Abstractions and Authorship
Clinical Advances in Hematology & Oncology
August 2015, Volume 13, Issue 8
August 4th, 2015 is the deadline for submission of abstracts for the American Society of Hematology (ASH) meeting that will be held in Orlando in December. On the ASH website is a detailed description of requirements for the abstract, which include: no more than 3800 characters (excluding spaces, tables, figures, and authors’ names, along with the names of their favorite pets), conflict of interest disclosures, a guarantee of no prior presentation of data to a meeting of 1000 or more participants, proper formatting, the withdrawal deadline, and responsibilities of the first (presenting) author.
There is a conspicuous absence of guidelines for the number or order of authors. Having been involved in the submission of numerous abstracts of late, including to the meetings of the American Society of Clinical Oncology, the European Society of Hematology, the International Conference on Malignant Lymphoma, and now ASH, I feel highly qualified to offer a primer on what seems to be the process, at least for pharmaceutical-sponsored or cooperative group studies.
Decisions regarding authorship are not for the faint of heart. Although all of the authors are credited in some way for their participation, when an abstract is cited, often the names of just the first 3 authors are listed—followed by the ubiquitous et al. Unless you are a Win, Place, or Show (to draw a comparison to horse racing), you are relatively anonymous. The decision process can be chaotic; politics often circumvent scientific contribution, feelings are hurt, and there is a sense that careers may be put in peril—all over the placement of a couple of words.
Let me give you some examples, with details changed to protect the innocents. A principal investigator on a study, the one who helped design it and who has been responsible for its conduct, expects to be the first author. This person also expects to be the one to present the data should the abstract be accepted. However, someone else accrued more patients to the study and, as such, stakes a claim to the number-one spot. For political reasons, the principal investigator is then relegated, somewhat begrudgingly, to last (senior author) place. However, another author, who has been placed in the close-but-no-recognition fourth spot, claims entitlement to the senior slot based on career duration and emotional grounds. The principal investigator, in an act of collegiality, acquiesces and is relegated to the penultimate spot. It is kind of like Matthew 20:16, in which “the last shall be first, and the first last,” except that in this case, the first becomes “next to last.”
The statistician is traditionally awarded the second position. However, this study had multiple statisticians over its duration, leading to a scuffle among them and the scientific coauthors regarding spots 3 through however many. All the while, nuclear medicine physicians are campaigning for the third spot. Additional authors are selected on the basis of their contributions to the study: accrual of a certain proportion of the patients, taking a leadership position in the group, or playing a role in study design. Where they all end up is rather arbitrary, but often contentious.
The worst situation is to be omitted altogether. Most clinical trials take years from concept generation to activation, completion, analysis, and abstract preparation. And then there are the long-term follow-up studies. A leadership role in drafting the abstract is granted to an eager, up-and-coming junior investigator with no prior involvement in the trial. The abstract is then circulated with the junior investigator’s name, while personnel from a bygone era who were critical to the genesis of the original trial are forgotten.
There should be a better way to determine authorship. The coauthors could have a civilized discussion and decide among themselves who ends up where; however, this assumes that none of them have a vested interest or feel as if a single abstract is crucial to an entire career. They could draw cards, or roll dice. Alternatively, there could be a winner-take-all cage match!
What we need are unambiguous guidelines, and somebody who has the authority and willingness to enforce them. That somebody needs to have a historical perspective. This should help solve the problem, until it comes time for what really revs up the tsunami: the order of authors in the actual manuscript.
Until next month . . .
Bruce D. Cheson, MD