Click for next page ( 106


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 105
5. Health Services Research Abstract Health services researchers have contributed to the understanding of factors affecting the effective- ness of health care, including methodological advances in the measurement of health status and in the conceptual underpinnings of cost-benefit and cost- effectiveness analysis. These efforts at quantifica- tion are relevant to the problems of constraining the rapid growth of health care costs and assessing the consequences of new forms of health care delivery. Support for health services research comes from diverse private, as well as government, sources. Philanthropic foundations have been a major source since the 1920s when the Committee on the Costs of Medical Care conducted studies of the incidence of disease, family expenditures for health services, incomes of physicians, and facilities for the provision of health services. Foundations with major health activities such as Robert Wood Johnson, Kaiser Family, Kellogg, and Hartford continue to provide support for the field. Research activities also are supported by professional societies and private business. Federal support of health services research has its principal focus in the National Center for Health Services Research and Health Care Technology Assessment, the Office of Research and Demonstrations of the Health Care Financing Administration, and the Veterans Administration. However, research programs and projects whose substance is health services research, but that are not identified as such, occur in NIT, the Department of Defense, and elsewhere in the federal government. State agencies for health and social services use the methodologies of health services research or contract with consultants and university-based researchers to evaluate health care practices within their jurisdiction. Efforts are needed to obtain systematic information on all public and private funds supporting health services research. Total federal expenditures were estimated to be $183 million in 1985; these expenditures are increasing but remain small relative to federal expenditures for biomedical R and D. 105

OCR for page 105
106 Unlike the behavioral, biomedical, and clinical sciences, data are not available that permit the committee to make quantitative estimates of the current supply of heal to services research personnel nor that support projections of future supply or demand. A much better base o f known ed ge is needed on the training, empl orient, and research activities o f heal to services researchers, and on the f and ing o f heal th services research . DEFINITION AND EXAMPLES OF CURRENT HEALTH SERVICES RESEARCH Health services research is a field of inquiry that addresses the structure and functioning of the health care delivery system. It is not a discipline in the sense of biochemistry or psychology, but rather a problem area in which are applied the theories and methods of the social and behavioral sciences, epidemiology, economics, biostatistics, and operations research. Some health services research is directly relevant to the evaluation of health programs and the development of health policy. Other research is focused on technology assessment or more theoretical studies addressing such issues as the optimal organization of health care delivery systems. Still other research has the aim of developing and improving data and methods for studying health services delivery. Investigators in this field employ a variety of research methods. Depending on the disciplinary background of the investigator and the aims of the research, a project might utilize, e.g., case analysis and randomized trials (medicine), interviews or questionnaire surveys (social sciences), observation studies (anthropology), empirical testing of theoretical models (economics), or experimental or quasi- experimental studies (behavioral sciences). Analytic techniques are drawn from biostatistics, epidemiology, econometrics, and statistics. Health services researchers also have developed research methods, of which health status measures (discussed below) constitutes an important example, and have made significant contributions to the development of cost-benefit and cost-effectiveness analysis. Health services researchers examine the influence of health care organization, methods of delivery, and health care financing on the quality, costs, and accessibility of health services. They also examine the development and deployment of health manpower. Ultimately their concern is with problems involved in the financing and provision of health services and with improving the effectiveness of those services as measured by improved treatment outcomes. Such lines of inquiry take on special importance during periods of major change in health care. The past two years have witnessed rapid and profound changes such as the institution of prospective payments for hospital payments by Medicare, the rapid growth of for-profit health care, the adoption of business-oriented goals by many health care providers, and the limitation of Medicaid patients' choices of providers by some states. Major employers such as General Motors have drastically changed their employee health insurance benefits to encour- age prudent use of health care resources. Physicians have started to form Preferred Provider Organizations and to enter other new organiza- tional arrangements such as free-standing surgical centers. These

