and governmental bodies, are intended to monitor providers' performance and to take any corrective action required. More generally, industries other than health care have developed a notion of quality improvement that entails companies' working with organizational and individual consumers to improve quality. The responsiveness of a company to consumers is an essential feature of quality control in these industrial programs and might be transferable to medical care delivery (68). The results of quality assessment may feed into quality assurance and quality improvement through the responses of hospitals and physicians, employers, third-party payers, and Federal and State governments to problems that are identified. Indeed, some experts regard how a provider responds over time to deficiencies in quality as a measure of that provider's quality (67). In its evaluation of hospitals, the Joint Commission on the Accreditation of Healthcare Organizations (JCAHO) has examined how institutions have dealt with deficiencies in performance or other problems that have arisen. As part of its effort to develop clinical and organizational indicators of quality, JCAHO plans to monitor on a continuing basis how hospitals respond to recognized problems (329). In this report, OTA assesses eight categories of potential indicators of the quality of care provided by physicians or hospitals (see table 1-1): Table 1-1.-Indicators of the Quality of Care Evaluated by OTA, by Type of Medical Provider Physicians Adverse events Formal State disciplinary actions PRO/HHS sanctions Malpractice compensation Evaluation of physicians' performance: hypertension Volume of services Physician specialization Patients' assessments Hospitals Hospital mortality rates Adverse events PRO/HHS sanctions Malpractice compensation Volume of services Scope of hospital services Patients' assessments SOURCE: Office of Technology Assessment, 1988. hospital mortality rates, for the institution overall, by department, and by condition or procedure; • adverse events that affect patients, as exemplified by nosocomial (institutionally acquired) infections in hospitals; • formal disciplinary actions by State medical boards against physicians, sanctions imposed by the U.S. Department of Health and Human Services (HHS) on the recommendations of utilization and quality control peer review organizations (PROs), and malpractice compensation; • evaluation of physicians' performance through their care for a particular condition, as exemplified by hypertension screening and management; • volume of services in hospitals and performed by physicians; • scope of hospital services, with particular reference to emergency services, cancer care, and neonatal intensive care units; physician specialization; and patients' assessments of their care. This report does not offer a comprehensive evaluation of the many quality indicators that have been suggested or used (185). Although OTA attempted to select the most promising indicators, without evaluating others, one cannot conclude that the eight categories of indicators considered by OTA contain the best measures in terms of validity and feasibility for consumer use. OTA chose indicators to reflect the perspectives of consumers and the medical, research, and policy communities. High priority went to indicators concerning aspects of care that matter greatly to consumers, such as humaneness and communication of information, and decisions that consumers are likely to face, such as selecting a hospital to provide emergency care. To reflect policy interests, OTA paid particular attention to indicators that quality assessors are using or considering, especially for public programs. The indicators also illustrate different approaches to measuring quality and cover different aspects of quality. To ensure the feasibility of its own analysis, OTA limited its choice to indicators for which sufficient information existed to support an evaluation. The remainder of this chapter summarizes the body of the report and presents policy options to address the problems identified. The summary first discusses the audience for information on quality, the assessment of quality from an individual consumer's perspective, and the dissemination of information to individuals. The summary then turns to the findings and conclusions regarding the specific indicators evaluated in this report. Based on the issues raised in the summary, the final section of this chapter analyzes policy options for congressional consideration. The body of the report considers the dissemination of information on quality, develops a framework to SUMMARY Many individuals and organizations that make decisions about the purchase and provision of medical services could use valid information about the quality of medical care to guide their choices. Individuals seeking medical care have historically relied on family and friends for advice and on physicians for referrals to other medical providers. All to one degree or another have lacked information on quality. Quality-of-care information is also important for employers and third-party payers who monitor the performance of physicians and hospitals or who selectively contract with certain providers. Unions would like to have information on quality to evaluate alternative plans and providers for their members, especially as cost-containment efforts have led insurers and employers to increase cost-sharing and to favor lower cost providers. With information on quality, these organizations could consider quality as well as costs in their selection of and arrangements with providers. Physicians, hospitals, and other providers themselves have lacked information on the quality of care and could benefit along with consumers from improved sources of information. Physicians could use valid information on the quality of care to select hospitals for staff appointment and for patient referral, to select other physicians for patient referrals, and to answer questions and interpret data for patients. Hospitals could also benefit from improved information about quality, in appointing physicians to staff and granting phy assess quality for individual consumers, and evaluates the eight categories of indicators. Appendix A describes the method used to conduct the study, and appendix B acknowledges the valuable assistance of many individuals. Appendix C presents the method that OTA used to analyze the reliability, validity, and feasibility of the indicators evaluated in this report. Appendix D discusses quality assessment activities of the Joint