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 238
Medicare: A Strategy for Quality Assurance - Volume I 8 Settings of Care and Payment System Issues for Quality Assurance Both the settings of care and systems of paying for care can profoundly affect quality of care. Differences across settings and payments systems need to be considered in designing and implementing quality assurance programs. For example, organizational structures, personnel, data systems, medical technologies, potential hazards, and patient roles vary substantially from hospital to home care. What is possible within the hospital may be infeasible or unreasonable in other settings. Likewise, different payment systems may involve different data collection capabilities and different incentives for overuse or underuse of care. The first part of this chapter focuses on sites of care with specific attention to the prevention, detection, and correction of quality problems in hospital, ambulatory care, and home settings. It asks how different characteristics of these settings (other than system of payment) may affect different elements of a quality assurance program. The second part of the chapter looks at fee-for-service and capitated systems for paying practitioners and institutions and considers the special quality of care concerns presented by each system. SETTINGS Overview Patients are seen in various health care settings, each having special characteristics that may contribute to the ease or difficulty of assessing and assuring quality. The hospital sector includes acute short-stay, long-term, specialty, and rehabilitation hospitals. The ambulatory care sector includes
OCR for page 239
Medicare: A Strategy for Quality Assurance - Volume I a broad arrangement of organizational settings. Office-based practice ranges from solo or small group practices to large multispecialty group practice. Ambulatory care can also be delivered in hospital-based outpatient departments, clinics, and emergency rooms, in hospital-related or independent ambulatory surgical centers, and in freestanding urgent care centers. Finally, community-based long-term care includes home care, adult day care, respite care, and care in hospice and residential nursing homes. This chapter focused on only one of each category of setting: namely, acute hospital care, office-based care, and home health care. We believe the issues raised by these settings and the lessons to be drawn about quality assurance will adequately reflect the challenges for developing a strategy for quality assurance for Medicare. In particular, we note four key aspects of these settings as they relate to the patient-provider encounter: (1) the clarity of the episode of care (how easily it is defined or identified), (2) the extent of information recorded about the intervention or encounter, (3) the relative emphasis on medical and health versus social (including family and environmental) components, and (4) the extensiveness, complexity, or formality of the organizational structure. The three settings of primary interest in this study differ on these four dimensions (Table 8.1). We discuss these issues in only very broad terms here, and are aware that there is great diversity within setting type. Specific methods of quality assessment and assurance are discussed in greater detail in Chapter 9 and by setting of care in Volume II, Chapter 6. The clarity of the definition of an episode of illness differs by setting. At one extreme is an inpatient hospital stay, which has a very specific beginning and end. By comparison, an episode of ambulatory care may be quite difficult to define, especially when it involves multiple visits for chronic care. When an episode is difficult to define, it becomes more difficult to identify quality problems as occurring within the episode. The larger and more complex a setting of care is, the more centralized, formalized, and standardized the patient care records and data systems are likely to be. When reimbursement depends on information on insurance claims (e.g., hospitals for Medicare), the insurance claims are likely to be more detailed. The more these conditions hold, the more records and data systems will support quality assessment methods that depend on retrieval of data from patient care, encounter, or financial records, and the more feasible it will be to feedback information about practice patterns to practitioners. Generally, hospitals will rank high on the extent of information recorded and maintained on patients, and ambulatory and home health care will rank much lower. Settings also differ in the importance of medical and social components. Services with a high medical or technical content may be more easily evaluated with objective measures, such as blood pressure for hypertension con-
OCR for page 240
Medicare: A Strategy for Quality Assurance - Volume I TABLE 8.1 Elements That Distinguish Settings of Care Setting Clarity of Episode Completeness of Recording Interventions Importance of Medical versus Social Components Degree of Organizational Structure Hospital inpatient High High High High Ambulatory office-based Very low Moderate Moderate Low Home health care Low Low Low Very low trol, whereas services with a strong social service component will require patient evaluations. To the extent that practitioners and facilities are members of a larger organization, they will generally be subject to more formal structures for review of credentials and reappointment, internal peer review, patient care policies and procedures, and documentation of care (Luke and Modrow, 1982; Scott, 1982; Rhee, 1983). Historically, organization and formalization of health care has developed first in the hospital, then in some areas of ambulatory care and home care, and minimally in office-based care. The presence of organizational structures generally provides a focal point for interventions to improve quality. Although we do not wish to overemphasize these differences among the various settings, they have important implications for the design of quality assessment and assurance systems. For example, one could use claims data to identify people undergoing certain inpatient procedures who then have a reoperation. This information could be fed back to the hospital medical staff for review and appropriate intervention. Such a trigger and response system is hard to translate to a home health setting. On the other hand, home care agencies have relatively stable populations who can be surveyed by mail or phone with relatively simple questionnaires. To structure our review of quality assurance approaches in various settings, we concentrate on three major needs: preventing problems, detecting problems, and correcting problems. The omission in this discussion of approaches to encouraging better performance (e.g., the continuous improvement model) reflects the scarcity of working models of this approach.
