of these involves a senior researcher who demands unwarranted co-authorship. Often, the dispute about the names to figure on the list of authors ends in a stalemate. Vagueness regarding authorship can be a symptom of an unhealthy research environment.

It may appear odd, or almost incomprehensible, that questions regarding authorship have the power to stir up such animosity among medical professionals and researchers in adjacent disciplines. Researchers in the humanities and social sciences shake their heads in disbelief – they still nurture the idea of single authorship. In these fields, the dictionary definition of an author still applies: A person who writes a text. This is not so in medicine and the natural sciences.

The sensitivity of this question should not surprise us. It involves the very currency of research. Increasingly, every researcher must defend his or her position by publishing articles. «Publish or perish» is taken literally. Careers and money are involved. There is much at stake for someone who is wrongfully excluded from the list of authors, or vice versa.

In medical research, the authorship criteria launched by the International Committee of Medical Journal Editors (ICMJE), also referred to as the Vancouver group, have gained wide acceptance. On paper, they appear straightforward: Authorship should be based on a) substantial contributions to conception and design, or acquisition of data, or analysis and interpretation of data; b) drafting of the article or revising it critically for important intellectual content; and c) final approval of the version to be published. All conditions (a, b and c) must be met. Even though the introduction of these rules has been important, they have failed to solve the problems of authorship by a wide margin. The rules are open to interpretation, and many fail to abide by them. Authorship abuse may come in many forms, including gift, guest and ghost authorships. In recent years, this journal has repeatedly brought up discussions on authorship issues (1 – 5). A glance into the database of the Committee on Publication Ethics (6) can provide an impression of the complexities and challenges involved.

The order of presentation on the list of authors is not randomly chosen. There appears to be a consensus that the first authorship is granted to whoever has made the greatest effort, thus earning the greatest merit. Who should be presented next remains less clear. Does it earn more merit to be presented as the second author than to be at the bottom of the list? In what order should the remaining authors be presented – according to their contributions (and if so, who should assess them) or in alphabetical order? Paradoxically, the Vancouver rules provide no guidance on such issues (7). Practices vary between various specialties, mostly on the basis of tradition and custom. In recent years, many journals have started to permit shared first authorship, i.e. the first authors on the list earn equal merit, and this is noted in the article. This also applies to this journal (from no. 6/2008).

When I first came to this journal in 2000, paper manuscripts were sent by mail between the editors and the contact author. When submitting a manuscript, the co-authors had to sign a letter of submission. There was no subsequent contact between them and the editors – we took it for granted that the contact author would keep them posted with regard to the editorial process. Following several scandals, such as the Sudbø affair, a scheme was introduced in June 2008 whereby all co-authors are automatically informed by e-mail when a manuscript is submitted, and of each subsequent contact with the editorial office until the final decision.

Some have proposed to do away with the concept of authorship as such, and rather use «contributorship» and «guarantors». This involves setting up a detailed list of the individual contributions to the manuscript. The proposal was launched in 1997, and the editor of The Lancet spared no praise. He claimed that this would have the potential to threaten the entire structure of modern science, and that everybody could freely ignore the authorship rules of the Vancouver group (8). However, not everybody shares this view. In my opinion, the system of contributorship and guarantors introduces nothing fundamentally new. The contributors will still be listed as authors in the medical data bases, and the system is just as vulnerable to fraud as the Vancouver rules. It is worth noting that the article that revealed the fraud perpetrated by Jon Sudbø was printed in exactly The Lancet. It might nevertheless be reasonable to inform the readers of what each author has contributed. Such a system is now being introduced in this journal. The identity of the article’s contributors is of interest not only to researchers and editors, it is also important to readers. This journal has already introduced (from no. 9 – 10/2011) a practice of including mini-biographies, where the authors note their year of birth, position, place of work and other information relevant to the article (9).

Traditionally it has been the custom that those who have contributed to a manuscript, but do not meet the criteria for authorship, are mentioned under the acknowledgements. However, such expressions of gratitude are of little value to those concerned. In order to implement large projects and maintain their interest in participating in years to come, it is essential that the contributors are rewarded for their efforts. Several of them would not qualify for authorship. As a solution to this, a few named authors are presented on behalf of a group of collaborators. In addition, the names of the collaborators are provided, and they are thereby credited for their contributions to the article (10). This journal is currently introducing such a practice.

These amendments serve a dual purpose: The readers are informed of the nature of the individual contributions and the main responsibility for the content of the article, and those who have contributed to the article earn their deserved merit – without diluting the concept of authorship.

Anbefalte artikler