OCR for page 105
107 changes have occurred within a context of the continued proliferation of medical technology, a greatly enlarged physician supply, and the increasing rate of growth in the nation's elderly population. Investigators in this field have made significant contributions to knowledge about the structure and function of the health system and have developed research tools to assess its effectiveness. Over the past 20 years, for example, substantial progress has been made in the development of measures and indices of health status. Investigators gators have developed aggregate indexes based on population mortality and morbidity. They also have developed measures that combine morbidity and mortality to construct a quality-adjusted life expectancy. These measures provide the means for monitoring health status in local, regional, or national populations. As an adjunct to economic measures such as per capita income or unemployment rates, health status measures provide indicators of population well-being. Individual health status measures also have been developed. For example, Katz and his colleagues at the Benjamin Rose Hospital developed the index of Activities of Daily Living (ADL), a-measure of patients' functional independence or dependence designed to study results of treatment and prognosis in the elderly and chronically ill (Benjamin Rose Hospital Staff, 1959~. More recently, investigators have developed measures that encompass a much broader range of physical, social, and psychological functioning (Brook, et al., 1979~. Even more subtle measures of health status are being introduced with the concept of individual preferences for specific health states or health outcomes (Lipscomb, 19821. Such measures are essential to the evaluation of health programs, the assessment of the effectiveness of alternative delivery modes, and the analysis of the outcomes of medical practice. This area of research continues to be important as the federal government, the states, and the private sector act to constrain the rate of growth in health care expenditures and new forms of health care delivery emerge that increase the need for tools with which to analyze the consequences. Health Maintenance Organizations and Health Care Costs Since the 1950s, health services researchers have made extensive study of Health Maintenance Organizations (HMOs) 2 to test the tThis discussion draws heavily on Bergner (1985) and Ware (1985~. 2The Health Maintenance Organization (HMO) provides a range of services to a defined population for a fixed annual or monthly payment. This form of medical care delivery, in contrast to fee-for-service, contains financial incentives to perform fewer services and to emphasize health promotion and disease prevention. Its proponents argue that, because of these incentives, the HMO offers the possibility for substantial cost savings in health care delivery (Luft, 1978~. Since 1973, the federal government has encouraged the development of HMOs with the dual objectives of (a) reducing costs through the widespread enrollment of a substantial fraction of the population in prepaid plans and (b) lowering costs more generally by competing with conventional insurers and providers (Luft, 1985~.

OCR for page 105
108 hypothesis that HMOs offer care at lower cost and to identify the sources of cost differences between HMO s and other providers. Much of this work has found lower per capita costs in HMOs than under conventional health insurance, although the evidence is much stronger for group-practice H~Os than for independent practice associations (IPAs)3 (Luft, 1985~. Lower hospital utilization has been shown to account for most of the difference. Enrollees in group practice HMOs have hospital utilization rates (days/1,000) about 30 percent less than those of comparison groups. The lower rates are due to fewer admissions rather than shorter lengths of stay (Luft, 1978, p. 1339~. However, the possibility that healthier individuals choose Amos could not be ruled out on the basis of these nonexperimental studies. This issue of self-selection has been addressed in two studies. A 1976 study compared the costs of providing services to members of a St. Louis HMO and a matched group who received care under fee-for-service and found similar rates of surgical utilization, significantly lower rates of non-surgical and overall utilization, and much higher rates of ambulatory utilization in the HMO members (Perkoff,-1976~. More recently, investigators at the Rand Corporation compared utilization among persons in Seattle who were randomly assigned to one of three health plans: a free fee-for-service plan, a fee-for-service plan with copayments, and free care in the Group Health Cooperative of Puget Sound (GHC). A random sample of voluntarily-enrolled GHC patients also was analyzed. The assigned GHC group had a somewhat higher rate of hospital utilization than the GHC control group {49 days per 100 compared with 38 days per 100~; however, imputed annual expenditures per enrollee were very similar for these two groups (Manning, et al., 19841. Whether assigned or voluntary, GHC enrollees had a rate of hospital admissions that was 40 percent less than the randomly assigned fee-for-service group. These findings suggest that self-selection has not markedly biased the results of earlier, nonexperimental studies and lend support to the group-practice model HMO as a lower-cost alternative to traditional methods of health care delivery. Health Status and Medical Care Utilization Expenditures for health care have grown enormously over the past 20 years. Between 1965 and 1983, current dollar expenditures grew almost tenfold. Adjusting for inflation, they more than doubled. Since the 1970s Victor Fuchs and others have called into question whether this increase has translated into better health for the American people. Not only have mortality and morbidity rates appeared not to be declining commensurate with the growth in national health care expenditures, but age-specific mortality rates in this country have compared unfavorably with other developed countries with lower per capita health expenditures. A major theme of Fuchs' book Who_ 3In an IPA, a physician group is paid on a capitation basis, but individual physicians are paid fee-for-service.

