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.
7 Policy Overview and Recommendations SUPPORT FOR HEALTH SCIENCES RESEARCH Since World War II, U.S. health science policy has led to an unquestion- ably successful health research enterprise. This growth and development has been nurtured by a unique mixture of research sponsors: the federal government, state governments, private foundations and voluntary health agencies, and corporations. This multifaceted support system has invested large amounts of resources into building an unequaled health research infrastructure. Such a diversified system of support has diminished the po- tential for centralized planning and has encouraged input from a wide range of views regarding the conduct of health sciences research ensuring that no one group could impose limitations on what ideas should be pursued or how the research should be conducted. In effect, this unique research system has preserved academic freedom and encouraged the creativity of health scientists allowing them to develop and test their hypotheses for improving our understanding of human disease processes. The commit- tee concluded that preserving this broad-based support system is essential to continue a vigorous US. health sciences research program. Opportunities in health sciences research appear to be growing almost exponentially. However, the number of excellent ideas far exceeds the available funding. In order to maintain the momentum for supporting the best science resources have to be allocated to the various components of the research enterprise. The committee emphasizes that the components of 162
POLICY OVERVIEW AND RECOMMENDATIONS 163 the research infrastructure are highly interrelated and dependent upon one another, and scientists will not be able to deliver optimum research results and train young investigators without adequate facilities and equipment. Likewise, overemphasis on increasing research space will be to no avail if the buildings are underutilized by scientists and mentors because of shortages of researchers and research funds. The charge to this committee was to analyze the entire research infrastructure (people, projects, facilities, and equipment) in a holistic fashion and develop a coordinated set of policies to ensure balanced allocations to the components of the research system: people, projects, and facilities. Although the committee acknowledges that many of the previous ac- complishments in the health sciences have been directly attributed to the magnitude of federal support for health research, the charge to the com- mittee did not include justifying a basis for increasing congressional ap- propriations. Undoubtedly, this phase of the allocation process is critically important to the continued success of the health sciences research enter- prise. Because of the overwhelming success of previous health research endeavors, new and heuristic research opportunities are emerging con- tinuously. However, such opportunities easily could consume substantial increases in funding. Although this study does not address the process for increasing congressional appropriations, the committee acknowledges that the level of future research support from Congress will be related directly to the potential societal benefits of health research. The primary goal of this committee in developing a balanced allocation process was to preserve the creativity of the individual investigator the most valuable asset to the research enterprise. Beyond the role of the federal government, the committee believes that better inter-sectoral communication among the government, in- dusay, and private nonprofit sponsors of health sciences research is necessary to meet this challenge. Patterns and Policies of Federal Health Sciences Support Federal support for basic research has been based on the following five principles: 1. Stable federal support for research in order to undertake long-range programs. 2. Peer review for evaluating scientific merit of all research projects paid for by federal funds. 3. Academically based scientific investigation. 4. Flexible scientific research management policies left to individual scientists and their institutions. 5. Accountability to the Congress, the President, and the American public.
164 FUNDING HEALTH SCIENCES RESEARCH Although many changes have evolved in the way science is supported by the federal government, these general principles have stood the test of time. Many private and nonprofit sponsors of health research have adopted them as well. Research Community Perceptions The committee closely analyzed federal funding trends and policies for health sciences research over the past two decades. Additionally, the committee reviewed the limited data available on the contributions from other governmental sources, private foundations and voluntary health agencies, and corporate research sponsors. These analyses revealed a continuous commitment to the support of the health sciences research enterprise in terms of both absolute and relative dollars. Funding patterns through 1989 revealed that more investigators and projects are receiving federal support than ever before, with the highest total allocations since the system began. Furthermore, despite federal budget cuts in many nonhealth domestic programs throughout the past decade, health sciences research has continued to receive annual increases in appropriations through fiscal year 1990. Regardless of these gains, there is a strong feeling within the scientific community that federal support for future health research is unstable and unpredictable. Much of the concern is based upon two often cited statistics: (1) the declining number of annual new and competing awards and (2) the decline in annual award rates (award rate = grants awarded/total approved applications). Adding to the confusion, NIH and ADAMHA recently have replaced the former system for awarding grants based upon raw priority scores with a more complex moving average percentile ranking system. In the new percentiling system the raw priority scores assigned to grant applications in the current review cycle of a particular study section are percentiled with priority scores from the two previous review cycles. In effect, percentiling diminishes the variance of the priority scores assigned by a particular study section over three review cycles. Once the slate of grant applications is presented to the various institutes, the institutes determine a certain percentile threshold for proposals to be awarded. By selecting a percentile funding cutoff for awarding grants, the variance in priority scores among different study sections is normalized. Since its inception, however, the percentile cut-off generally has been misconstrued by the scientific community as the award rate. For example, institutes may be funding to the 12th or 13th percentile but this translates into a 24-25 percent award rate. Thus, the low percentile cutoff has led to a false impression of "declining" support for health sciences by NIH and ADAMHA
POLICY OVERVIEW AND RECOMMENDATIONS 165 Adding to this misperception are other less explicit policy changes and disturbances within the health research environment reflected in award rates. For example, the policy to lengthen grant awards, which has increased the average duration of research grants from 3.2 years to approximately 4 years, implies that about 25 percent fewer grants will need to be renewed each year to sustain the same overall annual number of research projects supported by NIH and ADAMHA Since 1976, the number of grant ap- plications submitted for peer review has continued to outpace the growth in appropriations. This dramatic growth in the number of investigator- initiated research project applications reflects, in part, a surge in research opportunities as well as the growing practice of applying for multiple grant awards. Simultaneously, the approval rate for grant applications by study sec- tions has risen steadily, from 70 percent in the mid 1970s to nearly 95 percent. As a result, the growth in applications, combined with the in- creasing approval rate, has driven down the award rate throughout the 1980s (Figure 4-8~. Despite recent declines in annual new and competing awards, the total portfolio of NIH grant awards (competing plus noncom- peting continuations) has grown every year from the early 1970s until 1988. Although the total number of awards supported by NIH dropped by 200 In 1989, from 20,867 to 20,681, and further to 20,316 in 1990, the NIH investment committed to research in real dollar amounts has continued to grow. The total number of awards is expected to increase slightly by nearly 125 in fiscal year 1991. These policy changes have created substantial long-term pressures on the infrastructure of the research enterprise. Despite the overall long- term upward trend in research funding, these longer-term concerns became the primary motivation for this study. The committee has analyzed all of the available data pertaining to the overall level of support for research projects and has found no evidence to confirm the research community's perception of declining federal research support. To clarify these issues and to recommend corrections, the next section summarizes policy decisions, both implicit and explicit, that have contributed to the current anxieties within the research community. A Brief Review of Past Policy Decisions for Resource Allocation The First 20 Years: Balanced Growth in Research Support Between 1950 and 1968 NIH underwent rapid growth. Federal policies combined with growing congressional appropriations fostered considerable flexibility for funding research, with support being provided for all four
166 FUNDING HEALTH SCIENCES RESEARCH of the interdependent components of the research enterprise: (1) a well trained pool of researchers, (2) modern facilities, (3) modern equipment and (4) research project funding. During this era, various mechanisms for supporting research were established in an effort to control disease and improve health. While investigator-initiated research and development (R&D) grants have been the cornerstone of NIH and ADAMHA extramural programs during the postwar expansion, other mechanisms for investigator-initiated research support have included program project grants, center grants, and, more recently, cooperative agreements. R&D contracts have been yet another mechanism for supporting research, although contracts commonly are not investigator initiated. There has also been a strong federal commitment to train researchers and to build research facilities. Constrained Growth and Instability in the 1970s "Stabilization" Policy. Slower budgetary growth in the 1970s along with a dramatic inflation rate reduced the buying power of research dollars. One result of these forces was the fluctuating number of annual new and competing grants awarded in the late 1970s. For example, between 1975 and 1976 the number of new and competing awards dropped from 4,700 to 3,50~a drop of nearly 25 percent and then surged to 5,200 in 1978. These fluctuations led to uncertainty in the availability of ongoing support for health research. Consequently, the scientific community began to lose confidence in the future of federal research support. In response to these concerns, Congress, NIH, and ADAMHA agreed to a policy that established a minimum number of new and competing re- search project grant applications to be funded annually. Beginning in fiscal year 1981 and ending in 1988, minimum numbers of new and competing awards were established between NIH/ADAMHA and the congressional appropriations committees. This decision, in turn, reflected an explicit NIH/ADAMHA policy that investigator-initiated research project grants were the highest priority for their current research programs and that maintaining a minimum level of new awards would stabilize the health research base. However, the Administration's budget requests for NIH and ADAMHA also had to conform to the Department of Health and Human Services (DHHS) budgets. DHHS budgets were, in turn, highly influenced by budget balancing in the Office of Management and Budget (OMB). "Downward Negotiation" Policy. Despite added appropriations from congressional committees, the available funds were inadequate to fund fully the agreed upon number of awards. Thus, in order to comply, NIH and ADAMHA were forced into a policy of reducing ongoing research commit- ments (noncompeting continuing awards from previous years) as well as the
POLICY OVERVIEW AND RECOMMENDATIONS 167 amounts paid to new and competing awards in what commonly is referred to as downward negotiation. This is 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 negoti- ation is a euphemism, since 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 NIH/ADAMHA and OMB. This policy has placed addi- tional burdens on scientists, for they are expected to perform the proposed research with less than the recommended amount of funding. The committee concluded that the stabilization policy was a sound strat- egy to protect the research base. However, the necessary appropriations to support fully the ongoing research obligations of NIH and ADAMH~4 were not provided. As a result, NIH and ADAMHA were forced to make arbitrary ad- ministrative cuts in all grant awards to be able to fund the mandated new and competing grants. While the overall grant portfolio grew, these cuts contributed to instability in the research project support system as well as to an imbalance among support for the other components of the research enterprise. Further Constraints and Crises in the 1980s "Extended Duration of Awards" Policy. Although NIH and ADAMHA were increasing the numbers of new and 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, do they allow scientists sufficient time to achieve research goals. Additionally, these shorter-duration awards require too much emphasis on grant writing and administrative details. As competition intensified throughout the 1980s, the number of grant applications with very high priority scores increased. Nevertheless, despite high priority scores, any given ongoing research 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, and although amended applications might ultimately restore funding to the program, the research team may by then have been disbanded. As a consequence of these fears, multiple grant applications, with renewals in alternate years, were seen by many scientists as a means of providing continuity of funding for their research programs. 1b address these concerns, NIH and ADAMHA instituted a policy to increase the length of grant awards gradually. The intended results of increasing award periods were to provide more stability in research activities
