7
Health Services Research

Health services research provides the information needed to understand the strengths and weaknesses of our health care delivery system.1 Health services research has documented inadequate access to health care for uninsured, rural, and inner-city populations; failures of patient safety practices that kill as many as 98,000 Americans in hospitals each year;2 and poor-quality care for chronic diseases.3 Health services researchers have studied factors contributing to rising health care costs and clarified the contribution of new technologies and lack of incentives for efficiency.4 In general, the goal of health services research is to contribute to the health and well-being of individuals and populations. This requires health services researchers to go beyond disease outcomes and examine health status and health-related quality-of-life outcomes, as well as focus attention on prevention and health promotion services.

The contributions of health services research to policy, management, and clinical care have been diverse. Planners and policy makers, for example, look for ways to generalize findings from efficacy studies: persons recruited to randomized control trials testing new treatments typically are not representative of the larger population expected to benefit from the treatment. Thus, it is up to health services research to fill this information gap by assessing the impact of diagnostic and treatment technologies on patient outcomes and costs across practice settings and populations.

Health service interventions are inherently complex and usually involve multiple system levels, including patients, providers, health care organizations, financing, and community context (e.g., health resources, population sociodemographic and risk factors). The design, conduct, and analysis of complex interventions require the input of many disciplines, as well as advances in multilevel longitudinal statistical methods. The Agency for Healthcare Research and Quality together with the National Institutes of Health (NIH) and the Centers for Disease Control and Prevention (CDC) recently convened a conference on “Research Designs for Complex, Multi-level Health Interventions and Programs”5 to explore the design, implementation, and analysis of complex service interventions.

Translational research recently emerged as another important dimension of health services research design and analysis. Americans receive treatments consistent with the best scientific evidence only half of the time, which has raised questions as to what should be done to accelerate the adoption of “best practices.”6 This is a small but growing area of health services research and is likely to become a much larger and more important area in the future. Meeting the challenge of translational research can be expected to require additional disciplinary breadth, drawing on areas of marketing research, adult learning, and real-time decision support technologies.

Central to advances in any scientific field are measurement tools, and for health services research measurement tools span payment and financing, appropriateness of utilization, quality of care, and patient outcomes of care. In the past these tools have provided measurement systems to support financing innovations to increase hospital efficiency and to adjust capitation payments and in the future research will need to extend payment methodologies to provide incentives for quality performance, as well as efficiency and equity.

1  

Note: Many definitions of this multidisciplinary field are available in the literature, including those developed by previous NRC committees on personnel needs in the biomedical and behavioral sciences. For example, see National Research Council, 1977, 1983, 1989, and 1995. Other authors include the Institute of Medicine, 1995. A recent definition circulated within the community was developed by Kathleen N. Lohr and Donald M. Steinwachs (Lohr and Steinwachs, 2002).

2  

Institute of Medicine. 1999

3  

McGlynn, E. A., et al. 2003.

4  

Although spending more than other nations on health care, the United States has failed to achieve population health indicators equal to countries that spend substantially less (Reinhardt, 2002); furthermore, our quality-of-care indicators are not consistently higher (Hussey et al., 2004).

5  

NIH, Bethesda, MD, May 4–5, 2004.

