A New Paradigm for NIH Grants

Giving out smaller grants, but for larger periods of time, will fix a system in distress.

By | August 1, 2007

Credit: Mike Bentley
Credit: © Mike Bentley


Making Grants Go Further:
The Perils of Industrialization

The NIH grant funding system is in distress. The success rate for nonamended applications was less than 10% in 2005, 1 with the system overwhelmed by a large number of grant applications that are impossible to distinguish in terms of their significance and merit. Multiple resubmissions are common.

Although the NIH has introduced a new structure for review committees and some pilot changes, success rates have not improved. This crisis cannot be solved by tinkering with peer review, whose many problems I delineated in these pages eight years ago. 2 At the time, the NIH stated that it was seriously addressing these issues by emphasizing innovation, expediting the funding of recently reviewed grants, and improving study sections in its "Boundaries" process. Such minor changes have not improved the funding of biomedical research, and I therefore challenge the NIH to completely shift its paradigms, in the following ways, which are a revision of my original proposal.

Scientists opting not to play the grantsmanship game, and who have a track record of solid publications, evaluated by international panels of established scientists as well as their younger colleagues, will be awarded $300,000 per year for 10-year periods. These researchers will save precious time by not writing grant applications, avoid the anxiety of uncertainty, focus on their research, take risks, and be able to be creative. NIH will save millions of dollars and time on the review process, since investigators will not have to submit and resubmit applications every few years. The reviewers will save precious time to spend on their own research, and I would also suggest that the panels not be forced to travel to the NIH to participate in reviewing, to save the world tons of carbon footprint.

Funding 10,000 such grants will cost $3 billion in the first year. As the number of awardees is increased to 40,000 over a four-year period, the sum will increase to $12 billion per year. Assuming that 2,000 new long-term awardees are added each year, which is more than double the number of new R01s awarded in 2005, 1 the direct costs will increase by $0.6 billion per year.

The scientists on this program will not be eligible for traditional grants from NIH, but can participate in shared instrument grants and seek funding from nonfederal agencies, such as the American Heart Association and the American Cancer Society. The constant NIH funding will give them the leverage and productivity to obtain private funding.

The indirect cost rate will be reduced to 30% for all institutions, eliminating disparities between universities and saving large sums of money that will then go directly to laboratory research. The big research universities have already reaped the benefits of high-percentage indirect costs, and they can certainly adjust to the lower rate. With this rate, the total cost for the 40,000 grants in place four years following the initiation of this program will be $15.6 billion per year, slightly more than half the current NIH budget.

Young scientists with a proven postdoctoral track record, evaluated again by international panels, and starting out in a university or research institute position, will receive $50,000 per year for five years to establish their own research programs. This steady source of funding will facilitate productivity and help them obtain supplemental funding from nonfederal sources. The cost of 10,000 such grants is $650 million per year including indirect costs; this is about 2.3% of the current NIH budget. Even if 10,000 new grantees were added each year, the cost during the fourth year would be only $2.6 billion, less than 10% of the NIH budget.

The total costs in that fourth year would then be $18.2 billion, which is a bit more than 60% of the current NIH budget. That will leave about $3 billion for the NIH itself, and billions more for large program grants and R01s.

The productivity of scientists on the new funding system will be compared after 10 years with those who struggle through the traditional process. Regardless of the outcome of the comparison, the new system will be a more humane and reasonable way of funding biomedical research.

Nejat Düzgünes is chair of the Department of Microbiology at the University of the Pacific's Arthur A. Dugoni School of Dentistry.


1. H.G. Mandel, E.S. Vesell, "Declines in funding of NIH R01 research grants," Science, 313:1383, 2006. 2. N. Düzgünes, "Science by consensus: Why the NIH grant review system must be changed," The Scientist, 13:13, April 12, 1999.



