Health care outcomes assessment.

Although earlier assessments of medical care, frequently carried out by health care service researchers and quality assurance professionals, focused largely on the structural and process-oriented characteristics of medical care, during the past five years, a shift has occurred to study the outcomes of health care services. Relman has referred to the outcomes movement as the "third revolution in medical care,"(1) and Ellwood has stated that the outcomes movement can be expected to provide us with a "...central nervous system that can help us cope with the complexities of modern medicine.(2)

As U.S. health care expenditures continue to increase,(3) interest in assessing the effectiveness and value of health care services is rising. In conjunction with concerns regarding rising health care costs, interest in outcomes assessment is driven by research evidence suggesting that from 10 to 20 percent of U.S. health care services are unnecessary. This evidence is based on research findings from at least three sources: Wennberg's research demonstrating significant differences in health care utilization rates between small geographic areas, suggesting either overutilization in high use areas or underutilization in low use areaS4 ; the RAND studies of the appropriateness of utilization of six procedures in the Medicare population, demonstrating levels of unnecessary use of these procedures ranging from 14 to 32 percent'; and the research of Greenberg et al.1 demonstrating that approximately 20 percent of cardiac pacemaker implants performed in the study were done for inappropriate reasons. Current Activities

Increasing interest in outcomes assessment is now reflected in the activities of several organizations and researchers. The Joint Commission on Accreditation of Healthcare Organizations has initiated the Agenda for Change, shifting the focus of hospital accreditation from assessing structural characteristics of hospitals to the outcomes of hospital care.(7) Researchers are actively pursuing studies of outcomes of medical and surgical services based on information in large claims data bases,(l) and of outcomes of patients receiving care for chronic illnesses.(1-11)In addition, Ellwood has proposed a national strategy for outcomes management "...Outcomes management consists of a common patient understood language of health outcomes; a national data base containing information and analysis on clinical, financial, and health outcomes that estimate as best we can the relation between medical interventions and health outcomes, as well as the relation between health outcomes and money; and an opportunity for each decision-maker to have access to the analyses that are relevant to the choices they must make.(2)

The federal government has initiated outcomes assessment activities in two agencies. The Health Care Financing Administration has established the effectiveness initiative, designed to measure the outcomes of health care services provided to Medicare recipients,(12) and the newly formed Agency for Health Care Policy and Research has initiated funding for programs for outcomes assessment, and for using these findings of outcome research to develop practice guidelines." The increasing budget for this new Agency reflects the deep interest in outcomes assessment from $5.5 million in 1989 to more than $30 million in 1990. Studies funded by the Agency for Health Care Policy and Research to date include investigation of outcomes of surgical and nonsurgical interventions for benign and malignant prostatic disease, alternative therapies for acute myocardial infarction, surgical treatment for cataracts, surgical and nonsurgical treatment of back pain, total knee replacement, diagnostic and treatment strategies for ischemic heart disease (i.e., cardiac catherization, percutaneous coronary angiography, and coronary bypass surgery), and treatment of patients with biliary tract disease. (14)

Methodological issues

Methods used in outcomes assessment research represent a departure from methods used in more traditional clinical research. Significant areas of difference in outcomes assessment include study design, types of outcomes studied, information regarding patients prior to interventions, periods of patient follow-up after interventions, and sources of data. Ideally, all health care interventions would be assessed with the most rigorous and efficient of study designs-the randomized clinical trial. It is highly unlikely, however, that resources will be available to do so. As a consequence, quasi-experimental (cohort and case control studies) and nonexperimental designs (uncontrolled trials, case series, and case observations), coupled with appropriate statistical analyses designed to adjust for intervening and confounding variables not controlled for in the quasi- and nonexperimental study designs, are commonly used in outcomes assessment."

Outcomes assessment frequently includes measures other than the clinical and physiological outcomes common in more traditional clinical research. These outcomes include functional status, as well as other indicators of patients' quality of life, including pain relief, ability to return to preintervention activities, and general satisfaction. Conceptualization and development of measurement tools for these nonclinical outcomes are just now under development."

Outcomes assessment research requires information about patients' health status prior to their receiving the intervention. Although the specific types of pre-intervention patient information depend upon the outcomes being assessed, information is required regarding the patient' functional status, comorbidities, and severity of illness. The follow-up period in outcomes assessment extends for periods beyond the episode of care in which a particular intervention is provided. Data collection may occur six months to years after the intervention was provided. During long follow-up periods, however, intervening factors other than the intervention may influence the outcomes, complicating the study design and data analyses.

In more traditional clinical research, data are collected to answer specific questions, or to test specific hypotheses-primary data collection. In outcomes assessment research, however, in addition to use of primary data, analyses of information in large databases collected for administrative purposes are also used. Issues raised regarding the accuracy and completeness of data in these large databases must be addressed by researchers using the data in outcomesassessment. (16) Secondary analyses are generally designated as exploratory or case-finding, and as hypothesis generating as opposed to hypothesis-testing.