168 FUNDING HEALTH SCIENCES RESEARCH 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 does have 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. Despite these problems, the committee believes it is no longer justifiable to sacrifice the stability of support for productive scientists simply to maintain a given annual quota of grant awards. To this end, the committee supports this NIH and ADAMH~4 policy to extend award periods, even if it reduces the number of new and competing awards in any single year. Despite the consequence of a sharp reduction in new and competing awards in a correction year, the system will once again attain a balance in the out years Peer Review Process and Allocation Policy The committee also heard testimony from the scientific community about the effectiveness of the peer review process for evaluating grant proposals. As the proportion of approved proposals receiving funds has declined over the past decade, many scientists believe that the peer review process has favored the "old boy network" of mid to late career investigators who have been receiving continuous research support at the expense of young creative scientists just entering the competitive grant system. While there are data demonstrating that the average age of principal investigators is increasing, there is no evidence that these older scientists are less creative or that their grant applications are less meritorious. It is commonly believed that many scientists are reluctant to submit innovative or high-risk proposals because the review panels may be averse to recommending funding for less conventional research. Prevention and nutrition research are examples of proposals that may not fair well in the present structure of peer review for various reasons. Because of the confidentiality of unfunded research grant proposals, no data exist on the number of "novel" proposals not awarded. However, the committee acknowledges that a closer examination of the peer review system and its role in determining the effectiveness and efficiency of expended research funds may be warranted. Conclusions on Research Funding The most disturbing aspect of the scientific communi~'s perception of
POLICY OVERVIEW AND RECOMMENDATIONS 169 declining research project grant support is that it predisposes against sig- nificant corrections for other elements of the research base. The committee concluded that this has led to a climate where federal support for health sci- ences research has become focused too heavily on projects and not enough on developing career scientists and fostering creative environments. That is, over the last decade training, equipment, and facilities have become steadily and significantly underfunded in relation to research project support. Unfor- tunately, funding from other nonfederal sources, although substantial, has not compensated sufficiently for the accumulated loss of federal support for these long-term investments in the health sciences research enterprise. Before recommending the steps that can begin to address these accumu- lated imbalances, the committee reviews briefly in the next section some of the additional analytic factors that must be considered as propagating the specific status of the neglected health research components namely, training, equipment, and facilities. Support for Training Many of today's senior health science faculty members in colleges, universities, and medical centers can trace their careers to the various training programs underwritten by NIH/ADAMHA between 1950 and 1970. However, federal funds allocated for training new researchers have not kept pace with expanding research opportunities. As indicated in Figure 4-9, NIH support for training as a percent of the extramural budget declined from 17.2 percent in 1970 to 6.6 percent in 1978 and even further to 4.2 percent in 1988. Furthermore, inflationary pressures have been shrinking the real dollar value of stipends awarded to trainees and fellows. The emergence of unexpected health crises, such as AIDS, emphasizes the importance of maintaining a cadre of highly talented scientific personnel who can be redirected quickly as needed. According to a recently released report by the Office of Science and Engineering Personnel of the National Research Council, Biomedical and Behavioral Research Scientists: Their Raining and Sup, entrants into health sciences research have increased slightly while recruitment of Ph.D. scientists by the private sector has increased markedly. Furthermore, predictions of an increasing attrition rate among scientists trained during the postwar expansion is cause for concern about research personnel shortages before the end of this century. These and other factors will affect the pool of educators and mentors to train the next generation of scientists, in both academia and industry. The committee concludes that steps must be taken now to maintain the pool of career scientists by recruiting and retaining the best possible candi- dates. Resource allocation policies fostering health research careers will
170 FUNDING HEALTH SCIENCES RESEARCH require long-term investments from a variety of sources. Academia, gov- ernment, industry, private foundations, and voluntary health agencies need to play cooperative roles in developing and pursuing effective strategies for enhancing and renewing the nation's health sciences talent base. These strategies should not only focus on recruiting individuals into science ca- reers, but should also nurture these individuals to the level where they become independent young investigators. New approaches should include programs targeted at multidisciplinary and interdisciplinary training of sci- entists who are becoming increasingly necessary for addressing complex health questions. The federal government along with the scientific community must ac- knowledge the need for continued recruitment and take responsibility for developing new talent to ensure the future vitality of the health sciences enterprise. This committee acknowledges that funding for talent develop- ment may not be available from new congressional allocations for NIH and ADAMHA. The committee is also deeply concerned that any redistribution of existing funds from research project support will increase pressures on the funding picture. Nevertheless, the committee believes that the scientific community must show commitment to the long-term integrity of the overall system, even if that means short-tern sacrifices to research allocations in order to reinvigorate Gaining and replenish the scientific talent pool Support for Equipment and Facilities The committee concluded that inadequate or unsuitable space and ob- solete equipment have restricted the number and Apes of research projects that can be undertaken. Although the extent of the needs for construction, repair, and renovation of health sciences research facilities is difficult to determine, estimates run as high as $8 billion. The committee believes the long-term decline in federal programs for research facility construction and equipment renewal is partially responsible for deterioration of the nation's research laboratories. Consequently, the committee believes that these worsening conditions potentially could have adverse affects on research training and the productivity level of the nation's scientific work force. Federal support for research facilities has diminished drastically over the past two decades. Federal grant programs in the 1960s were very suc- cessful in expanding the nation's research capabilities, but several factors caused the NIH and National Science Foundation (NSF) facilities programs to be eliminated in the early 1970s. For instance, the increasing commit- ment of resources to the Vietnam War severely strained domestic programs, including facilities programs at academic institutions. Also, OMB (at that time known as the Bureau of the Budget) began pressuring NSF and NIH to justify continuing expansion of the nation's research facilities at a time when both college enrollments and the growth in federal R&D funds were
POLICY OVERVIEW AND RECOMMENDATIONS 171 leveling off. This forced a policy of supporting facilities solely through indirect cost recovery associated with research project grants. Except for some limited appropriations for AIDS research facilities, federal funds for health research facilities have been negligible over the past 10 years. This continued neglect comes at a time of escalating maintenance costs, increasing regulatory standards, increasing technological sophistica- tion, and a dramatic growth in scientific opportunities. Although some state governments and the private sector continue to make significant con- tributions to support the physical infrastructure for health research, they cannot be expected to meet the total demand. Clearly, there is a need to establish a national policy for renewal and expansion of the health sciences research infrastructure. The committee concluded that despite repeated shoddies calling for in- creased support for research facilities, no long-term federal strategy exists to restore the physical infrastructure for health research. There is no consensus within government or the research community on the need for expanded versus renovated facilities, the best mechanism for program support, or the respective roles of the interested sponsoring parties. Additionally, there is no mechanism to coordinate the various independent contributors support- ing facilities and equipment. Without a clear set of goals and a cohesive national policy, U.S. universities and research institutions will continue to decay and will be forced to seek short-term solutions to their facilities' needs by soliciting pork barreled appropriations from Congress. TOWARD A POSITIVE RESEARCH ENVIRONMENT The committee believes that the goals of health research can be achieved only by creating a positive research environment for health sci- ences. This environment should · identify and encourage young, talented individuals to pursue health research careers; · provide stable research support for talented scientists throughout their careers; · offer flexibility in allocating resources to foster creativity and meet changing demands; and · provide the modern laboratories and equipment necessary for sci- entific research and training. These characteristics, in turn, require effective coordination and leadership from the federal research agencies; competent, objective public and private sector administration; and responsiveness to the wishes of the American people through the political process. When the environment is positive, supportive, and reasonably pre- dictable, it nurtures innovative research. A congenial environment is
172 FUNDING HEAI7H SCIENCES RESEARCH equally important, for it encourages talented people to seek careers in health research while fostering the careers of established scientists. The continued vitality of health sciences research requires a system of stable support for scientists but with the flexibility to allocate resources in order to meet changing demands. The committee believes that while the scientific community must be held accountable for use of federal research funds, there has to be stable support and flexible policies to promote an optimal research environment. General Research Funding Guideline To place the existing research establishment into an economic perspec- tive, the committee analyzed each component in terms of capital investment relative to its productive life expectancy. The committee determined the following: (1) individual scientists are the most vital long-term investment in the research system; (2) capital investment in facilities is of a slightly shorter duration; and (3) individual research projects and the equipment used by researchers generally are the shortest and the most variable invest- ment relative to time. A certain degree of flexibility is necessary for supporting the compo- nents of the research enterprise. This fundamental principle implies that support for one component can be reduced for a brief period of time in order to provide funds to invigorate another component. The committee ascertained that those elements with the longest survival value (namely the research workforce and facilities) may be resilient enough to withstand temporary budget freezes or slight reductions in order to accommodate the immediate needs of components with shorter investment periods (re- search projects and equipment). Although short-term exigencies that favor support for one component over the others may be necessary for brief periods, continuance of such short-term policies will, in time, undermine the integrity of the entire system. In practice, emphasis on the short term needs of the research enterprise has led to underemphasis of funding for the training pipeline and facilities. Thecommitteeconcludedthattheg~idingprincipleformaintaininglon~ term balance within the system is a lequ ate support for each component. At any given level offederal finding, support for each component must be calculated relative to society 0cpectaiions of the entire research enterprise. Recommendation 1: The committee recommends that Congress, NIH and ADAMHA administrators, and scientists employ a priority-setting framework for allocating funds to meet long- and short-term research needs in order to correct and maintain the appropriate overall balance among the individual components of the research establishment (people, projects' and facilities).