6  

McGlynn. 2003. op. cit.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 79
Advancing the Nation’s Health Needs 7 Health Services Research Health services research provides the information needed to understand the strengths and weaknesses of our health care delivery system.1 Health services research has documented inadequate access to health care for uninsured, rural, and inner-city populations; failures of patient safety practices that kill as many as 98,000 Americans in hospitals each year;2 and poor-quality care for chronic diseases.3 Health services researchers have studied factors contributing to rising health care costs and clarified the contribution of new technologies and lack of incentives for efficiency.4 In general, the goal of health services research is to contribute to the health and well-being of individuals and populations. This requires health services researchers to go beyond disease outcomes and examine health status and health-related quality-of-life outcomes, as well as focus attention on prevention and health promotion services. The contributions of health services research to policy, management, and clinical care have been diverse. Planners and policy makers, for example, look for ways to generalize findings from efficacy studies: persons recruited to randomized control trials testing new treatments typically are not representative of the larger population expected to benefit from the treatment. Thus, it is up to health services research to fill this information gap by assessing the impact of diagnostic and treatment technologies on patient outcomes and costs across practice settings and populations. Health service interventions are inherently complex and usually involve multiple system levels, including patients, providers, health care organizations, financing, and community context (e.g., health resources, population sociodemographic and risk factors). The design, conduct, and analysis of complex interventions require the input of many disciplines, as well as advances in multilevel longitudinal statistical methods. The Agency for Healthcare Research and Quality together with the National Institutes of Health (NIH) and the Centers for Disease Control and Prevention (CDC) recently convened a conference on “Research Designs for Complex, Multi-level Health Interventions and Programs”5 to explore the design, implementation, and analysis of complex service interventions. Translational research recently emerged as another important dimension of health services research design and analysis. Americans receive treatments consistent with the best scientific evidence only half of the time, which has raised questions as to what should be done to accelerate the adoption of “best practices.”6 This is a small but growing area of health services research and is likely to become a much larger and more important area in the future. Meeting the challenge of translational research can be expected to require additional disciplinary breadth, drawing on areas of marketing research, adult learning, and real-time decision support technologies. Central to advances in any scientific field are measurement tools, and for health services research measurement tools span payment and financing, appropriateness of utilization, quality of care, and patient outcomes of care. In the past these tools have provided measurement systems to support financing innovations to increase hospital efficiency and to adjust capitation payments and in the future research will need to extend payment methodologies to provide incentives for quality performance, as well as efficiency and equity. 1   Note: Many definitions of this multidisciplinary field are available in the literature, including those developed by previous NRC committees on personnel needs in the biomedical and behavioral sciences. For example, see National Research Council, 1977, 1983, 1989, and 1995. Other authors include the Institute of Medicine, 1995. A recent definition circulated within the community was developed by Kathleen N. Lohr and Donald M. Steinwachs (Lohr and Steinwachs, 2002). 2   Institute of Medicine. 1999 3   McGlynn, E. A., et al. 2003. 4   Although spending more than other nations on health care, the United States has failed to achieve population health indicators equal to countries that spend substantially less (Reinhardt, 2002); furthermore, our quality-of-care indicators are not consistently higher (Hussey et al., 2004). 5   NIH, Bethesda, MD, May 4–5, 2004. 6   McGlynn. 2003. op. cit.

OCR for page 79
Advancing the Nation’s Health Needs A goal for health care access is the timely receipt of appropriate care, and criteria for evaluating the appropriateness of service use have been developed and are widely applied for utilization review. With fee-for-service financial incentives and comprehensive insurance coverage having the potential to encourage overuse of services, the measurement of appropriateness has been applied to identify unnecessary admissions and days of inpatient care. In contrast, those without health insurance are likely not to seek needed services or to delay doing so until severely ill, and methods have been applied to assess the consequences of inadequate access to primary care. Future work will need to provide better metrics of timeliness and appropriateness, and from a patient’s perspective, measuring appropriateness needs to take into account desired outcomes and social context. From another perspective, advances in measuring quality of care have provided tools to evaluate the processes of care and patient outcomes. Clinical research provides the foundation for establishing diagnostic and treatment criteria based on scientific evidence. In the absence of strong consistent evidence, clinical expert consensus has been used to set quality standards. One largely unanswered question is whether process standards can be expected to apply to 90 percent of patients, 80 percent, or less. Research is needed to improve our capability to match treatments to patients and minimize the need to try multiple treatments before achieving desired outcomes. FEDERAL SUPPORT OF HEALTH SERVICES RESEARCH In 1968 Congress recognized the emerging role of health services research for improved health care delivery in the United States and created the National Center for Health Services Research and Development (NCHSRD) in the U.S. Department of Health, Education, and Welfare. During those years, NCHSRD sought to develop research on issues of access, cost, and quality and to develop data systems to support research on utilization and cost of care.7 However, in the years that followed, the budget for NCHSRD declined and the future of the NCHSRD and its funding were uncertain. Private foundations played a critical role in sustaining the health services research field.8 In 1989 health services research once again found strong support in Congress, and a new vision for health services research was created in the authorization for the Agency for Health Care Policy and Research. Congress directed the agency, subsequently renamed the Agency for Healthcare Research and Quality (AHRQ), to undertake research on patient outcomes, develop practice guidelines, and disseminate research to change the practice of medicine.9 The agency placed greater emphasis than previously on the examination of clinical practices, decision making, and comparing the cost effectiveness of alternative approaches to diagnosis and treatment. While the National Research Service Award (NRSA) program included support for health services research from its inception (see, for example, NRC, 1977), Congress specified in 1989 that one-half of 1 percent of the NRSA budget for training be allocated for training health services researchers through AHRQ, subsequently expanding that allocation to 1 percent of NRSA funding in 1993. By August 2003, AHRQ had provided support for research training through the NRSA program to nearly 800 individuals in the form of predoctoral/postdoctoral traineeships and to another 80 individuals in the form of individual fellowship awards.10 It should be noted that in the early 1990s Congress authorized a 15 percent set-aside for NRSA training in service-related research supported by the National Institute of Mental Health (NIMH), the National Institute of Drug Abuse (NIDA), and the National Institute of Alcohol Abuse and Alcoholism (NIAAA) as part of the reorganization of the former Alcohol, Drug Abuse, and Mental Health Administration into the NIH. Nonetheless, AHRQ is seen as the lead agency in the area of health services insofar as NIH funding for health services research focuses on questions related to the delivery of health care related to NIH-specific diseases and disorders. CURRENT MARKET FOR HEALTH SERVICES RESEARCHERS Health services researchers work in a variety of settings, including academic health centers, the policy and planning offices of the federal and many state and local governments, throughout the health care delivery sector, and in the phar- 7   The center initiated large-scale demonstrations, including the Experimental Medical Care Review Organization (EMCRO) to develop tools for quality measurement and their evaluation. The EMCRO demonstrations provided the Medicare program with the methodologies it needed in the Professional Standards Review Organization (PSRO) to evaluate hospital use. The NCHSRD also competitively funded health services research centers in academic institutions and for Kaiser Permanente. 8   See the NIH/NLM-sponsored database, “HSRProj,” for details regarding health services research projects supported by various sectors: http://www.nlm.nih.gov/hsrproj. It should be noted that health services research in focused areas like mental health services, alcohol and drug abuse treatment services, and veterans’ health care continued throughout this time. Health services research funding also comes from the Centers for Medicare and Medicaid Services, the Centers for Disease Control and Prevention, the Department of Defense, and other NIH institutes. 9   In 2001 the reauthorization of Agency for Health Care Policy and Research led to a name change to the Agency for Healthcare Research and Quality (AHRQ). The word policy was dropped from the title and quality was added to reinforce the quality-of-care research mission of this agency. 10   These counts are based on the number of individuals who completed a minimum of 6 months of NRSA training, beginning their training sometime after August 1986 and completing their training by August 2003. P. Flattau, personal communication, September 2004.