Commission, the American Medical Association, and the PROS, and appendix E lists recent and ongoing research on quality assessment in selected public and private organizations. sicians admitting privileges and in monitoring their own performance and augmenting their quality assurance and risk-management programs. Quality From the Perspective of Individual Consumers Although many purchasers and decisionmakers can use information on quality, medical care is intended to benefit individual consumers. Thus, it is appropriate to evaluate the quality of care from the perspective of those individuals. The quality of medical care has many dimensions, a fact that reflects the diversity of acceptable outcomes for patients and the complexity of the medical care process. Medical care seeks to promote, maintain, and restore people's health (186). Health itself contains multiple dimensions, including physiologic health, physical functioning, mental health, and social functioning. Depending on their conditions, patients vary widely in the health outcomes that they desire, from increased longevity, mobility, and emotional wellbeing to reduced illness, deterioration, and suffering. The appropriate content of care varies accordingly, from prevention and screening to diagnosis, rehabilitation, counseling, and other therapy. Moreover, patients vary in their preferences; some prefer less-invasive, less-painful, or less-disfiguring technologies, even at the expense of a shorter life. Reflecting the complexity of the medical care process, prominent scholars have stressed the importance of evaluating both technical and interpersonal aspects of care (105,183). Both technical care, the application of medical science and technology to a problem, and interpersonal care, the personal interaction between patient and provider, enter into any episode of care and merit evaluation. Although consumers, providers, and the overall society from their own perspectives may emphasize different aspects of quality, all view both the technical and interpersonal aspects as important (183). Physicians have usually confined their evaluation to technical performance, while patients have shown more sensitivity to how they are treated (186). Society has more interest than individual consumers or providers in the equitable distribution and public health benefits of care, such as prevention of communicable disease. Besides encompassing the many dimensions of medical care and health outcomes, a definition of quality must take into account the limits and continuing evolution of medical knowledge. As knowledge expands, some technologies, such as gastric freezing to treat stomach ulcers, become obsolete and should be discarded, while others, such as cimetidine, are shown to be efficacious and should be adopted as appropriate therapies. The use of medical technology also entails some risk and cannot guarantee improvement in a patient's health. In a larger sense, the uncertainty surrounding patient outcomes stems from the fact that medical care is but one influence on the health of an individual or a population. In fact, an individual's genetic makeup, environment, and lifestyle seem to play a greater role than medical care in explaining the causes of death and illness that now predominate in the United States. The triad commonly used to assess the quality of care focuses on the structure, process, and outcome of care (183). Table 1-2 categorizes the indicators evaluated by OTA according to the assessment approach. The structure of care encompasses the resources and organizational arrangements in place to deliver care, such as medical personnel, facilities, and quality review committees. Assessing quality via structural indicators, such as physician Table 1-2.—Indicators of the Quality of Care Evaluated by OTA, by Assessment Approach Structure Volume of services Scope of hospital services Physician specialization Patients' assessments Process Adverse events Formal State disciplinary actions PRO/HHS sanctions Malpractice compensation Evaluation of physicians' performance: hypertension Hospital mortality rates Formal State disciplinary actions Malpractice compensation Evaluation of physicians' performance: hypertension Patients' assessments SOURCE: Office of Technology Assessment, 1988. specialization, presupposes that their presence increases the likelihood that providers will perform well and their absence, the likelihood that providers will perform poorly. This assumption in turn raises the question of whether specific structural characteristics are, in fact, associated with better performance. The process of care refers to the activities of physicians and other health professionals engaged in providing medical care. Although the appropriate care for a specific condition changes as knowledge expands, the thorniest problem with process measures of quality lies in the paucity of information about the efficacy of even wellaccepted medical procedures. One should limit evaluations of providers' performance to procedures likely to improve or harm patients' health and satisfaction. The problem is that the link between the process of care and patient outcomes has been established for relatively few procedures. Measuring quality via outcomes, namely changes in patients' satisfaction and health status, is the third approach. The problem with this method is that attributing changes in outcomes to medical care requires distinguishing the effects of medical care from the effects of the many other factors that influence patient health and satisfaction. In light of the conceptual difficulties just mentioned, process and outcome measures should be regarded as complements rather than alternatives to assess quality. Process measures gain validity as quality indicators only to the extent that they have been found likely to improve patient outcomes, and outcome measures gain validity only to the extent that they have been linked to the prior medical care process. Similarly, to acquire validity as indicators of quality, structural measures must be shown to be associated with efficacious medical processes or validated outcomes. Over the years, scholars have taken many different approaches to incorporating these complexities into a definition of the quality of medical care. This report examines several possible indicators of the quality of care provided by hospitals and physicians. Reflecting this task and the points above, this report uses the following definition of quality to guide its discussion: The quality of medical care is the degree to which the