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 25
2 Overview of Health Services Research The field of health services research is just one outgrowth of scientific discoveries and technological innovations that have, in this century, transformed the understanding of the determinants of health and disease and allowed health workers to diagnose, prevent, and treat many conditions that once promised only death, disability, or discomfort. This transformation in the capabilities of medical care has provoked complex and continuing changes in the organization, cost, financing, governance, and social significance of health care in the United States. Like higher income and education, greater access to health care has come to distinguish society's "haves" from its "have-nots." Health insurance has become the most valued employee benefit and a growing financial worry for many employers (IOM, 1993b). Efforts to assure universal access have created periodic upheavals in the political landscape, as have a succession of varied strategies to control spending on health care. These social, economic, and political developments have, in turn, helped to influence the direction and rate of changes in biomedical science and technology. Modern health services research was essentially born of demands for better understanding of how services are organized, financed, and delivered and with what consequences. Recently, the health care system has been dramatically restructuring itself. Governments and employers have intensified their efforts to direct people into managed care plans that control patient access to practitioners and services, pay providers fixed amounts for a defined set of services to a defined set of patients, and otherwise manage the use and cost of care. Health care organizations are 25
OCR for page 26
26 / HEALTH SERVICES SEARCH consolidating, expanding, shrinking, reorganizing, or otherwise changing in ways that are varied and often complex, and the power of specific institutions and professions is in flux. Health services research has played and will continue to play a central role in helping to document, assess, and explain these changes and their effects and to inform public and private decisions during a period of rapid change. A BRIEF LOOK BACK Although an extensive historical review was beyond the scope ofthis project, a brief look back is useful in understanding the field of health services research as it exists today. Not surprisingly, given the relatively imprecise boundaries of the field of health services research, any effort to trace its history involves some subjectivity, especially the further back in time one goes. Although the emphasis here is on the United States, it is clear that health services researchers from other countries have advanced the field in general and influenced American researchers. (The history of health service research education and training programs is considered in Chapter 4.) The following historical overview notes several landmarks in the development of the field, including both advances in methodology and changes in the problems targeted for study.' It points out some of the contributions of health services research to policymaking or at least to policy debates. The review also highlights the critical role that federal funding for health services research has played in the development of the field. Before 1965 Early instances of what may be broadly viewed as health services research tended to involve fairly simple descriptive surveys of personnel, facilities, or diseases. Starting with Michigan in 1883, states began to require the reporting of certain infectious conditions, and Congress in 1893 authorized the collection of such information from states and localities. Private organizations also became involved, as exemplified in the 1908 Chicago Medical Society birth record survey of midwifery, the first American Medical Association (AMA) survey of hospitals in 1909, and the 1914 AMA survey of state boards of health. Beginning in 1906 in Rochester, New York, dental inspections of school children 'This discussion draws on Somers and Somers, 1961; Anderson, 1968; PSAC, 1972; Flook and Sanazaro, 1973; NRC, 1977; IOM, 1979; Starr, 1982; Rorem, 1982; Shortell and Reinhardt, 1992; IOM, 1993b.
OCR for page 27
OVERVIEW OF HEALTH SERVICES RESEARCH / 27 generated data that helped build understanding of the prevalence arid variability of oral disease. These and similar kinds of data collection activities began to make clear the degree to which disease was associated with poverty. In the 1920s, the U.S. Public Health Service (PHS) continued investigations into the relationship between illness and income. During the early years of this century, Emest Codman pioneered empirical studies of the quality of medical care in hospitals. He argued that "hospitals, if they wish to be sure of improvement, must find out what their results are, must analyze their results, must compare their results with those of other hospitals" (Codman, 1914, cited in Fifer, 1990, p. 2943. Codman analyzed 337 cases of hospital care, identifying errors of knowledge, judgment, and skill. Codman's analyses and arguments were highly influential on organizations such as the American College of Surgeons, although a concerted emphasis on evaluating care by its results or outcomes did not really take hold until relatively recently. The work of the Committee on the Cost of Medical Care (CCMC) in the late 1920s and early 1930s became an early landmark in health services research, a truly multidisciplinary project in both its methods and its scope. With a research budget of $1 million, an astonishing figure for the time ($8.5 million in today's dollars), the 42-member committee and its 75-person staff produced 27 field studies and a final report between 1928 and 1932 (CCMC, 1932~.2 The CCMC studies and associated surveys provided the first extensive investigations of such matters as the incidence of illness and disability; the levels and kinds of care involved in treating specific illnesses and the implications for health work force and facilities planning; the distribution and organization of health care services; the level and types of health care expenditures; the uneven burden of medical expenses and access to medical care; and the development of group practice, prepayment, and other innovations in organizing and financing care. The scope and depth of the CCMC analyses have, arguably, not been approached again, and many subsequent analyses have come to generally similar conclusions, even several decades later. In the 1930s, the federal Works Project Administration undertook the National Health Inventory, which was intended both to document health status in urban areas and to provide jobs for unemployed Americans. Except for limited state and supplementary studies, nothing as extensive was attempted until 2Six private foundations helped initiate the work of the CCMC: the Carnegie Corporation, the Josiah Macy, Jr., Foundation, the Milbank Memorial Fund, the Russell Sage Foundation, the Twentieth Century Fund, and the Julius Rosenwald Fund. Other organizations, including the AMA, also helped support the study with funding, information, or other assistance.