OCR for page 79
Advancing the Nation’s Health Needs maceutical and health insurance industries.11 Unfortunately, no national statistical system reports on the size and composition of the health services research workforce, since national workforce statistics usually capture information about these scientists in terms of their primary discipline of training or employment and fail to identify the field of scientific inquiry as “health services research.”12 In other words, many scientists who work on health services research problems may do so while conducting research in other areas, such as social, biological, or health research. Furthermore, many health services researchers conduct research as part of their primary employment in nonacademic settings, such as those working in government policy and planning offices or those employed in the pharmaceutical or insurance industries. Therefore, this “part-time” involvement in health services research only further exacerbates efforts to estimate the size and composition of this workforce. Despite the absence of a national database, a number of studies over the years have yielded important insights into the nature and composition of this group of specialists.13 For example, the field of health services research draws talent several ways. In the early years of the field’s development in the 1960s and 1970s, clinicians and other health scientists simply redirected the focus of their research on the matter of improving health care delivery.14,15 Today, individuals enrolled in more traditional fields of science and engineering have the opportunity to focus their studies at the doctoral level specifically on problems related to health care services and delivery. They may do so as doctoral candidates in public health or health policy; in the social sciences, including health economics; in other health sciences such as epidemiology or biostatistics; or in health services research itself. Indeed, the NRSA program has made it possible for many individuals interested in receiving formal training in health services research at the predoctoral level to do so.16 Postdoctoral research training is another mechanism that has emerged over the past 20 years to foster the growth of a skilled health services research workforce.17,18 The AHRQ T32 NRSA program has proven to be an especially effective mechanism for attracting clinicians into a health services research career. Of the nearly 200 individuals who received postdoctoral NRSA support through AHRQ between 1986 and 1997, two-thirds had earned a doctorate in one of the clinical professions prior to NRSA training, primarily in internal medicine.19 Among those individuals holding a research doctorate and pursuing postdoctoral training, half had earned a doctoral degree in the social sciences.20 The best data available on the actual size of the health services research workforce comes from the membership roles of the professional organization, AcademyHealth. AcademyHealth draws its members from both health services research and health care policy and includes student memberships. While this database more than likely underestimates the total size of the workforce, it does provide some insights into its composition. The total AcademyHealth membership in 2004 was 3,745, and 1,688 members (45 percent) reported having a Ph.D., Sc.D., or other doctoral-level training in science. There were another 710 (19 percent) reporting an M.D. and only 15 with a D.D.S. The primary locus of employment for doctoral degree holders who are members of AcademyHealth is the academic sector. Table 7-1 shows the distribution of these individuals across all sectors. AcademyHealth membership includes roughly equal gender representation, females (51 percent) and males (45 percent), while student membership has greater female representation (64 percent) than male (30 percent). The ethnic mix of members is 13 percent from minority ethnic backgrounds, including Asian/Pacific Islanders (7 percent), African Americans (3 percent), and Hispanics/Latinos (1 percent); 24 percent are “unknown.” The remainder (63 percent) are Caucasians. Student membership shows greater diversity—24 percent coming from minority ethnic backgrounds, including 14 percent Asian/Pacific Islanders, 6 percent African Americans, and 2 percent Hispanics/Latinos, plus 54 percent Caucasians and 22 percent unknown. Table 7-2 shows the primary fields of interest of AcademyHealth members. Most members classify themselves into health services research, and only 15 to 20 percent, depending on their degree, list health policy. The exceptions are those with J.D. degrees, who are strongly oriented toward policy. 11   Based on data collected by the Agency for Healthcare Research and Quality, 2000. 12   As NIH moves more toward transdisciplinary research, the problem of lacking multiple classifications for both “discipline” and “field of application” may be faced by basic science and clinical researchers, as well as those working in health services research. 13   Institute of Medicine. 1995. 14   National Research Council. 1978. 15   Ebert-Flattau, P. 1981. 16   An AHRQ-funded review of the curricula vitae (CV) provided by former AHRQ trainees and fellows revealed that 73 percent of the AHRQ T32 predoctoral trainees had earned a doctoral degree by 1998, and the remaining 27 percent was either in training or ABD. Of those who earned a doctorate by 1998, about three-quarters had earned them in a health science field, including health services research, a related multidisciplinary health field such as health policy, health administration, or public health or in one of the other health sciences. Over three-quarters of the T32 predoctoral trainees had earned their baccalaureate degrees in one of the sciences, with 38 percent in the social sciences, 19 percent in the health sciences, and 22 percent in other scientific fields, including the physical and mathematical sciences. 17   In addition to federal programs of support, private foundations such as the Pew Charitable Trust have played a significant role in promoting training in health policy and health services research. See Institute of Medicine, 1997. 18   Institute of Medicine. 1997. 19   Other clinical postdoctoral T32 trainees had specialized in pediatrics (16 percent) or family practice (8 percent). 20   Ibid.

