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.
6 Accountability, Administration, and Leadership Public accountability and leadership are key aspects of the National Institutes of Health's (NIH's) stewardship of the biomedical enterprise because of the impera- tives to maintain public trust, reassure Congress that the public's interest is being served, and ensure that NIH's tactical and strategic objectives for its research and training programs are thoughtfully selected, effectively pursued, and responsive to NIH's research mission, national health concerns, and the need to prepare the next . . . . generation ot sclentlsts. Accountability is a critical and challenging aspect of leadership. It is especially challenging for an organization like NIH, which serves a broad array of constituen- cies and is devoted to research and training, in which outputs can be difficult to measure, market discipline is largely absent, and there is incomplete agreement on what metrics are the most appropriate. On the one hand, too mechanistic a system of accountability may fail to capture the nuances of scientific progress and indeed may stifle it or lead to an illusion of precision. On the other hand, too loose a system of accountability may lead not only to a potential loss of public confidence and trust but also to uncertainty about whether NIH's efforts are achieving, even in an . . . . approximate way, its ol electives. This chapter focuses on means by which NIH can enhance its public account- ability and ensure the continuing vitality of its leadership, both of which are influ- enced by and have a capacity to alter the agency's organizational effectiveness. Specifically, the Committee focused on the need for improved NIH-wide data gathering and coordination, increased attentiveness to hiring and review of senior leaders, and better use of the advisory committee system. Important additional aspects of NIH's ability to meet its scientific and public health mission are the 103
104 Enhancing the Vitality of the National Institutes of Health availability of sufficient resources in its management-support network to accom- plish its goals, and the ability to direct administrative functions in the best interest , . . . . . Ot its research ~ anc training mission. ANNUAL MECHANISMS FOR PUBLIC ACCOUNTABILITY Each year, NIH must complete two processes for accounting to Congress and the President on its progress in meeting its goals, conforming to its mission, and justifying its request for appropriations for the next fiscal year. Those processes, developing the budget and responding to the Government Performance and Results Act (GPRA) of 1993, are not sufficient to meet all accountability needs but they do provide useful starting points. The Budget Process: Congressional Justification Every fall, each institute or center (IC) with an appropriation must prepare a congressional justification budget (the CT) that details accomplishments of the pre- ceding year, current initiatives, and plans.) The process of preparing the CT is labor- intensive; programs in each IC are surveyed for data and material considered crucial for justifying programs and budgets. Each IC's information is submitted to the Office of the Director (OD), where it is reviewed and then compiled with all the other IC's CTs and submitted to the Office of Management and Budget (OMB). The exception to this process is the National Cancer Institute (NCI), which under the National Cancer Act of 1971 develops a "bypass budget" that goes directly to the President and OMB. The budget process is, perhaps, the most powerful accountability mechanism used by NIH, forcing the agency and its various units to justify their places in the programs of the President. It is both a planning and an accountability process. As in other federal agencies, the deadlines of the budget process drive the planning and priority-setting process. In addition, the capacity to influence the congressional appropriation process is one of the major ways in which NIH leadership can coor- dinate activities across NIH. Thus, development of the annual CT provides the perfect opportunity for NIH to respond to the Committee's call for more investment in trans-NIH initiatives, as described in Chapter 5. If all ICs were required to account for trans-NIH activities in their CJs, the NIH director, the Secretary of Health and Human Services, and Congress would have a better sense of progress in this category of work. iThe formal title is "Justification of Estimates for Appropriations Committees." The CJ has a section for each IC, consisting of a set of tables and budget narrative (see Department of Health and Human Services, 2002).