OCR for page 28
28 / HEALTH SERVICES SEARCH 1956 when the National Health Survey Act (P.L. 84-652) authorized a continuing survey program supplemented by methodological and special topic studies. In the private, nonprofit sector, the Health Insurance Plan of Greater New York and three Kaiser Pennanente regions pioneered health services research in the group practice setting after the end of World War II (Denser, Shapiro, and Einhorn, 1959; Greenlick, Freenborn, and Pope, 1988~. Beginning in the 1950s, some Blue Cross and Blue Shield plans also undertook applied health services research, for example to understand patterns of hospital use (Koen, 1965; Young, 1965). One of the early federal government efforts to apply health services research in a systematic way to a problem in medical care organization came in the 1950s when funding was provided for internal PHS studies in hospital administration and for demonstration projects to improve hospital organization and design. These projects had been authorized by the 1948 Hospital Survey and Construction Act (P.L. 79-725), more commonly known as the Hill-Burton Act. In 1959, the Hospital Facilities Study Section of the PHS became the Health Services Research Study Section. In 1963, the Bureau of State Services in the Department of Health, Education and Welfare (DHEW) consolidated several research efforts and began the first organized extramural research program in community health services. During this period, some health services research was also being sponsored by the National Institutes of Health (NIH). Indications that a coherent research field was emerging came with the creation of organizations, journals, and similar symbols. Within the American Public Health Association (APHA), the Medical Care Section (established by 1963) included a large contingent of health services researchers. The free- standing Association for Health Services Research (AHSR) was not, however, founded until 1981, when researchers recognized the value of having a focused voice for increased research Finding (Gray, 1992) and a forum for bringing those with an interest in health services research together to share ideas and research. Journals launched in the early 1960s included Medical Care (sponsored by the corresponding section of APHA), Inquiry (created by the Blue Cross and Blue Shield Association), and Health Services Research (established by Hospital Research and Education Trust, the research arm of the American Hospital Association). All are still being published today. In addition, more specialized journals often publish work that is based on health services research. These include the Journal of Health Economics, Journal of Clinical Evaluation, Journal of Quality Improvement, and Journal of Outcomes Management, as well as prominent clinical journals such as the New England Journal of Medicine and the Journal of the American Medical Association.
OCR for page 29
OVERVIEW OF [IEALTH SERVICES RESEARCH / 29 1965 and After Federally Sponsored Research In the 1960s, the federal government launched many significant initiatives in health care organization and financing, most notably Medicare and Medicaid. This expansion of government responsibility and spending focused policymakers' attention on the gaps in their knowledge about health care delivery and financing and about health behavior. It highlighted the need for efforts to evaluate the impacts of government programs and the factors contributing to the strengths, weaknesses, and unanticipated consequences of these programs. In addition, 1966 legislation authorizing the comprehensive Neighborhood Health Center component of the Office of Economic Opportunity (OEO) launched a large number of community health centers and a subsequent OEO effort to evaluate their community impact. National Center for Health Services Research. In 1965, the White House Conference on Health recommended the creation of a National Institute of Community Health Services to coordinate and stimulate health services research related to these and other federal health interests. Subsequently, the 1967 Report to the President on Medical Care Prices recommended a role for a national center in conducting research on cost containment, prepaid group practice, and community health care systems (Flook and Sanazaro, 1973~. In 1967, Congress enacted P.L. 90-174, which authorized the PHS to undertake a range of research, demonstration, and training activities. Under this general authority, the Secretary of DHEW established the National Center for Health Services Research and Development (NCHSR).3 NCHSR consolidated research activities from several units of DHEW, including those units concemed with community health service, medical care administration, and hospital and medical facilities. In addition, NCHSR established several health services research centers through contracts with academic or other outside organizations, including Harvard University, the University of California, Los Angeles; the University of North Carolina; and Kaiser Permanente in Northern and Southern California. Successor agencies have continued to use the "centers" concept in various forms with various participants. fin conjunction with various governmental reorganizations, the National Center was variously renamed the Bureau of Health Services Research, the Bureau of Health Services Research and Evaluation, and the National Center for Health Services Research (IOM, 1979~. The National Center was absorbed into the Agency for Health Care Policy and Research in 1989.