OCR for page 105
109 Shall Live was that--except for the very poor--"life-style" factors such as diet, exercise, smoking, and automobile driving were the major determinants of health and that therefore the marginal benefit from an additional dollar spent on health care was very small (Fuchs, 1974~. Investigations in the field of health services research have attempted to estimate statistically the relative contributions to health status made by health care and other factors. This issue is of great importance to policy deliberations because there is a range of strategies available for improving health--e.g., increasing the availability and accessibility of health services, encouraging health-enhancing behaviors (or discouraging health-detracting behaviors), improving environmental quality, and increasing job safety--and resources are constrained. In a 1969 study using state data, researchers found that a 1 percent increase in medical care expenditures per capita was associated with a small (0.1 percent) decrease in age-s~x-adjusted mortality for whites (Auster, et al., 1969~. Subsequent research (Silver, 1972) also suggested that medical care utilization exercised a negligible effect on mortality rates. However, a recent major study of small areas found higher health expenditures per capita to be associated with significantly lower mortality (Hadley, 1982~. SOURCES OF FUNDING FOR HEALTH SERVICES RESEARCH Non-government Sources From its inception in the 1920s, the field of health services research has received significant support from the private sector. Eight philanthropic foundations 4 supported the landmark work of the Committee on the Costs of Medical Care (1927-1932), which can be considered the principal origin of health services research in this country. The work of this committee included community surveys and other field studies aimed at producing a comprehensive picture of the incidence of disease and disability in the population, family expenditures for health services, the numbers and incomes of physicians and other service providers, and existing facilities for the provision of health services. This was the first time that such an ambitious attempt was made to establish a factual base for a broad consideration of health policy (Anderson, 1967, p. 19~. Foundations such as the Robert Wood Johnson Foundation, the Kaiser Family Foundation, the Hartford and Kellogg Foundations, continue to play an important role in the support of health services research. They have funded the work of major commissions whose work has included original research. They also have supported innovative health care programs as well as evaluation research to assess their effectiveness. 4The Carnegie Corporation, Josiah Macy, Jr. Foundation, Milbank Memorial Fund, New York Foundation, Rockefeller Foundation, Julius Rosenwald Fund, Russell Sage Foundation, and the Twentieth Century Fund.

OCR for page 105
110 They have funded pioneering investigations in medical care at univer- sities and provider organizations. In 1976, the contribution of foun- dations to health service research was estimated at over $26 million. Major professional associations support health services research in several ways; they collect, process, and disseminate data on their members which then serves as a resource for research by their own research staff members and by academic investigators. Major examples are the American Medical Association's Physician Masterfile and the data on hospital characteristics maintained by the American Hospital Association. Medical specialty societies have undertaken research on medical care quality and methods of assessing medical care. The American College of Surgeons (ACS), for example, developed a system for evaluating hospital surgical programs as early as 1918 (Flook, 1973, p. 100~. More recently, the ACS and the American Surgical Association jointly conducted a major study of surgical services and surgical manpower in the United States. The well-known SOSSUS study documented, among other things, the large number of non-surgeons performing surgery and the excessive numbers of physicians choosing surgical residencies (ACS and ASA, 1975, pp. 83-85~. In another instance, the American College of Radiology, with the support of NCHSR, conducted a pioneering study {1977) assessing the extent to which diagnostic radiologic procedures influenced medical decision-making. Industry involvement in health services research is increasing. The Blue Cross-Blue Shield Association, for example, supports research on health services utilization and financing. Large investor-owned health care firms such as the Hospital Corporation of America are providing funding for research in health care administration. Government Sources Significant involvement in health services research by the federal government dates from the 1930s. The first Health Interview Survey, covering over 700,000 households, was conducted by the Public Health Service in the winter of 1935-1936. Data from this survey, which continues to the present, were used by PHS staff to study aspects of the organization, financing, and evaluation of health services over a Am__ / T;l' ~1' ~ O70 _ ~ t\O ~ unease `r 'w~^ ~ '~ ~ ~ ~ ~ . eve J . Several researchers who served on the staff of the Committee on the Costs of Medical Care subsequently joined the Social Security Administration's Office of Research and Statistics (Fox, 1979, p. 29~.s This office developed estimates of national expenditures for health and became the principal locus for intramural and extramural research on the economics of health care. 5Agnes Brewster, I. S. Falk, Margaret Klem, and Louis Reed. This activity was first headed by Ida C. Merriam, the Assistant Commissioner for Research and Statistics.

