Predominantly middle authors in radiology: a bibliometric assessment

The participation in "team science" has led to the growth of co-authorship counts with resultant more middle authors in the medical and radiological literature. We sought to characterize predominantly middle author (PMA) academic radiologists and compare their bibliometrics with non-PMA (NPMA) faculty for advancement.

We performed a cross-sectional bibliometric analysis of 3,450 faculty who had ≥20 publications across 10 subspecialties at 143 U.S. institutions. PMAs were defined as authors with ≤10% combined first- and last-authored/total papers. A ≤20% sensitivity analysis was also performed. Mann-Whitney U tests, chi-square tests, and OLS regression (unadjusted, age-adjusted, and age+rank-adjusted) were performed.

139 (4.02%) of 3,450 faculty met PMA 10% criteria. 10% PMAs were significantly underrepresented at the full professor rank (19.4% vs 43.2%, p<0.001) and had shorter academic careers (median 16 vs 21 years, p<0.001) than NPMAs. PMAs had fewer total papers (median 34 vs 60) and lower h-indices (16 vs 19) (p<0.001). The number of citations/paper however favored PMAs (median 30.0 vs 24.6, p=0.040). After adjusting for academic age, total citations and h-index differences became non-significant, while the citations/paper advantage of PMAs strengthened (p<0.001). PMA prevalence correlated with subspecialty mega-author (>25 authors) paper participation (Spearman ρ=0.766, p=0.010). All findings were corroborated at the 20% sensitivity threshold.

PMA radiologists demonstrate lower total publications and h-indices than NPMAs, yet their papers attract significantly more citations, suggesting participation in higher-impact collaborative research. They are under-represented at full professor ranks despite their contributions.