Point of View: The NIH must reduce disparities in funding to maximize its return on investments from taxpayers
- University of Arkansas for Medical Sciences, United States
Figures
Figure 1
Heavily skewed distributions of NIH grant funding favor a minority and disfavor the majority.
A search of the NIH RePORTER database identified 25,674 investigators who received research project grant funding in FY2015. These individuals were ranked in descending order by the amount of funding they received, and then grouped into 52 bins, each of which contained 493 investigators (the remaining, lowest-funded 38 investigators were not binned). The same process was applied for amounts of funding to 2,038 organizations (39 per bin) and to 52 states, including Washington DC and Puerto Rico (1 per bin). Pareto plots display amounts of funding (histograms, left Y axis) to each bin. For example, the first bin of investigators got more than twice as many dollars as the second bin. Cumulative curves (right Y axis) display fraction of total funding to a given bin and all higher-funded bins (i.e., those to its left). Inset text (italics) in the top panel show the mean amount of funding (in $ millions, M) per investigator for select bins. The amount of funding per investigator that yields maximum productivity (the 'sweet spot' from Figure 2) is almost exactly the median amount of funding per investigator. The proposed lower and upper limits for support per awardee ($0.2M and $0.8M) would free up enough money to support about 10,500 additional investigators (21 additional bins) with mean funding at the productivity sweet spot.
Figure 2
Productivity peaks at about $400,000 per investigator and declines with lower and higher amounts of funding.
Each plot shows the marginal return (Y axis) as a function of annual NIH research project grant funding (total costs) per investigator (X axis); note that both axes are logarithmic, and that the range of the Y-axis varies from plot to plot. The marginal return for each amount of funding corresponds to the first derivative of the Cobb-Douglas production function, using relative citation ratios (RCRs) as the measure of production (Mongeon et al., 2016; Lauer et al., 2017). The RCR is a measure of article influence, developed by the NIH, that normalizes the number of citations received by a publication for the field of study and the time of publication (Hutchins et al., 2016). The three plots show the marginal return based on the maximum RCR (A), median RCR (B) and annual weighted RCR (C). The vertical dashed lines correspond to funding values of $250,000, $1 million and $2 million per investigator. Reproduced with permission and minor modifications (increased font size and line weights, repositioned panels and labels) from (Lauer et al., 2017) under a CC-BY 4.0 international license.
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Point of View: The NIH must reduce disparities in funding to maximize its return on investments from taxpayers
eLife 7:e34965.
https://doi.org/10.7554/eLife.34965
