Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4 Supporting Research Through NTH and ADAMHA As previously discussed, nearly half of all financial support for health sciences research comes from federal sources (see Chapter 2~. Of this, about 78 percent is disbursed through the National Institutes of Health (NIH) and the Alcohol, Drug Abuse, and Mental Health Administration ~- The NIH has become a world-renowned and highly respected biomedi- cal research organization with a mission to uncover new knowledge that will lead to better health for everyone.) Currently, NIH consists of 13 categori- cal institutes, 2 support divisions, 4 specialized centers, the Clinical Center, the Fogarty International Center, and the National Library of Medicine, all located primarily on a 300-acre tract in Bethesda, Maryland (Figure 4-1~. Unlike some foreign governmental support for medical research, only a small proportion of federally sponsored biomedical research actually is conducted in U.S. federal laboratories. Nearly 80 percent of the NIH budget is allocated to research and training at universities and other research institutions, both in the United States and abroad.2 Most of these funds are allocated through peer review processes that include the views of scientists and others throughout the country. Therefore, the NIH is a decentralized organization with scientific priorities determined by individual investigators, Congress, and other interested parties. The ADAMHA is responsible for advancing scientific knowledge to improve the understanding, prevention, and treatment of alcohol abuse and alcoholism, drug abuse, and mental health disorders.3 ADAMHA is 79
80 National Heart, Lung, and Blood Institute National Institute on Deaf ness and Communicative Disorders National Institute of Allergy and Infectious Diseases National Institute of Dental Research National Library of Medicine National Center for H uman Genome Research Division of Researc h G ran ts FUNDING HEALTH SCIENCES RESEARCH NATIONAL INSTITUTES OF HEALTH National Cancer Institute National Institute of Neurological Disorders and Stroke National Institute of General Medical Sciences National Eye I net i tu te National Institute of Diabetes and Digestive and K idney Diseases National Center for Nursing Research Division of Computer Research and Technology National Institute of Aging National Institute of Arthritis and M u scu loskeletal and S k i n D i ~ eas es National Institute of Environmental H eel t h S ci e noes National Institute of Child Health and Human Development Clinical Center National Center for Research Resources Fogarty International Center FIGURE 4-1 The institutes, centers, and divisions of the National Institutes of Health. composed of three separate institutes: the National Institute on Alcohol Abuse and Alcoholism, the National Institute on Drug Abuse, and the National Institute of Mental Health (Figure 4-2~. In addition to conducting and supporting biomedical and behavioral research and research training, ADAMHA is responsible for demonstrations, clinical training, treatment, prevention, and public information activities on public health problems
SUPPORTING RESEARCH THROUGH NIH AND ADAMHA ALCOHOL, DRUG ABUSE AND MENTAL HEALTH ADMINISTRATION 1 1 1 11 1 1 NATIONAL INSTITUTE NATIONAL INSTITUTE ON ALCOHOL ABUSE ON AND ALCOHOLISM DRUG ABUSE 81 NATIONAL INSTITUTE OF MENTAL HEALTH FIGURE 4-2 The research institutes of the Alcohol, Drug Abuse, and Mental Health Administration. related to its mission. The peer review process and programs for research and training support are in most ways identical to NIH. Federal funding programs for supporting research and research training outside of federal laboratories are primarily grants, cooperative agreements (financial assistance awards), and contracts (acquisition awards). At NIH and ADAMHA these are referred to collectively as "extramural" programs and fall into five major categories: research grants, research and develop- ment (R&D) contracts, research training awards, cooperative agreements, and construction authority (NIH only). R&D grants have been and con- tinue to be the cornerstone of NIH and ADAMHA extramural support for health research since the expansion of the NIH extramural programs began in the mid 1940s. ALLOCATIONS FOR NIH AND ADAMHA The policy change following World War II to advance basic knowledge by supporting civilian R&D in academic institutions stimulated steady increases in the NIH budget (Figure 2-6~. This growth has resulted because of the emphasis society has chosen to give to health research and because of the subsequent legislation that created numerous new institutes and expanded the extramural programs of NIH. The most rapid budget growth occurred between 1955 and 1965 a period of expansion. From the late 1960s to 1980, budget growth leveled off and may be referred to as steady state. During the 1980s, congressional appropriations to NIH increased an average of 10 percent per year, resulting in a 2 percent per annum real growth in the NIH budget (Figure 4-3. However, much of the increases of the past few years can be attributed to the growth in funding for AIDS research.
82 10 9 8 5 4 3 2 1 o FUNDING HEALTH SCIENCES RESEARCH DOLLARS (Billions) 77 78 79 80 81 82 Constant 1988 $ 7~ 83 84 85 86 87 88 89 90 91 YEAR ~+- Constant $ less AIDS ~~~ Current $ FIGURE 4-3 NIH appropriations with and without allocations for AIDS research from 1977 to 1991. (Appendix Able A-7) Note: Constant dollar calculations are made using the BRDPI deflator. Allocations among NIH extramural and intramural programs and pro- gram management have not changed significantly since the late 1970s (Figure 4-4~. Extramural programs account for nearly 80 percent of the NIH budget. The intramural program has remained at 10 to 12 percent of the budget over the same period. Program management, which includes the Office of the Director, Division of Research Grants, and the National Library of Medicine, has been receiving successively smaller percentages of the NIH budget, falling to 7 percent in 1989 from nearly 10 percent in 1977. Although appropriations for ADAMHA grew and paralleled those of NIH throughout the 1970s, the agency budget was cut in the early 1980s. Cuts in social sciences research and nonresearch community programs, recommended by the Reagan administration and enacted by Congress in the early 1980s, drastically reduced total ADAMHA appropriations.4 Only recently have appropriations for ADAMHA surpassed the 1979 level. The President's budget proposal for fiscal year 1991 requests more than $2.8 billion for ADAMHA (Figure 4-5~.5 Community programs have been and continue to be the largest portion of the ADAMHA budget, ranging from 53 to 61 percent of the ADAMHA budget over the past 10 years (Figure 4-6~. Obligations for research grew from 17 percent to 33 percent of the budget in the same period. When
SUPPORTING RESEARCH THROUGH NIH AND ADAMHA 83 measured in constant 1988 dollars, there has been an average annual growth of 3.5 percent in this part of the budget since 1977 (Figures 4-5 and 4- 6~. However, there have been wide annual variances in research support, ranging from a 12 percent cut in 1981 to an 18 percent increase in 1987.5 Recently, the rapid growth in ADAMHA's research budget reflects the government's priority for combating drug abuse through basic research. Since these two agencies are the primary federal sponsors of health sciences research, this chapter examines their research support programs. Although the support programs for research projects cannot be separated easily from training and facilities (Chapters 5 and 6), the committee tried to isolate them for the purposes of this review. 1b this end, this chapter explores the policies that have affected the levels of support as well as the number and types of research project support programs available from NIH and ADAMHA The chapter also examines the characteristics and trends of the scientists performing research sponsored by NIH and ADAMHN NIH DIRECT OPERATIONS Appropriations for research support divisions, extramural grant man- agement, and for the National Library of Medicine (NLM) have not kept pace with inflation. In constant 1988 dollars, funding for program support 1 00% 75~ 50% 25% 77 78 79 80 81 82 83 84 YEAR Extramural ~ Intramural 85 86 87 88 89 _ Operations FIGURE 44 Allocation of NIH budget from 19M to 1989. (Appendix liable A-8)
84 1 000 800 600 400 200 FUNDING HEALTH SCIENCES RESEARCH DOLLARS (Millions) - , ,+ O 1 1 1 1 1 77 78 79 80 81 82 : t 1 1 1 1 1 1 1 1 83 84 85 86 87 88 89 90 91 YEAR ' 8'' constant 1988 $ -+- current $ FIGURE 4-5 ADAMHA research allocations from 1977 to 1990. (Appendix liable A-9) 100% 75% 50% 25% 0% sir air Sir Mar air Wrier air ~ ~ ~ 1~ . d606~d 77 78 79 80 81 =3 Community Programs Training 82 83 84 85 86 87 88 89 YEAR Research Projects Program Support FIGURE 4 ~Allocation of ADAMHA budget from 1977 to 1989. (Appendix A-10)