Accountability, Administration, and Leadership The New Approach to the Government Performance and Results Act Under GPRA, every major federal agency must ask itself some basic questions every year: What is its mission? What are its goals and how will it achieve them? How can it measure performance? How will it use performance results to make improvements? GPRA forces a shift in the focus of federal agencies away from such traditional concerns as staffing and activity levels toward a single overriding issue: results. GPRA requires agencies to set goals, measure performance, and report on accom- plishments annually. Every NIH IC goes through an annual process in which program managers are asked to review their portfolios for scientific accomplish- ments that provide evidence of meeting goals and missions. Until 2003, NIH tracked the success of its programs through a comprehensive overview of its progress by reviewing and compiling scientific advances and external recognition of NIH-supported investigators. Starting in 2003, however, NIH is using more specific and transparent goals to measure research outcomes; the Com- mittee supports this change. The new framework being used by NIH characterizes goals on the basis of the likelihood that they will be attained and the time targeted for completion. For FY 2004, there are 27 goals arranged according to their likelihood of success and years to completion; the longest time line is set at 10 years (Table 6.11. According to NIH, the goals must be credible to researchers, the public, and other NIH stake- holders and should be as specific as possible to address definable problems. Goals that lend themselves to annual reporting and reports of incremental progress are encouraged. NIH goals must also coordinate with the overall Department of Health and Human Services (DHHS) plan, as well as those of the Centers for Disease Control and Prevention (CDC), the Agency for Healthcare Research and Quality (AHRQY, the Food and Drug Administration, the DHHS "Healthy People 2010" initiative, and the President's Management Agenda. COUNCIL OF PUBLIC REPRESENTATIVES NIH created the Council of Public Representatives (COPR) and Public Liaison Offices in response to an Institute of Medicine (IOM) report (IOM, 1998), which recommended that such bodies be created to formalize and systematize communi- cation with the public at the highest levels of NIH. The scope and activities of COPR are evolving, but its charge is broad. COPR is a federal advisory committee of 21 members of the public from across the country who are chosen through an open nomination and application process. Its role is to advise the NIH director on public input and participation in NIH's activities, research priority setting, and outreach programs and efforts. Thus, COPR provides a formal mechanism for public input at the level of the NIH director, but there are multiple additional opportunities for public input across NIH. In addition, each institute has public 105
106 Enhancing the Vitality of the National Institutes of Health TABLE 6.1 NIH GPRA Research Outcomes RISK 1-3 years 4-6 years 7-10 years High la Conduct medications 2a By 2007, demonstrate 3a Identify at least one development with use of the feasibility of islet clinical intervention that animal models, and begin transplantation in will delay the progression, to conduct Phase I and II combination with immune delay the onset, or prevent trials of two potential tolerance induction for Alzheimer's disease. treatments for alcoholism: the treatment of type 1 cannabinoid antagonist diabetes in human clinical 3b By 2010, develop one Rimonabant and studies. universal antibiotic effective corticotropin-releasing against multiple classes of hormone antagonist 2b By 2009, evaluate the biological pathogens. Antalarmin. efficacy of two novel approaches to prevent 3c Determine the efficacy of lb By 2006, develop one weight gain and/or treat using salivary diagnostics to or more prototypes for a obesity in clinical trials monitor health and diagnose low-power, highly in humans. at least one systemic disease directional hearing aid by 2013. microphone to help 2c Develop methods that hearing-impaired persons can classify at least 75% better understand speech of proteins from in a noisy background. sequenced genomes according to evolutionary origin and biological structure. 2d By 2007, develop an HIV/AIDS vaccine. continues
Accountability, Administration, and Leadership TABLE 6.1 Continued RISK 1-3 years 4-6 years 7-10 years 4a By 2005, develop two new animal models to use in research on at least one agent of bioterror. 4b By 2005, develop improved animal models that best recapitulate Parkinson's Disease (PD) 5a By 2007, evaluate the efficacy of three new treatments for HIV infection in phase II/III human clinical trials in an effort to identify drugs that are more effective, less toxic, and/or simpler to use than the current based on emerging scientific recommended HIV findings of genetic or treatment regimen. environmental influences, or interactions of genes and the environment on the development of PD. 4c By FY 2007, identify 20 small molecules that are active in models of nervous system function or disease and show promise as drugs, diagnostic agents, or research tools. 5b Establish the ,.. . . . ettlcacy ot statlns In . · , preventing progression or atherosclerosis in children with Systemic Lupus Erythematosis (SLE or lupus ) . 5c Expand the range of available methods to be used to create, analyze, and utilize chemical libraries, which can be used to discover new medicines. Specifically, use these chemical libraries to discover 10 new and unique chemical structures that could serve as the starting point for new drugs. 6a Identify the genes that control the risk for the development of age-related macular degeneration and glaucoma in humans. 6b By 2011, assess the efficacy of at least three new treatment strategies for reducing cardiovascular morbidity/mortality in patients with type 2 diabetes and/or chronic kidney disease. 6c By 2012, develop a knowledge base on Chemical Effects in Biological Systems using a "systems toxicology" or toxicogenomics approach. continues 107
108 Enhancing the Vitality of the National Institutes of Health TABLE 6.1 Continued RISK 1-3 years 4-6 years 7-10 years 7a By 2005, evaluate 10 commonly used botanicals for inhibition/induction of enzymes that metabolize drugs as a method of identifying potential botanical/drug interactions. 7b By 2006, integrate nanotechnology-based components into a system capable of detecting specific biomarker(s) (molecular signatures) to establish proof of concept for a new approach to the early detection of cancer, and, ultimately, cancer preemption. 7c By 2005, create the next generation map of the human genome, a so called "haplotype map" (HapMap), by identifying the patterns , . . . at genetic varlatlon across all human chromosomes. Low 8a By 2007, determine the sequence of an additional 45 human pathogens and three invertebrate vectors of infectious diseases. 8b Identify and characterize two molecular interactions of potential clinical significance between bone-forming cells and components of bone. Such interactions are defined as those having . ... . slgnlilcant Impact on t :le accrual of bone mass or the actual mechanical performance of bone (i.e., fracture resistance in laboratory animals). 8c Build a publicly accessible Collection of Reference Sequences to serve as the basis for medical, functional, and diversity studies. A comprehensive Reference Sequence Collection will serve as a foundation for genomic research by providing a centralized, integrated, non-redundant set of sequences, including genomic DNA, transcript (RNA), and proteome (protein product) sequences, integrated with other vital information for all major research organisms. 9a By 2009, assess the . , . impact ot two major Institutional Development Award (IDeA) programs on the development of competitive investigators and . . . t :lelr capacities to compete for NIH research funding. 9b By 2010, demonstrate through research a capacity to reduce the total years lost to disability (YLDs) in the U.S. by 10% by: 1) developing treatment algorithms to improve the management of treatment- resistant and recurrent depression and 2) elucidating the mechanisms by which depression influences at least two comorbid physical illnesses (e.g., heart disease, cancer, Parkinson's disease, or diabetes). Major depression is now the leading cause of YLDs in the nation. 9c By FY 2010, identify culturally appropriate, effective stroke prevention programs for nationwide . . . . . . Implementation In minority . . communities. Source: National Institutes of Health, 2003.