OCR for page 105
111 The National Center for Health Services Research and Health Care Technology Assessment The National Center for Health Services Research was established in 1968 with an explicit mandate to support health services research and research training. 6 Since its inception, the center has maintained an extensive program of extramural and intramural research and supported the training of researchers. Through these programs the Center n . . seeks to create new knowledge and better understanding of the processes by which health services are made available and how they may be provided more efficiently, more effectively, and at lower cost" (NCHSR and HCFA, 1985~. NCHSR is the primary source of federal support for research on problems related to the quality and delivery of health services. The NCHSR extramural program provides support for investigator- initiated projects in health services research conducted at universities, nonprofit organizations and institutions, and by industry. Priority areas for 1985 include: {1) Health promotion and disease prevention: health status measurement, organization, and provider studies, analysis of public and private program interventions, and methods to increase consumer knowledge and change health attitudes and behavior. (2) Technology assessment: studies of the safety, efficacy, effectiveness, and cost effectiveness of specific technologies, development of new methods for evaluating medical technologies, and diffusion of medical technology. (3) The role of market forces in the delivery of health care: market and industry structure, expenditure studies, strategies to enhance cost consciousness, and productivity studies. (4) Primary care: development and testing of better designs, measures, and analytic techniques to improve primary care research; evaluation and surveillance techniques to assess the quality of care and the effectiveness of health promotion and disease prevention efforts, studies of the medical decision-making process; and systematic evaluations of the effectiveness and costs of clinical care. 6Section 304 of the Partnership for Health Amendments of 1967 autho- rized support for research, experiments, and demonstrations related to the "Development, utilization, quality, organization, and financing of services, facilities, and resources" (Sanazaro, 1973, p. 152~.

OCR for page 105
112 (5) State and local health problems: methods for projecting the demand for service and related supply requirements, and forecasting health expenditures, studies to develop and evaluate decision models for allocating health care resources at various jurisdictional levels and among various programs in a cost-effective manner; and techniques to assess and project the impact of changes in health expenditures (NCHSR, 1984b). In its intramural research program the Center emphasizes four major health care issues. The Hospital Studies Program examines how competi- tion, reimbursement systems, and various types of regulation influence the use and costs of hospital care. The Health Services for the Aged Studies Program evaluates the impact of different reimbursement approaches on the admission practices and services of nursing homes, the feasibility of private, long-term care insurance, and the contribu- tion of informal support systems for the elderly. The National Health Care Expenditures Study, using information from a large national survey, examines how Americans use and pay for health care services. The Health Status and Health Promotion Studies Program focuses on measuring the level of health and on evaluating strategies to modify behavioral practices that have an adverse impact on health status. The National Medical Care Expenditure Survey, conducted in 1977, has provided a rich source of data yielding significant findings on the utilization of health care and how families finance their health care. This survey included interviews in approximately 14,000 households, complemented by additional surveys of physicians and health care facilities providing care to household members during 1977 and of employers and insurance companies about their insurance coverage. On the basis of data from the survey it has been estimated that three of every 10 dollars spent for health care are accounted for by persons whose activities are limited by chronic conditions, although these persons represent only 10 percent of the population. The Medicare and Medicaid programs pay a large share of their health care costs, including about half of their hospital care (NCHSR, 1984a) These data also have been used to estimate the proportion of the insured population that is underinsured, that is, that could face significant out-of-pocket expenses over and above their insurance coverage. For approximately five percent of the privately insured population under age 65, expected out-of-pocket expenses could exceed three percent of income (Farley, 1984~. These estimates highlight the large number of Americans who face substantial financial risk because they are uninsured for all or part of a year, or because their health insurance is not sufficient to cover certain expenses for which there is an appreciable statistical expectation. Center-funded research also has made significant contributions to the knowledge base for the design, implementation, and evaluation of Medicare prospective payment, including early support of efforts by Yale researchers to develop diagnostic groups that are homogeneous with respect to hospital resource use and continued efforts to improve measures of case mix/case severity. Research funded by NCHSR also improving data and .

