“But many that are first shall be last; and the last shall be first”

Abstract

In their recently published manuscript, Falagas et al. (1) purport to study at what age biomedical scientists do their "best work." Hence, based on the assessment of the total number of citations of 1500 articles published by 300 "randomly selected", highly cited scientists in six different biomedical areas of research, the authors suggest that "high quality scientific productivity" peaks at the ages of 30-35 years, with significant decreases from approximately 40 years of age and beyond. However, it must be noted that the authors chose to limit their analysis to articles for which the researcher was listed as first author. Therefore, I submit that the study merely reflects the fact that biomedical scientists, on average, are likely to work as post-doctoral researchers during the period of 30-35 years of age, and the study therefore does not necessarily reflect accurately on the association between age and "high-quality research output" or "landmark scientific contributions." Indeed, it is axiomatic in the biomedical sciences that the first author position on a manuscript is held by the graduate student or post-doc who performed the work at the bench (and contributed to the conception and development of the theoretical foundation of the study), whereas more senior researchers are usually positioned as last authors of the manuscript. The observation that "older scientists" (>40 years) also may contribute as first authors on some publications, according to the study of Falagas et al. (1), is likely explained by the fact that the order of authors is often reversed on review articles (as opposed to original articles); i.e., the senior scientist now becomes the first author. Such review articles are also frequently cited in the literature, and this could thus have contributed to the outcome of the above-mentioned analysis. We thank Dr. Fadeel for the interest in our work. The comment about post-doctoral fellows and the way order of authorship, in original and review articles, is sometimes decided in academic research projects is well taken. However, we believe that the degree of bias that this could have introduced to the results of our analysis is rather limited (1). One of the things that troubled us when we designed the methodology of our study was whether we should include last or middle authors' articles in our analysis. Generally, those listed first or last in the author list are considered to have contributed more to the work than middle authors. A recently published survey among promotion and tenure committee chairpersons assessed the perceptions of author contributions based on author list position and the total number of authors in a paper; in a five-author paper, the last author is perceived to have contributed from a low 16% of the total work (initial conception, work performed, and supervision) to a high 38% if he or she is also listed as a corresponding author (2). On the other hand, the first author is always perceived to have contributed more than his co-authors (2). Additional evidence that helped us decide on the methodology came from a study (3) investigating the number of named authors who do not meet the International Committee of Medical Journal Editors' (ICMJE) criteria for authorship (4) among articles published by three major medical journals. Up to 21% of all articles had an honorary author, and in 24% of these cases, the honorary author was positioned last in the author list. On the other hand, an honorary author was never the first author. Of interest, a study investigating researchers' self-reported contributions for articles published in The Lancet concludes that the number of contributions made by the first author is statistically, significantly greater than that made by the other authors, and it is unclear whether the order of subsequent names is related to the scientific contribution (5). Based on Dr. Fadeel's assumption, senior scientists may have contributed as first authors on some publications' articles that are mainly review papers. However, the scientific articles included in our analysis are the best work of the most prominent scientists and are truly landmark and historical contributions that received many of citations and made their authors well known in the scientific community or even offered them the Nobel Prize. Thus, it comes without surprise that only few review papers are among the most cited work of the highest-level scientists. Finally, in almost 20% of the top scientific work included in our analysis, researchers were older than 50 years old. In conclusion, we acknowledge that a relevant analysis of highly cited work of last authors would have a certain value, but we do not believe that it could serve the purposes of the present study. Disclosure: For the record, the first author of our article is 46 years old, and the last author is only 27 years old.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 12 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.