What Constitutes Grounds for Authorship?
Tonight in my ethics course, there was a case study offered in the area of authorship, and more specifically in the area of to whom authorship of a paper should be granted. The case involved a graduate student who had done the bulk of the work and the writing; her PI, who provided all things investigatory to the graduate student; an ancillary PI, possibly on the graduate student’s committee, who offered significant guidance in alternative experimental design and direction; a lab technician, who carried out some experiments but had no role in experimental design, data interpretation or analysis, or writing; and a final PI at another institution, who provided an antibody necessary for the experiment.
The question was: Who should be listed as authors on this paper?
Clearly, the former two players above should be listed as authors. The graduate student contributed the majority of the intellectual efforts to produce and report the data, and the main PI likely played a crucial role in directing the graduate student.
What is not so clear is whether any of the others should be listed as authors. For example, the lab technician contributed nothing intellectual to the project aside from her ability to transfer liquid from one container to another container ad nauseum. On the other hand, without her, the experiment could not have been completed.
Along the same lines, the third, antibody-producing PI was largely indispensable in the completion of the project, not for any intellectual contributions, but for provision of a crucial resource that would otherwise have taken a prohibitive amount of time and/or money to produce in-house.
Similarly, the ancillary, advice-offering PI provided much guidance necessary for the graduate student to have completed the project while not having contributed directly to its completion.
My feeling on this is that whether the technician and the ancillary PI are included or left off is a toss-up to me; I could go either way. According to the International Committee of Medical Journal Editors (ICJME), there are three basic requirements that an author should fulfill in order to be granted authorship:
- substantial contributions to conception and design, or acquisition of data, or analysis and interpretation of data;
and - drafting the article or revising it critically for important intellectual content;
and - final approval of the version to be published.
Whether the technician and ancillary PI have met these requirements still seems a bit cloudy to me, as each clearly contributed substantially, although in different arenas, to the guidelines covered in #1 above. #2, however, is what would hold me back from granting them authorship in favor of acknowledgment elsewhere in the paper, which is typically bestowed upon individuals who deserve recognition for their contributions to the project at a sub-author level. Obviously, then, the third PI would fall squarely into this category.
There are cases that tend to get a bit dicier than this, though. One that often comes up is the use of resources as a bribe of sorts in exchange for authorship. For example, the third PI could just as easily have refused to provide the antibody unless she were guaranteed to be listed as an author, and while doing this would be considered extremely faux pas and would likely toe the oft-ambiguous line between “furthering one’s career” and “committing an act of scientific misconduct,” it’s not as though it’s never happened before. Indeed, this would put the graduate student and her PI in a sticky situation: They need the antibody, but would they be willing to contribute to authorship misconduct in order to get it?
The list goes on. Authorship issues can come up anytime, especially in the context of misbehavior or poor communication and planning among the various parties, and they certainly demand some interesting ethical considerations when the qualifications for one’s authorship are equivocal.
For any researchers reading this: How would you handle this case study? Who should be an author, and who should not?
<p>From <a href=”http://drugmonkey.wordpress.com/” rel=”nofollow”>drugmonkey</a> (comment from a different post):<br />
“Easy. All are authors. I just HATE these bogus case study examples in ethics classes. Look this stuff is culturally determined. In some areas of science, all of these are authors. In others, only the prime two. So it boils down to, “what is the practice in your subfield”.
“But let’s examine this a little. We’re not talking about authorship position which is a whole ‘nother debate. We’re talking about add-on middle authors. who cares? I mean really?”<br />
“The issue REALLY boils down to one of fairness if you ask me. Authors only get ticked for adding more middles on their paper if they perceive that they are not being afforded similar opportunity to add to their CV with little effort. But this fairness issue has nothing to do with community level prescriptions for what constitutes authorship…”</p>
<p>I agree with you, drugmonkey, regarding the fairness issue, and I suspect the community level prescriptions you mentioned arose out of situations precisely like these in order to maintain fairness. The case above isn’t going to have effects on the entire field, or even an entire lab, but it could have direct implications on the CVs, for example, of those involved. What if the tech were applying for grad school and needed some publications? Certainly, if the PI could help it, she’d be on the paper for sure. What if the second PI were up for tenure? You can bet she’d be lobbying for herself to be on it.</p>
<p>Issues like this come up all the time as you probably know, and we should care about it, not because they have direct implications for us but because they at least have that potential. Incidentally, I heard a story recently of a PI who’s up for tenure and, without telling any of the other authors, made himself first author and bumped his own grad student who’d done pretty much 100% of the work. Clearly, this is more of an author placement issue, which indeed is a much different debate for a different time.</p>
<p>Also, to be clear, the case above was real. It happened. Not sure who got on in the end, though.</p>
sorry, original comment went elsewhere but thanks for placing it where it was intended.
this case is not only real it takes place about a hundred times a day! my question is, where’s the cost? who is hurt and how?
The real credit issues boil down to who is first author, who is last author and who is NOT an author who SHOULD be an author? The question of add-on middle authors isn’t anywhere near as important.
The PI who substituted herself as first author without mentioning it to the grad student is an interesting issue. The whole point of training is that someone needs to be…trained. So in many cases the grad student (or postdoc!) puts together a draft. and it may take a LOT of effort. but it may also not be really a paper yet and the PI may end up basically rewriting the thing. So what to do? The PI likely did most of the heavy conceptualization of the research area, funded the project and did the effective work on the actual submitted MS. In some views this is the first author. In other views, even if the PI does the whole design/direct/draft paper from start to finish s/he should be last author to show that it is his/her lab group. So the PI has to find a first author and upjumps a tech or grad student.
<p><blockquote>The real credit issues boil down to who is first author, who is last author and who is NOT an author who SHOULD be an author? The question of add-on middle authors isn’t anywhere near as important.</blockquote></p>
<p>100% agreement. But, that’s not to say it’s not a prevalent thing, as you’ve pointed out, nor an important issue; it is both of these.</p>
<p>I’d guess the harm stems from a perceived “dilution” of the importance of the first author’s work in this case, especially in the case of the tech being added on. All things being equal, a paper with two authors - a grad student and her PI - looks a lot better for her than a paper with six authors, four of whom did next to nothing.</p>
Depends how you look at it and what the eventual track record is. Sure, it is clear in the two author scenario that the grad student did the work (unless second author in which case????). But if someone has a CV with lots of multi-author papers this is interpreted as being the field that the person is in. If there are enough first authorships, well, that’s all the more to the credit that she pulled this off with so much internal competition for authorship…
Look I know trainees tend to obsess about 2nd, 3rd etc positioning on papers but this just doesn’t matter where it counts. There is practically no number of 2nd authorships that trumps one first authorship. You have to have pubs and people do care about pub/no pub so middle authorships count in this way. but as far as the question of “is this person on the track for independence?” goes? first authorship is everything.