SUPPORTING RESEARCH THROUGH NIT AND ~ . . 85 and management has declined by 30 percent. Likewise, the NLM budget has declined by 17 percent. Funding for the Office of the Director has remained fairly constant at 0.6 to 0.7 percent of the entire NIH budget. Intramural Research The largest portion of NIH funds for direct activities is allocated to the intramural research programs in the 13 institutes. Intramural program activities include basic research, clinical research, scientist training, com- munication of scientific findings, development of policies on biomedical research priorities, and translation of research findings into more effective medical care.6 Although none of these functions is unique to the intra- mural program, the intramural program is distinct in the federal portfolio of support for health research. The key features of the intramural pro- gram include freedom from competitive grant procedures; unique research resources, including the Warren Grant Magnuson Clinical Center; and research related directly to the individual institute missions. Whereas the intramural programs were allocated sizeable portions of the NIH budget during the early postwar expansion, allocations for the intramural program were stable throughout the 1980s at 10 to 12 per- cent of the total NIH budget (Figure 4-4~. Some institutes, such as the National Cancer Institute (NCI) and the National Institute of Environmen- tal Health Sciences (NIEHS), invest heavily in their intramural programs, whereas others, such as the National Institute of General Medical Sciences (NIGMS), have no significant intramural programs. In constant dollars, funding for the intramural programs has increased from $521 million in 1977 to $757 million in 1990 (Figure 4-7~. This reflects a 2 percent per year real growth in the intramural budget that parallels the overall growth in the NIH budget over the same period.6 During the 1980s, there was speculation that the intramural research program was not performing at the level of quality characteristic of it in the past. Whereas the NIH campus served as a primary training ground for health scientists in the 1950s and 1968s, there were signs in the 1980s that the NIH was beginning to have difficulty attracting and retaining outstand- ing basic scientists and clinical investigators. These deficiencies have been attributed to relatively low government salary scales, noncompetitive fringe benefits, and the other bureaucratic constraints of working in a federal agency. In response to these concerns and to the suggestion that the intramu- ral program could benefit by shifting to the private sector, the Institute of Medicine (IOM) conducted an in-depth review of the program in 1988.6 The IOM study committee concluded that the intramural program has
86 1 000 800 600 400 200 O FUNDING HEALTH SCIENCES RESEARCH DOLLARS (Millions) _ --_ _ ,/ , ~,0 - ,- ~,I,, +, -+ + ,,,,+-- 1 1 1 1 1 1 1 1 1 1 1 1 1 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 YEAR ' '' ' '' Constant 1988 $ ~+- Current $ FIGURE 4-7 NIH obligations for intramural research from 1977 to 1991. (Appendix A-8) previously made and continues to make valuable contributions to under- standing basic biological and disease processes. Despite NIH's difficulties in coordinating activities across institutes effectively and in responding efficiently to new challenges or crises, the IOM study committee also con- cluded that the federated organizational structure of NIH has helped meet the nation's biomedical research needs. In order to maintain the intra- mural program's excellence and credibility and to improve deficient areas, the study committee recommended some changes in NIH administration as well as in the scope of responsibilities of scientific administrators directing the intramural programs. The recommendations of the Institute's NIH intramural study commit- tee were as follows: · 1b increase administrative efficiency, the committee recommended that the Secretary of Health and Human Services delegate to the director of NIH the authority to make decisions on administrative matters without being subject to review by the Office of the Assistant Secretary for Health. · ~ increase NIH's ability to respond more effectively to emerging issues, new opportunities, and crises not confined to any single institute, the committee recommended that Congress annually appropriate a $25 million director's fund to be used to address these issues. · ~ enhance the quality of the intramural program, the committee
SUPPORTING RESEARCH THROUGH NIH AND ADAMHA 87 recommended that a formal review panel be established to evaluate each institute's scientific directors and intramural programs quadrennially. To be competitive, the committee believes that NIH has to exercise greater flexibility in the labor market and therefore recommended that Congress authorize NIH to develop and implement a demonstration project to overcome current staffing deficiencies. · To attract high-level scientists from outside government service, the committee recommended that Congress charter a foundation to permit private support for endowing 10 chairs for distinguished investigators. . . ~ attract junior-level investigators, the committee recommended that Congress authorize and appropriate funds to create an NIH Scholars Program in which outstanding young investigators at the assistant professor level are appointed on a competitive basis to independent, nontenured positions in the intramural program. This committee concurs with the intramural study committee that these measures will enhance the intramural program as the flagship of U.S. biomedical research. NIH Extramural Programs: R&D Grants R&D grants, particularly investigator-initiated research project grants (RO1), continue to be the cornerstone of the NIH extramural program (Bible 4-1 and Figure 44~. As growth in the NIH budget slowed during the 1970s, competition for grants intensified, and the number of new and competing renewal grants awarded by NIH fluctuated annually. Through the 1970s the number of proposals funded ranged from as few as 3,500 in 1976 to 5,900 in 1979 (Figure 4-8~; this number did not follow any particular pattern but depended on the cumulative grant portfolio and funds available in any particular institute. Initially, the 1976 Report of the President's Biomedical Research Panel brought to light the issue of fluctuating numbers of NIH and ADAMHA research project grant proposals being funded annually.7 By the end of 1970s, these fluctuations had increased and had caused even deeper notions of instability In the support of biomedical research. The 1979 and 1980 reports by the Health and Human Services (HHS) Steering Committee for the Development of a Health Research Strategy reexamined these concerns about the future of federal support for new as well as ongoing health research in light of impending federal budget constraints. These reports called for 5-year plans and evaluative procedures to be established for all of the health-related agencies in HHS.~9 The 1979 Steering Committee report also emphasized the need to stabilize the science base by making investigator-initiated research projects the first priority in the NIH and ADAMHA research budgets. As a result, Congress and the Executive
88 FUNDING HEAL:IH SCIENCES RESEARCH TABLE t1 NIH Research Grants by Kind and Type, 1988 (dollars in thousands) Kind of Grant and CodeNumberAmount _ _ _ Total25,754$4,727,320 Research projects20,&673,764,791 Traditional (R01)16,8712,564,198 Research program projects (P01)770634,809 New investigator research (R23)26011,920 SBIR (R43, R44)46054,334 MERll awards (R37)596140,829 Outstanding investigator (R35)7446,985 FIRSI awards (R29)1,227108,253 Other (R22, U01, P42)609203,463 Research centers621573,578 Specialized (P50)192164,531 Core grants (P30)176165,586 General clinical (M01)78102,159 Comprehensive (P60)3647,920 Biotechnology resource (P41 U41) 70 36,697 Primate research center (P5i) 7 33,300 Animal resource (P40, U40) 42 12,064 RCMIs (G12) 19 11,010 Other (P20) 1 310 Other research 4,266 388,952 Biomedical research support 1,110 90,918 & development (S03, S07, S10) RCPAs (K) 1,443 94,586 Cooperative clinical (R10, U10) 331 80,268 Minority biomedical support 102 83,407 (S06, S11, S14) Other 1,280 83,407 Represents awards, not obligations. SOURCE: U.S. Department of Health and Human Services. 1989. NIH Data Book 1989. National Institutes of Health Publication No. 90-1261. Bethesda, Md. Branch agreed on a policy that specified the minimum number of new and competing grants NIH and ADAMHA would be required to fund each year-"stabilization policy." The 1979 Steering Committee report suggested that the minimum number of competitive research grant awards for fiscal year 1981 be 5,000 for NIH and 569 for ADAMHA-4 From 1981 to 1988 increasing mini- mum numbers of new and competing research grants to be awarded were specified in either report or statutory language accompanying congressional appropriations bills. Whereas the initial NIH base was established at 5,000, the administration requested funding for only 284 awards for ADAMHA in 1981-only half the recommended level. However, this was modified