OCR for page 79
Advancing the Nation’s Health Needs TABLE 7-1 AcademyHealth Ph.D. and Professional Degree Members’ Employment Sector, 2004 Employer Total % Ph.D. % M.D. % D.D.S. % Pharm.D. % J.D. % Total 3,748 100 1,524 100 713 100 15 100 23 100 91 100 University 2,023 54 951 62 427 60 14 93 18 78 43 47 Research/policy center 444 12 182 12 90 13 0 a 1 4 6 7 Corporation 355 9 92 6 63 9 1 7 3 13 12 13 Government agency 292 8 118 8 54 8 0 a 0 0 6 7 Association 164 4 41 3 7 1 0 a 0 0 10 11 Consulting firm 148 4 44 3 11 2 0 a 0 0 6 7 Foundation 131 3 41 3 15 2 0 a 0 0 3 3 International agency 15 a 5 a 4 1 0 a 0 0 0 0 Unknown 176 5 50 3 42 6 0 a 1 4 5 5 aLess than 0.5 percent. SOURCE: AcademyHealth Membership Survey. The employment opportunities and careers in health services research vary widely. Academic careers may be in schools of medicine, nursing, public health, and other health professional schools, as well as engineering and traditional arts and sciences departments (e.g., sociology, psychology, economics, political science). To effectively manage interdisciplinary research, academic institutions usually have organizational structures such as centers or institutes for health services research. At some institutions there are multiple centers reflecting different areas of specialization and the availability of funding for specialized centers from federal and private sources. Private-sector health services research careers are available in many areas. Federal contract work evaluating major public policy initiatives is primarily done by private research firms. These organizations include RAND, Mathematica, Abt Associates, Westat, and others. These organizations are organized to do large-scale studies that are not as easily managed in most academic settings. Other private-sector health services research careers are in research organizations sponsored by health maintenance organizations and health plans, hospital systems, pharmaceutical firms, insurers, and other major stakeholders in health care. Health services research positions may involve directing research, translating research into practice and products, and management evaluation of health care operations. Associations for professional groups, manufacturers, and advocacy groups recruit people trained in health services research to strengthen their capacity to use information coming from health services research for their members. As efforts to translate science into practice accelerate, the demand for individuals skilled in health services research and communication to users is likely to grow. Government agencies recruit substantial numbers of health services research professionals to lead and manage research programs, to support policy analysis and development, and to work with managers and providers in the Veterans Administration (VA) and Department of Defense (DOD) health care delivery systems. New career paths for health services research professionals may emerge as research into effective translation of knowledge into practice grows. For example, the recently passed Medicare prescription drug legislation mandates comparative effectiveness studies of health care services, including prescription drugs, increasing the need for health services researchers trained in pharmaco-economics. The development of tools and techniques to support translation is TABLE 7-2 Primary Field of AcademyHealth Members by Ph.D. and Professional Degree, 2004 Primary Field Total % Ph.D. % M.D. % D.D.S. % Pharm.D. % J.D. % Total 3,748 100 1,524 100 713 100 15 100 23 100 91 100 Health services research 1,457 39 764 50 322 45 6 40 13 57 4 4 Health policy 709 19 177 12 102 14 3 20 2 9 42 46 Both 1,203 32 517 34 202 28 4 27 4 17 27 30 Unknown 376 10 66 4 87 12 2 13 4 17 18 20 SOURCE: Academy Health Membership Survey.