OCR for page 105
113 addresses the effects of DRG-based prospective payment on the quality of care and access to care. Annual research appropriations for the Center have leveled at approximately $15 million after declining for a number of years (Table In real terms, this represents a decline of 67 percent since 5.1) 1976. A small amount is allocated for the support of dissertation research. Between 10 and 20 grants of up to $20,000 are made annually to promising students whose dissertation topics are within Center priority areas. At present the Center has no other program for training of health services research personnel. . - TABLE 5.1 Annual Research Appropriations for the National Center for Health Services Research, FY 197~86 ($ millions) Research Appropriations Implicit GNP Fiscal Price Deflatora Year Current $ 1972 $ (1972 = 100.0) 1976 $26.0 $19.4 133.90 1977 24.0 16.9 141.70 1978 26.1 17.2 152.05 1979 26.1 15.8 165.46 1980 22.4 12.6 178.42 1981 ~ 21.5 11.0 195.14 1982 14.3 6.9 206.88 1983 14.6 6.8 215.63 1984 15.7 7.0 223.43 1985 14.8 6.4 232.29 1986 (proposed) 14.7 n/a n/a a From the U.S. Bureau of the Census. The deflator for 1985 represents the third quarter. SOURCE: National Center for Health Services Research. Health Care Financing Administration (HCFA) As the agency responsible for managing Medicare, Medicaid, the End Stage Renal Disease Program, professional review, and their accompany- ing statistical and monitoring activities, HCFA, through its Office of Research and Demonstrations (ORD), supports research and demonstrations related to these responsibilities. ~ ~ ~ _ . . Research areas include hospital `I=l~, e~`y~l~lun ~=yluenc, 1ong-cerm care, name health care, and alternative payment systems. In addition, this office supports program analysis and evaluation, including the development and analysis of program data and data from major health surveys, review of state Medicaid programs' management information needs, and the compilation and dissemination of state health activities (HCFA, 1984~.

OCR for page 105
114 To support and extend the activities of HCFA, it supports two health policy centers. The Brandeis University Health Policy Research Consortium, which includes the Boston University School of Medicine, the Center for Health Economics Research, and the Urban Institute conducts a broad array of analytic activities and recently has assisted the Office of Research in responding to a Congressional mandate for reports on the Medicare prospective payment system (P.L. 98-21~. A second center at The Rand Corporation/University of California, Los Angeles, primarily is providing expert consultation to ORD in planning, implementing, and evaluating demonstrations and experiments. This center also is supporting the analytic activities of the Office of Research in response to Congress. The enactment of prospective payment for hospital services represented a radical departure from historical methods of paying for hospital care. However, this legislation (P.L. 98-21) was based on research, demonstrations, and evaluations over a period of more than ten years. Most of this painstaking work was supported by HCFA. In the late 1970's ORD funded the development by researchers at Yale of Diagnosis-Related Groups, a classification scheme comprised of subgroups of patients that have similar clinical attributes and resource utilization patterns (Fetter, et al., 19801. The algorithm that was developed grouped patients in a manner that minimized within-group variation in length of stay while keeping the number of groups to a manageable levels HCFA also funded a demonstration of the use of per-case payment for hospital care under Medicare in the state of New Jersey using DRG's to define cases. All general acute care hospitals in the state were phased into the demonstration, starting in 1980. All hospital patients in the state and all third-party payers were included (HCFA, 1984~. Preliminary results from this demonstration formed the basis for a 1982 report to Congress on prospective payment under Medicare that led to the adoption of prospective payment in 1983. The ORD budget for research and demonstrations was $34 million in 1985 (Table 5.2), approximately evenly divided between research and demonstrations. In enacting prospective payment Congress called for HCFA to conduct studies and deliver a number of reports on the implementation and effects of this major change in payment for hospital services, as well as a report on the advisability and feasibility of incorporating physician payments into prospective payment. Fulfilling these Congressional mandates currently occupies about 30 to 40 percent of HCFA's research and demonstration resources. National Institutes of Health While NIH does not separately identify grants for health services research, such research activity can be found in a number of institutes, primarily in comprehensive centers and control programs for cancer, diabetes, arthritis, and cardiovascular and pulmonary diseases (IOM, 19791. The NIH biennially compiles information on federal obligations for the conduct of health research and development