POLICY OVERVIEW AND RECOMMENDATIONS 173 fib meet the health science research goals at any given level of overall support, relative amounts of support must be designated for the separate but interlocking components of the research system. The overriding objective of maintaining the integrity of a vigorous research system must be considered in all short-term decisions. Short- and long-term research goals must be defined not only by the amount of money allocated to perform specific research projects, but consideration also must be given to the number of researchers that will be necessary in the future, their equipment needs, and the adequacy of the facilities in which the research is to be performed. Thus, if the future requirements for investigators, facilities, and the amount of research can be estimated, the amount of funds necessary for each component to sustain a viable system can be calculated. Once projections of support needs for each component are determined, these estimates can be used to make a judgment as to what proportion of the total research budget should be designated for each component. Therefore, several interlocking levels of priority setting and decision making must be considered when allocating research funds. · The total appropriations to all federal agencies receiving funds for health sciences research, including NIH and ADAMHA; the allocations within each institute of NIH/ADAMHA for research and training needs; and · the allocations within specific research program areas (generally, disease areas such as leukemia within the National Cancer Institute [NCI] or acute myocardial infarction within the National Heart, Lung and Blood Institute [NHLBI]~. This can be interpreted to mean that each program division of each institute or agency will have specific programmatic needs to accomplish their mission at any given level of support. With established goals an estimated amount of investigators, research facilities, and research projects (with equipment as a proportion of project funds) will be required over a period of time. The desired balance among the components will differ depending on the area of research being supported. Considering that the research establishment is made up of a series of such long-term goals, it will be necessary to (1) replenish a certain percentage of talented investigators; (2) renovate or replace a certain percentage of buildings or renew equipment; and (3) support a certain level of research activity in order to preserve the integrity of the overall system and attain long-term goals. The committee did not attempt to focus on the substantive program- matic directions of the individual institutes nor their reasons for selecting, continuing, or redirecting their programmatic emphases. Each institute already has in place both a Board of Scientific Counselors that oversees the intramural program and a congressionally mandated National Advisory
174 FUNDING HEARTH SCIENCES RESEARCH Council that oversees the extramural program. Additionally, there are program advisory boards in many institutes that help define specific program objectives and research directions. In principle, the councils and associated advisory boards are responsible for extramural program planning and setting program priorities. Criteria for resource allocation at each of these levels should, however, be amenable to application of the principles and guidelines enunciated here. Each decision requires an estimate of the existing knowledge available to achieve the goal, the existing cadres of investigators needed to do so, and the availability of facilities and equipment to carry out these studies. Calculations must take into consideration the current capabilities of the research system (e.g., how much repair is necessary on the existing facilities, or, questions such as are there enough of the needed specialists to conduct the research). Next, projections must be made for annual targets toward achieving the determined goals within an established time period (for example, if 1,000 more investigators trained for a particular research field are needed by the year 2000, outlays for 100 more researchers annually for the next 10 years will be required). Once these objectives and programs have been funded, their implementation then devolves to the final two steps of the allocation process: (1) the allocation of awarded grant funds for a specific research project contributing to the goals of the research program and (2) the total allocation of funds to the specific universities, hospitals and research institutions that will assume fiscal responsibility for the funds, administer them, and provide the infrastructure for the research projects. The objective of this framework Is not to produce one overriding for- mula that can be applied across the spectrum. Rather, it is to allow for determining priorities among competing needs within different research areas. This framework seines as a guideline for determining particular research needs from the bottom up. The committee emphasizes the ~mpor- tance of designing a process that allows flexibility in order to meet a variepr of needs, both among various research programs as well as within specific research areas. The committee also emphasizes the need for continuous monitoring of talent development within research programs so that this information can be provided to coordinating advisory bodies such as the Federal Coordinating Council for Science, Engineering.. and Technology, (FCCSET) and a forum like the Government-University-Industry Research Roundtable (GUIRR) (see recommendations 6.1 and 6.2 below). Rebalancing of Health Sciences Research Funds The committee has conceded that allocation policies over the past two decades have forced an overall imbalance in the health sciences research sys- tem in which support for research project grants has been heavily favored at the expense of training and facilities. Reestablishing balance among research,
POLICY OVERVIEW AND RECOMMENDATIONS 175 training, and facilities is crucial in order for the United States to main- tain a vigorous research enterprise and sustain international preeminence in health research. In order to rebalance the system, the committee em- ployed the preceding framework to analyze the status of long-term capital investment among the components of the research base. In order to make up for past deficiencies in training allocations throughout the 1980s, and to meet higher personnel demands towards the end of the l990s, the committee feels that an accelerated growth of the training budget is necessary. The committee emphasizes that there is an integral relationship between research and training. Since an estimated one-quarter of NIH and ADAMHA support for research training is accom- plished indirectly through research project grants, allocation policy can not be separated easily into research and training components. However, for defining allocation policy, and in the absence of better data on research project grant funded training, these functions can be treated independently. The committee feels the research community must develop and implement corrective strategies now to avert a workforce crisis later in this decade. To address the funding imbalances, the committee developed allocation strategies under four budget scenarios for balanced funding through the l990s: (1) no real growth in the health sciences research budget (i.e., no growth beyond inflation), (2) 2 percent annual real growth, (3) 4 percent annual real growth, and, (4) possible allocation strategies for budgetary growth higher than 4 percent. Recommendation 2: The committee recommends that federal health research funds be reallocated over the next 10 years according to the suggested percentages in the growth scenarios outlined below. 1. No Real Growth: Even in the event of no average real growth in the health sciences research budget during the l990s, the committee recom- mends that funds for training the next generation of health scientists be increased incrementally from 4.20 to 5.75 percent of the total extramural research budget by 1995 and 6.75 percent by the year 2000. Concur- rently, the committee recommends that extramural construction funds be increased incrementally from the present 0.25 percent of the extramural budget to 0.50 percent by 1995 and that this level be maintained through the end of the decade. This redistribution of funds to training and facilities should come from increased congressional appropriations and not reduce the pool of funds for research (Figures 7-1 and 7-2 and Appendix Able A-22~. However, in real terms (dollars adjusted for inflation) there will be a slight reduc- tion of research funds under this proposal. The proposal calls for shifting 0.20 percent of the research budget (or about $12 million constant dollars per year) to the training budget each year for the next decade. With an
176 1 00% 80% 60% 40% 20% FUNDING HEALTH SCIENCES RESEARCH PERCENT OF BUDGET o% ~ 88 89 90 91 92 93 94 YEAR 95 96 97 98 99 2000 ~ Research ProJ Grts O _ Construction HI Contracts Centere/Other Grts ~ NRSA Training FIGURE 7-1 Percentage reallocation of the NIH extramural budget under both no real growth or two percent real growth scenarios. average cost per full-time training position (A lo P) equivalent of $24,000, this proposal would reallocate enough funds to increase ~ llPs by nearly 400 per year. The committee believes that this growth in the training budget will not enlarge the research project grant applicant pool; rather, the net effect of this gradual reallocation will be to replace the increasing number of scientists expected to retire later this decade. Furthermore, this recommendation parallels that recommended in the NRC report Biomedical and Behavioral Research Scieniists: Their Gaining and Supply.i The minor shift of funds for extramural construction will merely allow the NIH to meet the most urgent facilities crises. The committee cannot recommend shifting larger proportions of federal health sciences research funds into the construction category at a time when an increasing number of research grants are not funded fully. On the other hand, the complete absence of funds authorized for construction could jeopardize the building and renovation of facilities crucial to scientific progress. The committee recommends that a small percentage of funds be re- stored to the centers and other grants category over the next decade as well. The proportion of extramural funds committed to centers declined steadily throughout the 1980s, and continued decline in support could diminish the quality of the research conducted in these environments. It becomes all the more important to increase the support for centers that can serge as technology transfer sites for the translation of research results into clinical practice. Funds transferred to this category could be used for the growing
177 10 8 POLICY OVERVIEW AND RECOMMENDATIONS CONSTANT 1988 DOLLARS (Billions) o 88 89 90 91 ~ Res Proj Grts 92 93 94 95 YEAR ~ Ctra/Other Grts _ Construction ~ Contracts 96 97 98 99 2000 ~ NRSA Training FIGURE 7-2 Reallocation of the NIH extramural budget under no real growth scenario. number of interdisciplinary and multicenter disease prevention and epi- demiological studies. Also included in this budget category under other grants, are the funds for the Biomedical Research Support Grant (BRSG) program. Providing more funds through the BRSG program could enhance the abilities of research institutions to assist their young investigators at the local level and may help stabilize the research efforts of mid-career scientists if the traditional grant system becomes even more unpredictable (see recommendation 4.6~. Shifting funds away from research to training and facilities will have some negative ramifications. Over the next decade, the cost of these reallocations will be about $20 million (constant dollars) per year out of an annual $3.8 billion research project grant budget (1988 total). Since these funds would be reallocated from a varieW of research programs, the reductions in the traditional (R01) investigator-initiated research project grant pool would be minimized. 2. Two Percent Real Growth: In the event that the health sciences research budget grows, in real terms, an average of 2-percent annually, the committee again recommends that funds should be reallocated to training and facilities in the same proportions as in the zero growth scenario- training funds increased incrementally from 4.20 to 5.75 percent of the total extramural research budget by 1995 and 6.75 percent by 2000, and extramural construction funds should be increased incrementally from the
178 10 6 FUNDING HEALTH SCIENCES RESEARCH CONSTANT 1988 DOLLARS (Billions) o 88 89 90 91 92 ~ Res Proj Grts 93 94 95 96 97 98 99 2000 YEAR ~ Ctrs/Other Grts _ Construction ~ Contracts ~ NRSA Training FIGURE 7-3 Reallocation of the NIH extramural budget under a two percent real growth scenano. present 0.25 percent of the extramural budget to 0.50 percent by l99S and through the end of the decade. The real growth in the budget in concert with the reallocations will add more funds to training and facilities budgets without decreasing the research grant budget. Under this scenario, if the NIH budget grows by 2 percent annually in real terms (equivalent to the average annual real growth in the NIH budget throughout the 1980s), the committee feels that portions of the net increase also should be shifted to training and facilities (Figures 7-1 and 7-3 and Appendix Able A-23~. Since there has been no real growth in the training budget throughout the 1980s, the committee believes that these recommended percentages of reallocated funds will reestablish NIH's and ADAMHA's leadership in training. The net growth would allow for increasing the number of Pl lPs, but the committee feels that some of these augmented training allocations should be used to improve training programs and to address insufficient stipend levels (see recommendation 3 below). The percentage of the research budget allocated to facilities will not change from the zero-growth scenario since proportionately more funds will be available due to growth in the overall budget; and, in any case, the amounts needed to reach the estimated facilities construction requirement (see chapter 6) cannot be drawn from the existing sums. The committee emphasizes that these reallocations will preserve the same or higher level of research effort by not reducing the research portion
POLICY OVERVIEW AND RECOMMENDATIONS 179 of the budget in real terms. In fact, if the average size of research project grants remains constant ($184,000 in 1988) through the next decade, the total number of grants supported by NIH could potentially grow from the present level of 20,300 to nearly 24,000. Although the number of funded research grants will grow by about 360-370 per year over the decade, the success rate for applicants will remain relatively unchanged (presently about 24 percent) if the annual number of applications continues to exceed the present 19,500 level. 3. Four Percent Real Growth: In the event that the health sciences research budget grows an average of 4 percent annually, the committee recommends that funds for training be increased incrementally from 4.20 percent to approximately 5.4 percent of the total extramural research budget by 1995 and to 6.2 percent by 2000. Reallocation of funds for construction should follow the same pattern as the two previous scenar- ios: an incremental increase of construction funds to 0.50 percent of the extramural budget. The target percentages for funds to be reallocated to training under the 4 percent growth scenario are somewhat smaller than the figures in the 2 percent and zero growth scenarios (Figures 7-4 and 7-5 and Appendix Able A-24~. Although the overall percentage of the extramural budget committed to training is less under this scenario, the funding level actually would increase more rapidly because of the real growth in the overall budget. Obviously, faster growth of the training budget would eventually outpace the resources available to support the net increase in researchers. A 4 percent annual real growth in research funds would allow for a modest expansion of the research base over the next ten years. The net increase in available research funds would allow for the overall number of NIH research project grants to expand gradually, at a rate of about 1000 per year at 1988 grant sizes from the present 20,300 to about 29,400. In 1991 alone, this would raise the annual number of new and competing awards to approximately 6,000. However, with applications exceeding 19,500 and expected to go even higher, the annual success rate will only approach 28 to 30 percent. The committee believes that even at this pace of budget growth a large number high quality research proposals will go unfunded. 4. More Rapid Growth: The committee also considered the possibility that the NIH budget would grow at a more rapid pace, and what the longer term ramifications of such growth might be. The committee was convinced from the data and testimony it received that if all grant parameters (i.e., average grant size and duration, and the annual number of applications) were to remain constant the national health research effort could effectively utilize resources growing at a much higher rate. A larger research effort could build more effectively and rapidly upon the previous accomplishments
180 1 00 ear 80 40 20 PERCENT OFEXTR^~URALBUDGET of as as so g1 92 93 94 95 YE^R ~ Has Pro] Grts ~ Ctrs/Other Grts Construction ~ Contracts gG 97 98 gS 2000 ~ NOR Training MOUSE 7~ Pet ~ll~don ~ me NIH =1mmuml ~d~1 under a Cur at ~1 get ~a-. 10 8 6 4 2 CONST^NTlOBB DOLL^RS(ellIions) o ~ . as as go g1 92 93 94 95 YE^R Res Pro] Grts ~ Ctre/Other Grts Construction ~ Contracts oa 97 98 go 2000 ~ NRS^ Training nOU<E 7~ at ~ He NIH ~lmmum1 ~d~1 under a Fur petal ~1 s~nano.