OCR for page 30
3 0 / HEALTH SER VICES RESEARCH After 1969, NCHSR increasingly focused its resources on relatively short- term research and development (R&D) projects rather than on longer-term research. The goal was to identify, design, develop, introduce, test, evaluate, and replicate new health services strategies that met "specified performance criteria under realistic operating conditions" (Flook and Sanazaro, 1973,p. 158~. The initial target areas for R&D activities were physician and dentist extenders, · cost containment and financing, quality of care, medical technology, and health data systems. Experience acquired in two of these R&D areas had particular policy significance in the 1970s. In the quality arena, the Experimental Medical Care Review Organization program developed and partially tested the structures and processes for quality assurance that became the basis for the Professional Standards Review Organization program, created in 1972 mainly for the Medicare nro~ram. In the data systems area, the Health Services Foundation of ^ r . · · 1 ~ ~ ~ A: _ ~ ~ : _ ~ ~ 1 ~ 7 ~ ^ +~ ~ ~+ the Blue Cross Association carried out demonstration projects In one ~ ms ~o res~ a uniform hospital discharge data set. This work influenced the creation of a cooperative federal-state-local health statistics system overseen by the National Center for Health Statistics, which itself had been created in 1974 (NRC and IOM, 1992~. In 1989, NCHSR was absorbed into the Agency for Health Care Policy and Research (AHCPR). The legislation that established the AHCPR focused the agency's activities on effectiveness research, technology assessment, and guidelines for clinical oractice.4 A_ ~ The selection of these areas reflected policymakers' frustrations with continually escalating health care costs and their general awareness of two decades of health services research that had (1) documented wide variations in medical practices (e.g., Wennberg and Gittelsohn, 1982; Wennberg, 1984; Brook, Chassin, Park, et al., 1986), (2) suggested that some health care services were of little or no value (e.g. Chassin, Kosecoff, Park, et al., 1987; Eddy and Billings, 1988), and (3) examined various kinds of economic and organizational incentives to reduce inappropriate variation and utilization (e.g., Ellwood, 1975, 1988; Eisenberg, 1986; Enthoven, 1988~. Health Care Financing Administration. The Health Care Financing Administration (HCFA), which was established in 1974, also developed an extensive array of research activities, in particular through its Office of Research 4See Gray (1992) for an interesting account of the agency's creation.