OCR for page 105
115 TABLE 5.2 Research and Demonstrations Budget for the Health Care Financing Administration, FY 1979-86 ($ millions) Budget Fiscal Year Current $ 1972 $a 1979 $32.5 $19.6 1980 46.8 26.2 1981 38.9 19.9 1982 29.5 14.3 1983 30.0 13.9 1984 33.1 14.8 1985 34.0 14.6 1986 (proposed) 34.0 Ma a Deflated by the GNP Price Deflator, 1972 = 100.0. See Table 5.1. SOURCE: Office of the Director, Office of Research and Demonstrations, HCFA. and other health-related and to Congress (NIH, own health services {biomedical R and D, health services R and D, R and D) that are reported in NIH publications 1983b)., In developing the information on its _ research activities, NIH employs a keyword analysis of its computer- based CRISP system. For FY 1985, NIH obligations for health services research were estimated at $42.3 million (Table 5.3~. This amount has risen steadily since 1979. 7The instructions for reporting that are provided to federal agencies employ the following definition for health services R and D: The structure, processes, and effects of health services, and development and use of health resources. Examples of areas to be included are: (a) analysis of the organization, delivery, and impact of health promotion and disease prevention activities; (b) analysis of the factors underlying the increase in health care costs and the structural reforms and incentives which might modify these; (c) analysis of the implications of various health insurance and financing initiatives; (d) analysis of health manpower, such as education, requirements, distribution, utilization, and development (but excluding the actual training of such manpower); (e) analysis of technology-based approaches to modify the organization and delivery of health care services, with special emphasis on the uses of computer science and medical and information systems (excluding research on the effectiveness of diagnostic and therapeutic technologies); (f) relationship between the health services provided, and the health of the population; (g) analysis of emergency medical service system; (h) R and D on portable field units for emergency care, including adaptation of design and instruments for specific use; (i) analysis of long-term care services; (j) evaluation of health services R and D.

OCR for page 105
116 TABLE 5.3 NIH Obligations for Health Services Research and Development, FY 1977-85a ($ millions) Obligations . . Fiscal Year Current $ 1972 $b . _ 1977 $ 7.4 $ 5.2 1978 11.6 7.6 1979 30.3 18.3 1980 32.3 18.1 1981 37.3 19.1 1982 37.3 18.0 1983 39.7 18.4 1984 41.0 18.4 1985 42.3 fix ~ a See footnote 7 for definition of Health Services R&D. The FY 1985 figure is based on President's budget request. b Deflated by the GNP Price Deflator, 1972 = 100.0. See Table 5.1. SOURCE: National Institutes of Health. Other Federal Sources The National Center for Health Statistics, the primary agency for the production of national general purpose health statistics, conducts surveys and inventories that form the basis of both descriptive and analytic studies. The center also conducts research to enhance the quality of survey data and improve estimation methods. Other agencies of DHHS that fund health services research are the Alcohol, Drug Abuse and Mental Health Administration, the Health Resources and Services Administration, the Office of the Assistant Secretary for Health, and the Office of the Assistant Secretary for Planning and Evaluation. Outside DHHS, health services research is funded by the Department of Defense, the Department of Education, the International Development Cooperation Agency (AID), and the Veterans Administration. Estimated total federal obligations for health services research for 1985 were $183 millions (Table 504~. ~ As the information presented above indicates, a diversity of sources fund health services research, including foundations and industry as well as government. The major focal points for health services research in the federal government are the National Center for Health Services Research and the Office of Research and Demonstrations of the Health Care Financing Administration. The Veterans Administration also conducts a small health services research program. However, programs whose substance is health services research --but which are not called health services research--occur in other government offices and agencies. In order that this committee and others can properly assess historical trends in funding for health services research and to assess the outlook for its future funding, Win comparison, biomedical R and D obligations were over $5.6 billion.