SUPPORTING RESEARCH THROUGH NIH AND ADAMHA 89 upward by Congress to 345. Throughout the 1980s, the number of new and competing proposals to be funded became an integral part of the federal budget ritual. By 1987 grant awards exceeded 6,400 for NIH and ADAMHA awarded nearly 600. In 1988, the last year of stabilization, NIH funded 6,200 grant awards. Despite added appropriations from congressional committees, the funds available were never adequate to fund fully the agreed upon number of awards. In order to comply, NIH and ADAMHA were forced into a policy of reducing ongoing research commitments (continuing awards for already approved and funded grants) as well as the amounts paid to new and competing awards in what is commonly referred to as "down- ward negotiation" a recent practice for reconciling NIH and ADAMHA research grant commitments to annual appropriations by making across- the-board reductions in all grant awards. Downward negotiation is a eu- phemism, for little if any negotiation actually occurs between the scientist and NIH or ADAMHA Rather, these decisions concerning the overall proportions of the previously committed funds to be withheld in order to fund the required annual level of new and competing awards are made between the NIH or ADAMHA and the Office of Management and Budget (OMB). Actually, downward negotiations are administrative budget cuts In the grant awards. This policy has placed additional burdens on scientists, 21 18 1 5 12 9 3 o NUMBER (Thousands) 1. STABILIZATION · | 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 YEAR (~at ) Grant Applications Grants Awarded FIGURE 4-8 Number of grant applications submitted to NIH and the number of grants awarded from 1970 to 1991. (Appendix A-11)
go 100% 75% 50% 25% 0% FUNDING HEATH SCIENCES SUCH 70 72 74 76 ROD Grants _ Other 78 80 82 YEAR ~ RSD Contracts 84 86 88 89 ~ NRSA Training FIGURE 4-9 NIH extramural awards as a percent of extramural activity from 1970 to 1989. (Appendix A-12) for they are now expected to perform the research proposed with less than the recommended amount of funding. The committee concluded that the stabilization policy prevented ero- sion of the nation's scientific base by maintaining minimum annual numbers of investigator-initiated research grants. Research project grants increased from 51 percent of the total NIH extramural budget in 1978 to 67 percent in 1989 (Figure 4-9~. Funding for these grew from $2.5 billion in 1977 to $3.9 billion by 1989 when measured in constant 1988 dollars (Figure 4-10~. This is the only portion of the NIH extramural budget that has grown in constant dollars over the past decade. As competition for funding intensified throughout the 1980s, the num- ber of grant applications with very high-priority scores increased. Nev- ertheless, despite high-priority scores, any given ongoing project faced termination if its score in competitive renewal fell just below the pay line. With interrupted funding, individual scientists felt they would be forced to reduce staff below critical levels so that although amended applications might ultimately restore funding to the program, the research team may by then have already been disbanded. As a consequence of these fears, multiple grant applications, with renewals in alternate years, were seen by many scientists as a means to provide continuity of funding for their research programs. While NIH and ADAMHA were increasing the numbers of new and
5 4 3 2 o SUPPORTING RESEARCH THROUGH NIH AND ADAMHA 91 competing awards through the stabilization policy, the research community felt that the average 3-year award period for traditional research project grants (R01) was too short. Three-year awards do not allow for long-term research program planning nor, in many cases, sufficient time to achieve research goals. Additionally, these shorter-duration awards required more frequent renewals and therefore placed too much emphasis on grant writing and administrative details. Beginning in 1986, NIH and ADAMHA instituted a policy to increase the length of grant awards gradually. One intended result of increasing award periods was to provide more stability in research activities and scientists' careers and, possibly, to discourage the number of multiple grant applications by individual investigators. Additionally, longer award periods were viewed as a means to reduce the administrative workload for NIH and ADAMHA study sections by reducing the number of competitive renewal applications processed each year. Although increasing grant duration has a stabilizing effect on research careers, it also obligates NIH and ADAMHA appropriations further into the future. This policy of lengthening award periods, coupled with the phenomenon of increasing average award size, reduces the funds available for meeting annual targets of new and competing grant awards. In fact, obligations for noncompeting continuations have grown from 67-68 percent DOLLARS (Billions) NUMBER (Thousands) 91 79 80 81 82 8 3 84 85 86 87 88 89 90 YEAR R01 Grants -I- Current $ - Constant 1988 $ _ P01 Grants 13 Other Project Grts 25 20 15 10 5 o FIGURE 4-10 NIH allocations for research project grants and number supported from 1979 to 1991. (Appendix A-13)
92 100% 75% 50% 25% o% FUNDING HEA17H SCIENCES RESEARCH 82 83 84 85 86 87 YEAR ~ Noncompeting 88 89 90 91 Competing FIGURE 4-11 Percent of allocations for competing and noncompeting research grant awards from 1982 to 1991. (Source: National Institutes of Health, Office of Science Polipy and Legislation). Of the NIH extramural research budget in the mid 1980s to more than 76 percent in 1990 (Figure 4-11~. Although this is a small percentage shift, these growing obligations to noncompeting awards represent about $350 million that is not available for funding new and competing renewal grant applications. As a result, NIH awarded only about 5,400 new and competing awards in 1989; this figure is expected to drop further to nearly 4,600 in 1990.~° The total number of research project grants sponsored by NIH grew from 15,500 to 20,867 between 1977 and 1988. In 1989, however, the total number of grants dropped to 20,681, and the number is expected to drop yet even further in 1990 to 20,316. A small gain is expected in the proposed 1991 budget to 20,439. Similarly, the ADAMHA research grant portfolio grew from 1,250 to more than 1,900 over the same period. As a result of the growth of research project grants, other extramural categories have received less and less of the total extramural budget (Figure 4-9. SETTING PROGRAM PRIORITIES THROUGH PEER REVIEW Priority Scores Peer review of competitive proposals forms the core of the NIH/ ADAMHA extramural grants programed The process generally takes a
SUPPORTING RESEARCH THROUGH NIH ED 989 JAN FEB ~Grant Application Submitted MAR APR MAY ~Initial Review Group (Priority Score) JUN ~ Pink Sheet to Principal Investigator JUL AUG SEP ~National Advisory Council OCT ~Grant Award to Principal Investigator NOV ~(Revised application resubmitted). DEC ~Earl lest Pro ject Start Date 1 990 JAN FEB MAR APR MAY JUN JUL AUG Initial Review Group (Re-review) Pink Sheet to Principal Investigator National Advisory Council Project Start Date 93 FIGURE 4-12 Application time line for new RO1 research grant and resubmission of revised application. minimum of 9 months, from application deadline to final award, for new investigator-initiated grant applications (Figure 4-12~. The peer review process for research project grant applications is based on two sequential levels of review; the first level is review by a select group of scientist peers serving on the so-called study sections, and the second level is a review (consisting of approval or disapproval) by the National Advisory Council for the respective NIH institute, center, or division (Figure 4-13~.~3 Applications for research project grants are submitted to NIH's Divi- sion of Research Grants (DRG), which serves as the central receiving point for applications submitted to NIH and ADAMHA The DRG reviews each application for relevance to the overall mission of the Public Health Service (PHS) agencies and assigns the application to the most appropriate Initial Review Group (IRG), also known as study section (program project grant applications are assigned directly to IRGs within the respective institute). These IRGs perform the first level of scientific peer review, during which a priority score is assigned to the research proposal. The DRG has 71 chartered IRGs, with 12 to 20 members each. In certain cases special study sections or ad hoc groups are convened to review applications not falling into the purview of any particular study section. The study sections are grouped according to scientific disciplines and not by institute. Therefore, several study sections may refer grants to a single institute, or a study section may refer grant applications to different institutes.