OCR for page 79
Advancing the Nation’s Health Needs likely to become an industry that will require research skills in the design, evaluation, and testing of new technologies. Translation of knowledge for clinicians may be the initial priority, but priorities will likely expand to include managers, patients, and the public. The demand for well-trained health services researchers is currently strong and likely to continue to expand in the future. DEGREE PRODUCTION AND EMPLOYMENT Ideally, it would be useful to document the flow of individuals into the health services research market. However, graduate programs in health services research are not separately accredited and since the graduates could come from doctoral programs reflecting a wide variety of specialties, there is no accurate tally of doctoral students earning degrees in health services research. In 2004 AcademyHealth published a directory of health services research programs based on “self-identification,” which included 81 programs offering master’s or doctorate degrees. (Doctoral programs are mainly Ph.D. programs, including both disciplinary [e.g., health economics, medical sociology] and general training in health services research.) One source of information that reveals the complex educational history of the contemporary health services researcher workforce is an AHRQ-supported study of former NRSA trainees and fellows who received NRSA support from AHRQ between 1986 and 1997 and their employment situation as of December 1998.21,22 That study showed that these former trainees and fellows actively pursued research careers through a variety of employment paths. Most T32 predoctoral trainees who had completed their doctorates by 1998 did not pursue formal postdoctoral research training. Only 20 of the 102 individuals providing curricula vitae (CVs) reported pursuing postdoctoral research training following receipt of their doctorate. However, a majority (11 of 20, or 55 percent) had NRSA postdoctoral support. Employment information is available for 93 of the 102 individuals having received AHRQ T32 predoctoral support, and just over half (51 percent) accepted a position in an academic setting, another 29 percent took a first job in business/industry, and another 17 percent went to work in government following NRSA training. Of the 181 individuals who received AHRQ T32 postdoctoral research training support and provided CVs, 62 held research doctorates, and more than three-fourths only received AHRQ NRSA training support. By 1998, 58 of the former AHRQ T32 postdoctoral trainees with research doctorates were employed. The vast majority had accepted a first employment position in academia (44 of 58, or 76 percent), and the fraction grew slightly higher when their sector of employment in 1998 was analyzed (46 of 58, or 79 percent). A proportionately larger share of women held academic employment, 30 of 36 (83 percent) in 1998 than men, 16 of 22 (73 percent). On the other hand, of the 181 individuals who received T32 postdoctoral research training through AHRQ between 1986 and 1997 and provided CVs, 119 held clinical doctorates. Of these, two-thirds (78) relied solely on AHRQ for postdoctoral research training support. By 1998, 112 individuals had completed their postdoctoral research training. The first employment positions of these clinicians were chiefly in academia: 95 of 112, or 85 percent. However, it is interesting to note that another 10 percent took a first job in government (12 of 112). In 1998 the fraction working in government settings had grown to 20 percent (22 of 112) as had the fraction in industry/business (21 of 112, or 19 percent). Proportionately fewer individuals reported academic employment in 1998 (86 of 112, or 77 percent). Women holding clinical doctorates and having received AHRQ T32 postdoctoral training support were as likely (31 of 41, or 76 percent) as men (55 of 71, or 77 percent) to be working in academia. For AHRQ F32 fellows, 56 of the 57 trainees had been employed. The vast majority (42 of 56, or 75 percent) were employed in academia in 1998 regardless of degree type: 24 of 30 clinical doctorates (80 percent) and 18 of 26 research doctorates (69 percent). Over one-third of the T32 predoctoral trainees who were employed in 1998 reported having received grant/contract support (34 of 93, or 37 percent). Most of those with support reported having obtained it from private foundations (18 of 34, or 53 percent), although a number received support from NIH (14), AHRQ (9), or other U.S. Department of Health and Human Services (DHHS) agencies (13). By contrast, 60 percent of the postdoctoral Ph.D. T32 trainees who were employed by 1998 reported having had research grant/contract support (34 of 58). Besides having received support for research from private foundations (62 percent), some had received support from NIH (15 of 34), AHRQ (4 of 34), or other DHHS agencies (9 of 34). Similarly, 60 percent of the former T32 postdoctoral trainees with clinical doctorates who were employed in 1998 reported having received research grant/contract support (65 of 112). Most received support from private foundations (49 of 65). Other sources of research support included NIH (26 of 65), AHRQ (26 of 65), and/or DHHS (22 of 65). Of the AHRQ NRSA F32 fellows who were employed by 1998, over 70 percent received research grant and contract support (40 of 56). The majority of these grants and contracts were privately funded, as reported by 27 of 40 fellows. Of federal sources, most reported having received support from NIH (19 of 40), AHRQ (12 of 40), and other DHHS agencies (14 21   The AHRQ-funded study involved the collection of CVs from former NRSA trainees and fellows, the extraction of information about their educational and employment histories, and the analysis of career outcomes, including publication patterns. Information was summarized in The AHRQ Research Training and Career Development Databook: 1986 to 1998, (Agency for Healthcare Research and Quality, 2001). 22   The AHRQ Research Training and Career Development Databook: 1986 to 1998, 2000. Rockville, MD: Agency for Healthcare Research and Quality.