OCR for page 241
Medicare: A Strategy for Quality Assurance - Volume I The Hospital Prevention The hospital inpatient setting is the most structured of the health care settings. The premise is that meeting specified structural requirements strengthens the ability of an organization to deliver good quality care (or, alternatively, to prevent poor care) (Palmer and Reilly, 1979). Driven partly by risk management concerns and partly by requirements of the Joint Commission for Accreditation of Healthcare Organizations (the Joint Commission),1 systems for recommending medical staff for appointment and reappointment and for determining clinical privileges (and hence, for denying such appointments and privileges) are, by and large, well developed and set forth in medical staff bylaws. Similarly, data gathered on practitioner-specific performance for reappointment and privileging are increasingly refined. These structural characteristics are considered among the leading first-line defenses against poor care, because the implications for physicians of losing staff or clinical privileges are considerable.2 It is important to distinguish, however, between the ability to revoke privileges and the practical feasibility of undertaking such actions. Hospitals are typically divided into services, each with clinical directors or department chiefs. Each will have policies and procedures in addition to general hospital policies and procedures regarding professional and staff obligations and responsibilities. In recent years, the emphasis on accountability, lines of authority, and similar structural variables within the hospital setting has grown tremendously. Particular stress has been placed on the responsibilities of governing boards, especially since the landmark ruling in Darling v. Charleston Community Hospital (1965) that established a hospital’s direct corporate liability for medical staff quality problems about which it knows (or should know), even when the physicians are not employed by the hospital. Policies and procedures designed to protect hospital patients from medication errors, misidentification, and numerous other potentially adverse events have evolved over many years. There are well-developed systems for documenting the course of patient care in medical records, for instance, nurses’ notes, attending physician notes, operative notes, admission and discharge summaries, and results and interpretations of clinical tests, procedures, and examinations. Finally, issues of patient adherence to treatment plans are minimal in hospitals as compared with other settings of care; the physical environment is within the organization’s control and patient compliance is easier to monitor and ensure.