POLICY OVERVIEW AND RECOMMENDATIONS 181 in health research and further broaden our knowledge of human biology and disease. For example, simply to regain the 35 percent grant success rate that existed between 1980 and 1987, would require funds for approximately 7,000 new and competing awards annually. Using the allocation proportions described above, would require an 8 per cent annual real growth. The overall allocation of funds among extramural research projects, training, and facilities will depend upon the particular needs of the scientists performing research within various scientific programs and disciplines, and the granting mechanisms deployed to meet the goals of these research programs. The committee's suggested allocations are directed towards the overall distribution of funds in order to strengthen the research enterprise by ensuring adequate, but balanced support to all components of the research enterprise. The committee has not specifically examined the proportion of funds expended on intramural research within any given NIH/ADAMHA institute. This issue has been examined recently by another IOM study group. Growth in the intramural programs is guided by program objectives and advisory councils' oversight, and is constrained by space limitations and employment ceilings. Within these guidelines, the committee emphasizes that any funds to be redistributed should be drawn first from increases in the annual federal appropriations. However, even in the event of no real growth in the federal health research budget, the committee firmly endorses that incremental increases in training funds be reallocated from the nominal increases in the overall extramural budget (funds not adjusted for inflation). Under circumstances of real growth, the proposed training increases should come from the new funds so as to detract minimally from the ongoing research effort. Furthermore, the committee emphasizes the importance of making gradual reallocations in order to maintain research support stability. The committee is aware that this proposal may not be received favor- ably by the scientific community at a time when research grants are not funded fully and research careers appear to be in jeopardy. Although the committee recognizes that these short-term problems abound, it is making these recommendations for the long-term integrity of the research enter- prise. The earlier Institute of Medicine (IOM) report, Resources for Clinical , lnvesugai'on,~ has recommended that l,uUU clinical investigation training positions be made available. Additionally, the next biomedical and behav- ioral manpower report by the NBC scheduled to be released in 1992 is expected to review closely the need for increasing the number of physician- scientists as well as the scientific doctoral pool. If the federal research budget grows in real terms and if continued monitoring by the National Research Council (NRC) Committee on Biomedical and Behavioral Work- force Needs demonstrates an increasing demand for physician-scientists,
182 FUNDING HEALTH SCIENCES RESEARCH the proposed shift of funds to the training budget would make resources available for implementing these changes. Additionally, adjustments to the research granting system presented below are designed to stabilize re- search careers through additional steps and to ensure a vigorous, albeit constrained, health research establishment. TALENT RENEWAL It has been argued that there are already too many grant applicants not receiving federal funds and that training more will only exacerbate this problem. However, the committee believes that the United States must take a prospective view of the anticipated scientific work force demands for the next 20 to 30 years. There are strong indications that the failure to recruit new people into the health sciences and to compete with other more secure or appealing career lines will significantly hamper the United States' ability to confront future health research challenges. Furthermore, given the likelihood that retirements and other sources of attrition are on the rise, it becomes all the more imperative to address training needs immediately. Recommendation 3: The committee recommends an approach to re- store balance in the development of scientific talent through a broad spectrum of incentives and encouragements. The committee stresses that strategies must be developed to 1. recruit undergraduate students into the health sciences; 2. increase the interest of unde'Tepresented groups, including women and minorities, in careers in health science research; dents; 3. enhance the continuum of support mechanisms for graduate stu 4. improve the training of physician-scientists; and 5. devote additional attention to the needs of younger scientists. Increasing the Attraction of Health Science Research Careers for Undergraduates Recommendation 3.1: The committee recommends that programs be supported by the National Science Foundation (NSF) and other federal agencies, along with the private sector, to introduce undergraduates to career opportunities in health sciences research. The challenge of preparing and motivating students to pursue health science careers begins at the level of primary education. According to several recent studies, there has been a decline in the mathematics and science competencies of U.S. students from kindergarten through twelfth
POLICY OVERVIEW AND RECOMMENDATIONS 183 grade. Although the present study was not intended to address K through 12 education, the committee feels it is important to recognize that the development of future scientists begins in elementary school not at the time of college enrollment. Major losses to the science and engineering talent pool occur dur- ing the undergraduate years when career decisions usually are made. Throughout the l950s and 1960s, the primary source of federal support for undergraduate science education was the Science and Engineering Ed- ucation (SEE) Directorate of the NSF. In the early 1960s the SEE budget swelled to 46 percent of the entire NSF budget. However, by 1983 only 1.5 percent of the total NSF budget was committed to this directorate. Recent efforts to reemphasize the importance of federal support for science and mathematics education has resulted in the budget for SEE growing from $55 million in 1987 to a proposed $251 million in the 1991 budget nearly 10 percent of the NSF Budget. Students interested in the health sciences need to be introduced to research opportunities that will encourage them to continue these studies in graduate or medical school. While undergrad- uate science education falls within the purview of the NSF, the committee believes that research experiences for undergraduates pursuing careers in the health sciences cannot be addressed sufficiently by that agency alone. The committee concluded that programs sponsored by the federal government, foundations, and voluntary health agencies are necessary to encourage and support students pursuing careers in health research. These programs should include research experience in association with faculty members who can serge as role models and mentors. Several pro- grams, including NSFs Research Experience for Undergraduates (REU) and Research at Undergraduate Institutions (RUI) programs, Congress's National Scholars Program, and the Howard Hughes Medical Institute's donations to liberal arts colleges, have been developed and provide frame- works that could be expanded to improve the flow of undergraduates into the health sciences. These types of programs provide incentives for inves- tigators to become more involved in undergraduate research training and mentoring, and they offer stimuli for students to pursue careers in scientific research. Considerable discussion within the committee focused on establishing programs through NIH and ADAMHA to provide supplemental research grant monies to support the involvement of undergraduate students in research project grants similar to NSFs REU and RUI programs. Such programs would provide students with research experience while they are making career choices. This should increase the chances that many of the students involved in this type of research will go on to choose health research careers. The committee recognizes the merits of these types of undergraduate science programs, which have traditionally been within
184 FUNDING HEALTH SCIENCES RESEARCH the purview of the NSF and the Department of Education. Whereas a few members of the committee felt strongly that NIH and ADAMHA should invest more resources at the undergraduate stage of the pipeline, the majority of the committee concluded that in a no growth budget reallocation of existing funds for undergraduate programs could not be justified. However, in the event of real growth in the federal research budget, various undergraduate health research training models will merit additional attention. Enhancing the Attractiveness of Health Science Careers for Women and Minorities Recommendation 3.2: The committee recommends that programs be developed by the federal government and the private sector that are designed to encourage more women and minorities to pursue careers in the health sciences. Of particular concern with regard to undergraduate science enrollment is the underparticipation of women and minority students. Although these students may be sufficiently prepared for a science and engineering edu- cation when they begin their college studies, they choose these avenues of study in significantly lower proportions than similarly prepared white males. Undergraduate programs specifically designed to encourage women and minority students to pursue their scientific career aspirations could reverse this trend and significantly increase their numbers. Similar to the health sciences research opportunities program described in the above recommendation, a key element to the success of these pro- grams is undergraduate exposure to research under the guidance of faculty members or through collaboration with a graduate program or medical school. Currently, the federal government supports programs to increase participation of minority undergraduates in health sciences through awards to specific institutions. The NIH Minority Access to Research Careers (MARC) program for undergraduate experiences in research provides a model for introducing these students to careers in the health sciences. The committee believes that the current system neglects the diversity of individual needs of students in these underrepresented groups. The design of successful programs to address this issue should include support that is available to any qualified student, regardless of his or her choice of school. Support should be focused on the abilities of the individual and the research and training programs of the host institution including those institutions serving predominantly minority students. Thus, program expansion should emphasize fellowships and faculty involvement within every college and university. Programs that meet the needs of the individual students should
POLICY OVERVIEW AND RECOMMENDATIONS 185 help increase the number of students from underrepresented groups who go on to graduate studies in the health sciences. Reestablishing Competitive Predoctoral Support Recommendation 3.3: The committee recommends that NIH and ADAMHA reestablish a competitive predoctoral fellowship program for individuals. The committee believes that there is an imbalance in NIH and ADAMHA support for graduate studies in the health sciences. The current system heavily favors institutional training grants over individual fellowships. The committee believes, however, that a combination of mechanisms to support predoctoral students throughout their studies is important. In the 1960s the portion of the NIH training portfolio devoted to predoctoral fellowships accounted for more than ~ percent of the training budget. The National Research Service Act in 1974 almost eliminated this type of training support, shifting the emphasis to predoctoral training grants and away from fellowships awarded to individual students attend- ing institutions of their own selection. Currently, predoctoral fellowships account for less than 2 percent of all NIH training money (about 200 positions out of approximately 11,000 full-time equivalents). One variable in the committee's analysis is the extent of pre- and postdoctoral training support provided through research project grants. One estimate shows that the number of trainees supported on research project grants has nearly doubled between 1979 and 1987 from 2673 to 4426. Without accurate data, which is not collected by the NIH, the committee's recommendations can address only the mix of support within the training and fellowship programs. Most of the committee agrees that there would be a number of ad- vantages to reinstituting a predoctoral fellowship program. Students would be supported directly, allowing them more freedom to select the area of investigation they wish to pursue. Students would apply for these graduate fellowships in a national competitive process, similar to procedures used for NIH postdoctoral fellowships. Thus, student support would not be connected directly with renewal of investigator research project support. Most importantly, direct fellowship awards to students would provide a strong signal that the student is an integral and valued member of the health sciences research enterprise, which would enable more aggressive recruitment of students into postbaccalaureate education and training in the health sciences. The committee emphasizes that the underparticipation of minorities and women also must be addressed within this program as well.
