WAMErs:
I have an authorship order question which I would appreciate either feedback on, or direction to an appropriate quasi-official statement on the matter. The situation is a study coordinated and probably written up by staff of a private biotech company. The article is on a methodological side-study, not the parent study on the company product. All correspondence has taken place exclusively with company personnel (non medical), only one of whom is even listed as an author, and the first author has had no apparent input into revisions. The first author is an MD who helped oversee the parent study, but likely had minimal input into this methodological aspect, and, arguably, into the paper (hard to tell). Inquiry as to the various roles leads to the addition of 4 authors from the company, but no details about the first author except that his/her role fulfilled ICMJE criteria. Let us say, for argument's sake, that the MD indeed satisfies the criteria for authorship, somewhere on the list, but by no stretch of the imagination should they be considered the first author. Does a journal have any responsibility here?

Steve Goodman
Clinical Trials
Annals of Internal Medicine
___________________________

As far as I know, there are no official guidelines about the order of authors, yet, as several people have rightly pointed out, most readers/researchers believe the order indicates something about who did what, although I have found that these conventions vary between disciplines. My personal impression is that alphabetical listing seems to be declining—but I may be wrong. It would, however, seem a useful remedy in this case, so long as there is a note alerting readers to the fact that this system has been used. However, imposing an order on an author list might seem outside an editor's powers, especially if there is no mention of this in the journal's instructions.

On a slightly tangential issue, I was delighted to hear of a journal's action 'uncovering' deserving company authors. The Good Publication Practice guidelines for pharmaceutical companies (http://www.gpp-guidelines.org) state that, whatever criteria are used to determine authorship, they should be applied equally to employees of a commercial sponsor and external investigators. I guess one might also add that any system of determining the order of listing should also be applied fairly (ie, company employees should not automatically go to the bottom of the heap).

Liz Wager
Publications Consultant, Sideview
___________________________

What about having an "authors' contributions" section like this: http://www.ij-healthgeographics.com/info/instructions/default.asp#authorscon?

Maged N Kamel Boulos
Editor-in-Chief, Int J Health Geogr

___________________________

I agree with Liz Wager that imposing an order based on contribution to a manuscript is outside an editor's powers, as there is no hope of exercising control over what goes on beyond the virtual editorial "office" and is not visible in the text itself.

By way of illustration, I'll mention that as a text consultant I worked three times (and no longer do) with an author who unabashedly took turns with a friend to be first author on papers so that both could double their output. Guest authorship is not surprising, but the case of these two was extreme as their personal protocol literally required them never to ask the other to have anything to do with producing the other's paper. The "guest" might have looked through a microscope on some of the cases, but it wasn't a requirement to have done so. They kept count and made the decision on who was named first according to who was up for review. Thus, even the first author might be a "guest" and the second the writer. Guest authorship was not confined to a lofty supervisor at the end of the list.

Nevertheless, as Liz says, readers infer, romantically, that order is meaningful, at least in the life sciences and medicine. It would be welcome if journals could include information to correct erroneous inferences, merely for the sake of contributing to an accurate understanding of how research proceeds for historians, for instance asking that the first be the principal writer and at the same time encouraging statements about others' contributions in a special section.

Still, there is no hope of forcing people to tell the truth on a point that leaves few fingerprints. A linguist might be able to find evidence of degrees of influence in authorship in style and voice in some cases, just as an academic exercise, but it is wishful thinking that this could be done in peer or editorial review.

Regards,
M. E. Kerans
___________________________

At the South African Medical Journal, the first author is the person that puts pen to paper to write up the article. We also reserve the right to be told what each named author has actually contributed to the paper in question. But, as Dr Keran notes, the editor has no way of confirming what he or she is told by the authors, but I believe it is good practice to have some kind of rule.

Dan Ncayiyana
South African Medical Journal
___________________________

In my opinion, the primary author has overall responsibility for the manuscript and if he doesn't, then he should not be the primary author. We recently had an incident in which a co-author acted against publication policy without apparent knowledge of the senior author of a paper. We still addressed all communication to the senior author and, ultimately, held him responsible.

Norbert Gleicher
J Assist Reprod and Genetics

___________________________