Impact of Self-citation on the H Index in the Field of Academic Radiology

Rationale and Objectives: The Hirsch Index (H index) is widely applied as a metric of scientific productivity. The purpose of this study was to evaluate the role of self-citation on the H index in academic radiology. Materials and Methods: Through the National Resident Matching Program's Web site, one third (47/139) of radiology residency programs were selected randomly. All chairpersons and full professors were included. Using the Scopus database, we calculated the H index as well as the number of cumulative citations with and without inclusion of self-citations. We determined the proportion of academic staff in which H index increased by one, two, or greater than two integers. We also correlated the proportional increase in H index before and after inclusion of self citations with the number of publications. Results: A total of 487 academic staff (47 chair and 440 professors) was identified. Because of self-citation, mean +/- SD of the H index increased from 13.7 +/- 9.9 to 14.0 +/- 10.2; mean SD of cumulative citations increased from 1804 +/- 1889 to 1870 +/- 1971. H index numbers did not change in 376/487 (77%) authors as a result of self-citation. There was no correlation between number of publications and proportional change of H index. Conclusion: The effect of self-citation is minimal in academic radiology, as evidenced by the fact that cumulative citations increase by only 2% and the large majority of H index values do not change by even a single integer after inclusion of self-citation.