186 FUNDING HEALTH SCIENCES RESEARCH Improving the lining of Physicians for Research Careers Recommendation 3.4: The committee recommends that the number of physician investigators~ctive and in training be assessed. Assuming a real decline in the number of physician-scientists, the committee further recommends reallocating resources in order to create a more formal sys- tem for training physician-scientists, including cumculum requirements. In addition, experimental federally funded training programs in clinical research and public health research also should be established. The committee is concerned about the apparent inability to recruit sufficient numbers of physicians into scientific careers, especially clinical investigator careers. The physician-scientist is the critical link between the knowledge uncovered in the laboratory and the translation of that knowledge into clinical practice. However, various indicators, such as the proportion of NIH grants awarded to physician investigators and the numbers of physicians reporting research activities, have shown a steady decline in physician-scientist numbers since the early 1970s. The committee believes it is essential to have physician-scientists en- gaged in both basic and clinical research. Recruiting physicians into re- search careers is hampered severely by the length of time necessary for clinical training as well as the difficulty of conducting research during this training period. Additionally, the current unfocused structure of many physician research training experiences does not introduce trainees suffl- ciently to scientific project design, research methodology, and statistical analysis. Particularly troublesome is the apparent decline in the number of physicians pursuing clinical research. A recent IOM study committee closely examined the resources available for clinical investigation and the complex issues involved in attracting physicians into clinical research. The study group concluded that the data were inadequate regarding the nature of clinical research and level of support in this country. However, the study group was able to identify several barriers to individuals pursuing careers in clinical investigation. The committee believes there is a "triple threat" to academic physicians in the posttraining years: they are expected to be exceptional researchers, exceptional clinicians, and exceptional teachers and mentors. These pres- sures probably have discouraged many physicians from remaining actively engaged in research and will have to be alleviated in order to interest more physicians in research careers. The committee believes that formal training for physician-scientists should include experience in scientific research protocol design, research
POLICY OVERSEW ED COMMENDATIONS 187 methodology, epidemiology, and statistical analysis. A 1- or 2-year re- search experience, particularly when poorly focused, often leaves physician- scientists less competitive than Ph.D. scientists on peer-reviewed grant ap- plications. Therefore, a national program for training physician-scientists should encourage more physicians to pursue research careers, and improve their success in the competitive grant system. The committee supports the recommendation of the recent IOM re- port Resources for Clinical Investigation that federal funds be allocated for creating a national program for training clinical researchers. In addition, the committee believes this should include the aforementioned curriculum requirements and require matching funds from medical schools. This pro- gram should be assessed and analyzed periodically for its effectiveness in recruiting and retaining more M.D. scientists in clinical investigation. Possibly another way to enhance the training of physician-scientists is by expanding NIH's Medical Scientist Training Program (MSTP). This program facilitates obtaining both an M.D. and a Ph.D. degree in a health- related science. To encourage adequate training of physician-scientists, the committee encourages NIH to expand the MSTP program. The committee also supports ADAMHA in its efforts to initiate a similar program. The committee believes that the MSTP program provides valuable training experiences for physician-scientist trainees at a crucial period in their scientific career development. In addition, M.D./Ph.D. training support should be available through individual grants as well as through institutional training grants, analogous to doctoral training programs. Improving the Opportunities for Young Investigators Recommendation 3.S: The committee recommends that NIH and AI)AMHA modify their FIRST award programs to incorporate a formalized assessment of progress by a scientific panel in the third year. Once talented individuals have spent many years successfully training for a research career, they should have confidence they will be able to gain entry into the research system. Frequent proposal writing to obtain small project grants can prevent young investigators from developing a solid, long-term line of investigation. Also, the committee believes that the transitional period between postdoctoral training and established scientist is the most crucial in getting young, creative minds into productive research careers. As competition in the traditional research project grant system in- creased through the late 1970s and early 198Os, many in the scientific community felt that grant applications from young investigators were at a competitive disadvantage in peer review. Since many of these individuals had not established their own lines of investigation and were inexperienced
188 FUNDING HEALTH SCIENCES RESEARCH at writing concise and lucid grant applications, their applications may not have been received as favorably as those from senior scientists. 1b provide these young applicants access to the federal research grant system, NIH, and later ADAMHA, initiated a program aimed specifically at the posttrainee/prescientist career stage called the First Independent Research Support and Transition (FIRST award. The FIRST award pro- gram guarantees awardees 5 years of support and a maximum of $350,000. These longer award periods for young investigators provide stability and the additional time often needed to set up new laboratories. However, no funds are set aside regularly for these awards, which means there is no as- surance that FIRST grant applications will be awarded when the researcher is ready to apply for one. In effect, these young investigators continue to compete head on in peer review with other types of grant applications, namely traditional R01 investigator-initiated grant applications. The committee feels strongly that these awards are moving in the right programmatic direction for providing entry into the competitive traditional grant system (R01) for young scientists. However, modifications of the FIRST program may be warranted in order to provide more guidance throughout the award period as well as to establish greater similarities to the R01 grant program. Considering the nature of the FIRST award, the committee does not feel that the progress of these awardees should or could be comparable to that required in the traditional R01 system. This rereview should not be construed as a contingency for further funding. Rather, it should ensure that FIRST investigators are being indoctrinated properly into independent scientific investigation and prepared to compete for Rots. Furthermore, this should provide an opportunity to redirect the young investigator (if necessary) and ensure that the product of this research will, in fact, enhance the body of medical knowledge. A similar funding mechanism comparable to the FIRST award could be envisioned for midcareer scientists seeking additional training in a new field. IMPROVING THE RESEARCH PROJECT GRANTS SYSTEM 1b carry out this program of talent renewal without compromising the research base, the research project system needs, in the committee's view, some adjustments in order to preserve the existing pool of talented scientists as well. Policies that affect research project support should provide flexibility for responding rapidly to changing research needs but also should provide stable support for productive research scientists.
POLICY OVERVIEW AND RECOMMENDATIONS Modifying Research Project Grants 189 As the U.S. health research enterprise has matured, there has been increasingly more competition among the investigators applying for grant funding. The expanding realm of research opportunities, along with the increasingly sophisticated nature of research, has outstripped the effective- ness of the existing peer review process for determining priorities among the cluster of excellent grant applications. Whereas the original design of the peer review system operated well prior to the mushrooming of research opportunities in the 1970s, it is no longer reasonable to believe that mi- nor differences in priority scores are accurate measures for all or nothing funding. These inconsistencies have been highlighted by the differences in the spread of priority scores among NIH study sections. Consequently, NIH has attempted to address these variations by establishing pay lines according to percentile rankings as opposed to priority scores. Nevertheless, even with these modifications, the funding process still can deny funding for novel or otherwise excellent research applications that happen to fall just below the percentile funding point. The committee feels that these harsh cut-off points are allowing excellent research ideas to go unfunded and that they are potentially demoralizing established researchers whose applications may fall right at the margin for funding. Even after a research grant has been awarded, the amount of the final funding is subject to a completely arbitrary reduction in the process of downward negotiation, as described previously. This is the system NIH and ADAMHA have adopted to reconcile the funding of research grants to the amount of money received in annual appropriations. As the continuing obligations for NIH and ADAMHA increased by the series of policy changes during the 1980s, the percentage of downward negotiation ballooned from 1 to 2 percent to nearly 12 percent in many NIH/ADAMHA institutes. Step-Down or Rollover Funding Recommendation 4.1: The committee recommends that NIH and AI)AMHA, as well as other sponsors of research, develop pilot programs to evaluate step-down or rollover funding for selected grant awards. When a renewal application for an NIH or ADAMHA grant falls just below the cut-off point for funding, a mechanism should be available to allow these meritorious projects to continue for an additional year, possibly with a reduced budget. This would permit resubmitting a clarified proposal while not dismantling the laboratory and losing key personnel. Such a rollover mechanism need not be implemented automatically, and the peer review system could decide on whether research programs were no longer productive and should be stopped.