Accountability, Administration, and Leadership relations activities with a major focus on communicating with and receiving input from diverse stakeholders. COPR provides a relatively new opportunity for receiving public input and increasing accountability to a broad constituency. These and other institutional changes within NIH were designed to increase public accountability, although there has been no formal evaluation of the impact of these changes on indicators of public accountability. Furthermore, no criteria have been proposed to assess and monitor public accountability or the effectiveness of mechanisms to improve or assure it at different levels. CHALLENGES TO ACHIEVING PUBLIC ACCOUNTABILITY The committee heard anecdotal evidence that priority setting-criteria often are not transparent to the interested public (for example, voluntary health organiza- tions) and that efforts to follow indicators of success have been limited by lack of well-developed data systems suitable for tracking expenditures and research, train- ing, and other NIH-sponsored activities in relation to institutional goals or priorities. Because the American public is diverse, it is difficult to know what degree of public accountability is achievable at various levels, and how broadly public accountability should be implemented. Likewise, the public is likely to have divided views on what constitutes or fulfills public accountability. Indeed, in practice much of the monitoring of accountability may fall to advocacy organizations, which may or may not represent the diverse views of the spectrum of individuals suffering from various illnesses. Such organizations also might not advocate health research that cuts across diseases; this may lead to tensions over the accountability mechanisms used by Congress and the administration, including for example the level of appro- priations for particular ICs, the formation of new ICs, NIH leadership approval, and program and project mandates included in authorizations and report language. Given the barriers at both levels of translation (described in Chapter 3 and Sung et al., 2003), another challenge to addressing accountability is the weak link be- tween research results on the one hand and public knowledge and perceptions of their significance, potential impact, and NIH's role on the other. Several institute directors told the Committee that when NIH-supported extramural research results are reported, credit is given primarily to the investigators and their institutions without linking them to NIH's support. Giving credit to individual ICs may also minimize the importance of NIH's combined efforts. Thus, media acknowledgment of NIH's role is often minimal or absent, and this can create misleading impressions of NIH's public accountability. NIH officials told the committee that although there is no overarching (NIH-wide) communication plan, efforts are under way to develop one that mirrors the agency's "Roadmap," one that will review research plans from the perspectives of various audiences to determine whether NIH is communicating effectively. There is no doubt that communicating the depth and breadth of NIH's activities and missions is a challenging task, but doing so and doing so with credibil- 109
110 Enhancing the Vitality of the National Institutes of Health ity is essential to bolstering public confidence in the agency and enhancing its accountability. DATA GATHERING AND REPORTING A related problem in achieving accountability is that, given the large research and training portfolio of NIH, it is difficult to track NIH's many scientific contribu- tions and make information about them available in a manner that is understand- able to all constituents. One of the most common types of questions asked by Congress and the public is how much research on specific diseases is being funded? Such data provide only an approximation of the level of effort devoted to objectives, but individuals and groups concerned about specific diseases or health problems often use them as a measure of input and effort to assess aspects of the NIH research and training portfolio of most interest to them. Even NIH officials complained to the Committee that such data are difficult to gather and are usually suspect because each IC uses its own method to estimate its investment. Although senior management at NIH has long recognized that the absence of a standard method for "coding" is a serious problem, no concerted effort to develop one has yet produced results. NIH officials often point out that statistics on spending by disease may not be very useful precisely because there are no simple relationships between measures of the burden of disease and how NIH allocates funding. Health needs are an impor- tant factor, but there is rarely a straightforward one-to-one relationship between health needs and research funding allocations, let alone estimated incidence, preva- lence, or burden of disease. Furthermore, the scientific opportunities for progress vary greatly among diseases in sophistication of the current knowledge base, promising lines of inquiry, and availability of sufficient researchers and facilities. Therefore, the amount of research support that can be linked directly to a specific disease not only is difficult to establish but is not by itself an adequate measure of how much or how well NIH is progressing against the disease. Nor does it reflect the potential relevance of basic research to multiple specific diseases (IOM, 1998) in that it is sometimes difficult to know which research is most relevant to the health problem involved. Despite those challenges, the Committee concludes that the current lack of an information management method and infrastructure to collect, analyze, and report investment data in a timely fashion must be addressed. It is particularly important for NIH leadership to improve the quality and analysis of its data on the allocation of NIH funds by disease for planning and priority-setting purposes. The problem requires the development of an NIH-wide agreement on what to track and publish and a single method for coding data that uses consistent definitions and deals with the uncertainties inherent in counting research that is only related but not directly applicable to a specific topic. Once developed, the statistics should be kept current and their accuracy ensured through quality control. NIH information management
