The social value of gains in health and longevity are enormous, and a variety of evidence suggests that biomedical research has played a key role in generating these gains. The success of the rapid rollout of COVID vaccines is only the most recent example.
With an annual budget on the order of $47 billion, U.S. National Institutes of Health (NIH) funding plays a fundamental role in nearly every aspect of the biomedical research landscape. Basic research funding through NIH R01 awards are the bread-and-butter of funding support for most U.S.-based academic biomedical researchers, and around 50% of patents linked to drugs approved by the US Food and Drug Administration directly cite NIH-funded research.
Yet, there is broad agreement that the NIH faces a number of challenges as it continues to support biomedical research. The average age of an NIH principal investigator rose from 39 years old in 1980 to 51 years old in 2008. Principal investigators report spending 44% of their time doing grant-related paperwork and maintenance as opposed to active research. Despite large increases in overall scientific funding, the share of “disruptive” papers arguably continues to decline.
The path forward is unclear. There is genuine uncertainty over whether the NIH should support “people” (as in the NIGMS MIRA program) or “projects” (as in traditional NIH R01 funding mechanisms), and more broadly about whether traditional NIH funding mechanisms are flexible enough to fund the highest social impact research opportunities.
To begin to address this gap, this collection of short papers assembles scholars from diverse backgrounds to offer their insights on possible paths forward. Bhaven Sampat’s foundational piece delves into the history of debates surrounding the NIH’s initial design and subsequent reforms. These long-standing debates, such as the Bush-Kilgore debate over the focus of public research funding, still reverberate today.
Several scholars – including Pierre Azoulay, Chiara Franzoni and Paula Stephan (co-authored), and Lisa Larrimore Ouellette – contribute their thoughts on how research and experimentation could enhance the NIH’s ability to support high social-impact work and ensure funded discoveries reach beyond academia to directly benefit patients. Kyle Myers explores the potential for institutionalizing randomized experiments within the NIH to generate more robust evidence on key open questions.
Addressing the question of equitable distribution of NIH-funded advancements, Maya Durvasula offers insights on how to ensure diverse populations benefit from these discoveries. Finally, Robert Cook-Deegan and Rush Holt provide broader perspectives on what factors enable biomedical research to effectively improve health.
The NIH is arguably the most important institution shaping the rate and direction of scientific research, in an area – health – that has enormous implications for human well-being. Even very small improvements in the productivity of how NIH funding is spent could translate into massive improvements in social welfare. These short proposals are meant to open a conversation about practical, tractable reforms that could be considered in the coming years.
– Heidi Williams, Stanford University and Institute for Progress
Disclosure: Heidi Williams is a part-time, unpaid employee of the NIH through the Intergovernmental Personnel Act (IPA) Mobility Program and receives support from NIH grant P01AG005842 to the National Bureau of Economic Research (NBER). The author did not receive financial support from any firm or person for this article or, other than the aforementioned, from any firm or person with a financial or political interest in this article. Other than the aforementioned, the author is not currently an officer, director, or board member of any organization with a financial or political interest in this article.
Editor’s Note: Building a Better NIH is a project partnership between the USC-Brookings Schaeffer Initiative for Health Policy, the Hutchins Center on Fiscal and Monetary Policy, the Institute for Progress, and the Good Science Project. David Wessel and Richard Frank served as co-editors for the related publications.
Bhaven Sampat of Columbia University discusses how NIH historically has attracted bipartisan political support unrivaled in health or science policy. Despite broad, high-level support, there are important disagreements and uncertainties about NIH research policy. A specific focus of current reform efforts is on the agency’s peer review system, the machinery it uses to allocate the bulk of its funding. This paper provides historical context for these reform efforts by chronicling the origins, and evolution, and political economy of the NIH peer review process, and the complexities of NIH reform.
Pierre Azoulay of MIT reports on results on scientific risk-taking in the context of competitive renewal decisions for R01 grants at NIH and documents a pervasive and obdurate risk-taking penalty using 4 alternative measures of risk-taking. Second, he proposes a three-pronged research agenda for NIH as it considers how to encourage risk-taking. First, administrative data (which contains information about applications, not just funded grants) could be used to study whether the “risk-taking penalty” is present at the stage of a grant’s initial cycle. Second, the NIH could help validate metrics of scientific risk-taking. Third, NIH could perform randomized controlled interventions that would produce a solid empirical foundation for efforts geared at alleviating the risk-taking penalty.
Paula Stephan of Georgia State University and Chiara Franzoni of Polytechnic University of Milan describe how high-risk high-reward research is of critical importance for the long-term progress of science, but may be disadvantaged in peer review of grant proposals. The creation of special grants designed for high-risk high-reward applications is one possible approach to the problem, but this approach alone is unlikely to solve the problem. There is also a need to improve the assessment of risk-gain in evaluations of applications to all funding programs. The current protocol for assessing applications to R01 grants is likely vulnerable to cognitive biases. Specifically, it may easily lead to underestimating the potential expected value of research that has a small probability of leading to extraordinary results (high-risk high-reward), because the assessment of the potential value (e.g. the criteria “significance”) is conflated with the assessment of the probability of success. The authors propose an alternative protocol to test for evaluating the expected value of a research proposal designed to reduce this bias.
Lisa Larrimore Ouellette of Stanford Law School discusses reforms to policies to support commercialization of NIH-funded research. She explains the political and empirical challenges to reforms related to patenting and licensing policies and that the NIH may have greater social impact through reforms to its direct funding of commercialization. In particular, she proposes that the NIH design rigorous policy experiments to assess the impact of these programs, and that commercialization funding be more explicitly focused on addressing market failures.
Kyle Myers of the Harvard Business School argues that NIH could further its impact by partnering with social scientists and declaring an explicit objective to perform experiments and discover the most effective ways to evaluate and fund science.
Maya Durvasula of Stanford University discusses the relationship between clinical trial infrastructure and representative medical research. She argues that NIH is uniquely equipped to make targeted infrastructure investments that will facilitate the production of representative evidence across the clinical trial ecosystem.
Robert Cook-Deegan of Arizona State University considers improvements to the U.S. system of health research at large. He notes that NIH is a well-functioning fount of biomedical research, especially molecular, cellular, and clinical research. Its devotion to merit review based on peer review breeds conservatism that many hope will be addressed by the new Advanced Research Projects Agency for Health (ARPA-H). He argues that the largest gap in U.S. health research is relatively sparse support for health services research and social sciences; yet there is no clear path to support such politically perilous research, which nonetheless could have a major impact on health outcomes.
Former congressman Rush Holt (D-NJ) discusses the goals and challenges faced by the NIH in balancing the pursuit of fundamental scientific knowledge with the practical application of that knowledge to improve public health. The paper argues that there has often been a disconnect between the scientific community and the public, resulting in a lack of trust and understanding of scientific research. The author calls for a shift in the research paradigm, urging NIH to take a more active role in bridging the gap between science and the public.