Post by Former NIMH Director Thomas Insel: Funding Science
Several papers in the past few weeks have commented on the dismal state of funding for U.S. biomedical research1,2,3,4 After a doubling of the NIH budget between 1998 and 2003, NIH appropriations have stalled with little growth over the past decade. Indeed, corrected for inflation, NIMH has “undoubled” with a budget that is on par with 1999. As a result, less than one in five grants can be funded and many promising projects cannot be supported. As Alberts et al5 recently noted there is both a sense of hyper-competitiveness and despair in the academic community, with many young scientists deciding that a career in science is no longer practical, at least not in this country.
In an excellent review entitled the “Anatomy of Medical Research,” Moses and colleagues try to put all of this into a broader perspective citing three major trends. First, they note that for overall public and private investment, U.S. science investment increased 6 percent per year from 1994 to 2004 but only 0.8 percent per year from 2004 to 2012 with total investment ($117B) representing 4.5 percent of total health care expenditures. Remember that number of 4.5 percent. Second, they note that the investment in health services research is a small fraction of the total: $5B or less than 1/20th of the U.S. science investment. This figure is 0.3 percent of total health care expenditures. Finally, they describe how the U.S. is losing out to global competition, with U.S. funding declining from 57 percent to 44 percent of the global investment in science. While the U.S. was undoubling its investment in research from 2004 to 2012, China tripled its funding of science and now supports a larger scientific workforce. Expanding the global search for cures may seem like good news for patients and families, but history demonstrates that residents of the home country are the most likely to benefit from the health, economic, and social returns of new discoveries.
All of these recent papers decrying the state of U.S. research funding note the lack of growth. Some note the absence of any multi-year plan for supporting science in this country, in contrast to the multi-year strategies developed in Europe and Asia. Not only is U.S. science funded one year at a time, in most of the last 10 years the budget has not been approved until several months into the year, imperiling even one year planning. But none of these excellent and worrisome papers address two fundamental questions: (a) How much should we spend on research? And (b) how much of this investment should be public vs private?
The question of how much to invest on research has no single answer. Should investment be tagged to gross domestic product (GDP), to a percentage of previous funding, or to costs? Expressed as a percent of health care costs, research investment can be viewed as a potential solution to bring down expenses. Is 4.5 percent the right investment? Is this too much or too little? Is 0.3 percent the right investment for research on health services? In other industries, research and development investment shows a wide range; for instance, Moses et al cite research investment at over 10 percent of revenue for the information technology sector and about 2.5 percent of revenue for the auto industry. Health care is not so much a revenue source as a cost center, so research and development costs cannot be compared in quite the same way.
A better way to frame the question is by focusing on the cost of the problem to be solved. For NIMH, the problems can be defined as mortality and morbidity. The 40,000 suicides per year in this country are mostly related to mental illness. We know the costs and benefits of reducing homicides and reducing traffic fatalities, both of which have dropped at least 40 percent and now account for fewer deaths than suicide. What should we spend to reduce suicide by 40 percent? By 80 percent? This year about 100,000 adolescents will have a first psychotic episode, usually beginning on a long journey that is the reason that psychosis is among the top medical causes of years lost to disability. What should we spend to reduce this first episode and prevent chronic disability? In thinking about budgets for research, would it make more sense to begin with defining the problem to be solved rather than arguing for a 3 percent or 5 percent change in current funding or linking research funding to an economic indicator like GDP?
Why should the public support research? After all there is comparatively little public research funding for making computers or cars. What is different about health? The ecosystem for biomedical research has three major segments: the pharmaceutical industry ($36.8B), the expanding medical device and biotechnology industries (now at $31.1B), and government funding through the NIH ($30.9B).4 There are also other federal agencies ($7.1B), state and local governments ($6.3B), and philanthropic funds ($4.2B), but the big three are the major pillars of support for the $116B of U.S. investment in medical research, with industry leveraging NIH investment by two to one.4 The model has been that fundamental research supported with public money is further developed into new diagnostics and therapies by private industry, which spends more than double the NIH budget on research and development.
For NIMH this traditional model does not work. Most of our fundamental science does not have a simple hand off to industry. In contrast to most other institutes at NIH, NIMH is developing cognitive diagnostics and psychosocial interventions in addition to classical biomarkers and pharmacological treatments. But even within the pharmacologic pipeline, the two-to-one leverage of private to public investment enjoyed in other areas of medicine is no longer relevant for NIMH. Over the past five years, among the eleven major pharmaceutical companies producing drugs for brain disorders, portfolios have dropped more than 50 percent, from 267 projects in 2009 to 129 projects in 2014, with many of the remaining projects focused on Alzheimer’s disease, which is outside the NIMH mission.6 More important, our success is measured by changes in morbidity and mortality, not by private commercial success. Indeed, much of our research on interventions focuses on ways to make treatment less expensive and more accessible not more expensive and patent-protected.
Why should the public support research and development in mental illness? Other than some philanthropic contributions, no one else does. In contrast to heart disease or cancer, the public focus in mental health has been on services and policy, primarily on improving access to care. Unfortunately, today’s care is not good enough for many people with severe mental illness. More people than ever are receiving treatment for mental illness but the morbidity and mortality numbers are not dropping. Even with access and reimbursement, too many people do not recover and too many die from these disabling disorders. Investing in research is essential if we are to have better services to improve outcomes.
These recent papers on biomedical research funding have started an important national conversation. There is no question that the U.S. through NIH has led the world in the quality and quantity of support for research and discovery relevant to health. But our leadership is clearly in jeopardy; now is the time for a tough discussion about the future.
1 Collins FS. Exceptional opportunities in medical science: a view from the National Institutes of Health. JAMA. 2015 Jan 13;313(2):131-2. doi: 10.1001/jama.2014.16736.
2 Daniels RJ. A generation at risk: Young investigators and the future of the biomedical workforce. Proc Natl Acad Sci U S A. 2015 Jan 13;112(2):313-8. Doi: 10.1073/pnas.1418761112. Epub 2015 Jan 5.
3 Dzau VJ, Fineberg HV. Restore the US lead in biomedical research. JAMA. 2015 Jan 13;313(2):143-4. doi: 10.1001/jama.2014.17660.
4 Moses H 3rd et al. The anatomy of medical research: US and international comparisons. JAMA. 2015 Jan 13;313(2):174-89. doi: 10.1001/jama.2014.15939.
5 Alberts B et al. Rescuing US biomedical research from its systemic flaws. Prc Natl Acad Sci U S A. 2014 Apr 22;111(16):5773-7. doi: 10.1073/pnas.1404402111. Epub 2014 Apr 14.
6 Pankevich DE et al. Improving and accelerating drug development for nervous system disorders. Neuron. 2014 Nov 5;84(3):546-53. doi: 10.1016/j.neuron.2014.10.007. Epub 2014 Nov 5.