Accountability, Administration, and Leadership must also be improved to meet goals in tracking scientists trained and supported with NIH funds. NIH is currently instituting substantial upgrades in its business, grant tracking, and clinical research-protocol IT systems, but these upgrades will not address the problem under discussion here. The committee recognizes that developing an addi- tional informatics system to gather data on the nature of NIH's diverse research and training programs consistently and uniformly is likely to be expensive and time- consuming and will require substantial resources and personnel. However, the development of such a system would provide invaluable information to all parties interested in NIH's programs Congress, other Executive Branch agencies, the pub- lic, the research community, and the leadership of NIH itself. Indeed, the effective management of NIH's research and training programs require such a capacity, which therefore would constitute a worthwhile investment. It also would provide the most reliable information for consideration of proposals to add, merge, or eliminate institutes, centers, and offices. Congress, which is likely to be one of the main beneficiaries of an improved information system, should consider the need for additional resources for this purpose in the budget process. Recommendation 9: Standardize Data and Infor~nation Management Systems For purposes of meeting its responsibilities for effective management, account- ability, and transparency, NIH must enhance its capacity for the timely collec- tion, thoughtful analysis, and accurate reporting of the nature and status of its research and training programs and public health advances. Data shouic! be collected consistently across institutes and centers and submitted to a central- · · ~ ~ Alec information management system. BUILDING ACCOUNTABILITY THROUGH LEADERSHIP The vision of the NIH leadership regarding accountability and the procedures and structures that the leadership adopts to enhance it are perhaps the most impor- tant ingredients in the complex mix of policies and strategies that enable NIH to meet its responsibilities to all its constituents. Leadership and vision may influence the extent to which accountability is reinforced and implemented at diverse levels of the NIH system, from top management through staff and to individual intramural and extramural investigators. Although there have been performance plans for IC directors and senior scien- tists since Director Harold Varmus established them in the 1990s, the current administration has required a formal performance assessment for all supervisory personnel throughout the government. All supervisory personnel at NIH therefore are required to develop a "performance contract" listing the items each person is accountable for, which is the basis of an annual assessment. The NIH director must negotiate his or her performance contract with the Secretary of Health and Human 111
112 Enhancing the Vitality of the National Institutes of Health Services. Likewise the NIH director negotiates a contract with each of the more than 40 people who are his direct reports. The earlier system for performance plans was general in its approach, but the new contracts are more specific, for example, "increase the prevention and public awareness of diabetes." Relevant items cascade down from the top; for example, the diabetes goal is in the President's contract with the Secretary, who then delegates it to the NIH director, who assigns it to the appropriate senior staff. Thus, the performance contracts are a mechanism for senior management to convey what is important to their subordinates. Outreach and communication factors are major items in the contracts of IC directors. IC directors are also involved in two other types of review: Every 5 years, IC directors are reviewed for their overall performance, including scientific leadership, management of their institute or center, and outreach and communications. Every 4 years, IC directors who conduct their own research are evaluated scientifically, as are all intramural scientists. That means that the Board of Scientific Counselors of the institute in which the research is performed oversees the review. Most directors conducting their own research do so in another institute to avoid conflict of interest. Most senior government officials are in the top ranks of the government service system or in the Senior Executive Service (SES) and are covered by a variety of civil service provisions under Title 5 of the personnel law that protect them from dis- missal and loss of rights without considerable effort on the government's part. But salaries for SES personnel are capped. To be able to offer higher salaries and attract the nation's most distinguished scientists, NIH obtained permission 10 years ago to place its senior positions under Title 42 of the law. Title 42 allows NIH to offer higher salaries, although people hired under this authority lose many of the civil service protections provided under Title 5 and must accept 5-year terms as opposed to permanent employment. Most directors chose to move into the Title 42 pro- gram, so they are, in fact, already subject to 5-year renewable terms. Under this system the Secretary of Health and Human Services retains approval authority for appointments. Thus, the NIH director can only recommend, but not appoint, senior leadership in the agency. The committee concludes that this lack of authority hinders the ability of the NIH director to form a cohesive senior leadership team to achieve NIH goals. The committee also concludes that in the current NIH environment, reviews of leadership a form of public accountability are too informal and ad hoc to be effective. The processes and criteria for review are not obvious or well defined. One of the more obvious criteria for review, in addition to scientific excellence and leadership skills, should be an individual's performance in imagining and engaging in creative collaborations with colleagues in other institutes and centers, as such
