Surgeon Scientists Are Disproportionately Affected by Declining NIH Funding Rates

Abstract

Background: Obtaining National Institutes of Health (NIH) funding over the last 10 years has become increasingly difficult due to a decrease in the number of research grants funded and an increase in the number of NIH applications.

Study design: National Institutes of Health funding amounts and success rates were compared for all disciplines using data from NIH, Federation of American Societies for Experimental Biology (FASEB), and Blue Ridge Medical Institute. Next, all NIH grants (2006 to 2016) with surgeons as principal investigators were identified using the National Institutes of Health Research Portfolio Online Reporting Tools Expenditures and Results (NIH RePORTER), and a grant impact score was calculated for each grant based on the publication's impact factor per funding amount. Linear regression and one-way ANOVA were used for analysis.

Results: The number of NIH grant applications has increased by 18.7% (p = 0.0009), while the numbers of funded grants (p < 0.0001) and R01s (p < 0.0001) across the NIH have decreased by 6.7% and 17.0%, respectively. The mean success rate of funded grants with surgeons as principal investigators (16.4%) has been significantly lower than the mean NIH funding rate (19.2%) (p = 0.011). Despite receiving only 831 R01s during this time period, surgeon scientists were highly productive, with an average grant impact score of 4.9 per $100,000, which increased over the last 10 years (0.15 ± 0.05/year, p = 0.02). Additionally, the rate of conversion of surgeon scientist-mentored K awards to R01s from 2007 to 2012 was 46%.

Conclusions: Despite declining funding over the last 10 years, surgeon scientists have demonstrated increasing productivity as measured by impactful publications and higher success rates in converting early investigator awards to R01s.

Figure 1

(A) Total NIH funding. Total NIH funding from 2006 to 2016 is shown in $Millions. Total NIH funding has significantly increased by $212.7 million per year (p = 0.028). (B) Internal medicine funding. Internal medicine departmental funding from 2006 to 2016 is shown in $Millions. Internal medicine funding has increased by $4.93 million per year (p = 0.6942). (C) Surgery department funding. Surgery department funding from 2006 to 2016 is shown in $Millions. Surgery department funding has decreased by $2.79 million per year (p = 0.044). Linear regression analysis was performed for statistical significance.

Figure 2

(A) Number of funded RPGs and R01s. The total number of Research Project Grants (RPG) and R01s funded are shown from 2016 to 2016. Linear regression analysis is significant for both RPGs (p < 0.0001) and R01s (p < 0.0001). RPGs have declined at about 396 awards per year where as the number of R01s per year has declined at about 570 awards per year. (B) RPG/R01 Applications and success rates. The number of RPG Applications and R01 applications from 2006 to 2016 are shown in solid symbols and solid linear regression lines. The number of RPG applications has significantly increased by 933 applications per year (p = 0.001). The number of R01 applications has increased by 170 per year (p = 0.131). The RPG success rates and R01 success rates from 2006 to 2016 are shown in hollow symbols and dotted linear regression lines. The RPG success rate for funded applications has significantly declined by 0.35% per year (p = 0.020). The R01 success rate for funded applications has significantly declined by 0.43% per year (p = 0.019). Linear regression analysis was performed for statistical significance.

Figure 3

Grant application funding success rates. Research Project Grant application funding success rates are shown from 2007 to 2016. Internal medicine funding success rates have declined by 0.35% per year (p = 0.111). Surgery department funding success rates have declined by 0.21% per year (p = 0.425). All NIH RPG success rates have significantly declined by 0.41% per year (p = 0.020). Statistical analysis was performed using linear regression. Comparison of the 3 groups by one-way ANOVA (overall 10-year comparison) results in surgery success rates being significantly lower than internal medicine (p = 0.0001) and all NIH RPG success rates (0.011). Internal medicine success rates are not significantly different from All NIH RPG success rates (p = 0.223). IM, internal medicine; RPG, Research Project Grant.

Figure 4

(A) Surgery grant productivity. Surgery department R01 grant impact per $100,000 (n = 831) is shown in blue. The impact of surgery R01s has increased by 0.15 per $100,000 from 2006 to 2016 (p = 0.019). Surgery K-award to R01-equivalent award conversion has increased by 2.57% per year from 2007 to 2012 (p = 0.217). (B) MD only surgery. R01 grant impact. principal investigators with an MD were analyzed separately (n = 463). The grant impact per $100,000 of MD investigators has increased by 0.099 per year (p = 0.305).

Figure 5

(A) Number of grants awarded by division. The number of grants for which funding information was available is shown. Surgical oncology (n = 93), cardiac/thoracic surgery (n = 89), and transplant surgery (n = 74) had the highest number of grant awards. Acute care surgery (n = 7), surgery critical care, (n = 6) and breast surgery (n = 2) had the lowest number of grant awards. (B) Grant impact of all surgery divisions. The grant impact per $100,000 for each division is shown. Orthopaedic surgery (10.43) has significantly higher grant impact per $100,000 compared to general surgery (3.73; p = 0.005), urology (3.37; p = 0.011), vascular surgery (2.69; 0.001), and burn/shock trauma surgery (2.64; p = 0.438), using one-way ANOVA. Cardiac and thoracic surgery (6.55) has significantly greater grant impact per $100,000 compared to vascular surgery (2.69; p = 0.016). (C) Funding per grant by division. The funding per grant in $Million is shown for each division. Acute care surgery ($4.9 million) is significantly greater than cardiac and thoracic surgery ($2.3 million; p = 0.019), transplant surgery ($2.3 million; p = 0.015), otolaryngology ($2.2 million; p = 0.031), pediatric surgery ($2.0 million; p = 0.018), surgical oncology ($2.0 million; p = 0.004), urology ($2.0 million; p = 0.010), surgical research/surgical sciences ($2.0 million; p = 0.008), neurosurgery ($2.0 million; p = 0.005), endocrine surgery ($1.7 million; p = 0.042), and surgery critical care ($1.2 million; p = 0.016). Statistical comparison was performed using one-way ANOVA. *p < 0.05 and **p < 0.01 in comparison to acute care surgery.