OCR for page 79
Advancing the Nation’s Health Needs of 40). In summary, the majority of these former NRSA trainees and fellows were actively engaged in health services research careers. OUTLOOK FOR HEALTH SERVICES RESEARCH FUNDING The broad relevance of health services research has contributed to federal funding through multiple agencies, unlike the funding of most other areas of health research. AHRQ research is expected to address cross-cutting access, quality, and cost issues that are faced by the entire American health care system. Other funding sources seek to fund health services research in support of their organizational missions. The VA and DOD focus on their delivery systems, the Center for Medicare and Medicaid Services (CMS) on financing Medicare and Medicaid, CDC on prevention, and NIH on delivery of services for specific diseases. These funding sources are complemented by private sources, including major foundations (e.g., Robert Wood Johnson Foundation, Commonwealth Fund, MacArthur Foundation, Kellogg Foundation, Kaiser Family Foundation, California Wellness Foundation) and private corporations. The following discussion will be limited to federal funding of health services research. In 2001 the Coalition for Health Services Research, the advocacy affiliate of AcademyHealth, began an initiative to document health services research funding levels across the federal government. The first report was completed in 2003, and now there are annual updates. The 2004 report found: From information provided to us by the following federal agencies, we estimate that $1.5 billion was expended for health services research and related activities by the federal government in Fiscal Year 2003. This total is distributed to the following agencies: Agency for Healthcare Research and Quality (AHRQ) —$309 million; Centers for Disease Control and Prevention (CDC): —National Center for Health Statistics (NCHS) —$126 million; —Extramural Prevention Research Program —$14 million; Centers for Medicare and Medicaid Services (CMS) —$74 million;23 National Institutes of Health (NIH) (All Institutes) —$873 million; Veterans Health Administration (VHA) —$52 million; and The Department of Defense (DOD) —$15 million. Given that these agencies do not use a standard definition or uniform categories to report their expenditures, questions remain about what is included in these totals. From the data reported by AHRQ, CMS and NCHS we know that a total of $191 million was spent to support data systems used in health services and health policy research. We also know that the NIH expenditures include both health services research and dissemination activities. The health services funding of $1.5 billion when compared to total federal health research funding of $34.3 billion in 2003 shows that approximately 5 percent is being devoted to health services research, based on classifications used in each agency and institute. AHRQ allocates its funding by major programmatic areas (see Table 7-3, which shows the distribution of funding for fiscal year 2004). The largest categories in the budget are data development and informatics. Informatics is primarily in support of patient safety and the evaluation of electronic health records systems. Adding informatics, quality/safety, pharmaceutical outcomes, and chronic disease management together represents over 45 percent of the research portfolio addressing quality-of-care concerns. Data development funds include the cost of data collection and analysis of the Medical Expenditure Panel Survey, the primary source of data on medical expenditure patterns in the United States. AHRQ’s budget covers other major areas of health services research, including prevention, socioeconomics of health care, long-term care, and bioterrorism. TABLE 7-3 Health Services Research Funding by Major Programmatic Area, Agency for Healthcare Research and Quality, Fiscal Year 2004 AHRQ Portfolios as a % of the Total Fiscal Year 2004 Budget Budget Line/Portfolio % of Total Budget Quality/safety of patient carea 10.5 Infomaticsa 21.0 Data developmentb 22.3 Chronic care management 9.7 Prevention 9.5 Bioterrorismc 0 Socioeconomics of health care 12.9 Pharmaceutical outcomes 4.8 Training 2.6 Long-term care 5.6 Organizational support 0.9 Total 100 aThere is a significant link between the quality/safety of patient care and the informatics portfolios. This budget is primarily the patient safety earmark. b18.2 percent of the data development portfolio is devoted to the Medical Expenditure Panel Survey. cAHRQ’s bioterrorism research is funded through support from the Office of Public Health Emergency Preparedness. This funding is reimbursable and is therefore not part of the agency’s appropriated budget. SOURCE: Agency for Healthcare Research and Quality. 23   Most of the funding in CMS’s research budget actually represents congressional earmarks for activities that are only remotely related to CMS’s research and demonstration interests.