94 FUNDING HEALTH SCIENCES RESEARCH INITIAL REVIEW GROUP (STUDY SECTION) 0 Performs scientific merit review of grant applications 0 Makes budget and project length recommendations , . NATIONAL ADVISORY COUNCIL o Evaluates relevance of grant applications to institute program priorities and makes recommendation on funding o Advises institute on scientific priorities and policies FIGURE 4-13 Schematic diagram of the dual review system used for evaluating and awarding NIH and ADAMHA research grants. (Source: Modified from U.S. Department of Health and Human Sentences, Public Health Service. 1988. NIH Peer Review of Research Grant Applications. Bethesda, Md.: National Institutes of Health.) Study sections are composed mainly of nonfederal scientists represent- ing a wide range of specialties; these sections evaluate applications for scientific and technical merit. In most cases investigators have to demon- strate through preliminary studies that the proposed experiments can be completed if funds are awarded. Previous accomplishments by the inves- tigator, including publications, are considered during proposal review. If an investigator does not receive an award with an initial application, an amended or revised proposal can be resubmitted for another grant review cycle. Thus, the application-to-award period can be 18 months or longer (Figure 4-12~. Each proposal receives an in-depth reading from at least two study section members (primary and secondary reviewers) and limited review by the remaining members of the section. When the study section meets, the two reviewers discuss the proposal's scientific merit with their colleagues, and all reviewers in the section then vote on whether to approve or disapprove the application. If approved, all members of the review panel assign a priority score to the proposal; these scores range from 1.0 to 5.0
SUPPORTING RESEARCH THROUGH NIH AND ADAMHA 95 in tenths of a point, and lower scores indicate a higher priority. Individual scores are averaged and then converted into a scale ranging from 100 to 500 by the executive secretary of the study section. The raw priority score and the review panel comments are compiled into a critique (pink sheet) and sent to the applicant. A surge in the number of NIH grant applications in recent years has led to concern that certain study sections may be overloaded with proposals to review, although DRY Peer Review [tends (1986) indicates that, on average, study sections now handle fewer applications per member than in 1981.~3 However, averages may be misleading, because workloads vary and depend on subject matter and number of proposals received. Study sections also tend to be focused highly on specific research disciplines, and the committee believes that they may be ill-equipped to deal with the multidisciplinary nature of many research questions. Nonetheless, there have been concomitant increases in the number of special study sections and ad hoc groups to offset the increasing number of applications and changes in research directions. Award Rates The percent of NIH grant proposals receiving scores of 100 to 150 has more than doubled since 1978, with a concomitant drop in those receiving scores above 300 (Figure 4-14. Also, the approval rate for grant proposals has increased from 70 percent in the mid 1970s to nearly 95 percent in 1989 (Figure 4-15~. The improved scores and the increasing approval rates may be the result of several factors including the following: (1) improved grant writing skills as trainees spend longer periods of time in training programs, (2) new and better tools for conducting health research, (3) more amended or resubmitted proposals that receive better scores on the second or third attempt, and (4) behavioral changes by reviewers who know that even excellent proposals will not be funded if they are not ranked in the very top percentiles. Previously, grant awards were made by order of the priority ranking assigned by the study sections and approved by the advisory councils. However, a recent change to awarding grants by percentile rank has been enacted because of variances among the rankings of the various study sections. In the past, for example, two study sections may have submitted excellent grant proposals to a particular institute. However, because one study section may have ranked good proposals lower overall, the possibility existed that none of its reviewed grants would be funded. By contrast, study sections that ranked proposals higher had a better chance of being awarded in a system that relied on raw priority scores. The new system arranges the raw priority scores from the present
96 100 50 FUNDING HEALTH SCIENCES RESEARCH PERCENT of APPLICATIONS PRIORITY SCORE 75 77 ~ Score 100-175 79 81 83 85 YEAR ~Score 176-225 .. ~ Score 276-500 - Payline Score ~ Score 226-275 500 400 300 200 100 FIGURE 4-14 Distribution of priority scores for R01 research grant applications from 1975 to 1985. 100 80 60 40 A-- ok 1 PERCENT ~ / \ 20 '+ V~~~ - , · , ~, ~, ~, 1 ~ :~ STABILIZATION _ I A- - ~ , ~- -* ok, ~ ~ `.~ _ 1 1 1 1 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 YEAR (fist.) Approval Rate - Award Rate - +-- Success Rate FIGURE 4-15 Approval, award, and success rates for NIH research project grant appli- cations from 1970 to 1991. (Appendix Table A-11)
SUPPORTING RESEARCH THROUGH NIH AND ADAMHA 97 review cycle and the two previous review cycles of a study section into percentiles. Awards by the institutes are then made by determining a percentile cut-off point. Thus, percentiling normalizes the priority scores of each study section, thereby reducing the variances among the various review panels. However, this change has led to some confusion in translating percentile cut-off points into award rates. As a result of the increasing priority scores, the average pay line scores- the cut-off point for funding~ropped from more than 200 in the 1970s to less than 150 in 1989 (Figure 4-14~. Simultaneously, a decreasing proportion of these applications are awarded funding (Figure 4-15. ~ us, the average NIH award rate, the percent of approved applications funded, declined from 60 percent in 1975 to 29 percent for 1989, and it is expected to drop below 25 percent in 1990 (Figure 4-15~. However, there are large variances among the institutes-from 30 percent in the National Institute of Arthritis, Musculoskeletal, and Skin Diseases to 50 percent in the National Eye Institute for 1988. The dramatic decline in the overall award rate is moderated somewhat when the success rate, the percentage of total applications funded, is considered. Success rates at NIH have hovered between 35 and 40 percent over this same time period. Yet, as the approval rate edges closer to 100 percent, this discrepancy between award and success rates disappears. ADAMHA also has experienced increases in the priority scores in its proposal review system, albeit not to the extent of NIH. ADAMHA also experienced an increase in annual grant applications from about 1,800 in 1978 to 2,750 in 1988. The average award rate for research project grants has not declined as rapidly as that of NIH, hovering between 40 and 50 percent since the late 1970s.4~5 Although this is much higher than NIH, a smaller proportion of ADAMHA grant applications are approved by the peer review panels. When the success rates are considered, the average annual success rates for ADAMHA grant applications have not exceeded the 28 percent level attained in 1979. An NIH peer review committee was formed in 1987 to examine the cur- rent review process.~4 Among the major conclusions and recommendations from that review were the following: (1) innovative or multidisciplinary re- search should be reemphasized and enhanced by small grant mechanisms, (2) the number of pages in the experimental design and methods section of the grant application should be limited in order to ease the burden on the review process, (3) grant application submission via computer diskettes and electronic mail could accelerate the review process, and (4) applicants should be able to suggest several study sections to simplify the review pro- cess and use peer review resources more efficiently. Responding to these recommendations, in 1987 NIH implemented a 20-page limitation (with no more than 10 appendices) on grant applications. Also, NIH currently
98 FUNDING HEALTH SCIENCES RESEARCH has a trial project under way involving electronic submissions of grant pro- posals in hopes of shortening the 9-month grant cycle. An exception to the normal review period is the accelerated proposal review employed for AIDS-related research grants. In the spring of 1988, an experiment forcing reviewers to use incre- ments of 0.5 rather than 0.1 when assigning priority scores did not affect the priority score distribution significantly. Another suggested potential so- lution, that of increasing the number of readers for each proposal, has been met with strong opposition because of an anticipated increased workload on the study section members.~7 However, it was the committee's opinion that having more readers would include more scientific judgment in the final priority score. National Advisory Councils Once a priority score has been assigned, those proposals pertinent to a particular institute's mission are sent to a second review panel-the National Advisory Council. Each institute has a council that defines its program priorities and officially makes awards to investigators. Whereas the study sections evaluate proposals for their scientific merit, the councils concentrate more on a proposal's importance in terms of the institute's goals. The councils also advise the director and program managers on institute matters concerning overall program priorities and policies. The councils have a broad-based membership of scientific, profes- sional, and public sector leaders with expertise and interest in the institute's program areas. Generally, the councils consist of 12 scientists knowledge- able in the field and 6 volunteers from the nonscience community with a demonstrated interest in the discipline.~5 The councils do not have scien- tific support staff, nor do they have a budgetary allocation to investigate research issues affecting grants or other extramural awards. As a result, the councils generally approve a slate of grants proposed by program staff. Following council action, the grant proposals are returned to the DRG, which is responsible for administering the grant awards. However, decisions are made by the councils regarding proposals near the cut-off point and proposals submitted in response to requests to meet institute initiatives. In most institutes there is no official coordination between the Board of Scientific Counselors for the intramural program and the National Advisory Council, who oversees the extramural component.* Likewise, there is no ~NCI's intramural program has more direct control over the extramural program than do the other institutes.