Accountability, Administration, and Leadership collaborations will be an increasingly important aspect of moving some of the most needed new NIH initiatives forward. And the informal review process has changed and depends on the particular policies and practices of the sitting NIH director and the personnel system in which a given IC director resides. The committee also believes that a healthy degree of turnover in leadership is critical for sustaining the vitality of a research organization. It also provides the opportunity for leading scientists across the nation to leave their positions for a set period and to come to NIH as a form of public service and, in part, to provide effective scientific leadership to critical elements of the nation's biomedical enterprise. Recommendation 10: Set Terms and Conditions for IC Director Appointments and Improve IC Director Review Process a. All IC directors should be appointed for 5-year terms. The possibility of a second and final term of 5 years should be based on the recommendation of the director of NIH, which should include consideration of the findings of an external review of job performance. The authority to hire and fire IC directors should be transferred from the Secretary of Health and Human Services to the NIH director. b. The director of NIH should establish a process of annual review for the performance of every IC director in terms of his or her effectiveness in fulfilling scientific and administrative responsibilities. The results of such reviews should be communicated, as appropriate, to the Advisory Committee to the director and/or the Council of Public Representatives. By communicating, as appropriate, the results of reviews to the NIH director's advisory groups, the IC directors can demonstrate an additional level of account- ability. While some aspects of a review should be held as confidential, those elements that relate directly to the mission and objectives of NIH should be made available to the director's advisors. The committee concluded that review and revitalization of the OD is an essen- tial prerequisite for accountability and leadership. The committee noted that the National Science Foundation Act of 1950 (42 U.S.C. 1861 et seq.), as amended, creates a term of 6 years for the NSF director and concluded that this has been a good mode! for creating a system of accountability and periodic review that has the possibility of transcending changes in administrations. Recommendation 11: Set Terms and Conditions for the NIH Director Appointment The NIH director, appointed by the President, should serve for a term of 6 years unless removed sooner by the President. The possibility of a second and final 113
114 Enhancing the Vitality of the National Institutes of Health term of 6 years should be based on a positive external review of performance and the recommendation of the secretary of Health and Human Services. Finally, the Committee believes that the special status granted NCI by the National Cancer Act should be re-examined. The National Cancer Act of 1971, in addition to making the NCI director a Presidential appointee, created the President's Cancer Panel, composed of two scientists and one management specialist who provide progress reports to the President on the status of NCI's research. The act also replaced the National Cancer Advisory Council with an 18-member National Cancer Advisory Board composed of scientists and laypersons offering guidance and advice to NCI on all major initiatives. In addition, the act allows the NCI director to submit the institute's budget directly to the President, bypassing the NIH director in the process. Because the President appoints the NCI director and the NCI budget bypasses the NIH director, it is possible that an unnecessary rift is created between the goals, mission, and leadership of NIH and those of NCI. NCI is by far and has been for some time the largest NIH institute (approximately 17% of the total NIH budget). It is not in the interests either of NIH's overall research and training programs, or of NCI, for the NIH director to have such limited authority. In addition, as the biologi- cal and cellular basis of cancer becomes more widely understood, the basic science underlying cancer research has direct implications for the etiology and progression of numerous other diseases, for example the autoimmune, infectious, and even cardiovascular diseases. Thus, for scientific and administrative reasons, NCI's special status should be reconsidered. Recommendation 12: Reconsider the Status of the National Cancer Institute Congress should reassess the provisions of the National Cancer Act of 1971, particularly as they affect the authority of the NIH director to hire senior management and plan and coordinate the NIH budget and its programs in their entirety. It should be noted that the requirement that NCI prepare a bypass budget every year has some positive aspects in that the institute must undertake an annual strategic planning process. This useful exercise should not be dropped if NCI changes its administrative status as recommended above. Rather, all ICs should be required to develop an annual strategic plan, if they are not already doing so. THE ADVISORY COMMITTEES Like other federal science agencies, NIH makes extensive use of advisory com- mittees of nonfederal scientists, health advocacy representatives, and others to ensure the best possible input of expertise and additional perspectives on the evaluation of programs and the development of policies and priorities. NIH had over 140 char-
