Inequity of NIH cancer funding in the United States: an ecological study predicting funding based on disease burden from 2008 through 2023

Abstract

Background: Disease burden has been used to predict National Institutes of Health (NIH) funding but included diseases with little underlying relationship. Here we focus on cancers to create a more appropriate model to allow for more targeted scrutinization of funding allocation.

Methods: An ecological study using NIH funding data (2008-2023) was performed. Inclusion of cancers was based on their presence in the NIH Research Portfolio Online Reporting Tool and the 2021 Global Burden of Disease (GBD) study. Disability-adjusted life years (DALY) were collected and to evaluate the impact of public interest, Google Trends data was used. Multivariable linear regression determined appropriate funding based on disease burden and public interest. To quantify how each cancer's funding differed from model predictions residual values were used to calculate the percent over/under funding.

Findings: Fifteen cancers met inclusion criteria. Neuroblastoma had the greatest ratio of funding to DALYs per 100,000 people (US$14,000,000) while lung cancer had the lowest (US$300,000). Stomach cancer was the most underfunded (197.9% [95% CI: 136.0%, 276.2%]) while brain cancer was the most overfunded (64.1% [95% CI: 53.8%, 72.1%]). Even at their lowest funding values in the study period brain, breast, and colorectal cancer all had greater than 40% overfunding. Contrarily, the lowest annual funding for leukemia, uterine, and stomach cancer received less than 150% of expected funding. Despite its overfunding brain cancer had an increase in DALYs in the study period.

Interpretation: Modeling by disease category demonstrated disparities in funding indicating the need for reevaluation for possible funding inequities. The year-by-year approach taken in this study will drive the ability for future research to better understand NIH funding decisions. Additionally, the role of public interest in research funding needs to be further evaluated to ensure that popularity does not override disease burden, in funding decisions.

Funding: No Funding.

Keywords: Burdens of disease; Cancer; DALY; Equity; Global health; Grant funding; Health disparity; Health economics; Health policy; Medical ethics; NIH; National institute of health; Research funding.

Fig. 1

Percent error of residuals for years with maximum funding, average funding, and minimum funding. Positive values indicate overfunding compared to predicted while negative values indicate underfunding. The three values presented are the maximum funding year (Blue), average funding (Black), and minimum funding year (Red). Values represent residual (actual funding–predicted funding) divided by actual funding. Confidence Intervals (CI) were made by determining the uncertainty of predicted values based on the standard error of estimate of the predictions. Following exponentiation these CIs were subtracted from the actual funding. These values were then divided by actual funding to represent CIs in the figure. For overfunding the theoretical maximum was 100% overfunded. Therefore percent funding discrepancies should be interpreted in the context of actual funding values. Cancers are presented in descending order based on average NIH funding dollars.

Fig. 2

Annual funding discrepancies from predicted values. Values represent actual funding–predicted funding divided by actual funding. Positive values indicate overfunding (blue values) compared to predicted while negative values indicate underfunding (red values). Confidence Intervals (CI) were made by determining the uncertainty of predicted values based on the standard error of estimate of the predictions. Following exponentiation these CIs were subtracted from the actual funding. These values were then divided by actual funding to represent CIs in the figure. For overfunding the theoretical maximum was 100% overfunded. Therefore percent funding discrepancies should be interpreted in the context of actual funding values. Cancers are presented in descending order based on average NIH funding dollars.

Fig. 3

National institutes of health funding shown by A. Number of Grants per Year and B. Proportion of total grants and total funding from the National Cancer Institute (NCI). Values represent funding data ranging from 2008 through 2023 although not all cancers have full data. Cancers are presented in descending order based on average NIH funding dollars in descending order. NCI = National Cancer Institute.