SUPPORTING RESEARCH THROUGH NIH AND ADAMHA 250 200 150 100 50 O 77 78 79 80 81 82 99 DOLLARS (Thousands) ,+ - 83 84 85 86 87 88 89 90 91 YEAR Constant 19813 $ -ok- Current $ FIGURE 4-16 Average size of research project grant awards from 1977 to 1991. (Appendix liable A-14) mechanism for coordinating priorities among the councils of the 13 separate institutes, which can create inconsistencies in the overall NIH program. lYends in Grant Size The average size of NIH research project grant (RPG) awards has grown from $82,200 per year in 1977 to $198,100 per year in 1990 (direct and indirect costs included) (Figure 4-16~. In 19~ constant dollars, the true increase in research project grant size has been from $167,000 to $177,000, reflecting a 5 percent increase. However, the growth rates of different types of research grants vary widely. Between 1977 and 1988 the average size of RO1 grants grew from $67,400 to $155,000. Adjusting for inflation this reflects an 11 percent growth, or about one percent per year. By contrast, the average size of program project grants declined by nearly 26 percent in real terms over the same period. Since its inception in 1983, Small Business Innovation Research grants nearly doubled in real terms by 1988, although they account for a small portion of the extramural research budget. Several factors may have contributed to these award increases, the most significant being personnel costs. Personnel costs, which account for 63 to 72 percent of the research project grant direct costs, rose throughout
100 FUNDING HEALTH SCIENCES RESEARCH the award period. The General Accounting Office (GAO) has speculated that higher wages, more experienced personnel, increased numbers of personnel through project expansion, or less voluntary help may account for this increase. The cost of new and more sophisticated equipment also may have added to the increasing grant awards size. However, this committee noted that the percent of research project awards allocated for permanent laboratory equipment dropped from nearly 12 percent in 1966 to less than 4 percent by 1989.~7 Several other factors may directly affect the size of research project grants including the following: the Biomedical Research and Development Price Index (BRDPI) adjustment for inflation used by NIH; the mix of new, competing renewal, and continuation grants funded in a particular year; indirect cost recovery; increasing complexity of research; increasing regulations on animal use; and increasing use of human subjects. RESEARCH SCIENTIST TRENDS Applicant fiends The percentage of applicants funded on their first attempt varied in the 1970s: from about 24 percent in 1973 to more than 40 percent in 1975. However, in the 1980s the rate stabilized at just under 30 percent. Thus, about half of all applicants will receive a grant award if they apply persistently every year. This is demonstrated in Figure 4-17 which shows the percent of applicants for NIH research projects grant support by year of initial applications from fiscal year 1970 through 1986. This cohort analysis of first-time applicants was limited to traditional research project (R01), new investigator research award (R23), and First Independent Research Support and Transition (FIRST) award (R29) applications. As shown in Figure 4-8, the number of NIH grant applications has increased considerably in recent years; Ph.D. applicants largely account for this growth. This is particularly true for first-time applicants of tradi- tional research grants (R01) (Figure 118~. The number of first-time Ph.D. applicants has grown from 1,200 in 1965 to nearly 2,000 in 1985. Fifty- four percent of the first-time R01 applicants in 1965 had a Ph.D. degree compared to 80 percent in 1985. Over the same period, the number of first-time R01 applicants with M.D. degrees has declined 15 percent, from 950 in 1965 to 800 in 1985. This translates into 28 percent of the first-time applicant pool having an M.D. degree in 1985, compared to 41 percent in 1965. Approximately 100 M.D./Ph.D.s are first-time applicants for NIH funding each year, which also translates into a smaller percentage of the total first-time applicant pool. First-time M.D./Ph.D. and Ph.D. applicants have a slightly higher
SUPPORTING RESEARCH THROUGH NIH AND ADAMHA 1 00% 75% 5096 25% 0% 101 (Thousands) · · 1 1 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 YEAR _ Award 1st year Award after 1st year Number of applicants ~ No support received FIGURE 4-17 Percent of applicants for NIH research project grant support by fiscal year of first application from 1970 to 1986.11 approval rate than M.D. applicants, but the approval rate trends for all three subgroups are parallel. The trends of first-time RO1 recipients parallels that of the first-time applicants (Figure 4-19). Investigator Turnover As NIH funds more new and competing awards, it supports more prin- cipal investigators. However, the half-life of NIH support for investigators, a measure of time in which one-half of the grant recipients leave the NIH- supported system, has declined continuously since the 1940s. From 1945 to 1954, the median survival time for an investigator in the system ranged from 13 to 21 years. By 1960 the half-life of investigators had dropped to less than 7 years, and it currently stands at less than 6 years. Thus, it appears that as entry rates go up, median survival time goes down as investigators have more difficulty with their competing renewal applications. Many factors contribute to the higher turnover rate of principal in- vestigators. Much discussion and analysis have focused on award periods. Although the average length of ROls has increased to nearly 4 years since 1977 (Figure 4-20), it has been suggested that young investigators need more than the usual 3-year award to set up and become productive. NIH
102 2500 2000 1 500 1 000 500 FUNDING HEAL:IH SCIENCES RESEARCH NUMBER OF APPLICANTS 65 67 - 69 71 73 75 77 79 81 83 85 YEAR OF FIRST APPLICATION ~MDJPhD -+- MD PhD FIGURE 4-18 First-time R01 grant applicants by degree from 1965 to 1985. (Source: NIH Deputy Director for Extramural Research, and Division of Research Grants) 1 600 1 400 1 200 1 000 800 600 400 200 NUMBER OF AWARDEES , A ~ ..................................................................... O < ~in- -a 1' 1 1 'i' 65 67 69 71 73 75 77 79 81 83 85 YEAR OF FIRST AWARD * MD/POD -+- MD PhD FIGURE 4-19 First-time R01 grant recipients by degree from 1965 to 1985. (Source: NIH Deputy Director for Extramural Research and Division of Research Grants)
SUPPORTING RESEARCH THROUGH NIH AND ADAMHA 4 3.75 3.5 3.25 103 AWARD LEN GTH (Years) - 3 , 1 1 1 1 1 1 1 77 78 79 80 81 82 83 84 85 YEAR 86 87 88 89 FIGURE 4-20 Average award length for RO1 research project grants from 1977 to 1989. (Appendix Table A-15) has responded to this concern by establishing the FIRST program, which provides 5 years of support for young investigators. For midcareer inves- tigators NIH's Method to Extend Research in Time (MERIT) award can extend the period of grant support up to 10 years. NIH and ADAMHA study sections and advisory councils also have been encouraged to fund proposals for the requested length of the project rather than cutting grant periods. In doing so, future budget obligations of NIH and ADAMHA increased. However, providing stability for investigators over extended pe- riods of time must be weighed against a more flexible system with shorter periods of support. Multiple Awards The committee discussed at length principal investigators having multi- ple grant awards. An analysis of principal investigators receiving RO1 grant awards revealed a slight trend toward multiple awards. Between 1978 and 1989 the total number of RO1 research project grants grew from 11,929 to 16,084 (Table 4-2~. However, of the 4,000 net increase in awards, about 25 percent were awarded to principal investigators already receiving RO1 grant support. Thus, the number of principal investigators grew only by
104 FUNDING HEALTH SCIENCES RESEARCH 3,000 from 10,200 to 13,200. In this same period, the average number of awards per awardee grew as well from 1.168 to 1.218. The reasons for multiple awards are complex. However, the committee feels that, in many instances, investigators may need more than one grant for their research programs. Additionally, the committee believes that intense competition and downward negotiation have forced many investigators to apply for multiple grant awards to avoid funding gaps and to fund their research program fully. The committee also heard speculation that senior investigators with a proven track record of grant support may be listed as the principle investigator to ensure a steady stream of NIH or ADAMHA research grants. Another confounding aspect of research support is identifying the total number of scientists supported on NIH grants. The NIH data base on grant recipients only includes data on principal investigators. Thus, no data are collected on co-investigators. It is likely that the percent effort on an NIH grant by a co-investigator exceeds that of the principal investigator, who may be dividing his or her time commitments among two or more research grants. Also, junior scientists who may be co-investigators on several projects, but are not principal investigators, may be overlooked by the NIH data base and also by tenure review committees. There is evidence that the average age of principal investigators is increasing. Between 1979 and 1987 the mean age of investigators supported by ADAMHA increased from 41.5 to 44.6 and from 41.8 to 43.8 for NIH.~9 It is not clear, however, if this has any connection to multiple awards, especially from senior scientists. This may simply reflect the aging of the U.S. population overall. The committee Is concerned however, that large blocks of grant funds could be controlled by a few elite scientists, essentially closing the door on young scientists trying to get into the grant system. R&D CONTRACTS The NIH enters into contractual agreements to fund R&D in various private and public institutions. Generally, contract proposals are submitted directly to a specific institute in response to Requests for Proposals (RFPs) to develop such research resources as highly specific animal colonies or cell culture lines. The review process for R&D contracts differs from peer review for research grant applications.3 The requirements of the contract are determined by the institute prior to the announcement of the REP. Once submitted, the proposals are evaluated by a technical review group within the institute. Finally, senior scientific and contracts management staff evaluate merit and analyze cost considerations. In 1977 R&D contracts accounted for nearly 20 percent of the extra- mural budget, but by 1989 allocations for contracts had declined to less than