Accountability, Administration, and Leadership tered advisory committees as of May 2002, more than any other federal agency.2 The Public Health Service Act (PHSA) authorizes appropriate scientific and techni- cal peer review of biomedical and behavioral research grant and cooperative agree- ment applications, research and development contracts, and research conducted at NIH through its advisory committees. As described in greater detail in Chapter 2, NIH uses several types of advisory committees. Those groups can be located in the Center for Scientific Review (CSR) (the study sections) or the councils and boards created and used by individual institutes that choose not to use CSR for review of particular initiatives. National Advisory Councils and Boards perform the second level of peer review for research grant applications and offer advice and recommendations on policy and program development, program implementation, evaluation, and other matters of impor- tance for the mission and goals of the IC; and they provide oversight of research conducted by IC intramural programs. The dual review system, which separates the scientific assessment of proposed projects from policy decisions about scientific areas to be supported and the resources to be allocated, permits a more objective evaluation than would result from a single level of review. NIH can make awards only if they have been approved by a national advisory council and the Secretary, and this helps to insulate NIH from pressure by a member of Congress or the administration to fund a particular project. The national advisory councils are also charged with providing advice on policies and programs, although several studies have found that members of the national advisory councils do not always fee! they play a strong role in policymaking.3 The dual system of review provides the respon- sible NIH officials with advice about both scientific and societal values and needs (NIH, 1992b).4 In the appointment process, the President generally follows the recommenda- tions of the Secretary, and the Secretary generally follows the advice of the NIH and IC directors in filling positions, although they add their own candidates from time to time. During the 1972-1974 period, when the Nixon Administration was trying to exert greater control over the NIH budget, there was conflict with the scientific community over the perceived politicization of the advisory committee appointment process; this issue re-emerges from time to time and is of current concern to the scientific and health advocacy communities (e.g., Bass et al., 20031. Moreover as a general matter, the success of any scientific enterprise is dependent on the encourage- 2They have 4,298 members, 75% of whom are members of initial review groups that evaluate appli- cations for research funding. See overview and list of committees by appointing officials at http:// wwwl .od.nih.gov/cmo/about/index.html. 30ne study was conducted by the Institute of Medicine's Committee for a Study of the Organizational Structure of the National Institutes of Health in 1984. The other was conducted in the mid-199Os by a committee appointed by the NIH director. Neither report was made public. Copies are in the posses- sion of the authors. 4Contracts are subjected to a similar peer review process, except that the second level of review is by senior IC staff. 115
116 Enhancing the Vitality of the National Institutes of Health ment of a wide variety of independent views. The Committee believes that it is essential that members be appointed to these advisory groups because of their ability to provide scientific or public health expertise to the review and approval of awards and policies. They should not be selected to advance political or ideological . . positions. Several related issues emerged during the committee's deliberations with respect to NIH's advisory council system. First, there are important differences in the use and roles of the councils among ICs. Some councils are actively involved in setting institute goals and planning. In other cases, council actions are pro forma, with little advice or involvement sought from council by institute personnel. In still other cases, council members might also be grantees of the institute, and thus might fee! constrained in expressing strong views or views that differ from than those held by institute or program staff. Those issues highlight a missed opportunity for NIH. Advisory councils should routinely and consistently be consulted in the priority setting and planning processes of an institute. They should have active involvement in decisions regarding issuance of program announcements and requests for applications, which often reflect an institute's priorities and responses to emerging opportunities or demands. They should be working to ensure that the IC is held accountable in reaching its goals and communicating with the public. They should understand and be supportive of relevant trans-NIH initiatives. Finally, a criterion for review of every institute director should be how he or she interacts with and uses the expertise of his or her advisory council. Under Section 406 of the PHSA, national advisory councils have up to 18 members appointed by the Secretary and nonvoting ex officio members from NIH and other federal agencies. Two-thirds of the appointed members are to be "from among the leading representatives of the health and scientific disciplines (including not less than two individuals who are leaders in the fields of public health and/or social sciences) relevant to the activities of the national research institute" and one- third "from the general public and shall include leaders in the fields of public policy, law, health policy, economics, and management." The Committee believes that the advisory council system should guarantee that ICs receive independent and qualified advice. Their members therefore must be reasonably free of conflicts of interest. In addition, if NIH is to achieve the goal of increased trans-NIH collaborations, it will be important to have cross-fertilization of institutes through advisory council mem- bership. For example, it would be useful to have a cancer researcher (who receives funding from NCI or the American Cancer Society) serve on the council of the National Institute of Environmental Health Sciences or the National Institute of Child Health and Human Development. Recommendation 13: Retain Integrity in Appointments to Advisory Councils undo Refrain Advisory Council Activity undo Membership Criteria a. Appointments to advisory councils should be based solely on a person's