OCR for page 31
OVERVIEW OF HEALTH SERVICES RESEARCH / 31 and Demonstrations. This office built on the work of the Office of Research and Statistics in the Social Security Administration, which initially administered the Medicare program. In the late 1970s, HCFA was the major funder of research on health insurance and health care expenditures (IOM, 1979~. The agency also initiated demonstration projects to examine enrollment, payment, benefit design, and other issues involved in extending health maintenance organization (HMO) coverage to Medicare beneficiaries and to test a model of capitated acute and long-term care for frail older persons with a special emphasis on geriatrics (Langwell, Rossiter, Brown, et al., 1987; Harrington and Newcomer, 1991; Kane and Blewett, 1993~. In addition, HCFA (sometimes building on work initially funded by NCHSR) has supported research to develop methods to pay for services to Medicare and Medicaid beneficiaries in ways that would encourage cost-effective health care by discouraging unnecessary use of inpatient care, expensive technologies, and specialist services. The initials for the resulting tools in particular, DRGs (for prospective payment for hospitals based on diagnosis- related groups), RBRVS (for physician payment based on a resource-based relative value scale), and AAPCC (for a risk-adjusted, capitated payment for health care plans~have become part of the everyday vocabulary of policymakers and managers. The development of better methods for adjusting payments for services to account for a patient's health status or risk of incurring expenses is a major issue on HCFA's current research agenda. Department of Veterans Affairs. As directed by Congress, the Department of Veterans Affairs (VA) has had a health services research and development office, referred to as HSR&D, since 1976. In 1981, the office was "reoriented to emphasize health services research as a management tool" to improve health care for veterans (Goldschmidt, 1986, p. 798~. As part of this initiative, the VA set up research field offices in each of its six regions. It also established several Centers for Cooperative Studies in Health Services, which provide support to investigators planning large-scale studies and promote collaborative use of common research protocols and data sets. In addition, the VA has funded the Management Decision Research (:enter, which provides consultation, technical assistance, and research support to improve system level management and the capabilities of NIA senior staff. Recently, in conjunction with the Foundation for Health Services Research, the VA launched a new periodic publication, Forum: Translating Research into Quality Health Care for Veterans. Other federal agencies. In addition to the units cited above, several other federal agencies developed health services research agendas. For example, the Office of Economic Opportunity initiated (but only briefly oversaw) the national
OCR for page 32
32 / HEALTH SERVICES RESEARCH Health Insurance Study at the RAND Corporation in 1971. This project became the Health Insurance Experiment (HIE), which "remains one of the largest and longest-running social science research projects ever completed" (Newhouse and the Insurance Experiment Group, 1993, p. vii). It continued into the late 1980s with support from the Office of the Assistant Secretary for Planning and Evaluation in the Department of Health and Human Services. Several institutes of the NIH have also supported health services research. Most notable are the institutes concerned with mental health, drug abuse, and alcoholism and alcohol abuse, which are currently mandated to set aside 15 percent of their budgets for health services research. Other parts of the NIH and the Departments of Agriculture, Defense, Education, and Energy also have sponsored some activities that could be counted as health services research, although they may not be explicitly described as such.5 Other Sponsors of Health Services Research Private foundations, state governments, industry, and other sources have also supported health services research, although specific data on the level and distribution of such Finding are largely unavailable. The HSRProj database maintained by the National Library of Medicine, however, provides some information on foundation funding of health services research studies. An analysis of roughly 900 of the 1,200 studies6 listed in the database late in 1994, showed that six private foundations (the Robert Wood Johnson Foundation, W.K. Kellogg Foundation, Pew Charitable Trusts, John A. Hartford Foundation, Commonwealth Fund, and Henry J. Kaiser Family Foundation) provided approximately 20 percent ($70 million) of the Finding for the studies listed; the federal government provided the other 80 percent. HSRProj does not contain information on the research supported by state governments or industry. Several private foundations, including those that funded the work of the CCMC between 1928 and 1932, have a long history of funding health services research. Examples of major foundation-sponsored research activities include projects on the appropriateness of medical care (e.g., work at RAND sponsored by the Commonwealth Fund and others), indicators of health status (e.g., IOM conferences on advances in health status assessment fended by the Henry J. 5For example, the National Institute of Nursing Research technically does not fund health services research. Some of the nursing research projects they support do, however, seek to increase knowledge of the structure, processes, and effects of health services specifically, nursing services for individuals and populations. 6Data elements were missing on 300 of the files. The committee could not determine if the missing files contained information on foundation- or government-funded studies.
