The H-index in academic radiology

Abstract

Rationale and objectives: The h index is a recently developed tool to assess the impact of an author's publications. The purpose of this study was to apply and evaluate the h indexes of US academic radiologists.

Materials and methods: Radiology programs that participated in the residency match in 2009 were identified through the National Resident Matching Program's Web site. One third of programs were randomly selected. The academic ranks (instructor, assistant professor, associate professor, professor, or chairperson) of faculty members were determined on the basis of information from the programs' Web sites during October and November 2009. One third of radiologists at each randomly selected institution were randomly selected for detailed analysis. For each radiologist, an automatically computed h index was obtained through the Scopus database. The h index was compared across ranks using analysis of variance. A multivariate logistic regression analysis was also performed to determine the best predictors (number of publications, number of citations, h index, and number of citations per publication) of academic rank.

Results: Sixty hundred eighty-three radiologists from 47 programs were included in this study. The mean h indexes were 1.1 +/- 2.7 for instructors, 2.3 +/- 4.1 for assistant professors, 6.2 +/- 7.2 for associate professors, 12.5 +/- 10.8 for full professors, and 12.0 +/- 9.5 for chairpersons. There was a significant relationship between h index and academic rank (P < .0001). Multivariate logistic regression analysis demonstrated that h index (P < .0001) and number of publications (P < .0001) were the best predictors of academic rank.

Conclusion: There exists a significant relationship between h index and academic rank, with h index increasing with academic rank. These results offer a benchmark for comparing a given academic radiologist to national averages.