Relative Value of Adapted Novel Bibliometrics in Evaluating Surgical Academic Impact and Reach.

Abstract

The Hirsch index, often used to assess research impact, suffers from questionable validity within the context of General Surgery, and consequently adapted bibliometrics and altmetrics have emerged, including the r-index, m-index, g-index and i10-index. This study aimed to assess the relative value of these novel bibliometrics in a single UK Deanery General Surgical Consultant cohort. Five indices (h, r, m, g and i10) and altmetric scores (AS) were calculated for 151 general surgical consultants in a UK Deanery. Indices and AS were calculated from publication data via the Scopus search engine with assessment of construct validity and reliability. The median number of publications, h-index, r-index, m-index, g-index and i10-index were 13 (range 0-389), 5 (range 0-63), 5.2 (range 0-64.8), 0.33 (range 0-1.5), 10 (range 0-125) and 4 (range 0-245), respectively. Correlation coefficients of r-index, m-index, g-index and i10-index with h-index were 0.913 (p < 0.001), 0.716 (p < 0.001), 0.961 (p < 0.001) and 0.939 (p < 0.001), respectively. Significant variance was observed when the cohort was ranked by individual bibliometric measures; the median ranking shifts were: r-index - 2 (- 46 to + 23); m-index - 6.5 (- 53 to + 22); g-index - 0.5 (- 24 to + 13); and i10-index 0 (- 8 to + 11), respectively (p < 0.001). The median altmetric score and AS index were 0 (range 0-225.5) and 1 (range 0-10), respectively; AS index correlated strongly with h-index (correlation coefficient 0.390, p < 0.001). Adapted bibliometric indices appear to be equally valid measures of evaluating academic productivity, impact and reach.