OCR for page 242
Medicare: A Strategy for Quality Assurance - Volume I Detection Quality assessment efforts have been hindered by the virtual absence of sound clinical evidence of the efficacy of much medical care (Komaroff, 1983; Eddy and Billings, 1988; O’Leary, 1988). Quality assurance programs have had to construct standards for reviewing care based on a combination of the best available information, expert opinion, locally developed consensus of practitioners, and guidelines provided by specialty societies. In recent years, hospitals have adopted systems that track indicators of performance. Using methods of epidemiologic surveillance, they identify patterns of nosocomial infections and other unusual occurrences by practitioner, unit, shift, or procedure (Istre et al., 1985; Lynch and Jackson, 1985; Buehler et al., 1986; Sacks et al., 1988). Incident reporting systems are designed for rapid reporting of problems, although the value of these systems is debated (Craddick and Bader, 1983; GAO, 1987). Occurrence screening (discussed in Chapter 9) is widely used by hospitals to detect adverse events. Quality measurement in hospitals tends to be driven by the requirements of the Joint Commission. In the past its requirements focused heavily on structure and process—specifying quality assurance activities for the medical staff, nursing service, and so forth. More recently, Joint Commission standards began to emphasize “monitoring and evaluation” of high-volume, high-risk, or problem-prone activities, and hospitals, in turn, have reoriented their quality assessment and assurance activities. The Joint Commission is now moving toward a considerably more complex focus on patient outcomes and the effective use of patient outcome data in quality assurance efforts (Joint Commission, 1987, 1988). These efforts address specific conditions and physician specialties such as anesthesia and obstetrical care as well as hospital care generically. The success of problem detection activities in the hospital setting rests heavily on the adequacy of patient records and data systems. In the hospital, the dates of admission and discharge define the episode of care. The individual patient admission is the unit of analysis for quality assessment. Patient care assessments, interventions, and responses are recorded in a single medical record. Although the completeness, legibility, retrievability, and other characteristics of patient records can vary widely across hospitals, Joint Commission standards for medical records as well as the function of these records as a primary legal document both dictate their form and constrain their content. After discharge, the hospital prepares a discharge abstract summarizing the main features of the medical care provided and patient status at the time of discharge. Admission and discharge diagnoses are routinely assigned based on a standard coding system (ICD-9-CM).3 A code specifying the diagnosis-
OCR for page 243
Medicare: A Strategy for Quality Assurance - Volume I related group (DRG) is derived from the ICD-9-CM codes as part of the uniform billing system (UB-82 and for Medicare the HCFA-1500), and considerable research and administrative work have gone into trying to assure the accuracy of diagnostic coding (GAO, 1988). Since the inception of the prospective payment system, which tied reimbursement to DRG (principal diagnosis), the coding has become more precise and (some feel) more accurate, although the most recently reported error rate of 20.8 percent is still fairly high (Hsia et al., 1988). The present coding scheme is not adequate to determine the severity of illness, the reason for death, the sequence of diagnoses, or sometimes even the precise location of a procedure. For example, whether myocardial infarction preceded or followed coronary artery bypass surgery during a given admission will not be detectable from coded data; neither can the specific site of a hip replacement (i.e., right or left hip) be known from such data. Records of medical and nursing interventions are typically extensive in the hospital chart, but the accuracy of the information may be compromised by concerns about confidentiality, utilization review, and malpractice liability (Burnum, 1989). Bearing in mind that the test data themselves may not be wholly accurate, results of tests available in the record form a relatively comprehensive picture of events occurring during an inpatient episode. Thus, for quality assessment purposes, medical, nursing, and other clinical data are available, although retrieval may entail manual abstracting of voluminous amounts of (sometimes poorly organized) material. The field of health information systems is growing rapidly. Increasing numbers of community hospitals have clinical-care-based management information systems, and investment in sophisticated software is expected to rise with internal and external demands for information and technical capability, all of which tend to take for granted the suitability of medical record-based information. Feinglass and Salmon (forthcoming, pp. 3–4)) optimistically describe the near horizon in medical informatics: The development of advanced telecommunications, fiberoptic cabling and local area networking, new magnetic and optical data storage technologies, new print graphics, optical scanning and video display capabilities, speech recognition and voice synthesis methods, and new structured query languages and integrated operating system software have the potential to revolutionize the practice of medicine and the medical work place.4 Perhaps within a few years, most major hospitals will have bedside computer terminals, with voice input devices and a host of logical audit functions for interpreting medical staff orders, storing clinical data, and providing uniform standards of quality assurance. These systems will not solve all problems. Virtually no data on previous ambulatory or inpatient care at other hospitals are available in the hospital