OCR for page 33
OVERVIEW OF HEALTH SERVICES RESEARCH / 33 Kaiser Family Foundation), and regional data projects (e.g., work supported by the John A. Hartford Foundation). The committee is aware of few states that undertake or sponsor significant health services research on an ongoing basis, and much of the research on state programs and problems is actually sponsored by the federal government or private foundations. States, however, have undertaken analytic work to support policy decisions on a variety of health problems or issues including infant mortality, use of managed care arrangements for Medicaid beneficiaries, nursing home regulation, substance abuse, and access for underserved populations. They often contract on an ad hoc basis with consulting firms, research organizations, or universities for studies of specific programs or issues. Some notable examples of state-supported health services research include the New York State Department of Health's analyses of physician-specif~c mortality rates for cardiac surgery (Hannan, Kilburn, Racz, et al., 1994; Green and Whitfield, 1995) and the work of the Oregon Health Resources Commission to develop a priority list or rank ordering of combinations of medical conditions and treatments as a basis for allocating limited resources for health care (Eddy, 1991~. In the past 50 years, various private organizations have created (and sometimes abolished) research arms to conduct both externally and internally sponsored research. These include professional associations (e.g., the American Hospital Association, the AMA), insurers (e.g. individual Blue Cross and Blue Shield plans), community health groups (e.g., Greater Detroit Area Health Council), and prepaid group health plans (e.g., the Health Insurance Plan of Greater New York, Group Health Cooperative of Puget Sound, Kaiser Permanente). The research funding pattern found in early prepaid group practice plans appears typical of this larger group of organizations. That is, the organizations provided seed money, but the majority of the funding for research came from the federal government. Over a 30-year period, for example, Kaiser Permanente contributed approximately $25 million, or 16 percent, of the $150 million in total funding for its Northwest region's Center for Health Services Research. How much private organizations, including competing health plans, will invest in health services research in the future is an important question. When they invest, business considerations will guide their choices of topics and their decisions about making information publicly available. For example, health plans that support technology assessments to inform coverage or influence practice patterns may conclude that subjecting their scientific analyses to peer review and public scrutiny provides essential credibility. Similarly, in some situations, the competitive advantage is in having early access to information for decisionmaking, and once this time-limited benefit has been reaped, the information may be made public. In other cases, commercial considerations may be weighed differently. For example, some systems to assess severity of illness have kept some of their algorithms proprietary while making other elements
OCR for page 34
34 / HEALTH SERVICF,S RESEARCH public (IOM, 1989~. Firms that provide services such as technology assessments on a subscription basis typically have to limit public availability of their analyses or find other ways of generating income for the activity. Market research on consumer attitudes and decisionmaking is traditionally proprietary, although purchasers are now demanding that some information about consumer satisfaction with competitive health plans be made public. Public and private universities have received much of the extramural health services research funds provided by federal agencies, and they also have obtained research funding from private foundations and industry. In addition, universities have sometimes provided seed grants or have self-funded a small amount of health services research from their own resources. The committee was not aware of any systematic information on the extent of university-sponsored (as opposed to university-conducted) research. Several freestanding private research organizations, such as the Urban Institute, Lewin-VHI, Abt Associates, Mathematica Policy Research, and Battelle Memorial Institute, have been major recipients of public and private research funds and have made significant contributions to health services research. The RAND Corporation, as already mentioned, conducted the multi-million-dollar national HIE as well as several of the studies of appropriate medical care cited earlier. As far as the committee is aware, however, these organizations are able to undertake little if any self-funded research. HEALTH SERVICES RESEARCH TODAY AND TOMORROW As a multidisciplinary field, health services research has drawn on and combined concepts and methods from many disciplines to provide frameworks for analyzing the structures, processes, and outcomes of health care and for informing decisionmaking. The agenda of topics for future research is quite lengthy and challenging. It both builds on longstanding questions about the availability, effectiveness, and cost of health care and incorporates new emphases that reflect the complex changes now occurring in the health care system. The extent to which this research agenda can be implemented depends significantly on the level and stability of federal funding, which has been and continues to be uncertain. The discussion of health research issues below is quite selective, but it illustrates the central themes of the field: the organization and financing of health services; access to health care; practitioner, patient, and consumer behavior; quality of care; clinical evaluation and outcomes research; informatics and clinical decisionmaking; and the health professions work force. Although categorizations of research areas inevitably emphasize separation, cross-cutting inquiry is an important characteristic of the field.
