With a flat budget and success rates for research grant applicants now just 18%, the National Institutes of Health is weighing options to revise its research portfolio and help increase the number of grants, a top agency official told a AAAS audience.
Sally J. Rockey, deputy director for extramural research at NIH, said that some changes in grants management already have been proposed as part of the president’s budget submission for the 2013 fiscal year.
These include an across-the-board funding cut of 1% for continuing grants; negotiating the budgets for new competitive grants to avoid growth in the average size of award; eliminating increases for inflation in multi-year grants; giving additional scrutiny to researchers who already receive in excess of $1.5 million a year; and continuing to fund early-stage investigators at the same rate as established investigators for new grant applications.
But significantly increasing the annual number of research project grants, or RPGs, would require additional steps, Rockey suggested.. That might particularly be required if the NIH budget were to suffer significant cuts during the congressional appropriations process, she said, or if Republicans and Democrats fail to come up with a long-term deficit reduction plan, automatically triggering deep discretionary spending cuts on 1 January 2013.
“I think it is very prudent for all organizations to think about how they are going to manage under these austere times,” Rockey said.
Rockey spoke at the 37th annual AAAS Forum on Science and Technology Policy, which convened in Washington, D.C., on 26-27 April. More than 400 elected officials, government and business leaders, researchers, foreign embassy staff, educators, and journalists attended.
Along with flat budgets, the biomedical research price index is continuing to rise 3% or more a year, she said, leaving grantees with a purchasing power equivalent to what it was in 2000. The average NIH competitive research grant now is $414,000 for 4.3 years, according to Rockey. About 50 institutions, mostly large research universities, get 70% of the funds, and about 20% of the investigators get about half of the total funding for research project grants of about $15.4 billion.
One option for the future is to do nothing, Rockey said, and let the grant success rate continue to decline (it was 40% or more in the 1980s) as competition for survival among researchers intensifies. She called that the Darwinian option. It also would likely mean a reduced emphasis on innovation as grant applicants play it safe in order to get through peer review.
But Rockey outlined other options meant to improve success rates and increase the number of awards. They include:
- Reduce the average size of each award. If each competing research project grant for the fiscal year 2010 had been reduced by $25,000, there would have been 9903 awards that year instead of 9287.
- Limit the number of awards per principal investigator (PI). If the number had been limited to no more than five competing awards per PI in 2010, there would have been an additional 15 awards. If the limit were three per PI, there would be an additional 264 awards. And if the limit had been no more than two per PI, there would have been 956 additional awards.
- Limit the amount of money per principal investigator. Limiting the total grant support per PI to $1 million would have yielded about 2000 additional competitive grants at an average cost of $400,000. Limiting the total to $800,000 would have yielded 2400 more new grants. Limiting it to $400,000 would have meant 4400 additional grants.
- Limit the salaries of principal investigators. Between 30% and 50% of the NIH grant money goes to salaries for research investigators and support staff, Rockey said. That percentage is higher than it was 25 years ago. Congress mandated last year that the top salary under a research grant should be no more than $179,000, down from $199,000. If salaries were limited even further, Rockey said, it could have “a pretty dramatic effect” on existing grants but also free up additional money.
Each of the options has drawbacks, however.
While reducing the average size of new awards by $25,000 would have yielded $232 million in savings in 2010, Rockey said, the redistribution of that money into additional grants would have had only a marginal impact on the overall applicant success rate: up from 20.5% to 21.7%.
Sharply limiting the amount of money per principal investigator “could be devastating” to researchers who are doing ambitious, multi-year projects, Rockey said. It also could have a substantial impact on commitments to new investigators just starting out.
Limiting the number of awards per principal investigator could call into question whether the process supports a true meritocracy, she said.
And limiting the salaries of principal investigators is a problem, she said, given that academic institutions have come to rely on resources from NIH grants to help pay salary costs for faculty and support staff. Rockey said a major research university reported that it lost $10 million after the congressionally mandated cut in grant-funded salaries and had to replace the money from other sources.
Institutional dependence on NIH funding, in whole or in part, for staff salaries “may not be sustainable over the long term,” Rockey said. But she argued that any changes to the system should be done in a thoughtful way, through discussions by NIH with its academic partners, rather than through quick, mandatory changes.
NIH has been soliciting advice about the future of its extramural research grants at the NIH Extramural Nexus. That site also is home of Rockey’s popular “Rock Talk” blog, where she discusses issues related to the research grant program and posts relevant information. Rockey said her office has received comments from 348 groups and thousands of individuals.
“Many people thought we should keep the current system,” Rockey said. “Just keep the Darwinian approach. Don’t try to go in there and socially engineer anything.” Others weighed in on the merits of the various options for change, including some approaches beyond those discussed by Rockey. These could include limiting payment for indirect costs associated with grants, limiting large project grants, and providing more support to small labs and individuals by limiting grants to large labs.
Institutions also could help manage the demand for grant money by reducing the number of applications submitted by their faculty, Rockey said. And NIH can examine its research priorities, seeking to reduce support for less innovative ideas and eliminating some of the duplication of effort. It also could encourage its peer reviewers to think about innovation as one of the key criteria for a successful grant application, she said.
Rockey expects a continuing and lively discussion of options. “We’re going forward,” she said. “We’ll see out this plays out in the future.”
See the full program for the 2012 AAAS Forum on Science and Technology Policy.
See more AAAS.org coverage of this year’s Forum.