Avatar of: Sui Huang

Sui Huang

Posts: 3

August 5, 2007

Düzgünes makes a very interesting proposal for a long-due paradigm shift in the NIH review system. It will not only make the funding decision process more humane and efficient, but as he briefly alluded to, also address a central problem in the evolutionary dynamics of the reviewing process. Düzgünes states that the system is ?overwhelmed by a large number of grant applications that are impossible to distinguish in terms of their significance and merit?. This epitomizes an interesting universal phenomenon observed in many complex dynamical systems doomed to enter a terminal crisis: The system is locked-in a self-sustained dysfunctional state that resists correction from within. \n\nThis fateful evolutionary dynamics works like this: The low funding rate stimulates multiple and repeated submissions of (admittedly, similar and mildly modified) proposals which swamp the reviewing system and impedes its proper operation so that the evaluation scores given are near stochastic. Who has not encountered a lowering of the score for an objectively improved proposal (e.g., after inclusion of significantly more preliminary data)? As researchers become aware of the stochasticity of the decision process, they adapt their strategy by mass producing applications and bombarding agencies with submissions, for in a random process the number becomes the key determinant of the probability of success. This further overwhelms the system, increases the stochasticity of scoring, which in turn stimulates more quicky, opportunistic applications. \n\nThis self-perpetuating vicious cycle not only costs time and energy and nerves of applicants and reviewer that could be spent for more creative tasks. The time pressure also lowers the overall quality of both, submissions and comments. But a far reaching and insidious consequence is that in a stochastic system, naturally, the average reigns. And average, necessarily, means mediocrity. In other words, the entire review process has over the years evolved a state that actively and robustly promotes mediocrity, and it is locked-in this state. \n\nWhile the review system was primarily designed to weed out cranks from the rest, it now not only fails in doing so, but certainly, also cannot identify the Mozart?s among the Salieri?s and enrich for the former. The king of mediocrity, as Salieri has been portrayed in the popular culture, reigns. Accordingly, at the NIH, average is king. There is little chance for outliers - on both side of the distribution. Innovation is not favored at a probability higher than chance. Perhaps Düzgünes? proposal may at least end the reign of mediocrity by breaking the self-re-enforcing futile cycle driven by scores that are not significantly different from pure chance.
Avatar of: Dr.Shanthi Raam

Dr.Shanthi Raam

Posts: 43

August 14, 2007

Ten year grant with 300,000 per year is a very good idea to give the recipient scientist stability, financial security and best of all the emotional security that comes along with it. However, restricting scientists who receive these grants barring them from applying for any other federal grants will stifle the scientist. For example, NCI periodically announces "requests for proposals" or RFP focussing on a theme which is of high importance to unravel a certain elusive phenomenon and to answer pivotal questions in certain areas. A scientist, recepient of one of these long term grants may be working in the field in which the RFP is announced and therefore might be able to provide solution to the problems that are being focused in the RFP. By not allowing the scientist to respond and participate will stifle science and the scientist. I am talking from personal ecperience because I found that RFPs are the best way to bring together scientists working in an area to find solutions to a problem by bringing in their own individual perspective. \nNCI already has a way to fund scientists long term (10 years) through SPORE (specialized projects of research excellence) programs but these are given only to established tenured scientists. However, the senior scientist is given salary support to hire other scientists for long periods. \nAnother point that is suggested is reduction in ever escalating indirect costs. By providing the scientist stable 10 year grant, the cost for basic salary and the laboratory supplies are taken care of by the grant. Will the institutions therefore agree to a reduction in indirect costs ? Let us see how many university Provosts participate in this discussion.\nThe most important issue is to come up with a plan that will provide steady support for the scientists. The suggested plan will provide that. It needs to be modified to allow the scientist freedom to grow and their research to expand.\n\nIt is going to be interesting to see how this discussion shapes up. I will be reading the responses every day.

September 4, 2007

What's wrong with competition? Too many have grown accustomed to the NIH budget as a form of scientific welfare. Not all science deserves to be funded and not all scientists deserve to be funded. Too many universities have grown filthy rich from indirect costs from NIH budgets. Its times like these that shake off the dead wood and force investigators and universities to focus and prioritize and manage their funds more efficiently. Although academic curiosities are interesting and potentially rewarding, in this day and age they need to be paid by someone else besides the taxpayers.

Popular Now

  1. Unstructured Proteins Help Tardigrades Survive Desiccation
  2. What Budget Cuts Might Mean for US Science
    News Analysis What Budget Cuts Might Mean for US Science

    A look at the historical effects of downsized research funding suggests that the Trump administration’s proposed budget could hit early-career scientists the hardest.  

  3. Opinion: On “The Impact Factor Fallacy”
  4. Inflammation Drives Gut Bacteria Evolution
Business Birmingham