Savitz, David A.
Career progression and authorship follow a fairly predictable course. The first papers epidemiologists write often reflect opportunities provided by established investigators, frequently using existing data. The junior partner typically is the first author of such papers, with the research mentor listed second or last as a reflection of their valued, important role.
As researchers progress, the mentoring role becomes more prominent. Coauthorship (second or last) comes to predominate, with many more publications being facilitated than written. This progression provides vital learning opportunities for new investigators, blending the energy of younger researchers with the wisdom of their elders, each party receiving the credit they deserve. The comments that follow are not intended to discourage senior epidemiologists from providing these opportunities.
My argument is that it is not enough just to mentor others as authors. Senior epidemiologists should also continue to be first authors, with all that implies. In this issue of the journal, there are first-authored publications by Anders Ahlbom1 and Noel Weiss,2 and in recent issues by Gladys Block,3 Aaron Folsom,4 and Kyle Steenland.5 These senior members of our profession merit particular credit for taking the time and initiative to serve in this role, on top of their many other administrative, mentoring, and teaching obligations. The motivation for my plea is not only that such authors enhance the prestige of the journal but, more importantly, that first-authorship by senior epidemiologists benefits the science and the profession of epidemiology. My reasons are as follows:
1. Senior researchers are sometimes the ones most able to formulate and articulate a vision that transcends isolated (albeit important) research findings. Someone needs to connect the dots, challenge the simplistic interpretations, and point the way to the next steps. Although less experienced investigators sometimes do so, the richer perspective of experience is often needed. The breadth and depth of insight that senior scholars bring to expert panels or review committees are no less valuable in academic journals.
2. The proactive role inherent in first authorship requires struggling with the nuances of data analysis and composition of manuscripts. This helps to keep the senior researcher current in the literature and research methods. Ultimately, it makes them better mentors for the many times when they coach rather than play. Even in routine data-analysis papers, the first author infuses a distinct perspective that will be subtly different if led by the senior investigator. Remaining in the fray by having one's papers rejected, doing battle with fussy editors, and having to carefully format references sustains character and can be therapeutic.
3. The profession benefits when senior researchers continue to put out their ideas for reaction and criticism. The same candor and strong, informed point of view that constitutes leadership of research teams, academic units, or professional societies is needed in published work. Rather than limiting such views to immediate colleagues or those who happen to attend the author's seminars, first-authorship allows a wider audience to reflect on the senior member's insights (and to disagree when necessary) in the more structured, open, scholarly process of publication.
I have shared this perspective and plea over the years with many senior colleagues whose insights I value. Not surprisingly, to the best of my knowledge, I have had no effect whatsoever on their behavior. The very achievements that would make their perspective so valuable in print are an indication that they have no reason to heed my unsolicited advice on setting priorities. The most frequent counterargument I hear is that the mentoring role is more important for them than the first author role. If they could only do one or the other, I would support that priority. However, these are typically the individuals who are fulfilling a dozen different roles with great success already. I am doubtful that those senior researchers who write more mentor less.
Another response is that they often fulfill the role of first author but allow their junior colleagues to receive the credit. However, in these collaborative efforts, an admittedly hazy line could be crossed at which the senior researcher should publicly acknowledge their lead role as first author. In my first opportunity to work with a research mentor toward publication, we discussed authorship early in the process. I remember well her reasoned approach to the issue; I would work with her in conducting the analyses and I would take the lead in developing a draft with her guidance. If that draft turned out to be reasonably close to a publishable paper, I would be first author, whereas if she had to completely rework the document, she would be first author and I would be second. Allowing for such a contingency made great sense then and now. There are certain expectations associated with first authorship on the part of journal editors, readers, and those who are evaluating the accomplishments of the junior investigator, and these expectations call for truth in labeling.
Primarily, I am expressing my appreciation to the specific senior researchers whose first-authored work appears in this issue and in past issues of the journal. I hope to inspire at least some others of their stature to consider emulating their achievements. Because arguing with senior colleagues one at a time has failed (perhaps analogous to an individual clinical intervention), then maybe I can be more influential using the population approach. If these words can nudge many people a small amount, a few could fall over the precipice to dig into their files, literally and figuratively, and share their insights. We would all be the better for it.
1. Ahlbom A, Feychting M, Gustavsson A, et al. Occupational magnetic field exposure and myocardial infarction. Epidemiology. 2004;15:403–408.
2. Weiss NS, Dhillon P. Case-control studies on the efficacy of cancer screening: overcoming bias from non-random patterns of screening. Epidemiology. 2004;15:409–413.
3. Block G, Mandel R, Gold E. On food frequency questionnaires. The contribution of open-ended questions and questions on ethnic foods. Epidemiology. 2004;15:216–221.
4. Folsom AR, Anderson JP, Ross JA. Estrogen replacement therapy and ovarian cancer. Epidemiology. 2004;15:100–104.
5. Steenland K, Deddens JA. A practical guide to dose-response analyses and risk assessment in occupational epidemiology. Epidemiology. 2004;15:63–70.
© 2004 Lippincott Williams & Wilkins, Inc.