190 FUNDING HEALTH SCIENCES RESEARCH NIH had "phase-out" procedures in the past that allowed investigators to receive funding during an interim year while an amended competing re- newal was being rereviewed. Reestablishing transitional-year awards would reduce the threat that research teams and laboratories will be dismantled completely if support temporarily lapses because of an unsuccessful com- petitive renewal. A pilot program could evaluate the utility and risks of a transitional funding period during grant renewal. Into possibilities for implementing this concept are 1. Rollover funding: This first transitional scenario would apply to research project grants awarded for periods of 5 or more years. An NIH/ADAMHA review of competing renewal applications would be con- vened two years before grant termination (e.g., in year 4 of a 5-year grant) and would lead to one of two possible outcomes: · An accepted application would allow the research project to con- tinue for an additional 5 years. Thus, the renewal award would provide funding for the fifth year plus an additional 4 years, extending the project to 9 years. · An unsuccessful competing renewal in year 4 would require that the investigator submit an amended competing renewal application in year 5. If the amended application is approved, funding would be continued for years 6 through 10. 2. Step-down funding: Another possible mechanism would be to extend funding for an additional year for those renewal applications that fail to merit adequately high percentile rankings and for which revised renewal applications would be invited by the review committee. In such cases the extension year would be funded at a fixed level, such as 60 percent of the last fully funded award period. This type of program would allow investigators to retain key research staff while a revised grant application was being considered. These are examples of mechanisms that would allow investigators to par- ticipate in two consecutive review cycles prior to losing funding. A Sliding Scale of Support for Approved Awards Recommendation 4.2: The committee recommends that NIH and ADAMHA consider modifying the traditional investigator-initiated grant system (ROT) to fund grants on a sliding scale based on percentile ranking. The compression of grant applications receiving high- priority scores and the necessity of determining a single pay line for funding does not necessarily take into consideration the benefits that could be derived from those grant applications at the margins. Furthermore, the committee heard
POLICY OVERVIEW AND RECOMMENDAT ONS 191 concerns that novel research applications may fall frequently at the margins of the pay line, potentially denying breakthroughs in the medical sciences. The committee believes that the scientific community has to cast a wider net in order to capitalize on excellent opportunities that may fall below the funding cut-off point. In an environment of extreme fiscal constraint and across-the-board downward negotiation, expanding the research base seems unlikely. A sliding-scale funding mechanism could reinforce and protect the best research projects and reduce the pain inflicted throughout the system from downward negotiation. It also would increase the opportunity to sponsor high quality research proposals that are increasingly falling just below an arbitrarily established pay line. In effect, the adminzs~ative cut would be presented to the scientist at the time the grant Is awarded not episodical) and unpredictably through out the award period. One suggested plan would scale down the award duration or funding level based on some criteria such as the percentile ranking. For example, funding, · those applications in the top decile would receive full funding, those applications in the second decile would receive 90 percent · those application in the third decile would receive 80 percent funding, and · those applications in the fourth docile would receive 70 percent funding. A proposal such as this would not be warranted in an environment where 50 percent of grant applications were funded. However, with award rates at or below 25 percent in most institutes, the committee believes that few options exist to expand the research base. For illustrative purposes, if the R01 research budget were $2.0 billion and the average grant size and length was $175,000 for four years, the total number of R01 grants that could be supported in any one year would be 11,428. The turnover per year would be 2,857 assuming 100 percent funding. If funding were 90 percent, which is in line with the current downward negotiations, the total would grow to 12,698 with an average of 3,175 turning over each year or a gain in projects receiving support of slightly more than 300 per year. By using the proposed sliding scale in this example, about $100 million would be available for funding about 3,850 grants per year or an expansion to about 15,360 total. For scientists, the security of knowing how much funding actually will be received is far superior to a progressive downward negotiation that slowly compromises all research endeavors and complicates administrative planning by research institutions. This proposal would encourage investiga- tors to set priorities in their own programs according to their funding level,
192 FUNDING HEALTH SCIENCES RESEARCH since those with lower percentages of funding would have to choose which aspects of their research to pursue. This would preserve scientific talent by not forcing investigators out of the system in the case of a fund/no fund decision. Furthermore, this strategy might also increase the opportunity for young investigators with novel ideas to gain initial access to the grant system despite inexperience in grant writing. A Dedicated Mechanism Specifically for Novel Research Proposals Recommendation 4.3: The committee recommends that NIH and ADAMHA consider revamping the Small Grants program (R03) for funding innovative, high-risk ideas. The committee is concerned that the peer review system may not effectively identify novel ideas that have the potential for making significant breakthroughs in medical knowledge. Because the system is geared toward building on the accepted body of current knowledge, grant proposals that seek to explore tangential or contradictory theories may not fare well in the priority rankings. As funds have become more constrained, the committee believes that study sections and institutes have become even more disinclined to fund high-risk research proposals. The committee suggests that NIH and ADAMHA adopt the model of NSFs pilot program called Expedited Awards for Novel Research. This program, begun in the engineering directorate in 1986, was expanded in 1989 to include modified peer review. Awards of up to $50,000 are available to principal investigators with especially innovative ideas. The committee emphasizes that this system should not be viewed as an alternative to the peer review system. Rather, it should be used as an opportunity to support exciting but high-risk research that would otherwise go unfunded. Changes in Research Management NIH Director's and ADAMHA Administrator's Emergency Funds Recommendation 4.4: The committee endorses the recommendation by the IOM group studying the NIH Intramural Research Program that Congress annually appropriate to the director of NIH a discretionary fund of no less than $25 million. A discretionary fund also should be apprm printed for the ADAMHA administrator. (The committee acknowledges that the proposal for an NIH director's fund has been submitted in the President's 1991 budget.) The committee concluded that the dynamic nature of the health re- search environment frequently requires that monies be available to address emerging problems and/or research needs. The committee found that the
POLICY OVERVIEW AND RECOMMENDATIONS 193 directors of the various institutes at NIH and ADAMHA are in a unique position to determine specific areas that require urgent attention and that cannot necessarily wait until the next congressional appropriations cycle. This policy change would strengthen the leadership of NIH and ADAMHA by allowing the directors to address emerging issues and special interinstitute research opportunities. This approach also would improve flexibility and provide the directors with the resources to initiate intramural activities across institute lines, without intruding on the independence of the individual institutes. This proposal has surfaced many times in recent years but has never been approved by OMB. However, the President's bud- get for fiscal year 1991 includes a $20 million fund for the NIH director. There also is a provision to allow the director to reprogram up to 1 percent of the NIH budget without congressional approval. Multiple Grant Awards and Grant Sue The committee had lengthy discussions about principal investigators having multiple grant awards. The committee felt that, in many instances, investigators may need more than one grant for their research programs, but the committee also was concerned 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. However, after considerable discussion, the majority of the committee concluded that the system should not impose arbitrary limits on the number of grants per investigator for fear of denying potentially exciting research. Likewise, artificial ceilings on the size of grants may adversely affect the quality of research by requiring large research projects to be broken down into subparts. Regulation of either or both of these issues has the potential to stifle creativity and inhibit scientific advancement. However, in an era haunted by concerns for domestic spending constraint, it is incumbent on individual scientists and the peer review system to support the best workers to perform the best science in the most productive environment. Implicit in these evaluations are the amounts of significant effort an investigator has available to devote to the needs of a given project and the staffing required. Clearly, if the scientific community appears to be taking unfair advantage of these loosely regulated areas, congressional or administrative controls are inevitable, as evidenced by the recently implemented salary cap on NIH/ADAMHA sponsored Investigators. Federal Demonstration Project Recommendation 4.5: The committee recommends that the Federal Demonstration Project be expanded as additional experience becomes avail- able.
194 FUNDING HEALTH SCIENCES RESEARCH The Florida Demonstration Project (FDP) was intended to reduce the administrative burden on grantees by streamlining procedures and reducing costs in the sponsored project system. The primary objectives of the project were to · standardize postaward administration of federal research grants among the federal agencies to the extent possible; eliminate most federal prior approvals for budget reallocation; simplify research project management procedures; and · allow an investigator's collective research program to be treated as one administrative and accounting unit rather than each project as a separate unit. Initial reactions to the Florida Demonstration Project have been gen- erally quite favorable. As of October 1988, the project was redesignated as the Federal Demonstration Project and was expanded to include 26 institutions. This creative approach is likely to continue to be extremely valuable by allowing scientists to concentrate more on their research than on administrative details. Research Institutions Research institutions have always shared the support of health sciences researchers with the federal government and other research sponsors. The recommendations of Vannevar Bush emphasizing federally funded research within academic settings was based, in part, on the preexistence of academic laboratories, research career paths, and the close linkage of research and training. As federally sponsored research programs have expanded, the com- mittee believes some institutions have exploited federal resources as a means to enlarge their faculties by creating positions that rely entirely upon "soft" finding through research project grants. Many research institutions have been resourceful in finding additional institutional monies or philanthropic support for investigator salaries or research support. Despite these efforts, the committee ascertained that many universities and research institutes have been unable to secure ade- quate flexible resources in order to create stability for their faculties. This situation increases the pressure on scientists to obtain federal support as a foundation for a career in research. With the growth of health sciences research funds slowing, the com- mittee believes that universities and research institutes should strive to strengthen their commitments to career scientists. ~ recruit, retain, and augment their research faculties, academic research institutions may need to allocate more internal "hard" funds for the career development needs of young research faculty as well as for those scientists in midcareer. In
POLICY OVERVIEW AND RECOMMENDATIONS 195 addition, funds also need to be made available for retraining of older inves- tigators and to offset temporary lapses in external research grant support for established scientists. The committee recognizes that research institutions and universities may find it very difficult to decide how to trade off faculty development for health scientists against other competing institutional needs. However, the committee also believes that a clarification of the institution's objectives would be constructive for the future vitalibr of these organizations and the realistic appraisals by their faculty of their future opportunities. Biomedical Research Support Grant Recommendation 4.6: The committee recommends that NIH continue to fund the BRSG program to universities and research institutions in order to continue flexible program development under institutional control. Furthermore, the committee suggests that the universities and research institutions disburse BRSG funds through faculty peer review groups to support new research initiatives, especially those of young investigators. The ability of universities and research institutions to reward young talent and preserve ongoing projects increases the sense of career security among researchers. The committee believes that the Biomedical Research Support Grant (BRSG) provides flexibility to university faculty and admin- istrators to support new and ongoing initiatives within their own institutions. The size of these awards is related directly to the amount of project funds received from NIH. These funds are disbursed through various mecha- nisms at the institutional level. In many cases faculty peer review groups decide the utilization of these funds. The committee believes that the BRSG program has played a significant role in funding young scientists and other institutional initiatives crucial to their overall research and training programs. The committee recommends that funding for the BRSG program should be maintained to allow universities to make decisions regarding their own faculty research needs and that creation of other block grants for developing and preserving scientific talent should be examined as well. However, funding for BRSG and similar grant programs at NIH and ADAMHA has been a continual target in budget cuts. Early in the 1980s the BRSG program was slated for total elimination by OMB. Between fiscal years 1989 and 1990, the BRSG program suffered a cut of $11 million, falling from $55.2 to $44.4 million, and the proposed 1991 budget intends to reduce this program further to $17 million. The committee feels that this small commitment to flexibility and researcher security is crucial for promoting stability in the careers of health scientists.
196 FUNDING HELLISH SCIENCES RESEARCH RESTORING THE PHYSICAL INFRASTRUCTURE Increasing regulatory standards are putting added stresses on the ability of institutions to improve facilities and upgrade equipment. Sophisticated equipment for protecting the health of laboratory workers using deadly pathogens and hazardous chemicals sometimes requires large budgetary outlays. As federal regulations for animal care and facilities requirements increase, research institutions must invest heavily to comply. Estimates of the average costs to meet new federal animal regulations run as high as $40,000 per grant. Inadequate facilities and equipment will have to be corrected gradu- ally, for commitment of a substantial portion of existing federal funding to facilities at this time would create another imbalance in the support mechanism for people and projects. The most direct approach to the in- frastructure crisis is increased federal funding for health sciences research facilities and equipment. Many believe that renewed federal support for construction and renovation is necessary and that such a program would help stem the flow of direct appeals by individual institutions to Congress for pork barrel appropriations for specific facility development. Many cre- ative solutions will be required to fill the enormous need to modernize the physical research infrastructure. Direct Grant Program Recommendation 5.1: The committee recommends that Congress au- thorize and appropriate funds for a competitive matching fund construction program to renovate or construct health sciences research facilities, bear- ing in mind the increased costs of updating facilities to meet recently enacted regulations. Federal construction programs should focus on renovating existing space as well as funding new construction. Initially, a program could be established without additional appropriations by creating a scientific con- struction authority and appropriating a portion of the nearly $300 million funnelled by Congress to certain institutions through ad hoc pork barrel amendments. These monies would be subject to a comprehensive merit review, taking into consideration both scientific criteria and appropriate so- cioeconomic and political criteria. The committee feels strongly that pork barreling does not serve the best interests of the nation in the long run and thus should be avoided. It is unlikely that any new program will be funded at the same level as the Health Facilities Construction Authority was during the 1960s. The proposed program should allocate matching funds to act as an inducement for attracting private and corporate monies as well as state appropriations.