Accountability, Administration, and Leadership scientific or clinical expertise or his or her commitment to and involvement in issues of relevance to the mission of the institute or center. b. The advisory council system should be thoroughly reformed across NIH to ensure that these bodies are consistently and sufficiently independent and are routinely involved in priority-setting and planning discussions. Councils should be effectively engaged in discussions with IC leadership to enhance accountability, facilitate translation of goals and activities to the scientific community and the public, and provide feedback to the IC director. To achieve sufficient indepen- dence and avoid conflicts of interest, a substantial proportion of a council's scientific membership should consist of persons whose primary source of research support is derived from a different institute or center or from outside NIH. RESEARCH MANAGEMENT AND SUPPORT Although administrative or overhead costs are often suspect in the eyes of those who would like to see more money going directly to research or training, at appro- priate levels they are essential to the effectiveness of any organization, including those that sponsor research and training programs. Similarly, the effectiveness of those responsible for the wide array of necessary administrative services depends on their leadership and management capabilities and their ability to keep administra- tive and overhead costs deployed in a manner that best supports the primary missions of the organization. In the case of NIH, the resources for administrative and overhead functions flow through the Research Management and Support (RMS) budgets of the various units that make up NIH. These budgets, collectively, support all the administrative costs of operating NIH, including management of extramural activities (planning, receipt, peer review, and awards), some intramural research program costs, pro- gram development, priority setting, education and outreach, acquisition and main- tenance of new information technology systems, professional development, and facility management. Given the structure and funding mechanisms of NIH, the aggregate RMS budget is composed of 25 budget line items, one from each of the ICs that receive separate budget appropriations from Congress. RMS is functionally distinguished from the NIH OD, which is responsible for strategic leadership and receives a separate appropriation. The administrative costs of NIH have been scrutinized regularly over the last few decades. In the early 1990s, congressional limitations were imposed that restricted inflationary and program growth of the RMS budget. A 1997 manage- ment study by Arthur Anderson (National Institutes of Health, 1997b) led to many management improvements, including · Centralization and improvement of purchasing programs · Conversion of the mail service to an outsourced performance-based contract 117
118 Enhancing the Vitality of the National Institutes of Health · Development of generic position descriptions · Hiring of a Chief Information Officer, and · Creation of a Central Service Review Committee to review the budgets of central service organizations (NIH, unpublished draft report). In FY 1996, the NIH appropriation contained language that reduced RMS by 7.5% below the FY 1995 level. Despite growth in the overall NIH budget, the RMS reduction was not made up in FY 1997 nor was any growth provided; in FY 1998, an increase of only 1% was allowed. In contrast, from FY 1995 to FY 1999, the extramural program grew at a rate seven times that of RMS. In the FY 2001 budget, RMS represented 3.3% of the total NIH budget, down from 4.5% in 1995. Since FY 1993, the RMS share of the total NIH budget has decreased every year. From FY 1984 through FY 1999, inflation, based on the Gross Domestic Product, increased 58.3%. During the same period, RMS grew 96.2%, while overall NIH grew 217.6%, more than four times the growth rate for RMS (NIH, unpublished draft). At the same time, the growth in the NIH budget and the rise in congressionally mandated activities have increased the administrative requirements needed to operate a growing and diverse research organization; for example, GPRA is a labor- intensive and expensive annual exercise required by Congress. Other new program- matic requirements have involved the establishment of centers, registries, and other funding requirements, all of which add costs for which RMS must be further stretched. To accommodate RMS reductions, many institutes have implemented measures to reduce costs, such as introducing modular grant applications and awards, stream- lining reviews, and converting to electronic-based research administration. Those are laudable goals under any circumstances, but adverse consequences of the restricted RMS budget seem to be growing. Many are concerned that the strain on the system harms the peer review system, stretches staff too thin, limits business oversight and scientific review, and hinders the ability to respond to increasingly complex research programs and conduct trans-NIH initiatives. NIH's own assess- ment of the negative impact of the restricted RMS budget found seven areas being adversely affected: stewardship of public funds; scientific advice and program devel- opment; public health education and community outreach; information technology acquisition, maintenance, and training; staffing issues; professional development; and facilities management (NIH, unpublished draft). There may have been good reason in the past to celebrate the containment of costs, but the Committee feels that the effectiveness of NIH is now imperiled by the lack of adequate resources to provide appropriate support both for its primary research mission and for meeting . . . . . . . its accountal ~1- sty response AL 1tles. Other groups have also suggested that RMS funding be raised to provide adequate means for accomplishing NIH's primary goals and to ensure a capacity for strategic planning and evaluation of its programs. In 1998, an IOM committee