105 on Go on in C, ._ - s .; *= Ct A i_ U. - Ct o ._ ._ as Ct z o - .= ._ m .g - ._ o ID be .= c g o z - £ - U) ~4 o ~ ~ Z U) o A: £ O Z U) C) ¢< U) Ct - a ~ ~<a to to hi oo ~ ~ o~ o oo ~ o~ oo oo ox ~ o ~(~(~1 ('1 ~<~ ~0 ..... U)U)V) ~4 ~ 0 o oo os ~o ~ ~oo ~o ~oo oo r ~oo oo oo o~ r ~c] o ~oo oo oo oo oo et 0. O. C ~O. ~0 oo oo oo oo oo _ r ~0 ~r ~0 0 oo oO C ~d ~O ~o ~oo ~ oo ~o^ o^ o ~o^ o^ o^ o^ o^ o^ ~0 0 0 ~ < ~<~ ~4 K ~rn oo ~cr ~d- ~00 oxox ~oo ~o ~o ~i ~u) ~\0 \0 - 4 ~ox o^ ~ici ~i <~i ooox o ~ko ~oo ox o Y oS ox ~o ~c ~oS ox - ce ~> u, o o . - u) .5 . - s~ - - c - o ~q - , ~> ~ - - ~. - - ~ ) : ~q ~o : - . - - ~ ~ z . - - .. o ~ s:: u) ~- o ~ a 0 * ~n
106 1 000 800 600 400 200 FUNDING HEALTH SCIENCES RESEARCH DOLLARS (Millions) - - o Ll I I I I I I I I I 77 78 79 80 81 82 83 YEAR - Constant 1988 $ 84 85 86 87 88 ~+~ Current $ FIGURE 4-21 NIH obligations for research and development contracts from 1977 to 1989. (AppendLx liable A-12) 10 percent (Figure 4 9).11 In constant dollars, NIH obligations to R&D contracts have been halved over this period (Figure 4-21~. This shift may have been caused by the stabilization policy's emphasis on grants, where many R&D contracts may have become grants to reach the congressionally mandated quotas. R&D CENTERS The NIH supports nearly 600 centers designed to consolidate related research efforts and resources into a single administrative and program- matic structure. Centers serve as well as a vital institutional resource for multidisciplinary research. The funds provided through center grants are used for salaries of key staff, operation of shared resources and services, and center administration. These funds also may be used to recruit new talent to the center, to fund investigators who previously have not obtained competitive peer-reviewed federal funding, to provide interim research support for center investigators, and to obtain new shared resources. Specialized Centers, Center Core Grants, and General Clinical Re- search Centers (GCRCs) comprise the bulk of the centers program (Table 4-1~. Whereas these centers are primarily for research, other centers support research resources, such as primate and other specialized animal
SUPPORTING RESEARCH THROUGH NIH AND ADAMHA 107 colonies, as well as biotechnology resources. Centers and other research grants, including Biomedical Research and Support Grants, Career De- velopment Awards, and Cooperative Clinical Grants, now account for 18 percent of the NIH extramural budget compared to 22 percent in 1978 (Table 4-1 and Figure 4-9. The application and review processes for clinical and research centers differ slightly from those of investigator-initiated grant applications. The applicants generally submit a letter of intent. Although the center applica- tions are subjected to peer review by the appropriate institute, the review process also commonly involves site visits. Criteria for review include inter- disciplinary coordination, commitment of parent institution, qualifications of the director, impact of the center on the knowledge in the field, staff expertise and needs, shared resources needs, and the quality of the research protocols that will be performed in the center.20 The GCRC program is designed to support defined areas within aca- demic medical centers dedicated to patient-related research. These centers can be composed of specialized in-patient and out-patient facilities, labo- ratories and equipment, and mainframe computers. These are staffed by specialized personnel, such as biostatisticians, computer systems managers, research nurses and dieticians, and research laboratory technicians. The GCRCs have been instrumental in collecting and analyzing data in vaccine-related clinical research for several bacterial and viral immunogens, and they have provided the infrastructure for the major efforts underway to perform clinical investigations on AIDS. In fact, budget increases since 1986 for the GCRC program were earmarked almost entirely for AIDS research (Judith Vaitukaitus, personal communication). However, the number of GCRCs supported dropped from 93 in 1970 to 78 in 1988, and budget allocations to the GCRCs over this period do not reflect any real growth (Figure 4-22~. In 1989 the IOM undertook a study on NCI's Center Core Grant program.20 These core grants provide about 20 percent of NCI's grant support in cancer center institutions. By maintaining cohorts of specialized research scientists and clinicians, the centers are successful in obtaining funds from other federal agencies and nonprofit organizations. Whereas the number of centers supported by NCI has stabilized since the late 1970s, budget allocations have declined continuously when measured in constant 1988 dollars (Figure ~23~. The IOM cancer study committee concluded that the NCI centers program would be in serious jeopardy if measures were not taken to reverse the continual erosion of funding for centers. This study committee recommended the following: . The NCI should strengthen its core support of cancer centers in
108 FUNDING HEALTH SCIENCES RESEARCH order to exploit fully the application of these advances in the prevention and treatment of cancer and its consequences. · The director of NCI should take immediate steps to avert a crisis in the funding of the program during the 1989 fiscal year. The committee recommended further that the directors of NCI and NIH, with the secretary of DHHS, work with the appropriate committees of Congress to develop an adequate budget for the program's 1990 fiscal year. · The NCI should develop a systematic program plan to ensure adequate fiscal, managerial, and organizational resources; coordination with related programs; and effective scientific oversight for the cancer centers program. · The director of the NCI should consider how best to increase representation of the cancer centers program in NCI planning and decision- making processes, including regular representation of the centers at the NCI executive committee meetings and creation of an external advisory committee to review their multidisciplinary programs. The director of the NCI should strengthen substantially the man- agement capabilities of the cancer centers program unit. That unit must be able to plan, monitor, evaluate, and implement the cancer centers program adequately. 700 600 500 400 300 200 100 DOLLARS (Millions) NUMBER of CENTERS ,+ it' - .. . ,/- - . . . . . . . . . .. ~ ~ ~ ~ ~ ~ ~ ~ ~ ::. ~. * ~ ~ ~ O 1 1 1 1 1 1 ~ 1 1 ~1 1 1 1 1 1 O 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 YEAR ~ GCRCs - +-- All Centers 700 600 500 400 300 200 100 Constant 1988 $ FIGURE 4-22 NIH support for research centem from 1970 to 1988. (Appendix lible A-16)
SUPPORTING RESEARCH THROUGH NIH AND ADAMHA DOLLARS (Millions) 140 120 100 80 60 40 20 109 NUMBER of CENTERS ' '/ ,, . -- 4~ 1 -.~-Y' / ./ . . . . . .... .. O , , , , , , , , . . 1 1 1 1 1 1 1to 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 8889 YEAR -+- Number of Centers - Constant 1988 $ 80 70 60 50 40 30 20 10 FIGURE 4-23 National Cancer Institute support for cancer centers from 1972 to 1989. (Appendix Bible A-17) RESEARCH ADMINISTRATION Grants Management The Florida Demonstration Project As the federally sponsored research system has grown and aged, it has acquired a myriad of administrative, managerial, and financial proce- dures. These procedures seemed necessary and appropriate at the time the sponsored programs began or when they were recommended by Congress, federal agencies, state governments, universities, and participating scientists and engineers. Although all of these groups contributed to the accretion of bureaucratic requirements, there has been a recent movement by them to try to increase efficiency and productivity in all sectors of the American economy, including the research sector. A model program, designated Phase I of the Florida Demonstration Project (FDP), therefore was designed to test the efficacy of standardizing and simplifying the financial and administrative requirements of grants as a means of enhancing research productivity.2i 22 This program was intended to reduce the administrative burden on grantees by streamlining procedures and reducing costs in the sponsored project system. The demonstration began in April 1986 and ended in December 1987. Nine campuses of the Florida State University System and Miami University participated in the demonstration along with five federal agencies: the Department of