POLICY OVERVIEW AND RECOMMENDATIONS 197 After reviewing past policy, the committee felt that the federal contribution could be highly leveraged by requiring matching from private and state sources on the order of 1:4 (federal:other). Matching could be done on a sliding scale based on the economic need of the institution as determined by a comprehensive merit review, but it should not exceed a 1:1 ratio. Facility needs do not necessarily conform to the categorical divisions of the individual institutes at NIH or ADAMHA With limited resources available, coordination of the federal biomedical facility renewal efforts can be accomplished only by the directors of NIH and ADAMHA Compre- hensive review should include input from the institute directors, the White House Office of Science and Technology Poligy, Congress, and the scientific community. Such a process should prevent unfair political competition for resources and stem appeals for pork barrel funds. Indirect Cost Adjustments Recommendation 5.2: To allow greater flexibility for institutions to address their own facilities needs, the committee recommends that the sponsors of health research modify indirect cost (IDC) calculations in the following ways: 1. The federal government should change federal grant accounting procedures to allow negotiation of separate line items in the IDC recov- ery rate for facilities renovation and construction separate from that of administrative and library costs. 2. The federal government should increase IDC use allowance to reduce amortization periods for buildings and equipment. 3. Private foundations, voluntary health organizations, and corpm rations should observe more closely the true costs of the research they sponsor, including the IDC portion. There are also indirect means by which the federal government and other sponsors of health research can renew the health research infrastruc- ture. Within the federal system the IDC recovery rates for health research conducted at universities and research institutions are negotiated on an individual basis with federal agencies. The allowable depreciation costs for facilities and equipment do not accurately reflect replacement costs. The current version of OMB Circulars A-21 and A-110 allow for building amortization over 50 years (or 2 percent per year) and equipment amor- tization for 15 years or 6.67 percent per year. From the data available, the committee concluded that this portion of IDC recovery allowances is unrealistic and inadequate. Another facet of this complex problem is the underpayment of indirect costs by foundations and corporations. By placing caps on the amounts
198 FUNDING HEALTH SCIENCES RESEARCH of indirect costs allowable on sponsored research, these sponsors com- pound IDC recovery problems for colleges and universities. As a result, institutional funds are consumed to support the indirect costs associated with these projects. This constraint may force some institutions to refuse support from these sources if the indirect costs cannot be recouped in a fair and equitable manner. Many nonprofit sponsors are very concerned about the high IDC rates at the top private research institutions. Paying these high IDC rates easily could consume much of these sponsors' resources available for the direct costs of performing research. However, if universities are forced to transfer indirect costs into direct cost categories, the required funding will remain the same. Most research buildings become obsolete for conducting sophisticated research in 20 years, and equipment is often obsolete between 4 to 7 years after purchase. The committee feels that sponsors of health research should link support for particular facilities with individual research projects to allow faster recovery of institutional funds used to maintain facilities and to repay loans used for construction or renovation. In order to accomplish this, research institutions need to have options available to recoup previous expenditures for renewing their research physical plant. This could be done by changing the annual IDC allowance for building amortization from the present 2 percent to 5 percent and by raising the allowance for equipment amortization from 6.67 to 20 percent. This would allow research institutions to depreciate their buildings over 20 years rather than 50 and to depreciate equipment in 5 years rather than 15. The committee emphasizes that this policy change must not reduce the pool of funds available for direct costs and strongly urges universities and other research organizations to keep down the administrative portions of overhead. This seems impossible in light of the increasing federal bureau- cratic regulations, but failing to keep these costs in check will inevitably lead to IDC caps and subsequent loss of institutional control over these fi- nances. However, this policy change could allow research institutions more flexibility in setting their own priorities within their budgets for IDC recov- ery. The committee also emphasizes that some assurances must be made on the part of the grantee institutions that these funds be sequestered and utilized only for facilities and equipment renewal and not for administra- tive overhead. Inaction now will only exacerbate the growing infrastructure problems at colleges and universities.
POLICY OVERVIEW AND RECOMMENDATIONS Creative Financing 199 Recommendation 5.3: The committee recommends that rules be ad- justed so that indirect costs can be applied to direct rental costs of leased facilities. Alternatives to the traditional forms of capital formation are beginning to reshape the way academia raises money for capital improvements. State and local governments are investing in academic facilities for education and garnering possible economic advantages by providing a sound scientific base in their locales. Partnerships with industry (although limited) are pro- viding an alternative method for capital formation. Patenting and licensing intellectual property also are bringing financial returns that can be invested in facilities at some institutions as well. The committee heard suggestions that institutions should attempt to offset some of the high costs of research facilities and equipment by entering into cooperative agreements to share resources. Some research institutions recently have developed innovative approaches to develop research facilities by creating long-term arrangements with private developers. By combining off campus IDC rates with direct rental payments, research institutions can enter into such lease arrangements. It is conceivable that in such cases rental payment may provide a means of eventually purchasing the proper by the research institution. In some cases research institutions may wish to lease land to a devel- oper who will construct a research facility. The developer may, in turn, lease the space in the research building back to the research institution. In such cases maximum flexibility should be provided so that the building can be leased or purchased through direct or indirect costs associated with research conducted in the facility. Developer interest in these types of projects may be predicated upon tax accounting rules, which may require some accommodation with regard to how rental or overhead funding is provided. ESTABLISHING AN ONGOING PROCESS FOR RESEARCH PROGRAM MANAGEMENT AND OVERSIGHT Federal priorities for health sciences research are determined by the federal budget process through a complex system of interactions among the Executive Branch, Congress, the scientific community, industry, the public, and special interest groups. Ultimately, the federal agenda is set by the funds allocated by Congress through its authorizing, budgeting, and appropriating mechanisms and the recommendations made by Congress in report language. The committee concluded that the present system is
200 FUNDING HEALTH SCIENCES RESEARCH becoming increasingly stressed by short-term corrective actions whose long- term consequences have not been Ally assessed. Growing federal deficits, earmarking of funds to meet specific health needs, and rigid allocation policies within the health sciences establishment have reduced flexibility within the system. These problems emphasize the need to review federal priorities and to coordinate federal health sciences research efforts. Integration of scientific priorities, as determined by peer review or other review mechanisms, with sound policy will lead to more effective resource allocation, thus improving the overall environment of health sciences research. Although the committee endorses an open forum for discussing pnorii:ies and manners of addressing the problems facing health research, it also emphasizes that top~own research directives will be counterproductive to research. Priorities in the private sector are determined in equally complex processes. Whereas many of the larger foundations and voluntary health agencies have boards, steering committees, or a standardized peer review mechanism, others may not have a coordinated means of making deci- sions according to scientific or other objective criteria. Likewise, corporate R&D decisions are based upon financial determinants in accordance with directives from the boards of directors and stockholders. The committee does not believe that corporate sponsors should be forced to subject their decision making to open peer review, nor would it be possible. However, foundations and corporations should be able to consult with federal poli- cymakers in order to arrive at sound decisions that may complement the federal effort and meet their own needs as well. Failure to maintain constructive policies that integrate the efforts of government and private and nonprofit sponsors of research will limit scientific progress, jeopardize our continued leadership, and imperil our economic strength. It is imperative that review and oversight of the bal- ance among the research components be conducted on an ongoing basis. Therefore, the committee focused on developing mechanisms whereby the sponsors of health sciences research could work cooperatively to monitor progress, develop solutions, and make recommendations to address the problems facing health research highlighted in this report. The objectives of this process are (1) to optimize the use of resources from all sponsors of health sciences research; (2) to improve the nation's capacity to respond to health crises and capitalize on new research opportunities; and (3) to restore balance in the components of the system and resource allocation between support for people, projects, and facilities. Improving Communication Among the Federal Agencies Recommendation 6.1: The committee recommends that a Federal
POLICY OVERVIEW AND RECOMMENDATIONS 201 Coordinating Council for Science, Engineering, and Technology (FCCSET) subcommittee for Health Sciences be established to review federal priorities and coordinate federal health sciences research efforts on a continuing basis. . Because of the impact that health-related decisions have on the Amer- ican public, the committee believes it is essential to continue having high- level health sciences research advice available to the President through the Office of Science and Technology Policy (OSTP). The committee believes that effective mechanisms are necessary for developing cross-cutting health science policy among the federal scientific agencies. As such, the FCC- SET provides an excellent model for interagency coordination. FCCSET is composed of the science and technology advisor to the President and one representative from each of the 13 federal agencies sponsoring research. The FCCSET can establish various committees composed of appropriate high-level federal agency representatives to provide a direct link among government agencies, and it can coordinate federal activities. While the committee believes that the FCCSET will address inter- agency coordination of research, the White House also needs a formal mechanism for obtaining broad scientific advice from nongovernmental sci- entists. The current director of OSTP has established a President's Council of Advisors on Science and Technology (PCAST), composed of nongovern- mental science experts. This is the kind of advisory body the committee had in mind, and it is pleased to note the establishment of PCAST as a means of providing the President and FCCSET with advice from nonfederal scientists. The health sciences FCCSET committee should use the framework for assessing science and technology budgets proposed in a recent Academy report to evaluate support for health sciences research across federal agen- cies. The committee should develop guidelines for federal research prior- ities by considering the following categories from the National Academy of Sciences report Federal Science and Technology Budget P,ionties: New Perspectives and Procedures as they apply to the health sciences: · research related to the sponsoring agency's mission; · health research activities of individual agencies that contribute to the overall science and technology enterprise (including the components of training, fundamental research, and infrastructure); · cross-cutting research activities of several agencies that contribute to broad national objectives given priority by the President and/or Congress; and activities that constitute significant health research initiatives by virtue of their considerable cost. The committee was pleased to note the appointment of a health scientist last year as the associate director for life sciences within the OSTP
202 FUNDING HEALTH SCIENCES RESEAP~H and recommends that this associate director be chairman of the proposed FCCSEI' for Health Sciences. The committee also recommends that this post continue to be held by individuals with experience in health research and research administration. Under the chairmanship of the associate director for life sciences, a special FCCSET committee for health sciences research would bring together the NIH director; ADAMHA administrator; NSF director, health research directors in the Departments of Veterans Affairs, Defense, and Energy; and the heads of the other government agencies sponsoring federal health sciences research programs. These federal agencies would use the guidelines provided by FCCSET to set agency priorities as they pertain to their individual missions. Subsequently, these priorities would be used for budget development by the agencies. The science advisor, in cooperation with the President and director of OMB, then would match program priorities with budget requests to meet the nation's health science research needs. The committee believes that advice obtained through this mechanism will improve intergovernmental coordination for defining national health sciences research priorities. Ultimately, this will lead to more effective policies for allocating resources for project support, training, equipment, and facilities. Improving Communication Between Federal and Nonfederal Health Sciences Research Sponsors Recommendation 6.2: The committee Commends that a forum such as the Government-University-Industry Research Roundtable (GUIRR) be established to review the support of health sciences research on a regular basis and to facilitate communication among the various sectors that support health sciences research. The vitality of the health sciences research enterprise depends not only upon federal government activities but the cooperation of all parties involved in health sciences research: universities and independent research institutes, as well as the private sector (foundations, voluntary health or- ganizations, and corporations). Each must recognize the interdependence of the various sponsors of health science research in order to maximize its own contributions. These venous participants should have a mechanism for open dialogue to facilitate the efficient use of the limited health science research resources. The GUIRR provides a model for developing a forum to address these issues. The GUIRR was established by the National Academy of Sciences, National Academy of Engineering, and the IOM to address crosscutting