OCR for page 105
117 TABLE 5.4 Federal Obligations for Health R and D, by Agency and Type of Research, FY 1985 ($ thousands) Health Biomedical Services Other Health- Agency Total R&D R&D Related R&D TOTAL, All Agencies 6,274,776 5,615,477 183,177 476,122 Dept. of Health and lIuman Services, Total 4,930,294 4,720,833 128,166 81,295 National Institutes of Health 4,345,429 4,303,159 42,270 Other Public Health Service Agencies, Total 541,865 417,674 43,191 81,000 Alcohol, Drug Abuse, and Mental Health Adnun. 355,563 332,792 22,771 Centers for Disease Control 83,982 83,982 Food and Drug Admin. 81,000 - 81,000 Health Resources and Services Admin. 3,920 900 3,020 Office of the Assistant Secretary for Health (including NCHS and NCHSR) 17,400 17,400 Other DELIS, Total 43,000 42,705 295 Health Care Financing Admin. 35,000 35,000 Office of the Secretary 8,000 7,705 295 Other Agencies, Total 1,344,482 894,644 55,011 394,827 Dept. of Agriculture 147,558 27,417 296 119,845 Dept. of Commerce 4,073 733 - 3,340 Dept. of Defense, Total 473,059 410,351 13,340 49,368 Dept. of the Army 332,499 273,396 12,016 47,087 Dept. of the Navy 69,570 67,113 176 2,281 Dept. of the Air Force 50,059 50,059 - _ Defense Agencies and Service Schools 19,383 19,383 Other DOD 1,548 400 1,148 Dept. of Education 30,821 30,821 Dept. of Energy 178,116 106,942 71,174 Dept. of the Interior 16,977 16,977 Dept. of Labor 5,075 5,075 Dept. of Transportation 7,924 7,268 656 Consumer Product Safety Commission 709 466 243 Environmental Protection Agency 51,295 30,777 20,518 International Development Cooperation Agency (AID) 36,992 32,451 4,541 National Aeronautics & Space Admin. 113,883 34,951 78,932 National Science Foundation 83,500 68,804 14,696 Veterans Admin. 194,500 174,484 6,013 14,003 SOURCE: National Institutes of Health. these government bodies are encouraged to identify those programs and projects that are health services research. Efforts also should be made to develop an approach for systematically obtaining information on health services research funding by private industry, foundations, and state governments. THE MARKET OUTLOOK FOR HEALTH SERVICES RESEARCH PERSONNEL Data are not available that would allow the committee to make quantitative estimates of the current supply of health services research personnel, nor that would support projections of future

OCR for page 105
118 supply or demand. This situation is markedly different from that in the biomedical, behavioral, and clinical sciences. In those areas, (a) the participants fall into distinct disciplines which enable them to be identified and counted, (b) data on sources of funds supporting research are routinely available, (c) employment of researchers is concentrated in well-defined academic departments. In addition, the federal government and organizations such as the ALEC and APA have made major investments to develop data on research personnel and the institutions that employ them. The committee has been fortunate to be able to draw on these data sources in order to analyze supply and demand for these fields. The committee encourages the development-of a base of knowledge on the training, employment, and research activities of health services researchers, and on the funding of health services research. Such data are necessary for the quantitative assessment of the market for these investigators by this committee and others, and also could contribute to a qualitative assessment of the "match" between the problems addressed by health services researchers and the qualifications of members of the field. The research agenda will of necessity have to take into account the diversity of training among health services research personnel, the multiple sources of research funding for the field, and the nature of employment that includes government and private industry as well as academia. These characteristics set health services research apart from the other fields for which this committee makes recommendations and greatly complicate the development of systematic information. At the same time, they are characteristics that are not unique to health services research but are common to applied, multidisciplinary areas such as area studies, urban studies, and population studies. The research agenda should be developed with the participation of a broad representation of interested organizations such as the major federal funding sources for health services research (NCHSR, HCFA, NIH, VA), private foundations that have provided significant support for the field, academic and non-academic employers of health services researchers including health services research centers, the Association for Health Services Research, and other relevant professional organizations. The research agenda can draw upon the past work of this committee's health services research panel, which gave considerable thought to the merits of various approaches for improving the information base in health services research. Health Services Research Centers In some universities, a center serves as the focal point for health services research. In 1984 there were 38 academic health services research centers" according to the Association for Health Services 9Defined as "an organization or entity whose primary mission is the conduct of health services and policy research by a multidisciplinary staff, which is either based in or formally affiliated with an academic institution" (AHSR, 1983~.

OCR for page 105
119 Research (Table 5.5~. These centers vary widely in size, organiza- tional location, sources of funding, and training capabilities. Anecdotal evidence suggests the need for an assessment of center funding and study of the factors characterizing effective centers. Of particular concern is support of research which has a long-range orientation, especially research on methods and concepts and research that illuminates fundamental health-related behaviors of institutions and individuals. Individual investigators as well as enclaves of health services researchers are found in departments of political science, social and behavioral science, economics, epidemiology, biostatistics, operations research, nursing, medicine, and surgery. They are also found in departments of community medicine, maternal and child health, health education, health policy and management, and health administration, departments that often are multidisciplinary and may share characteristics in common with centers. TABLE 5.5 Characteristics of Health Services Research Centers, 1983 Number of Centers Number of Full-Time Employees Budgeta Organizational Location Office of the President, Chancellor, or Vice President Graduate School of Business, Management, Public Administration, Social Welfare School of Medicine School of Public Health Other Sources of Funding Federal Government State/Local Government Private Foundation Corporation Parent University Endowment Income Other Training Capability None Internships Predoctoral fellows Postdoctoral fellows Other - a Based on 36 centers reporting. 38 3-71 $120,00~$5.5 million 6 5 9 8 10 28 12 27 18 16 6 3 12 9 13 15 SOURCE: Association for Health Services Research (1983).

