Retraction Watch

Tracking retractions as a window into the scientific process

Unintended consequences: How authorship guidelines destroyed a relationship

with 14 comments

It started as a simple email exchange over authorship. But it angered one researcher so much that it ended a 20-year collaboration.

In January 2017, a chemist based in Mexico had finished writing a paper describing the structure of a molecule. Sylvain Bernès, at the Instituto de Física Luis Rivera Terrazas, asked his co-author—the head of the lab where the molecule had been synthesized 10 years ago—to review the draft and include any co-authors involved in the initial work.

The researcher added three co-authors to the paper. Bernès became concerned. He wanted to follow the International Committee of Medical Journal Editors (ICMJE) authorship recommendations as strictly as possible. As far as Bernès could tell, none of the new authors had actually contributed to the work, potentially violating the recommendation about authorship contributions.

The current ICMJE authorship recommendations, which were updated in August 2013 to include an author’s responsibility to ensure the accuracy of the work, are as follows:

  1. Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; AND
  2. Drafting the work or revising it critically for important intellectual content; AND
  3. Final approval of the version to be published; AND
  4. Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Bernès emailed the director of the lab and the co-authors to voice his concerns:

I feel that 3 co-authors never participated in that work

He then asked for the authors to explain what “they actually contributed to the work”:

This was probably a bad idea.

His colleague responded:

de inmediato no hay ninguna duda al respecto: se suspende toda relación de trabajo contigo“. (“there is no doubt about it: any working relationship with you is cancelled”).

In one brief email exchange, a 20-year collaboration had ended over authorship guidelines.

Bernès noted that part of the problem is that his colleague may not have been aware of the guidelines. The bigger issue, however, is that “local patterns and customs regarding scholar publications” may trump recommendations:

…in Mexico, guest authorship is accepted within a lab where members have been collaborating for a long time, and questioning for the real participation of individuals may be seen as an offense

Bernès told us:

…any matter about authorship in Mexico is very very sensitive. Frequently, the content of an article to be submitted is not really discussed between the researchers, while the list of co-authors, the order, and who is the author for correspondence, may be debated over months.

We don’t know how often authors face unintended consequences from guidelines, but according to Elizabeth Wager, a publications consultant at Sideview and a member of the board of directors of The Center For Scientific Integrity, our parent non-profit organization:

While including senior or long-term colleagues is viewed as unacceptable guest authorship in some countries (and according to ICMJE criteria) this may be viewed as professional courtesy in others, or at least essential for career progression.

Darren Taichman, secretary at ICMJE, told us:

The goal of the ICMJE’s recommendations is to help promote best practices and ethical standards in the conduct and reporting of medical research. The answer to instances where best practices are not followed is not to lower the standards. Rather, we should all – investigators, authors, universities, funders and editors- continue our efforts at promoting what we believe to be best practices.

Regarding the “substantive contribution” recommendation, Taichman said:

The ICMJE’s criteria for authorship aim to ensure that in addition to providing credit for substantive contributions, authorship requires responsibility and accountability for what is published. “Honorary” or “guest” authorship accomplishes neither of these aims. It is unfortunate that in some environments, pressure to list individuals who have not made substantive contributions as “guest” or “honorary” authors continues. We believe that the solution is for the leaders of research groups and institutions where such practices continue to set examples and make clear that appropriate standards are to be met, rather than for the standards to be lowered to meet current local custom.

But, Wager noted, there can also be “grey areas of what constitutes ‘substantial’”:

… especially in multicentre and large studies, there is often a degree of arbitrariness between those who meet the 1st criterion (ie play an active part in the research) and those who are invited or selected to take part in the publication and therefore also meet the 2nd criterion.

That is why Wager prefers the “contributorship model”—where the contribution of each author is clearly delineated:

Then if somebody is listed as head of the lab or department despite not having made much contribution to this particular project, this is transparent.

In this case, the paper was only in its second round of peer review when the dispute over authorship arose, so Bernès simply withdrew the submission.

Unintended consequences of another recommendation

Bernès ran into another issue with the ICMJE authorship guidelines in January 2017. In this instance, Bernès spotted inaccuracies in a paper he co-authored after it had been published. Bernès caught the issue late because he had never received the manuscript to review prior to publication. According to Taichman, secretary at ICMJE:

ICMJE recommends that editors send copies of all correspondence to all listed authors. A group of authors needs to ensure that each member has reviewed and approved the final manuscript.

Again, Bernès contacted the lead author to correct the issues and asked why he had not been sent the manuscript to approve.  Bernès learned that the PhD student who had done the bulk of the work was under the pressure of a short deadline. In this case, Bernès was able to resolve the issue without ruining the relationship. According to Bernès:

we will continue to collaborate in a good spirit.

Still, Wager said:

In my experience the criterion that all authors must approve the submission (and revisions) causes the fewest problems… And one reason it’s useful is to let authors review so they can spot problems.

But, Wager noted, the only way to ensure enforcement is to create an authorship agreement:

Many pharma companies now track publications using special software packages and they won’t submit until all authors have said they approve. I think it is up to authors to explain that they expect to approve the final (and revised) submissions and to be consulted about timing.  I reckon it’s OK to email authors with a reasonable deadline (of at least a few days) and say that if you haven’t heard from them by that date you will submit. But I’d certainly expect to approve any article with my name on it.

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.