OCR for page 35
OVERVIEW OF HEALTH SERVICES SEARCH / 35 Organization and Financing of Health Services The recent debate over health care reform has highlighted concerns about the organization and financing of health care that health services researchers have long studied. Although it may have been largely invisible to the public, the media, and many political figures, three decades of health service research clearly contributed to the formulation of recent proposals for health care reform. This contribution is reflected in the prominence in reform proposals of provisions involving managed care, consumer choice, outcomes and performance monitoring, and risk selection in health insurance even though researchers might sometimes disagree with the policy conclusions set forth in specific proposals. Health services research has helped to clarify the features and effects of health insurance and market-based strategies of health care (e.g., requiring patients to bear some of the economic cost of their health care decisions and offering choices among health plans). In particular, it has shed more light on a dilemma long familiar to actuaries: that consumers' continuing exercise of choice in the purchase of health insurance can seriously diminish the degree to which the burden of health care expenses is shared among the well and the ill and thus undermine the very notion of insurance. Various studies (many sponsored by HCFA as part of its effort to direct Medicare beneficiaries into HMOs) have underscored how the financial benefits to health plans of avoiding high-risk individuals could be far greater than the rewards from cost-effective management of health care and have highlighted how difficult it is to compensate for this risk selection dynamic (IOM, 1993~. Although other considerations and values certainly shaped the debate over health care reform, concern about insurance dynamics was central to controversies about minimum or standardized benefit packages, health insurance purchasing cooperatives, community- versus experience-based pricing of insurance, statewide reinsurance pools, medical savings plans, and regulation of marketing practices. Health services research has illuminated the incentives of traditional fee-for- service and cost-based mechanisms for paying for health care. It has also devised tools and techniques that have facilitated the development of various alternative methods of paying for health services. As described earlier, these mechanisms include case-based payment for hospital services related to broad diagnostic groups, relative value scales for individual physician services, and adjusted per capita payments to health care organizations responsible for enrolled populations. As more health services have shifted from the hospital to other settings, researchers have attempted to evaluate the effects of these shifts on the quality, cost, and availability of care. Similarly, researchers continue to investigate the extent to which care provided by dental hygienists, nurse practitioners,
OCR for page 36
3 6 / HEALTH SER VICES RESEARCH occupational therapists, physical therapists, physician assistants, and other "mid- level" practitioners can safely, effectively, and efficiently substitute for physician and dentist care. These discussions have focused attention on conceptual and methodological questions about what constitutes effective care and who experiences the costs or savings of alternative modes of care. Access to Health Care In the health care context, access may be defined as the timely receipt of appropriate care (IOM, 1993a). From the field's earliest days, access has been an important focus of health services research, which has drawn attention to differences in access across income, racial, and other groups. Although it has helped to identify the cultural, organizational, and other nonfinancial barriers to access, it has particularly highlighted the financial barriers including the lack of insurance coverage. ~7 Research has delineated who is less likely to have health insurance and how coverage affects access to care, utilization of services, and health outcomes. It has clarified the central role played by employers, especially larger employers, in insuring full-time, full-year workers and their families. Each year, new survey results are closely examined for changes in the numbers and proportion of uninsured in the population. Researchers continue to explore how lack of coverage affects health, for example, by investigating the impact of cutbacks in Medicaid coverage (Lurie, Ward, Shapiro, et al., 1984; Hadley, Steinberg, and Feder, 19919. Such research can inform both the public and policymakers and shape governmental or political decisions. Quality of Care With rising costs have come demands for greater accountability from health care practitioners and institutions and for monitoring tools or systems that will document the quality of care. Quality has been defined in another IOM study as "the degree to which health services tor ~na~v~auats and populations Increase the likelihood of desired health outcomes and are consistent with current . .. . . . . . . professional knowledge" (199Ob. c. 21). Health services researchers attempt to ~ ~ , . , define and identify quality problems such as unnecessary or inappropriate care, underuse of appropriate care, and poor technical or interpersonal care. Drawing on concepts and techniques from industrial quality management, statistics, informatics, and operations research, health services researchers have been investigating strategies for monitoring performance and strengthening accountability. They have developed methods to measure health outcomes and to link variations in outcomes to characteristics of health care. Researchers have
OCR for page 37
OVERVIEW OF HEALTH SERVICES SEARCH / 37 also sought to educate potential users consumers, purchasers, and policymakers about the strengths and limitations of specific monitoring and accountability mechanisms. One critical need is for continued development and testing of practical and economical mechanisms for collecting, synthesizing, and disseminating valid, reliable, and usable health data. For example, researchers have investigated how "report cards" might be designed and implemented to help purchasers and con- sumers choose among health plans or providers. Laying a data-driven foundation for such judgments involves applied, theoretical, and philosophical challenges for research and policy in both public and private spheres. A particular challenge involves methods for assuring that performance comparisons reflect appropriate adjustments for different patient populations. Despite considerable progress, developing techniques to adjust for differences in severity of illness, comorbidity, and other factors continues to be among most difficult technical problems facing health services research today. These problems are, in many respects, another manifestation of those facing health plans when they enroll sicker populations that make their costs higher than those of competitors with healthier enrollees. Health services researchers can also inform (but not decide) the debate over the merits of governmental versus market strategies for monitoring and improving quality. Research in the 1970s contributed to skepticism about the effectiveness of community health planning, professional review organizations, and other regulatory strategies for improving efficiency and quality in health care. Now that market-driven strategies predominate, the focus of research is shifting. One concern is whether adequate tools and the data are at hand to assess the effects of a restructured health care system, especially when the incentives for providing too much care are diminishing and those for providing too little care are increasing. Although some role for external monitoring is generally conceded, many believe that quality improvement must, for the most part, be internally motivated and managed (IOM, 1990b). That means that large health care organizations need not only leadership but also people and continuing processes for establishing objectives, designing strategies for meeting those objectives, implementing the strategies, collecting and analyzing evidence about their impact, and redesigning activities as appropriate. In this environment, the boundaries of research, management, and marketing can become blurred. j Clinical Evaluation and Outcomes Research New medical technologies are often put into practice without valid evaluations of their effectiveness or their cost-effectiveness (OTA, 19943. Moreover, some long-standing practices depend on unverified claims of efficacy
OCR for page 38
38 / HEALTH SERVICES RESEARCH (expected benefit under ideal conditions of use) and effectiveness (expected benefit under average conditions of use). Dissatisfaction with these circumstances has given rise to clinical evaluative studies and outcomes research on the benefits and harms of different strategies for preventing, diagnosing, or treating illness. Relevant studies have focused on the assessment and comparison of alternative interventions for a given clinical problem and the identification of short-term and long-term outcomes of interest to patients, practitioners, and policymakers. In addition, researchers have devised instruments or techniques for measuring a variety of health outcomes, comparing the performance of health care organizations, and accounting for patient characteristics and other variables that may complicate comparisons of outcomes. The scope of clinical evaluations and outcomes studies is wide. Examples include evaluations of oral health outcomes in countries with different approaches to preventing and treating oral disease and organizing oral health services, assessments of the effectiveness of home-based intervention for people caring for a family member with dementia, and studies of whether treatment for skin ulcers in people with diabetes varies according to the type of doctor seen. More generally, the health outcomes of interest to practitioners, patients, and other decisions are quite diverse. They include not only mortality and morbidity but also health status, functional capacities, quality of life, patient and family satisfaction with health services, and professional satisfaction. The measurement of health status and functional outcomes comprises a major area needing further theoretical development, which could include development of a "production function" for health. economists, clinicians, and others to collaborate on the design of integrated health systems that will economically produce the health outcomes that are desired by patients and their families. There are few data or theories to clarify the trade- offs between costs and quality. The results of clinical evaluations and outcomes research may be used in formulating clinical practice guidelines to assist patients and providers in making decisions about appropriate medical care (IOM, 19923. More generally, translating evaluative and outcomes knowledge into improved medical care and its outcomes will require contributions from other areas of health services research including those concerned with practitioner and patient decisionmaking, medical informatics, and quality improvement strategies. That would require
OCR for page 39
OVERVIEW OF [IEALTH SERVICES RESEARCH / 39 Informatics and Clinical Decisionmaking The country has made large investments in medical information systems that are supplying health care providers, managers, and researchers with quicker and easier access to more complete health care information about both individuals and groups. The information provided by computer-based information systems is proving particularly useful for clinicians, administrators, and researchers who are attempting to measure and improve the quality and cost-effectiveness of health services through better design and management of clinical and administrative systems. Important challenges for health services researchers, computer scientists, and others remain in such areas as linking patient records across inpatient and outpatient care settings; establishing protocols for ensuring the confidentiality and accuracy of information; and developing more user- friendly hardware and software for entering, retrieving, or analyzing information. In these activities, the federally sponsored National Information Infrastructure initiative will permit a wide variety of parties, including health services researchers, to transmit, store, process, analyze, and display data in different forms such as text, still images, sound, and video (Lasker, Humphreys, and Braithewaite, 1995~. Clinicians have access to a variety of on-line and other information resources. They may, for example, subscribe to on-line article retrieval services such as Medline and Grateful Med. as well as the other specialized medical information services provided by the National Library of Medicine. They may also purchase CD-ROM or other software packages that allow them to search the medical literature for information on specific clinical problems, to locate clinical practice guidelines, and to work through quantitative algorithms to identify probabilities of various benefits or harms of alternative clinical strategies for a given patient's problem. Patients also have access to much of this information and may use it to guide their own decisions with or without consulting a clinician. The impact of these information resources on attitudes and behavior will present new research questions in an area of long-standing interest to health services researchers. In addition to providing information, computer-based information systems can incorporate a variety of tools to assist clinical decisionmaking. These tools include automatic reminders or alerts that are triggered when certain patient information is entered or obtained or when certain pharmacy, laboratory, or other tests are ordered. Thus, a physician or nurse may be reminded that a diabetic patient is due for an opthalmalogic test or that penicillin is contraindicated for a patient with a past allergic reaction to the drug. The development and evaluation of these decision aids pose important opportunities for health services researchers.