110 FUNDING HEALTH SCIENCES RESEARCH Agriculture, the Department of Energy, NIH, the Office of Naval Research, and the National Science Foundation. At the outset there were four primary program objectives: (1) stan- dardize postaward administration of federal research grants among the federal agencies to the extent possible; (2) eliminate most federal prior ap- provals for budget reallocation; (3) simplify research project management procedures; and (4) allow an investigator's collective research program to be one administrative and accounting unit rather than separate units. However, many elements of the sponsored project system were to remain intact, such as the basic framework for federal stewardship and account- ability, and there were to be no changes in the federal agencies' project proposal, proposal review and evaluation, and project award mechanisms, and methods for reimbursement of direct and indirect costs. The Government-University-Industry Research Roundtable (GUIRR) of the National Academy of Sciences was primarily responsible for develop- ing the FDP. The FDP was evaluated initially in August 1986, followed by two questionnaire evaluations in November 1986 and April 1987.22 These evaluations of the project showed the following results: (1) less time was needed for final action on such project management items as extensions, budget changes, and travel changes; (2) paperwork and administrative tasks decreased throughout the system; (3) principal investigators had more flexi- bility, responsibility, and control, which the investigators believed increased their own laboratory productivity; (4) federal sponsoring agencies showed a greater trust in the universities' administrative capabilities; (5) relationships between the principal investigators and university administrators as well as between university and federal administrators improved; and (6) scientific and financial accountability were maintained. The project now has entered Phase II and has been redesignated the Federal Demonstration Project. TNenty-six institutions have been added to the original ten for further evaluation of this project, which began October 1, 1988. Biomedical Research Support Grants One program that is tied closely to NIH research project grant awards is the Biomedical Research Support Grant (BRSG) program sponsored by the National Center for Research Resources (formerly known as the Division of Research Resources). Unlike the investigator-initiated research projects that are awarded through a competitive system using peer review, BRSGs are awarded to institutions according to a formula. Rather than requiring institutions to submit proposals for specific projects, the BRSG program provides funds to those research institutions engaged heavily in biomedical research. Thus, the BRSG program provides proportionately
SUPPORTING RESEARCH THROUGH NIH AND ADAMHA 111 more funds (up to a maximum of $500,000) to institutions that previ- ously have demonstrated the strength of their research efforts through the competitive grants system. The flexibility accorded the institution in determining the use of the funds is the unique characteristic of the BRSG program. Recipient in- stitutions are required to have a designated program director, to have an advisory committee to oversee use of the funds, and to advertise the availability of these funds within the institution. With the exception of a few minor restrictions, the use of the funds is left to the discretion of the institution. Many institutions establish peer review panels to evaluate proposals requesting BRSG funds for research or shared equipment. The latitude granted to the institutions for the use of these funds gives them more flexibility to fund emerging areas of research or new investigators be- fore they can be fully competitive in the traditional research project grant (R01) system. The BRSG program has come under increasing attack in the budget preparation process over the past few years and has been identified as one program to trim in order to reduce the NIH budget. In fact, the program was trimmed from $55 million in 1989 to $44 million in 1990. It is slated for yet deeper cuts in the 1991 budget, in which only $17 million is allocated to the program. It is not clear to the committee why there have been repeated attempts to eliminate this program. The committee can only speculate that NIH may not favor local control of these funds, which are not subject to national peer review. However, the committee strongly endorses this program because it allows all institutions to enhance their own health research programs according to institutional needs. The BRSG program may be an increasingly valuable grant mechanism for maintaining career stability for mid level scientists if the number of research grants awarded becomes more unpredictable. INCREASING COSTS OF ANIMAL USE The cost of acquiring and caring for laboratory animals has continued to increase since 1978. In large institutions the costs of maintaining cen- tralized animal facilities usually are included as part of per diem charges for animals. For example, at one university per diem charges for mice rose from 5.5 cents in 1978 to 14 cents in 1987 an increase of over 150 percent while the Consumer Price Index rose by 74.1 percent during that time. Over the same period, the per diem charges increased for dogs (from $3.60 to $8.61), for monkeys ($1.05 to $2.71), and for cats (from $1.30 to 3.20~.23 ADAMHA estimated that new regulations concerning animal care would cost in the neighborhood of $40,000 to $70,000 per grant on the care of primates and dogs.24
112 FUNDING HEALTH SCIENCES RESEARCH The scientific community is very concerned about the increasingly stringent federal regulations for animal experimentation and the effect these regulations have on the increasing costs of doing research. Pressure from animal rights groups to tighten these regulations has had and will continue to have a profound effect on the numbers and kinds of animals used in health sciences research. Although scientists are actively involved in seeking alternatives to animal testing, they feel certain types of systems biology experimentation must be performed in animals. The committee fears as well that curbing all animal experimentation directly or indirectly by imposing unrealistic regulations will slow scientific progress in fighting human disease. INDIRECT COST RECOVERY During World War II, the federal government entered into contractual agreements with research universities in mutually beneficial partnerships. As a result, the government was able to capitalize on research results by supporting scientists at these institutions. Not only were the direct costs of performing the research supported, but the indirect or overhead costs of the research were reimbursed as well. Thus, it became federal policy to reimburse institutions for the ancillary costs of performing federally sponsored research.25 The institutional indirect cost rate is negotiated annually on an indi- vidual basis with one of the sponsoring federal agencies, and this rate is honored by other agencies sponsoring research at the institution. The Office of Management and Budget circulars A-21 and A-110 set the government- wide accounting principles for direct costs and indirect costs of sponsored research at colleges and universities.26 The indirect costs are calculated as an average cost of research on a prorated share of all overhead costs in proportion to the ratio of sponsored research expenditures (all sources) to the total expenditures of the university, rather than the marginal costs of a research project. Indirect costs are subdivided into the following seven categories: (1) operation and maintenance expenses, (2) use charges for buildings and equipment, (3) library expenses, (4) sponsored projects administration, (5) general administration, (6) student administration and services, and (7) departmental administration. Whereas the costs associ- ated with the first three categories can be documented, the latter four administrative cost components are the least definitive and most difficult to evaluate in terms of individual research projects. Although the costs as- sociated with operations, maintenance, and use allowance are more easily documented, they are no less controversial in these extremely tight fiscal times. Before 1966 indirect costs were fixed by Congress. Since then the