OCR for page 105
120 The committee applauds the survey of investigators associated with academic health services research centers that is being conducted by the Association for Health Services Research. This survey represents a significant step in that not only will it provide a picture of a very important subpopulation of health services researchers, but also be invaluable in the development of a broader research strategy. The data from this survey should become available in 1986. Training for Health Servicm Research As this committee stated in its 1983 report, "A competent principal investigator in health services research must have two sets of qualifications. The first is an adequate grasp of a discipline or professions [and] the second...is an understanding of...the delivery and financing of health care and a mastery of suitable research methods (IOM, 1983b, p. 121~." An indication of the diversity of disciplines/professions among health services researchers can be gleaned from the results of the 1978 survey of former principal investigators on NCHSR research grants (Table 5.6) and NCHSR trainees (Table 5.7~. The distribution of health services researchers by discipline probably would be different today. Too, these data did not represent investigators who received support from HCFA, ADAMHA, NIH, or other sources. Based on the 1985 estimate (Table 5.4) of $183 million in federal obligations for health services research, NCHSR support represents approximately 8 percent of the total. The second set of qualifications can be acquired (1) through formal coursework and research experience during predoctoral training, including dissertation research, {2) through formal postdoctoral training, or (3) informally, through research experience gained after completing graduate training. Early contributors to health services research came from this latter group, ~switching" to health services research from clinical medicine, public health, or the social sciences. On the basis of the committee's 1978 survey of health services researchers, it appears that newer entrants to the field are more likely to have had formal training in it. Part of a research agenda on health services research personnel should be an assessment of the appropriateness of training. lathe committee listed as examples anthropology, sociology, psychology, economics, political science, biomedical and clinical sciences, public health, epidemiology, biostatistics, operations research, health administration, health education and public administration.

OCR for page 105
121 TABLE 5.6 Field of Highest Degree Reported by NCHSR Principal Investigators, FY 1978 Field of Highest Degree TOTAL Number of Individualsa 398 Total Behavioral Sciences Anthropology Psychology Sociology Total Social Sciences Economics/Econometr~cs Political Science Other Total Biomedical Sciences Biometrics/Biostatistics Other Total Medical Sciences Public Health and Epidemiology Nursing Other Total Other Fields Bioengineering Operations Research Public Administration Other Total Medical Doctorates 60 3 14 43 41 30 5 6 20 13 7 36 11 7 18 08 o 2 2 104 133 aExcludes full-time degree candidates. SOURCE: National Research Council (1977a). The committee reaffirms its position that health services research is an important field that offers significant potential for increasing understanding of health care. The field's importance is even greater in this time of rapid and profound changes in the organization and financing of health care and the proliferation of medical technology. To maintain an adequate pool of qualified investigators to address questions of the quality, cost, and effectiveness of health care in the future, the committee recommends that NRSA awards be made specifically and explicitly for health services research training at levels of support set out in Chapter 1.

OCR for page 105
122 In the early 1970s, the federal government provided support to over 800 health services research trainees and fellows (NRC, 1975-81~. By 1981, this number had dwindled to zero--neither the NTH, ADAMHA, nor the HRSA was supporting any extramural training in health services research. This committee has recommended that these training programs be restored to about the 1976 level. In addition, the committee continues to endorse the dissertation grant program of NCHSR as an effective means for increasing the pool of health services research personnel. TABLE 5.7 Field of Highest Degree Reported by NCHSR Trainees, FY 1978 Field of Highest Degree TOTAL Total Behavioral Sciences Anthropology Psychology Sociology Total Social Sciences Economics/Econometrics Political Science Other Total Biomedical Sciences Biometrics/Biostatistics Other Total Medical Sciences Public Health and Epidemiology Nursing Other Total Other Fields Bioengineering Operations Research Public Administration Other a Excludes full-time degree candidates. SOURCE: National Research Council (1977a). Number of Individualsa 565 236 18 32 186 76 36 16 24 21 20 1 131 45 10 76 101 1 6 5 89