Accountability, Administration, and Leadership recommended that Congress "adjust the level of funding for RMS so that NIH can implement improvements in the priority-setting process, including stronger analytical, planning, and public interface capacities" (IOM, 1998~. In 2001, the Federation of American Societies for Experimental Biology also recommended in- creasing RMS. The growing mismatch between the most essential or mandated administrative requirements and the RMS resources available to pay for them must be addressed. Recommendation 14: Increase Funding for Research Management and Support Congress should increase the appropriation for RMS to reflect more accurately the essential administrative costs required to effectively operate a worId-ciass $27 billion/year research organization. Moreover, when additional congres- sional mandates are imposed on NIH through the appropriations process, they should include funds to cover necessary administrative costs. SUMMARY NIH uses resources in various ways to enhance public accountability, leader- ship, and management efficiencies. However, improvements can be made. First, NIH must commit to developing an improved system for gathering, managing, and reporting data to facilitate public engagement, strategic planning, management of the research and training portfolios, congressional justifications, . ... . . ant scenic communication. Second, increased attention to the system of hiring and periodic and systematic review of IC directors will revitalize the leadership, invigorate the overall scientific community, and facilitate change and evolution of NIH's mission and goals. The NIH director should have the authority to appoint IC directors, including the direc- tor of NCI, with the goal of building a team that shares a vision and a plan. Congress should revisit the special status of NCI to determine whether it continues to meet the needs of the current NIH organization and structure. Third, leadership must make better use of the advisory committee system, which should be a consistent source of independent advice. Fourth, in order to operate a worId-class research agency, NIH must be provided sufficient resources to support its management needs. 119
fellow of AAAS ant! many other professional societies. He has received numerous awards from both professional and lay groups for his national leadership in science, mental illness and mental health, and substance abuse ant! addiction. Gilbert S. Omenn, MD, PhD, University of Michigan (IOM) Gilbert S. Omenn is Professor of Internal Medicine, Human Genetics, ant! Public Health at the University of Michigan. From ~ 997 to 2002 he was also UM Executive Vice President for Medical Affairs and Chief Executive Of ficer of the University of Michigan Health System. Previously he was professor of medicine and environmental health and Dean of the School of Public Health & Community Medicine at the University of Washington. He served as Associate Director of the White House Office of Science and Technology Policy and then the Office of Management and Budget in the Carter Administration, and chaired the Presiclential/Congressional Commission on Risk Assessment and Risk Management from 1994- 97. He has been a National Institutes of Health Research Career Development Awardee, a Howard Hughes Medical Institute Investigator, and founding director of the University of Washington Robert Wool! Johnson Clinical Scholars Program. His research is focused on proteomics and cancer prevention, as well as health promotion for older adults, science-based risk analysis, and the ethical, legal, and public health policy aspects of genetics. Dr. Omenn holds an MD from Harvard and a PhD in genetics *om the University of Washington. Franklyn G. Prendergast, PhD, Mayo Cancer Center Franklyn Prendergast is Director of the Mayo Clinic Cancer Center in Rochester, Minnesota, and Professor of Biochemistry and Molecular Biology. His research focus is in structural protein biology and bioimaging. He is a recipient of the E.E. Just aware! of the American Society of Experimental Biology. He is a member of the American Association for the Advancement of Science, the American Society for Biochemistry and Molecular Biology, and Sigma Xi. He earned his PhD in biochemistry from the University of Minnesota and his medical degree from the University of the West Indies. Stephen J. Ryan, MD, University of Southern California (IOM) Stephen Ryan is Professor of Ophthalmology and Dean, Reck School of Medicine of USC and Senior Vice President for Medical Care, University of Southern California. His research relates to macular degeneration, ocular trauma, retinal detachment, ant! other retinal diseases. He previously served as Chairman of the Department of Ophthalmology at USC and as a member of the National Advisory Eye Council for the NE! of the NIH. He is a member of the Institute of Medicine and currently serves as President of the National Alliance for Eye and Vision Research. He earned his MD from Johns Hopkins University. Samuel C. Silverstein, MD, Columbia University (IOM) Samuel C. Silverstein is John C. Dalton Professor of Physiology ant! Cellular Biophysics and Professor of Medicine at the Columbia University College of Physicians and Surgeons. His research focuses on structure and function of polymorphonuclear and mononuclear leukocytes ant! endothelial cells in innate immunity, in diseases associated with chronic inflammation such as atherosclerosis ant! Alzheimer's disease; and in host defense against infectious micro- organisms inclucling Legionella pneumophila and M. tuberculosis. He has server! on the Councils of the American Society for Cell Biology (! 988-92), and the National Institute of Allergy ancl 120