SUPPORTING BESEECH THROUGH NIH ED 1 00% 75% 50% 25% 113 of 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 YEAR Direct Cost ~ Indirect Cost FIGURE 4-24 Proportion of direct and indirect costs for NIH research grants from 1970 to 1988. (Appendix Bible A-19) policy changed, and the average rate has climbed from 25 percent to more than 31 percent for NIH-sponsored research (the range is 7 to 100 percent) (Figure 4-24~. This reflects a shift in federal policy from direct support for facilities through facilities grants to a policy of indirect cost recovery. In effect, universities and colleges have been forced to increase indirect cost rates in order to maintain or update facilities. However, individual investigators perceive rising indirect cost rates as a threat to the available pool of funds for the direct costs of health research. Indirect costs for facilities renewal is covered in more detail in Chapter 6. SUMMARY AND CONCLUSIONS From this review of the funding trends for the various programs of research support, the committee concluded that the stabilization policy was successful in maintaining support for investigator-initiated research project grants. As a result of the emphasis on research project grants, this portion of the extramural budget grew from 51 to 67 percent between 1977 and 1990, rising from $2.5 billion to $3.9 billion. Consequently, in order to meet increasing annual targets of new and competing grant awards, funds had to be taken from commitments to ongoing research projects as well as by reducing new awards through downward negotiation. Funds also were
114 FUNDING HEALTH SCIENCES RESEARCH diverted from other programs specifically research centers and contracts- and resources for training declined steadily throughout the 1980s. The committee further concluded that scientists responded to these growing problems by increasing their number of applications to maintain research programs. Although the total number of grants supported has grown from 15,500 to more than 20,000 since 1977, the often-cited award rates and pay lines have dropped steadily over this period. These declines have been caused by several factors including the substantial increase in the volume of applications; and lengthening award periods. These declines also have been caused in part by the fact that study sections have been approving steadily increasing percentages of the applications and assigning increasingly higher (lower-numbered) priority scores. Additional conclusions were drawn by the committee from an analysis of the effects of the lengthy application and review process for research grants. The review process for new NIH/ADAMHA grant applications takes a minimum of 9 months. Without intermediate feedback steps, those applications requiring modification and resubmission can take as long as 18 months. A misunderstood or poorly written proposal, even if it involves meritorious work, may lead to loss of support for a successful laboratory with important ongoing projects, and although it is technically possible to restore funding by resubmitting amended proposals and clarifying the parts that were unclear, in reality, support has since disappeared. The committee believes that, in these cases, people are fired and the projects halted, and by the time support is later restored, it is often too late to salvage talent and other resources. The committee was concerned that the frailty of this system hinders long-term research planning and can affect career investigators adversely. It also prevents timely work in new areas and in those sectors in which health crises require faster-paced research activities. Thus, despite the historically unbroken trend of increasing annual funding, the committee concluded that the episodic nature of funding (in terms of interruptions), downward negotiations, and the progressively de- creasing likelihood of receiving awards have fostered perceptions within the academic community that careers in health sciences research are unstable and unpredictable. These perceptions are exacerbated by delays in deter- mining federal budget allocations, fluctuating funding patterns within NIH and ADAMHA, and reductions in awards for ongoing projects. Perhaps partly because of these problems, the length of time that principal inves- tigators receive continuous RO1 grant support from the NIH/ADAMHA sponsored research system has been cut in half, from more than 13 years In the l950s to less than 6 years in the 1980s. In addition, as the federally sponsored research system has grown and aged, it has acquired a myriad of administrative, managerial, and financial
SUPPORTING RESEARCH THROUGH NIU ED ~ 115 procedures that have increased bureaucratic requirements. The Florida Demonstration Project is commendable in its goal to reduce administrative burdens and increase laboratory productivity of the scientific work force. The committee concluded that health sciences research requires an environment that identifies, encourages, and develops creativity. Such an environment requires stable support for scientists and flexibility in allocating resources to meet changing demands. When the environment is positive, supportive, and reasonably predictable and optimistic, it encourages the recruitment and retention of talented health sciences researchers. REFERENCES 1. U.S. Department of Health and Human Services; Public Health Service. 1989. NIH Almanac. Publication No. 89-5. Bethesda, Md.: National Institutes of Health. 2. U.S. Department of Health and Human Services; Public Health Service. 1989. NIH Data Book Publication No. 90-1261. Bethesda, Md.: National Institutes of Health. 3. U.S. Department of Health and Human Services; Public Health Service. 1988. ADAMHA Funding Mechanisms for Grants and Awards. Rockville, Md.: Alcohol, Drug Abuse, and Mental Health Administration. Seggel, R.L. 1985. Stabilizing the Funding of NIH and ADAMHA Research Project Grants. Washington, D.C.: National Academy Press. 5. U.S. Department of Health and Human Services; Public Health Service. ADAMHA Data Source Book, F Y 1988. ADAMHA Program Analysis Report No. 89-18. Rockville, Md.: Alcohol, Drug Abuse, and Mental Health Administration. 6. Institute of Medicine. 1988. A Healthy NIH Intramural Program: Structural Change or Administrative Remedies? Washington, D.C.: National Academy Press. 7. U.S. Department of Health, Education, and Welfare; Public Health Service. 1976. Report of the President's Biomedical Research Panel. DHEW Publication No. (OS)76-500. Washington, D.G 8. Institute of Medicine. 1979. DHEW's Research Planning Principles: A Review. Washington, D.C.: National Academy of Sciences. 9. Institute of Medicine. 1980. DHEW Health Research Planning, Phase II: A Review. Washington, D.C.: National Academy of Sciences. 10. Moskowitz, Jay; Associate Director for Science Policy and Legislation, National Insti- tutes of Health. 1989. Presentation at ALAS Symposia on Research and Development in the FY 1990 Federal Budget. Washington, D.C.: April 1989. 11. U.S. Department of Health and Human Services. 1988. Extramural [lends, FY 1778-87. Bethesda, Md.: National Institutes of Health, Division of Research Grants. 12. U.S. Department of Health and Human Services; Public Health Service. 1985. DRG Peer Review [bends; Workload and Actions of DRG Study Sections, 1975-1985. Bethesda, Md.: National Institutes of Health. 13. U.S. Department of Health and Human Services; Public Health Service. 1986. DRG Peer Review [lends; Member Characterisitcs: DRG Study Sections, Institute Review Groups, Advisory Councils and Boards, 1977-1986. Bethesda, Md.: National Institutes of Health. 14. U.S. Department of Health and Human Services; Public Health Service. 1988. The health of biomedical research institutions: Report of the regional meetings. Proceedings of the 57th Meeting of the Advisory Committee to the Director, National Institutes of Health, Bethesda, Md. June 27-28, 1988.
116 FUNDING HEALTH SCIENCES RESEARCH 15. U.S. Department of Health and Human Services; Public Health Service. 1988. NIH Advisory Committees: Authority, Structure, Function, Membem. Publication No. 88-11. Bethesda, Md.: National Institutes of Health. 16. U.S. General Accounting Office: Biomedical Research: Issues Related to Increasing Size of NIH Grant Awards. Report No. GAO/HRD-88-90BR; May 1988. 17. U.S. Department of Health and Human Services; Public Health Service. 1985. Academic Research Equipment and Equipment Needs in the Biological and Medical Sciences. NIH Program Evaluation Report No. 85-2769. Bethesda, Md.: National Institutes of Health. 18. U.S. Department of Health and Human Services; Public Health Service. 1984. The extramural awards system. Proceedings of the 50th Meeting of the Advisory Committee to the Director, National Institutes of Health, Bethesda, Md. November 19, 1984. 19. U.S. Department of Health and Human Services, Public Health Service: Age [lends of ADAMHA Principal Investigators. ADAM HA Program Analysis Report, No. 88-10. Rockville, Md.: Alcohol, Drug Abuse, and Mental Health Administration. 20. Institute of Medicine. 1989. A Stronger Cancer Centers Program. Washington, D.C.: National Academy Press. 21. Hively, W. 1988. Getting rid of red tape. American Scientist 76:241-244. 22. National Academy of Sciences; Government-University-Research Roundtable. 1988. Unpublished evaluation documents. 23. National Research Council. 1988. Use of Laboratory Animals in Biomedical and Behavioral Research. Washington, D.C.: National Academy Press. 24. Holden, C. 1989. A preemptive strike for animal research. Science 244:415-416. 25. Association of American Universities. 1988. Indirect Costs Associated with Federal Support on University Campuses: Some Suggestions for Change (Draft). Washington, D.C.: AAU Ad Hoc Committee on the Indirect Costs to the Executive Committee of the AAU. 26. Office of Management and Budget: Principles for Determining Costs Applicable to Grants, Contracts, and Other Agreements with Educational Institutions. OMB Circular A-21. Washington, D.C. (Revised February 1979.)
