I realize I am a few days late to this discussion, but it was my birthday. I was busy getting older.
Over at the Drugmonkey blog and out in the Twitterverse, the PIs of the blogosphere are haranguing over some guidelines released by the British Medical Journal on the criteria for authorship. Here is what the guidelines say:
The uniform requirements for manuscripts submitted to medical journals state that authorship credit should be based only on substantial contribution to:
- conception and design, or analysis and interpretation of data
- drafting the article or revising it critically for important intellectual content
- and final approval of the version to be published.
All these conditions must all be met. Participation solely in the acquisition of funding or the collection of data does not justify authorship.
We want authors to assure us that all authors included on a paper fulfil the criteria of authorship. In addition we want assurance that there is no one else who fulfils the criteria but has not been included as an author.
Many of the commenters debating the esteemed Dr. Monkey on his blog have gotten wrapped up in this idea that courtesy authorship is RUININGZ TEH JOURNALZ!!! I say unto them that this is tomfoolery! I see no data-based evidence beyond the stories of a few folks that this is a substantial problem. And I happen to agree with Drugmonkey that the addition of one author does not dilute the contributions of all the others on the types of papers most of us are publishing – the kinds with a half dozen or so authors. Meh, I say. I also agree with him that when you Pubmed stalk someone (as we all do) and you see that they have 300 total publications with 40 as last author, it’s easy enough to conclude that bullshittery is afoot and move on.
Now, I have no particular insights into the rationale behind BMJ’s statement. But, as I pointed out at Monkey’s joint, I suspect that the BMJ is trying to deal mainly with a very limited problem that is specific to a subgroup of physician ass clowns doing large-scale clinical trials where the staff and director of every data collection center are listed as authors. Having participated in that sort of trial, I can tell you that I had a minimal role in the design, no real role in the interpretation, and no role in the crafting of the manuscript. We were basically a fee for service site. I collected data and sent it back to the main study center. It would have been wrong for me to have demanded authorship there, yet the center director whose contributions were equally minimal was adamant that he be listed. Some clinical trials are released with copious numbers of authors.
Need I remind anyone of this?
Figure 1: A publication from the Human Genome Project. A partial list of authors required a full page in Nature to publish.
Just the partial authorship list of the above paper had more than 250 authors. The full list had to be published in a supplement. So, I conclude that in the face of such shenanigans, the addition of a sixth author to a manuscript isn’t doing dick to the future of science.
But, there is one group I worry might be hurt from overly stringent interpretation of these sorts of guidelines – undergraduates and other more junior investigators. No new investigator is going to leave a super high level PI with the ability to squash them like a bug off of the author list. If this is indeed a problem (which I do not acknowledge), does anyone really see it going down like this?
New PI: Can I have some of your new antibody?
BSD PI: Yeah, just include me in the author list.
New PI: Hmmmm…your contribution does not seem to meet the BMJ criteria for authorship.
BSD PI: Why, by jolly! You’re right! How silly of me to have considered it! I shall have one of my 30 graduate students send your antibody along post haste! Best of luck to you, tally ho!
More likely, folks might interpret these guidelines to mean that very junior folks, like undergraduates, don’t meet the criteria for authorship.
The last paper I submitted for review had two undergraduates listed in the middle. Each of them worked on a different aspect of the research. How much did they contribute to the design of the project? Well, if you compare their contribution to mine, in terms of what ultimately ended up being the design, it looks sort of like this:
Figure 2: Relative contributions to my most recent manuscript without considering the contributions of other authors.
But I think you have to consider the meaningfulness of that relatively small overall contribution relative to their capacity to contribute. For me, as someone who is familiar with the literature and has a battery of techniques already in my arsenal, my contributions required a more minor dedication of my time and talent. For them, these contributions required a much more substantial dedication of their time. Troubleshooting a portion of a method is no joke to them. So, are their contributions meaningful? I say, absolutely.
I admit that I did the majority of the writing, but we discussed the data and it’s meaning and presentation. And, I will admit that in one case I decided to display the data differently after talking to them. I also used some of the SOPs they had written to draft the methods section. There is little doubt in my mind that my mentoring of them affects how I think about my science. So, did they make a contribution to the design of the manuscript? Absolutely, even though they never added text to the formal manuscript document. And, certainly, each of them approved the final product before it was submitted.
I think we need to take great care to not over interpret these guidelines or to create too much of the problem of “courtesy authorship” at the exclusion of those who warrant authorship. Especially given the fact that the benefit afforded to a very junior scientist far outweighs the benefit of authorship or detriment of not awarding for a high level PI. Novice investigators are the ones more likely to actually be hurt by the overly judicious use of these new standards.
