In the world of bioscience, whether you are a scientist, funding agency, prize jury, promotions committee or dean, you are constantly seeking better ways to understand the impact of research advances. This may appear straightforward: find evidence that the work is original, reproducible and—by virtue of its acceptance and adoption by others—has led to major changes in a given field. Sometimes the evidence appears obvious to educated observers and can be confirmed by considering the opinions of recognized subject matter experts.
Often, however, those involved in such assessments seek to supplement this time-honored approach by quantifying how frequently published findings are cited by other researchers. While this data-driven approach for assessing a publication’s impact through citation counts and related metrics will likely never fully replace expert opinion, the use of informatics is increasing as the number and diversity of publications also increase and generate greater competition for the rewards of high-impact research. This use of informatics to evaluate publications has become something of an academic discipline. It is in part a response to the difficulty of separating signal from noise in the sea of published information and to concerns about the poor reproducibility of many published findings, even those found in highly regarded and “high impact” journals.
There is much more that can be said about this important issue, especially as it is highly relevant to the academic life of institutions such as Harvard Medical School. But for the moment, I would like to call your attention to an article titled “A list of highly influential biomedical researchers, 1996-2011,” published last year in the European Journal of Clinical Investigation. The authors, who are leaders in this particular area of inquiry, used various sources to identify those 400 biomedical researchers—out of a pool of 15 million authors—who had the highest total citations or the highest h-index (a commonly used indicator of publishing impact) during a 15-year period. Their overall goals were to provide data that would enable authors and others to identify potential limitations and caveats to the informatics approach, to further explore the relationship between such indices and other indicators of scientific impact and to identify potentially useful applications of the data.
It is worth noting that of the 400 most influential biomedical researchers identified in this paper, no fewer than 47 are Harvard faculty; of these, the greatest number are HMS faculty, representing the Quad and our affiliated institutions. No other university, school or research institution comes close. Although I believe this is an imperfect and limited measure to judge the overall research influence of an institution, I view the data as an indicator of the astonishing success of Harvard Medical School and as a meaningful indicator of the depth and accomplishment of our remarkable faculty.