Written by Victoria Stern

July 5th, 2017 at 8:00 am

Comments
  • BG July 5, 2017 at 9:03 am

    “Unintended consequence” seems to imply some unfortunate aspect of the outcome. A falling out between two researchers (one of whom apparently ignored the ethical question to the point of indignation) is not ideal of course. It would be better if there had been a debate, or at the very least, some indication both parties were prepared to seriously address these concerns. But if the guidelines have helped one researcher identify and act on another’s ethical failings, this seems like a positive outcome as a result of the guidelines. I don’t know the motivations of those who put the guidelines together, but one could imagine they might have anticipated and even hoped for a situation like this. After all it defies the very meaning of the term “author” if you are including people who have contributed nothing or very little merely because you wish to game the system in favour of your colleagues by including them.

  • rfg July 5, 2017 at 9:12 am

    I’ve been in some rather intense discussions over criterion 4, which I fully support.

    Of course, all authors are responsible for insuring that the data in a paper is accurate.

    These days however it’s questionable if even the submitting author will accept responsibility for all the data in a multi author paper.

    • anon July 5, 2017 at 7:26 pm

      The problem is of course that in many cases (for example papers that have experimental, theoretical, and/or computational parts) it’s absolutely impossible for any single person to know if everything in the paper is accurate.

  • TL July 5, 2017 at 9:56 am

    The ICMJE guidelines do not in fact state that once you are even tangentially connected to the data collection, you get guaranteed co-authorship for the next 100 years without any further intellectual input required.

  • Volga Tronovic July 5, 2017 at 10:14 am

    Mexico? “local patterns and customs regarding scholar publications” ? Those “local patterns” are widespread in the US and European nations as well. Big names or FOSA (Friends of Senior Author) are commonly added as honorary authors. People in glass houses…..

    • Rob Knell July 5, 2017 at 4:07 pm

      Not in my experience in the UK, USA or Australia – but I can only speak for my own personal experience of course.

      • anon July 5, 2017 at 7:17 pm

        Maybe not quite “widespread”, but in my experience honorary authorship isn’t very uncommon in the UK or US either.

  • kmf July 5, 2017 at 12:09 pm

    I worked for a US government agency for many years. At least for my own papers and for many others that I knew about, authorship credits were pretty routinely given to people who had contributed almost nothing to the articles. They might read through a draft and make minor editorial comments, which would be considered as “revising it critically for important intellectual content.” Or not even that. Adding non-contributing authors might actually benefit the first author because it would show that the first author was engaging in collaborations, which would help his or her performance ratings, even though the first author might be doing 99.9% of the work. It was pretty much part of the culture to extend authorship to a lot of people who had contributed almost nothing, often simply because they were colleagues who had some kind of ‘ownership’ of the topic area.

    • ab July 5, 2017 at 3:20 pm

      I agree, I’ve worked in over five labs on three continents and in various institutions, universities and research institutes, and it is the rule to gift authorship to some people that had nothing to do with the data; just being director, making decision about equipment, bringing in core funding, it meant ‘should be being invited as co-author’, and the contribution is typically a couple of minor corrections, no intellectual input on the paper.
      Old habits die hard. It’s not going to go away soon, no matter the ICMJE recommendations.

    • kmf July 6, 2017 at 12:37 pm

      As a follow-up, I am pretty sure ‘authorship’ rules are pretty deeply embedded in the culture of many, perhaps most, universities and agencies, and the ICMJE recommendations make almost no difference to the culture. People are awarded honorary or semi-honorary authorship because they are part of the ‘group’ or because they run the show or because they indicated some casual interest in the project in passing or whatever. And there are slightly grayer areas for example if someone oversaw all the lab work for a data set, does that mean they are coauthors even if they had zero input into the analysis and never even read the paper? Where is any enforcement of the ICMJE recommendations going to come from? It’s not really in anyone’s interest to limit authorship, and as shown in this case, can lead to trouble.

      • the dude July 6, 2017 at 1:11 pm

        So you will perpetuate the problem because it’s not in anyone’s “interest?” What about doing what’s right? Nobody cares anymore, I guess.

        • kmf July 6, 2017 at 5:54 pm

          I am retired and not perpetuating the problem, but just raising the question of how the ICMJE recommendations might get followed and who is in a position to do something about this.

        • Volga Tronovic July 6, 2017 at 5:59 pm

          You are welcome to try and change the world if you like. If you get beaten up in the process by entrenched forces who hold the levers of power and want to maintain the status quo, don’t come crying to me.

  • John Minter July 8, 2017 at 11:06 am

    The ICMJE recommendations are quite reasonable and section 4 outlines the proper approach for handling people that don’t fit the criteria for authorship: acknowlegements.

    I ran into a situation a couple of years ago where a coworker and I had done some microscopy and microanalysis on a couple of specimens from a scientist from one of the local non-profit labs. A couple of years later, we received notification that he wanted to submit the work for a conference and wanted to list us as co-authors. We had not been involved recently, had not seen the manuscript and other data, and therefore could not defend the work. Our solution was to respond that we were pleased that he was able to complete the work (he had with many other labs as well) and were honored that he thought of us. We suggested that we did not qualify as authors under current accepted guidelines and suggested that an acknowlegement for results would be sufficient. His response was pleasant and I did not detect any offense.

  • Post a comment

    Threaded commenting powered by interconnect/it code.