Pediatric Surgical Research Output in Germany in the Last 30 Years - An Assessment and International Comparison of Three Dedicated Paediatric Surgical Journals

Abstract

Purpose: Research output of once-leading countries in surgical journals is decreasing despite an overall increase of scientific publications by 8% per year. We aimed to assess research outputs of German, Dutch, and Israeli pediatric surgeons in dedicated pediatric surgical journals in order to get insight into trends in pediatric surgical research. Methods: We collected bibliographic information on all original articles in the Journal of Pediatric Surgery, European Journal of Pediatric Surgery, and Pediatric Surgery International in 1985-1988, 2000-2003, and 2015-2018 that had a German, Dutch or Israeli last author from a department of pediatric surgery. Citation counts were obtained from the Web of Science. Results: Research output of German pediatric surgery decreased from 19 manuscripts in 1988 (0.1/surgeon/year) to eight manuscripts in 2017 (0.02/surgeon/year), whereas those of the Netherlands increased from two manuscripts in 1985 (0.08/surgeon/year) to 12 manuscripts in 2016 (0.3/surgeon/year). The declining German research output negatively correlated with increasing numbers of specialist pediatric surgeons for total (τ = -0.54; P = 0.0156) and manuscripts per surgeon (τ = -0.79; P = 0.0001), resulting in a negative trend over time (χ² = 11.845, P = 0.0006). Analyses of citation patterns revealed that manuscripts by Dutch pediatric surgeons and those published in the Journal of Pediatric Surgery had higher absolute citation counts than the reference category of a German manuscript in the European Journal of Pediatric Surgery. Age-corrected citation rates resembled this result by increasing from 2000 to 2003 ( $\tilde{\text{x}}$ = 0.799, range: 0-3.368) to 2015-2018 ( $\tilde{\text{x}}$ = 2, range: 0-5) (P = 0.035) for the Netherlands. Assessment of manuscript types revealed that the proportion of prospective studies increased in the German sample (χ² = 5.05, P = 0.0246), but remained the lowest among the comparators. Surprisingly, the proportion of non-clinical manuscripts from Germany also increased over time (χ² = 4.001, P = 0.0455), whereas it remained constant in both the Netherlands and Israel. Conclusion: German pediatric surgical research output decreased in the last thirty years based on the sample of dedicated pediatric surgical journals, while Dutch productivity increased. Citation rates-as a measure of scientific impact-were associated and increased with Dutch manuscripts. The involved factors remain to be determined and whether this represents a shift toward other journals or mirrors a general development.

Keywords: academic surgery; bibliometrics; citation analysis; geographical research distribution; institutional research distribution; research productivity.

Figure 1

Absolute and relative number of manuscripts published in three pediatric surgical journals by examined countries. (A) Absolute number of manuscripts per country. (B) Relative research output in manuscripts per hospital-based pediatric surgeon specialist by country. Research output per surgeon declined over time (χ² = 11.845, df = 1, P = 0.0006) in Germany, but neither in the Netherlands (χ² = 0.607, df = 1, P = 0.436) nor in Israel (χ² = 0, df = 1, P = 1).

Figure 2

Hospital-based board-certified specialist pediatric surgeons in Germany by year according to the data published by the Bundesärztekammer (General Medical Council). They were highly correlated (R = 0.98, 95% confidence interval: 0.96–0.99, P < 0.0001) and showed a linear increase of 10.4 (95% confidence interval: 9.6–10.4) hospital-based board-certified pediatric surgeons per year (F_{(1, 32)} = 684.8, P < 0.0001).

Figure 3

Correlation of hospital-based pediatric surgeons in Germany and scientific output in the three pediatric surgical journals. (A) The absolute number of manuscripts decreased with a concomitant increase in the number of hospital-based surgeons (τ = −0.54, P = 0.0156). (B) Highly negative correlation of the relative research output and the number of hospital-based pediatric surgeons (τ = −0.79, P = 0.0001).

Figure 4

Treemap visualization of the cumulative share of manuscripts of the respective pediatric surgical center in the observed periods 1985–1988, 2000–2003, and 2015–2018. (A) Total manuscripts of all three journals. (B) Manuscripts in the Journal of Pediatric Surgery. (C) Manuscripts in the European Journal of Pediatric Surgery. (D) Manuscripts in Pediatric Surgery International. Preference of several pediatric surgical centers for some journals is indicated by their different representation among the sample (P = 0.0073), in which both the European Journal of Pediatric Surgery and Pediatric Surgery International differ from the Journal of Pediatric Surgery (P = 0.0063 and P = 0.0123, respectively), but not among each other (P = 0.414).

Figure 5

Manuscripts published by academic and non-academic pediatric surgical centers in Germany. The number of manuscripts from non-academic centers experienced a negative linear trend (χ² = 10.9, df = 1, P = 0.001).

Figure 6

Absolute citation counts of all included manuscripts. Citation data obtained from Clarivate's Web of Science.

Figure 7

Journal- and country-wise comparisons of absolute citation counts in the three investigated timeframes. Data represents mean and 95% confidence intervals calculated from the zero-inflated negative binomial regression presented in Table 7. Non-overlapping confidence intervals indicate significant differences. P-values were corrected for multiple comparisons using Tukey's method.