OCR for page 40
40 / HEALTH SERVICES RESEARCH Practitioner, Patient, and Consumer Behavior Health services and behavioral science researchers have spent many years exploring health-related human behavior and studying how practitioners and patients make decisions about health and health care. Investigators continue to search for the demographic, cultural, economic, and other factors that shape individual actions such as seeking medical or dental care, selecting among treatment options, following treatment recommendations, and purchasing health insurance (or selecting from alternative insurance plans). Much, however, remains to be learned about ways of encouraging desired patient behaviors and decisions on such diverse subjects as adherence to dietary protocols for diabetes, timely and continuing use of preventive services, and appropriate requests for antibiotics. Research on practitioner behavior likewise continues. Investigators ask how education, training, professional socialization, practice environments, economic incentives, and other factors affect decisions. They raise challenging issues such as how to reconcile dramatic variations in rates of health care interventions with the notion that there is a common base of professional knowledge accessible to all practitioners. Do management models that stress the reduction of such promise-particularly in a competitive market-to support or undermine the role of the clinician as an expert agent and advocate for patient interests? In a restructured health care system, managers can look to past research on physician responses to economic incentives, guidelines for clinical practice, data on practice patterns, and other management or policy tools. variations rib r- , Health Professions Work Force Some of the most sensitive questions asked of health services researchers involve the education and supply of health care workers. Are there too many ~ ~ Does the country have too few generalist practitioners or too many specialists or both? Are health professionals located nurses or are there too few? How will the changing medical force? appropriately across geographic areas? marketplace affect demand for health personnel and for neaten pro education programs? Should the content of medical education change? What attracts qualified people to particular professions? What are the obstacles to increased participation by underrepresented minorities in the health care work 1 1 '_ 1_ _ _1 ~ 1_ __~: ~ ~ Efforts by analysts and policymakers to forecast and plan the supply of health personnel and services have not proved particularly successful (Fell, Welch, and Fisher, 1993; Kindig, 1994; Capilouto, Capilouto, and Ohsfeldt,
OCR for page 41
OVERVIEW OF HEALTH SERVICES RESEARCH / 41 1995; IOM, 1995~. Health services researchers are, however, continuing efforts to improve their estimating tools (NRC, 1994~. These difficulties and efforts, which obviously are significant for this study, are discussed further in Chapter 3. CONCLUSION The discussion above has several implications for the education and training issues that this IOM committee was asked to examine. First, health service researchers face no shortage of important, contentious questions and methodologic challenges. Second, despite some skepticism, public and private decisionmakers have for decades been influenced by health services research (although they may not have known the source for what has become common knowledge), and they have increasingly looked to it for tools to help measure the effectiveness of health services and organizations. This implies some trajectory of rising demand for research and people to design, conduct, and report it. Third, demands for internal management information and external accountability should provide employment opportunities for health services researchers as large medical care organizations, integrated health systems, and accrediting agencies establish health services research units to provide information and analyses. Fourth, public resources for health services research are vulnerable when attempts are made to pare government budgets, which complicates efforts to determine how many health services researchers should be trained. Fifth, research questions that have a public good aspect may be less likely be addressed by market organizations if they cannot readily capture the major benefit of their research investment. The next chapter of this report discusses the health services research work force in more detail. It also presents what the committee learned about public and private demand for health services research and researchers.
OCR for page 42
Representative terms from entire chapter: