Authorship disputes are one of the most common conflicts in research groups around the world.
Last June 12, the PRBB Good Scientific Practice (GSP) working group organised an event to learn about some of the underlying reasons for authorship conflicts and about some international initiatives that are trying to prevent and solve them, as well as existing tools and resources to deal with them. The event also included discussions about what we can do from our centres to ensure authorship and credit are given in a fair way.
The event kicked off with Jana Selent, coordinator of the Committee for the Integrity of Research and Good Scientific Practice (CIR) at the Hospital del Mar Research Institute, as well as the the instute’s representative at the PRBB GSP group. After an overview of this group’s aims and the previous initiatives it has organised, Selent introduced some general authorship rules, as per the PRBB GSP code and the ICMJE (International Committee of Medical Journal Editors) guidelines. According to what seem very logical and clear criteria, to be considered an author a researcher must comply with these four requisites:
- Having done a substantial contribution to the conception, acquisition, analysis or interpretation of the work; AND
- Having drafted or reviewed the text; AND
- Having approved the final version; AND
- Having agreed to be accountable for the work.
But of course, the devil is in the details, as was later discussed in length.
Current situation at the PRBB
Selent presented the current authorship guidelines and training at the PRBB and its centres.
All of the PRBB institutions have authorship rules – they all follow those in the PRBB GSP code, and some centres also have their own more specific ones. Also, the great majority of the PhD students at the park receive information about authorship rules via the “Science in action” course at the UPF PhD Program.
Selent also gave an overview of the most common issues the centres usually encounter regarding authorship:
- Authors order disagreement.
- Group versus individual recognition in large scale consortia projects, where there are limited spaces for authorship.
- Issues with several conflicting ‘main authors’, in the case of long-term projects where some researchers leave the lab and others come.
- Issues regarding the authorship status of staff that selects samples or others in cohort studies and similar.
The need for a research assessment reform
Next presented Paula Samsó, from the Scientific Coordination Office at IDIBAPS and member of the Spanish chapter of CoARA, the Coalition for Advancing Research Assessment.
She started setting the stage about the publish or perish culture and the way research and researchers are evaluated. Beyond affecting unethical authorship practices, this culture may also lead to publication bias, with only positive results being published, and a lower research quality due to the focus on quantity rather than quality; to bad practices such as data manipulation, fake affiliations or predatory journals to artificially increase the number of publications; to decreased innovation, since risky and long-term projects are less likely to be engaged in, and to reduced collaboration and open science practices, for fear of competition. Last but not least, also concerning were inequalities, specially for those with non-linear career paths or care giving roles that can affect the quantity of papers, and mental health issues due to the pressures.
But not all is gloom. Despite the dire situation, there are – and have been for the last years – several initiatives that aim to decrease this pressure to publish by reforming research assessment.
DORA, the San Francisco Declaration on Research Assessment, was launched in 2012 in recognition of a “need to improve the ways in which researchers and the outputs of scholarly research are evaluated”. CoARA, the Coalition for Advancing Research Assessment, follows the same principles, aiming to “recognise the diverse outputs, practices, and activities that maximise the quality and impact of research through an emphasis on qualitative judgement in assessment”. For that, they acknowledge there’s a place for quantitative indicators, such as the Impact Factor, but only as support for other more qualitative and holistic indicators.
Importantly for the focus of this event, CoARA recognizes that research is based on teamwork, and therefore calls for a broader recognition of all meaningful contributions, not only papers and not only first or last authors – although of course those who have contributed most need to be acknowledged appropriately.
The King’s experience
Serena Mitchell, research integrity officer at King’s College in London, also presented in this event their experience in dealing with authorship conflicts. King’s has a Department of Research Governance, Ethics and Integrity, with around 20 professionals who, among many other things (training, policies, engagement, etc), deal with about 20 cases of authorship conflicts a year. Similarly to what Selent showed about the PRBB, most of the conflicts involved a junior researcher (PhD or postdoc) and a PI, and they were about the order of authors, or authorship in papers after someone had left the lab. Mitchell emphasized the importance of ensuring that, whatever the conflict, the research results ended up published. “There’s an ethical push to publish, due to the use of public fundings, or the use of animals or human volunteers. and publishing is also a requirement of most funders”, she reminded the audience.
The King’s representative talked about their existing policies (e.g. on FAIR data) and the ones that are in development (e.g. on authorship and use of CRediT) and recommended as a good resource the materials regarding authorship disputes available at UKRIO.
When asked about how to enforce the recommendations of the office, Mitchell said that it helps “having senior leaders owning the recommendation”. Sometimes, her office also outlines “the institutions’ responsibility to publish and the researcher’s contractual obligation to publish”.
The authorship project
Eva Casamitjana, research manager at the Barcelona Institute for Global Health (ISGlobal) and one of the centre’s representative at the PRBB GSP working group, closed the first part of the event talking about “The authorship project”, developed by researchers at the University of North Carolina at Charlotte Graduate School.
Based on the idea that the best thing to deal with authorship conflicts is to prevent them, the project suggests signing an “Authorship agreement” before starting a project. This living document should be agreed on as early as possible and can be modified as needed under the agreement of everyone involved. Importantly, the document includes a section on what to do about non-responsive co-authors, setting in writing how many months these can be unresponsive before they can be excluded from the authors list. If, despite all the written agreements, conflicts still arise, the document directs to the institution’s own procedure for solving authorship problems.
An interactive workshop
Participants were few but very involved, and the second part of the event consisted of a group discussion about what our institutions can do to tackle authorship challenges.
It was clear that, despite the formal training – only at PhD level – and the obligation to adhere to the PRBB GSP code, there is low awareness about the guidelines and about the procedures to follow in case of conflicts.
It was also apparent amongst the participants the role of hierarchies and power dynamics and that, despite the rules, authorship decisions are very much dependent on the Principal Investigators (PIs). They should therefore be targets for training and awareness so they can introduce changes in their groups, like the authorship agreement mentioned above. But even if some changes can be done at the microenvironment level (within each research group) it’s essential that changes come from the institutional level and the research environment at large. Indeed, there was general agreement that a key issue to prevent many authorship conflicts was having an evaluation scheme that does not put the focus exclusively on the number of papers, and on who is first, corresponding or last author. And that, for this to really happen, it would be important to align these criteria in all institutions at the national level – something that CoARA would like to implement.
The inclusion of data or sample providers in the acknowledgements, rather than in the author byline, was also suggested to avoid cases where authors were included in lots of papers they hadn’t even read. In that sense, calls were made for integration of discoverable acknowledgements in repositories like PubMed, as well as for a better use of taxonomies like CRediT.
There was also discussion of how international and interdisciplinary projects can introduce asymmetries in authorship expectations and communication. In cross-country partnerships barriers like language, access to resources, and differing authorship cultures, as well as the structural disadvantages of low- and middle-income countries (LMIC) researchers, were mentioned. In the case of basic-clinical collaborations, an added dilemma was that the clinicians’ salary partially depends on how many papers they author, which adds an extra layer of pressure that can lead to inappropriate authorship. But, as in the case of data or sample providers, being included in a paper they have not really participated in raises ethical issues regarding accountability of all authors and trust in the presented research.
A participant also pointed out the fact that authorship rules haven’t changed for the last 40 years, while the way research is done has changed a lot, with a huge increase in large-scale collaborations between centres; and how this disagreement can also lead to frictions.
“There are more and more articles with over 200 authors. We are lying massively when we say all those 200 comply with all authorship requirements. It’s unreasonable”
Participant
Some specific suggestions of things to do at the PRBB level were:
- Improving the community’s awareness about the GSP code and the integrity committees.
- Doing an anonymous survey to collect the type of authorship problems present in the community.
- For the centres to write their own guideline/policy regarding authorship, which some centres have already done, or are in the process of doing.
- Stimulate the use of authorship agreements and discuss the possibility of including these in the protocols sent to ethics review committees.
A dialogue to be continued
The workshop was just a first step to describe and discuss the main issues regarding authorship in our community; unexpectedly, there was strong consensus that implicit norms and hierarchical expectations often distort fair authorship assignment and that better visibility and sharing of the existing documents and guidelines within the centres would be positive to ensure everyone knows who to address in case of conflict.
The insights will be shared among participants and the leadership of the centres, with the idea of continuing the discussion at each institution to take further steps towards better and more fair authorship practices.