process of care increases the probability of outcomes desired by patients and reduces the probability of undesired outcomes, given the state of medical knowledge. Under this definition, the quality of a hospital's or physician's care is judged against the likelihood that the care will achieve the desired patient outcomes. Which elements of patient outcomes (health and satisfaction) predominate depends on the individual patient or condition. As emphasized above, valid assessments of quality require linking the medical care provided (the process of care) with the effects on patient health and satisfaction (the outcomes of care). This definition of quality also incorporates the notion that there are different levels of quality: a minimum level below which quality is unacceptable and levels of acceptable quality, including some levels in which important concerns about the quality of care remain and improvement is possible. Quality assessment and information systems take on different purposes that correspond to the different levels of quality: to identify unacceptable providers, so that they can be helped to improve and, as a last resort, be removed from practice; and to identify gradations among good quality providers, so that people can gravitate to the better ones and perhaps ultimately improve the general level of care. Since consumers vary in the importance that they attach to different aspects of care, information systems could also identify discretionary aspects of practice, so that people could act on their preferences. A framework to assess quality from a consumer perspective starts with the technical and interpersonal aspects of care that influence desired outcomes, namely improvements in the various dimensions of health and in patient satisfaction. Such a framework should also address the choices that people face and the care that they receive during an episode of care. Surveys of individual consumers and the literature indicate that the following aspects of the medical care spectrum have importance for patient health and satisfaction: • responsiveness to urgent and emergency situations; referral to the appropriate level of care; • humaneness; • communication of information; • coordination and continuity of care among providers; Although this report generally excludes issues. of access, two aspects of access clearly overlap with quality of patient care and have such strong implications for patient outcomes that they are included in this report: providers' responsiveness to urgent or emergency care and providers' referral of patients to the appropriate level of care. Inclusion of the next two aspects in the framework reflects that people place high priority on being treated with respect and on receiving pertinent information from their physicians, including information to prevent disease and promote health (392). The last five categories, from primary prevention to management, relate to steps in the medical care process. Coordination of care receives separate mention to emphasize that, even if each step in the process is performed appropriately, poor-quality care can result from lack of coordination among providers. Continuity of care improves patient satisfaction and compliance (177), although its importance, like that of other aspects of care, varies with the situation (183). The relationship between these aspects of care and the indicators evaluated by OTA is summarized in table 1-3. There is limited evidence on how quality-ofcare information is likely to affect people's choice of providers. No empirical study addresses directly the effects of such information on consumers' choices or the elements of an effective strategy for disseminating such information. But drawing on principles of health behavior and studies in related fields, one may hypothesize that the following elements are necessary for consumers to receive information and to incorporate it into their choices of physicians and hospitals: ⚫ stimulate consumer interest in the quality of care, • provide information easy to comprehend, ⚫ use many media and formats to present the information, • use respected sources to interpret the information, • make the information readily accessible, and • provide consumers the skills to use and physicians the skills to provide the information. These elements, like the studies from which they were drawn, relate mainly to mass communication. Although mass media have a role to play in raising consumers' awareness of quality-of-care issues and information, approaches that also included social support and skills training are likely to prove more effective in stimulating people to apply quality-of-care information to their interactions with providers and their choices regarding particular medical problems. Findings Regarding Specific Although none of the indicators evaluated in this report convey definitive information about the quality of an individual hospital or physician across the range of medical care, several of these indicators can provide useful information to organizations and individuals. For those consumers who consider physicians' character as well as skills in judging the quality of care, formal disciplinary actions by State medical boards can be accepted as valid indicators of poor-quality physicians. Consumers and others would be well advised to use many of the other indicators as initial screens for possible quality problems and to combine. information from several indicators to decide whether further exploration is warranted. Information about unacceptable care merits more attention than information that ranks good-quality providers because of the more immediate concerns raised by poor quality and the state of quality assessment techniques. Used as screens, certain indicators can identify physicians or hospitals about which there are reasonable grounds for concern. Armed with this information, individuals could then question their providers and evaluate whether a quality problem exists. A hospital whose unadjusted mortality rate exceeds expected levels, for example, may house a regional trauma center; this factor rather than poor quality might account for the high mortality rate. Similarly, that a hospital recommended by a surgeon has a low volume of cardiac surgery may reflect accounting conventions and not be related to the quality of care. Consumers would also be well advised to combine information from more than one indicator of quality, to increase the likelihood of learning whether a quality problem was or was not present. A cardiac surgery patient could gain confi |