Who writes, whose rights, and who's right? Issues in authorship
Authorship is personally and professionally sensitive: collaborative relationships, careers and research team sustainability depend on demonstration of output, and publications are the easiest metric. One of the most common ethical issues editors encounter is authorship disputes. Most frequently, disputes arise because someone thinks that he/she has been omitted from a list of authors; however, there are also claims that individuals listed as authors are unwarranted. In addition to authorship disputes amongst research teams, it is increasingly common to encounter disputes between post-graduate research degree students and supervisors, or between the supervisors themselves.
Omissions and inclusions of authors are covered by several agencies. In the UK, the Research Councils (Research Councils UK 2009) address this matter under the heading of ‘Unacceptable research conduct’, where they identify it as:
Misrepresentation of involvement, such as inappropriate claims to authorship and/or attribution of work where there has been no significant contribution, or the denial of authorship where an author has made a significant contribution. (p7)
With respect to who should be included on the ‘byline’ of an article, the UK Research Integrity Office (UK Research Integrity Office 2009) suggests that:
Authorship should be restricted to those contributors and collaborators who have made a significant intellectual or practical contribution to the work. No person who fulfils the criteria for authorship should be excluded from the submitted work. Authorship should not be allocated to honorary or “guest” authors (i.e. those that do not fulfil criteria of authorship). Researchers should be aware that anyone listed as an author of any work should be prepared to take public responsibility for that work and ensure its accuracy, and be able to identify their (sic) contribution to it. (p19)
What counts as work and ‘contribution’
The key issue is how ‘significant contribution’ is defined in each context. In terms of the above, ‘work’ could mean both the research project and the publications arising. There are instances where authors have been omitted despite having contributed as co-investigators but not included in dissemination or consulted at the publication stage. With post-graduate student research, there are examples where one or all supervisors have been excluded from publications even though they have contributed directly to supervising thesis work and facilitating student intellectual development. In addition to student supervision meetings, ‘work’ could mean several additional supervisory activities to support production of the thesis such as supervisors sharing knowledge, experience and intellectual commentary. Supervisors may embed students in their teams to expose them to a relevant intellectual learning context and introduce students to key people in the field to enhance their thinking and understanding. They may create opportunities for dissemination and feedback to enhance intellectual credibility, which would not otherwise be offered to the student, thus facilitating relevant employment opportunities. Denying an opportunity for co-authorship despite having made a significant contribution to a piece of work still constitutes unacceptable research conduct.
The ICMJE guidance
In the US, the National Institutes of Health (NIH) refer scientists to the International Committee of Medical Journal Editors (ICMJE, www.icmje.org), which concludes that an author is someone who had ‘made substantive intellectual contributions to a published study’ (International Committee of Medical Journal Editors 2013). ICMJE has recommended that the following criteria for authorship should be met:
(1) substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data; (2) drafting the article or revising it critically for important intellectual content; and (3) final approval of the version to be published
The BMJ, however, takes the view that the definition of authorship, produced by ICMJE, has some serious flaws as it ‘does not make clear who has contributed what to the published study, nor does it clarify who is responsible for the overall content. It also excludes those whose sole but often large contribution has been to collect data’ (BMJ 2013). The BMJ, therefore, distinguishes between authorship and ‘contributorship’, with an additional contributor section where some contributors may not be authors. Disputes about authorship will still happen, as contributors have to decide where to draw the line between those contributors who will be authors and those whose contribution will be acknowledged. These judgements are likely to vary between teams and institutions. Smith (1997) cites Rennie et al. in suggesting that:
Contributors should include all those who have added usefully to the work. They might include somebody who suggested the idea and design for the study but did nothing further, or somebody who collected many of the data but was not concerned with design or analysis
JAN adopts an even more flexible variation of ICMJE guidance, requiring that all authors have agreed on the final version of the paper and met at least one of the first two criteria. This more flexible approach acknowledges, for example, the contribution of research officers to the collection of data and subsequent co-authorship.
Defining contributorship and what constitutes ‘substantial contribution’ is more easily worked out in advance. It is important that research and supervisory teams agree on a publication strategy and policy on authorship from the start of the project. However, this should be sufficiently flexible to align with target journals: although authorship and contributorship guidelines have common principles, they do vary on key thresholds for authorship and do not distinguish between student and other types of research. Furthermore, institutions have a role in ensuring correct and complete authorship attribution on papers, given that misconduct affects institutions as well as individuals (Watson & Hayter 2013). Any unresolved pre-submission authorship disputes should be managed by following the COPE (Committee on Publication Ethics; www.publicationethics.org) principles, with the proposed Chair being independent of the authors and devoid of conflicts of interest.
What can editors do?
As editors, we are unable to judge whether listed authors have met authorship criteria and cannot know if there are omissions to authorship. Whilst some journals ask for the contribution of each author, this can only prompt authors to identify at least some input from each and is unlikely to influence omissions. Thus, we select papers for publication purely on the basis of merit and the information provided by listed authors. This may result in publication of papers that omit true authors as well as papers with ‘gratuitous’ authors, who have not contributed to the work but lead the laboratory, department or clinic, or with whose patients the work occurred. This can result in claims of unacceptable research conduct or misconduct.
As editors, we review claims of unacceptable authorship conduct carefully. A variety of circumstances serve as ‘red flags’ to possible authorship issues, including requests for changes of authorship during the process of review. That committee on Publication Ethics (2008) provides a set of flowcharts to help editors investigate any requests regarding authorships, and lists alerting circumstances as including requests for addition of extra authors or removal of an author before or after publication. Advice is also provided on how to spot other authorship problems, such a guest, ghost or gift authorship.
As editors, we would like to be proactive in addressing authorship issues. However, the circumstances listed above may not always emerge and we may be unaware of author disputes until after we publish a paper. Unfortunately, authorship issues are impossible to rectify once a paper is in the public domain. Errata and corrigenda can be issued, but the authorship of a paper, as finally published, is immutable regardless of the erroneous inclusion or exclusion of rightful authors. Therefore, it is vital that the authorship of any piece of work is established using ethical principles in advance of submission. Careers are made on the basis of publications and careers can be halted by accusations of inappropriate authorship.
The solutions lie with authors instituting an agreed ethical approach to authorship and contribution at the start of the project and, where appropriate, being sufficiently flexible to meet specific journal requirements and authorship guidelines. According to the UKRIO:
Researchers should address issues relating to publication and authorship, especially the roles of all collaborators and contributors, at an early stage of the design of a project, recognising that, subject to legal and ethical requirements, roles and contributions may change during the time span of the research. Decisions on publication and authorship should be agreed jointly and communicated to all members of the research team. (p18)
Authorship order will also need to be agreed, but presents issues of precedence rather than potential misconduct. The NIH takes a similar approach, with the NIH Ombudsman suggesting that collaborators begin with a written partnership agreement, likening this to a ‘prenuptial’ where the responsibilities and benefits of each partner are explicitly identified.
Authorship issues and disputes arising from them are not remote. As editors, we are frequently asked to investigate and arbitrate; however, this is inevitably at the stage – post-publication – where nothing can be done to rectify any injustices. You will note that this editorial is multi-authored and the irony of an argument between the editorial team of JAN about order of authorship was not lost on us. We are not proposing a new method to address this, but, to save us from having to decide on relative contribution – we all contributed to it – or to enter a dispute about which position was really the one indicating prestige, we decided to go by accident of birth or marriage in ordering authorship.