POLICY OVERVIEW AND RECOMMENDATIONS 2~3 issues that affect all areas of science and technology. It is composed of scientists, engineers, administrators, and policymakers from all sectors with the objectives to understand ~sues, to inject imaginative thought into the system, and to provide a setting for discussion and the seeking of common ground. 1b ensure that the balance of support among components of health sci- ences research is reestablished and maintained, this review would include evaluation of the relationships among support for research projects, the number of researchers being trained compared to the nation's needs and scientific opportunities, and the status of research facilities. This proposed committee should include representation from the executive and legisla- tive branches of the federal government, pharmaceutical and biotechnology industries, state governments, academic research institutions, private foun- dations, and voluntary health agencies. The committee recommends that such a forum initially identify the special responsibilities, interests, and contributions of each of these support sources and explore means to achieve health sciences research goals through greater interaction. The committee also suggests that this group consider sponsoring meetings and workshops or holding public hearings on issues such as · the special roles and responsibilities of government, industry, and nonprofit organizations in supporting health sciences research; the necessity of devising long-term plans to meet next century's research training needs; · ways to finance the escalating costs for facilities and equipment; · the appropriate balance of support for research projects, training, and facilities; · the impact of reallocating resources on the various components of the research enterprise; · cooperatives among research institutions and the private sector, in- cluding review of successes and failures in order to improve new initiatives; · ways to foster communication among scientists, health practitioners, and corporations to increase technology transfer; and · long-range planning for health sciences research including formu- lating a framework to assist establishing 5-, 10- and 20-year goals for individual participants. After careful consideration of issues that affect all supporters of health sciences research, the proposed GUIRR-like forum should be able to pro- vide information and advice about the needs and activities of the scientific community and their supporters to the proposed FCCSET committee out- lined above. This advice would be particularly useful in formulating federal guidelines that include consideration of the need to balance commitments
204 FUNDING HEA17H SCIENCES RESEARCH to support investigators directly, to restore facilities and equipment, and to provide training opportunities. Unresolved Issues to Be Addressed by These Forums The committee heard a number of complaints about the disease ori- entation and traditional disciplinary emphasis of federal support for health sciences research, with too little money available for newer fields such as nutrition and prevention research as well as for interdisciplinary projects that do not fit easily into current health research categories. In contrast, the committee also heard strong support for the current system, with the belief that some of these other areas could be handled easily within the existing organizational structure. However, the charge to this committee did not include an evaluation of the allocation of resources between or among the many topics within the disciplines of health sciences research. In a time of intense competition for available resources, where the potential for a national health crisis exists at any time, choices must be made where some fields of research receive more support at the expense of others. Vigorous advocacy by particular special interest groups has had enormous benefit in our democratic decision-making system. However, conflicting views by these groups can confuse decision makers in both the administration and Congress. Such conflicts have made priority setting among competing scientific initiatives extremely difficult. The committee recognizes that it would be advantageous to employ formal and explicit criteria in setting national health priorities and allocating scarce federal re- sources. Both government science administrators and nonfederal scientific advisory groups could benefit in their decision-making considerations from such criteria. The science advisors in the White House OSTP, along with the pro- posed health FCCSET, should work closely with P CAST, NIH/ADAMHA administrators and advisory groups, and the proposed GUIRR-like com- mittee to determine appropriate criteria for setting priorities among fields within the health sciences disciplines and for evaluating new initiatives. This would not be a means to rate competing disciplines but rather to evaluate scientific initiatives. Procedures should be developed that would permit scientific advisory committees and peer review panels to compare competing initiatives and reach unambiguous recommendations about pri- ority, based on criteria such as scientific significance, breadth of interest, potential for new discoveries and understanding, possible contributions to the improvement of health, and the feasibility and logistics of the proposal. The committee believes that this should not be perceived as an attempt to empower government administrators to be central planners; rather, it is
POLICY OVERVIEW AND RECOMMENDATIONS 205 the committee's intent to establish a high-level priority-setting process with a wide range of input from all sponsors and performers of health research. Many issues were brought before the committee that were outside of the committee's charge. Many of these dealt with structural aspects of the Executive Branch that the committee was not able to address. These included: (1) the role of the research components of ADAMHA and their relationship to the research institutes in NIH, (2) the role of the Assistant Secretary for Health in oversight of research sponsored by NIH and ADAMHA, and (3) the role of federal laboratories in facilitating technology transfer. The committee believes that these items should be on the agenda of OSTP and a health FCCSET to encourage open debate. The committee also deliberated extensively on the issue of 2-year congressional budget appropriations for the federal agencies that sponsor R&D. Whereas assessment of research needs would be conducted annually, including appropriate congressional testimony and progress reports from the various agencies, a rolling 2-year funding cycle could set minimum bud- get levels for particular institutes and programs. Under these circumstances, funding could be initiated at the beginning of each fiscal year without long delays, and this would lessen researcher anxiety about the priority level at which grants will be awarded. Such planning would also diminish the requirement that agencies arbitrarily reduce the level of previous awards because of unanticipated changes in funding levels. Although Congress would retain the option to reduce funding in the second year of a 2-year cycle, the actual history of overall budgetary stability suggests that such de- creases would be unlikely. Congress is likely to avoid such midcycle changes in the interest of stability in the research environment. Although there are many positive aspects to a 2-year budget cycle for federal research agencies, the committee did not believe it was within its charge to recommend such a policy change. Improving Communication and Cooperation Among Research Sponsors Recommendation 6.3: The committee recommends that sponsors and researchers explore ways to share facilities and equipment among research institutions, industry, and government. The committee also heard suggestions that institutions offset some of the high costs of research facilities and equipment by entering into cooperative regional agreements to share these expensive resources. Even if this cannot be done on a widespread basis, limited cooperation can further advances in health research and possibly can reduce unnecessary duplication of capital investments. As equipment and facilities costs continue to soar, cooperative sharing should reduce the need to duplicate investment in physical infrastructure.
206 FUNDING HEALTH SCIENCES RESEARCH Understandably, there are difficulties with proprietary rights and maintain- ing intellectual freedoms. However, the model of cooperation employed by the National Institute of Standards and Technology sets a precedent for the success of these types of ventures. This could be achieved by employing the GUIRR structure proposed. All sponsors and performers of health research should explore ways to increase sharing of facilities and equip- ment. Even though conflict of interest must be avoided, the committee is convinced that cooperative agreements can arise without compromising the integrity of researchers or institutions. Foundations and Voluntary Health Agencies Recommendation 6.4: The committee recommends that foundations and voluntary health organizations maintain their support for new lines of investigation and research projects that, for political or structural reasons, NIH and ADAMHA cannot fund. Traditionally, foundations and voluntary health agencies have been key supporters of interdisciplinary or innovative projects or of those projects that, for political or other reasons, are difficult to support with federal funds. Although nonprofit organizations will never have the resources to rival federal funding for health sciences research, they can respond to new lines of inquiry faster than the government bureaucracy allows. Fur- thermore, the disease-specific nature of voluntary health agencies provides them with a greater focus for supporting innovative ideas in specific areas of investigation as well as for funding trainees. Although the committee believes that foundations and voluntary health agencies are integral to the health research enterprise, it emphasizes that these organizations must not be considered substitutes for federal support. Rather, these organizations should supplement federal efforts and fill in gaps in support in specific areas of research. SCIENTIST RESPONSIBILITIES Federal health research allocation policies often have emerged piece- meal out of the continuing political process. Policy decisions largely reflect scientific, political, and economic influences. The sponsors of health re- search need to work toward common goals with the research community in order to provide an optimum environment for health research. The committee's recommendations to now have focused primarily on the re- sponsibilities of the sponsors. Little has been said about the role of research scientists and their responsibilities to the research system. Indeed, the key to a viable system is the active participation of scientists in all aspects of the research enterprise, including priority setting and allocation policy.
POLICY OVERVIEW AND RECOMMENDATIONS 207 The committee concluded that research scientists could take actions that would help to improve the future success of the enterprise beyond their own commitment to specific research projects. Scientists should assume a more active role in the policy decision-making process and should champion the overall needs of the research establishment. Health research is a long- term investment, and scientists need to express their views to governmental representatives so that Congress and the Executive Branch can set national research priorities. Scientists also have a responsibility to serve on peer review panels; to review journal articles; and to provide advice on policy boards of the federal government, private foundations, and charitable organizations. The committee believes that scientists should become more involved in improving the public's understanding of science. Negative publicity about science and scientists seems to be uppermost in the public consciousness in recent years. A very small number of highly publicized cases of alleged scientific misconduct and fraud are cited by some to be the tip of an iceberg of deception and misconduct pervading the scientific community. On the other hand, members of the scientific community have argued that the high degree of methodological reproducibility establishes the sound basis of scientific observation. Researchers must continue to show high regard for animal welfare and the proper handling of toxic wastes in order to avoid negative ramifications on the research establishment. 1b improve the public's opinion of science, the committee believes that scientists must strive to rid the system of misconduct; they must cooperate fully with their institutions and research sponsors in cases of suspected wrongdoing. Also, scientists need to help prevent overreaction to these unfortunate incidents that could easily stigmatize the field. The committee endorses the recommendations of a recent IOM study group report, The Responsible Conduct of Research in the Health Sciences,4 which includes rec- ommendations that scientists, individually, as well as through professional societies and other organizations, promote high ethical standards in the conduct of research. Failing to address these concerns in the rapidly paced and highly competitive realm of modern biomedical research could have serious consequences, for each new case of scientific misconduct increases the possibility of federal regulation. The committee is concerned that leg- islatively mandated guidelines for ethical conduct and scientific reporting could impede research activities and increase research costs. A CALL TO ACTION Many of the problems, issues, and opportunities considered by this committee have been tackled before by the scientific community and by advisors to and within government. Despite numerous recommendations by
208 [ENDING HEALTH SCIENCES RESEARCH those various groups, no decision to act has been made, and the basic prob- lems therefore have persisted. The present analysis has sought to include all of the sources of health sciences research support in order to provide a more comprehensive overview of current trends for all components of the research establishment The committee concluded that an imbalance in support among the components of the research enterprise needs to be addressed immediately to ensure a viable system into the next century. Effective and longer-term corrections will be made only when those who are examining the issues have the authority to act on their conclusions as well. Therefore, the committee believes that in order to begin resolving the problems discussed in this report and to make the best use of available research funds, ongoing communication among all research sponsors and the whole of the scientific community is vitally important. Only in this way can the wisdom invested in the enterprise be applied in a continuing effort of self-regulation and success. REFERENCES 1. National Research Council. 1989. Biomedical and Behavioral Research Scientists: Their Training and Supply. Washington, D.C.: National Academy Press. 2. Institute of Medicine. 1988. Resources for Clinical Investigation. Washington, D.C.: National Academy Press. National Academy of Sciences, National Academy of Engineering, and the Institute of Medicine. 1989. National Issues in Science and Technology: V. Federal Science and Technology Budget Priorities, New Perspectives and Procedures. Washington, D.C.: National Academy Press. 4. Institute of Medicine. 1988. The Responsible Conduct of Research in the Health Sciences. Washington, D.C.: National Academy Press.