Potential Uses

Outcomes assessment is likely to be used in several key health care activities, including quality assurance, the development of recommendations or guidelines for use in the clinical practice of medicine," technology assessment," and the development of explicit review criteria to be used in utilization review." Quality assurance is central in the outcomes assessment movement, including integration of outcomes assessment in the continuous improvement model of quality assurance.(19) Once largely qualitative and descriptive in content, currently outcomes assessment brings quantitative data regarding motality, morbidity, and quality of life to quality assurance activities. Such data are increasingly reviewed by payers as they select providers of health care of acceptably high quality.

If the findings of outcomes assessment research are not used in the clinical practice of medicine, their value to society will be significantly reduced. To better inform the day-to-day practice of medicine, therefore, government, professional, voluntary, and proprietary organizations are now involved in developing recommendations and guidelines to ensure that physicians use medical and surgical services appropriately.(20) Findings from outcomes assessments, together with expert consensus formulated using group judgment methods, serve as the basis of these recommendations. In addition, payers and insurers increasingly use findings from outcomes assessments as the basis for criteria to review the appropriateness or necessity of services being regulated or purchased.

Expectations

As the outcomes assessment movement emerges, it faces many challenges. The tools of outcome measurement, including conceptualization of the nonclinical outcomes to measure, study designs, methods for data collection, and appropriate statistical analyses, all require further development and refinement. The ultimate success of the outcomes assessment initiative depends on the willingness of providers, payers, insurers, and patients to work collaboratively. A critical issue with regard to the future of this movement is the extent to which outcomes assessment will be found to reduce health care costs and, if it does not, the degree to which payers, including the federal government, will continue to support it.

Although the future of outcomes assessment remains uncertain, it has already set in motion changes in the way health care is viewed in the United States. The patient has come to be viewed as far more central in decisions involved in patient care, and the importance of assessing and ensuring the quality of health care services, particularly in the environment of cost containment, has been strongly emphasized. c3

References 1. Relman, A. Assessment and Accountability:The Third Revolution in Medical Care.' New England Journal of Medicine 319(18):12202, Nov. 3, l988. 2. Ellwood, P. Outcomes Management: A Technology of Patient Experience.' New England Journal of Medicine 318(23):1549-56, June 9, 19". 3. Levit, K., and others. National Health Care Expenditures: 1988.' Health Affairsg(2):171-84, Summer 1990. 4. Wennberg, J., and others. Hospital Use and Mortality among Medicare Beneficiaries in Boston and New Haven.' New England Journal of Medicine 321(17):1168-73, Oct. 26,1989. 5. Chassin, M., and others. Variations in the Use of Medical and Surgical Services by the Medicare Population.' New England Journal of Medicine 314(5):285-90, Jan. 30,1986. 6. Greenspan, A., and others. The incidence of Unwarranted implantation of Permanent Cardiac Paoemakers in a Large Medical Population.' New England Journal of Medicine 318(3):158-63, Jan.21,1988. 7. O'Leary, D. The Joint Commission Looks to the Future.' JAMA 258(7):951-2, Aug. 21, 1987. 8. Wennberg, J., and others. Use of Claims Data Systems to Evaluate Health Care Outcomes: Mortality and Reoperation following Prostactomy.' JAMA 257(7):9"6, Feb. 20, 1987. 9. Stowart, A., and others. Functional Status and Well-Being of Patents with Chronic Conditions: Results from the Medical Outcomes Study.' JAMA 262(7):907-13, Aug. 18, 1989. 10. Wells, K., and others. Me Functioning and Well-Being of Depressed Patients: Results from the Medical Outcomes Study.' JAMA 262(7).914-9, Aug. 18,1989. 11. Tarlov, A., and others. The Medical Outcomes Study: An Application of Methods for Monitoring the Results of Medical Care.' JAMA 262(7).,925-30, Aug. 18, 1989. 12. Roper, W., and others. Effectiveness in Health Care: An Initiative to Evaluate and Improve Medical Practice.' New Engl and Journal of Medicine 319(18):1197-202, Nov. 3,1988. 13. Lee, P., and others. The Physician Payment Review Commission Report to Congress.' JAMA 261(16):2382-5, April 28,1989. 14. Research Activities. Rockville, Md.: Agency for Health Care Policy and Research, Aug. 1990. 15. Moses, L. Framework for Considering the Role of Data Bases in Technology Assessment.' International Journal of Technology Assessment in Health Care 6(2):183-93, 1990. 16.Institute of Medicine. Effectiveness Initiative: Setting Priorities for Clinical Conditions. IOM-8904. Washington, D.C.: National Academy Press, April 1989. 17. Fuchs, V., and Garber, A. The New Technology Assessment.' New England Journal of Medicine 323(10):673-7, Sept. 6, 1990. 18. Institute of Medicine. Controlling Costs and Changing Patient? The Role of Utilization Management. Washington, D.C.: National Academy Press, 1989. 19. Berwick, D. Continuous Improvement as an Ideal in Health Care.' New England Journal of Medicine 320(l):53-6, Jan. 5,1989. 20. Brook, R. "Practice Guidelines and Practicing Medicine: Are They Compatible?' JAMA 262 21 :3027-30, Dec. 1, 1989.