OCR for page 79
Advancing the Nation’s Health Needs NIH institutes report funding health services research as shown in Table 7-4. NIMH, NIDA, and the National Cancer Institute have the largest programmatic commitments, ranging from 17 to 23 percent of their budgets. Other institutes report smaller budget commitments to health services research. Overall, AHRQ provides 20 percent of all health services research funding as reported by federal agencies. Other federal agencies support more focused program-specific and disease-specific health services research. Private funding of health services research is substantial, but no comprehensive source of information is available on nonfederal sources. THE NATIONAL RESEARCH SERVICE AWARD PROGRAM IN HEALTH SERVICES RESEARCH: THE AGENCY FOR HEALTHCARE RESEARCH AND QUALITY The NRSA program provides support for training in health services research. AHRQ receives funding equal to 1 percent of all NRSA funds for NIH. In 2003, AHRQ supplemented NRSA funding with $500,000 to fund a total of 82 predoctoral candidates and 69 postdoctoral fellows, plus six pre- and postdoctoral minority positions. In terms of success rates for training grant applications, the agency in recent years was able to fund only about 55 percent of the requested training positions. This is very similar to the rate for all NIH training awards. In addition, several NIH institutes provide TABLE 7-4 NIH 2004 Health Services Research Budget Estimate (Dollars in Thousands) Institute Health Services Research Budget Total % of Total Institute Budget % of NIH’s Total National Institute of Mental Health $208,543 15.10 23.20 National Institute on Drug Abuse $153,572 15.50 17.10 National Cancer Institute $151,094 3.20 16.90 National Institute on Aging $74,800 7.30 8.40 National Institute of Diabetes and Digestive and Kidney Diseases $70,936 3.90 7.90 National Institute on Alcohol Abuse and Alcoholism $65,000 15.20 7.30 National Heart, Lung, and Blood Institute $53,940 1.90 6.00 National Library of Medicine $23,606 7.60 2.60 Other National Institute of Health Services Research $94,442 30.30 10.50 Total $895,933       SOURCE: Coalition for Health Services Research (2004). Federal Funding for Health Services Research, Washington, D.C. NRSA awards in health services research, including NIMH, NIAAA, and NIDA and probably others. Overall, the total number of trainees is likely less than 2 percent of all NRSA training positions. No data are available on graduates of doctoral programs who plan to pursue health services research careers and are not funded by NRSA. It would be expected that these numbers far exceed NRSA recipients, as they do in other health research fields. While there is incomplete information on all individuals with training in health services research and those supported by NIH, since they are part of a larger training activity within several institutes, there is some information on NRSA trainees supported by AHRQ. In particular, the AHRQ-commissioned outcome study cited earlier24 documented AHRQ-supported 383 predoctoral and postdoctoral trainees through 24 university-based or university-affiliated T32 training sites. Of these, 160 represented T32 predoctoral institutional trainees and another 223 T32 postdoctoral institutional trainees. Another 67 AHRQ F32 individual NRSA fellowships were awarded, for a total of 450 trainees. The majority of AHRQ-supported NRSA trainees and fellows between 1986 and 1997 were female (265 out of 450, or 59 percent), a difference especially evident among T32 predoctoral trainees (115 of 160, or 72 percent) and F32 fellows (41 of 67, or 61 percent). There were slightly more males (114) than females (109) among T32 postdoctoral trainees during this period. No data are currently available on the characteristics of trainees since 1997, but it is known that about 450 have been supported by AHRQ through individual awards and 27 institutional training grants. Information on the trainees before 1997, which came from this review of current CVs, revealed that 377 (or 84 percent) of the 450 trainees provided their CVs and that 139 had T32 predoctoral support, 181 had T32 postdoctoral support, and 57 had F32 fellowships. By 1998, 102 (or 73 percent) of the AHRQ T32 predoctoral trainees had earned a doctoral degree, and the remaining 37 were still in training. Of those who earned a doctorate by 1998, about three-quarters (74) earned them in a health science field, including health services research (23 of 102, or 23 percent); related multidisciplinary health fields such as health policy, health administration, or public health (30, or 29 percent); or in one of the other health sciences (21, or 21 percent). Over three-quarters of the T32 predoctoral trainees had earned their baccalaureate degrees in one of the sciences, with 38 percent in the social sciences, 19 percent in the health sciences, and 22 percent in other scientific fields, including the physical and mathematical sciences. At the postdoctoral level, T32 training has proven to be a mechanism that attracts clinicians into a research career. Of the 181 AHRQ postdoctoral trainees providing CVs, two- 24   AHRQ, 2000. op. cit.

OCR for page 79
Advancing the Nation’s Health Needs TABLE 7-5 Field of Research Doctorate for Agency for Healthcare Research and Quality T32 Postdoctoral Trainees by Cohort Cohort (Start Year) N Social Sciences Health Sciences Other Research Doctorates 1986–1989 11 6 3 2 1990–1993 24 13 8 3 1994–1997 27 17 8 2   SOURCE: Tabulation for the Agency for Healthcare Research and Quality. thirds (119, or 66 percent) had earned a doctorate in one of the clinical professions prior to NRSA training, the majority of which were in internal medicine (62, or 52 percent). Other clinical postdoctoral T32 trainees specialized in pediatrics (16 percent) or family practice (8 percent). Among the 62 research doctorates pursuing postdoctoral research training through AHRQ NRSA T32 awards, half earned a doctoral degree in the social sciences. Furthermore, it appears that AHRQ T32 postdoctoral research training has increasingly attracted individuals from the social sciences since 1986 (see Table 7-5). The AHRQ NRSA F32 fellowships attracted individuals from both the clinical professions and the sciences into careers in health services research. Nearly half (26, or 46 percent) held a doctorate in a research field such as epidemiology, health policy, sociology, or psychology. The remainder (24, or 42 percent) earned a clinical doctorate in such fields as internal medicine or pediatrics and held both an M.D. and a Ph.D. degree prior to pursuing AHRQ fellowship training (7, or 12 percent). The majority (two-thirds) of individuals holding AHRQ F32 fellowships between 1986 and 1997 worked prior to receiving NRSA support. More than half (55 percent) had been employed in academia prior to F32 training, especially those with research doctorates. CONCLUSION Health services research relies on the knowledge and understanding of a broad spectrum of research fields, and the field is growing as both public and private research funding increase and as the breadth of research expands to include greater emphasis on intervention and translational research. Data show an increasing demand for researchers in this area and the need for increased training to meet the demand, both for students in doctoral programs and health professionals who bring specialized skills to the field. This training should include the knowledge and skills needed to function effectively as a member of an interdisciplinary team, and the NRSA program is the appropriate vehicle for this training. RECOMMENDATIONS Recommendation 7-1: Health services research training should be expanded and strengthened within each NIH institute and center. Biomedical research has created a growing gap between research advances in biomedical science and the ability to apply them effectively to improve the health of the public. Thus, there is a need for more effective health care delivery practices to ensure effective and evidence-based care and to reduce waste and unnecessary risk to patients. These issues are not particular to just a few NIH institutes and centers where training support for health services research is now focused, and the health services research would be better served if training occurred more broadly across NIH. Recommendation 7-2: AHRQ training programs should be expanded, commensurate with the growth in total spending on health services research. Recognition of the rising costs of care, together with concerns about quality and consistency, has driven increases in health services research. Health services research has established an important evidence base to enable patients and health care organizations to evaluate the benefits and risks of diagnostic and therapeutic intervention and to compare relative values of older and newer approaches as choices proliferate. This field can also evaluate different approaches to health care delivery and financing, which will allow the nation to benefit optimally from the dramatic advances in biomedical science. Training programs should grow proportionately with the need for individuals who have a wide range of disciplinary skills to conduct this research effectively.