OCR for page 244
Medicare: A Strategy for Quality Assurance - Volume I record beyond a typically brief admitting history that may omit significant past events and illnesses. Patient satisfaction questionnaires may be distributed at the time of discharge, but are not usually part of the medical record. No information about the outcome of inpatient care is gathered routinely after discharge. Thus, assessment of the quality of care must depend on data collected during what is usually a short hospital stay. In most moderate and large hospitals, patient care is very visible to peers because of the many practitioners who see both the patient and his or her medical record. Major procedures are generally observed by an array of physicians, nurses, and others, who are expected to use the incident and screening systems to report adverse occurrences. Teaching hospitals have substantial numbers of trainees and systems of supervision and accountability. There are multiple opportunities for consultation and patient care conferences, especially in the teaching hospital. The situation may be quite different in small rural hospitals relying on a handful of physicians. Ideally, guidelines in the form of expert systems could be contemporaneous with care and help the practitioner in decision making. Some prototype interactive systems now provide computerized clinical reminders (McDonald et al., 1984; Tierney et al., 1986, 1988), predictive value of tests, and warnings of drug-drug interactions or misuse of antibiotics.5 Implementation of these systems is likely to occur first in the hospital environment. Correction Medical data systems tend to be more useful for detecting poor technical quality rather than overuse or underuse. Where data permit the development of performance profiles, results can be distributed to providers and performance followed over time. These data can figure prominently in decisions about clinical and staff privileges, which are the chief means of interventions exercised by hospital officials and directed at physicians who are considered to provide substandard care. Assuming that firm standards of care can be developed and applied for quality assessment, peer pressure from colleagues and pressure from chairpersons of departments and others with clinical authority can be brought to bear on those individuals not performing according to expectations. Nevertheless, these arrangements can be weak instruments of problem correction, especially in smaller institutions where personal and social ties may be stronger than organizational procedures or leadership or where medical staff are intimidated by legal action from sanctioned practitioners. During site visits for this study, people responsible for hospital quality assurance repeatedly noted extreme difficulties in dealing with problem staff members
OCR for page 245
Medicare: A Strategy for Quality Assurance - Volume I who threaten costly and personally bruising legal suits. Of particular concern is the recent U.S. Supreme Court decision in Patrick v. Burget that opens peer review activities to challenge under antitrust statutes. Although the legal implications of this case are not yet clear, it appears already to have had a chilling effect on some peer review activities (Meyer, 1989), and it has added to the difficulty of recruiting clinical chairpersons who are responsible for annual physician review and reappointments. Apart from quality problems relating to individual staff is an array of systems and interpersonal problems, including cumbersome or irrational bureaucratic procedures, disputes across medical specialties and medical-nonmedical domains, strong hierarchies, and separate medical and nursing quality assurance systems. These system characteristics may (1) create patient care problems themselves, (2) complicate the correction of problems, or (3) prevent the resolution of disagreements about appropriateness of practice (Flood and Scott, 1978; Shortell and LoGerfo, 1981; Hetherington, 1982). During site visits we heard about routine delays of several weeks in receiving autopsy results; surgical specialists not calling in medicine specialists soon enough; disputes between the departments of medicine and surgery on appropriateness standards for endoscopy; inordinate delays in initiating drug therapy because of pharmacy problems; delayed admissions to the intensive care unit (ICU), an increasing patient:nurse ratio in the ICU resulting in demonstrably increased morbidity; long trips to the hospital for magnetic resonance imaging tests for patients with head injury; and delayed patient discharges because plans for prostheses were not timely and forms were difficult to complete. Hospitals have been buffeted by major changes in their internal operating environment. Efforts to identify and to correct problems are complicated by the shrinking length of stay, a higher patient acuity level, staffing shortages, rapid turnover of employees, and the increasing use of part-time and hourly workers. Because individual clinicians or managers may find it harder now than previously to recognize patterns of problems, they may need to rely more on formal systems of quality assessment and assurance. In addition, the hospital is increasingly subject to external, community factors that can dramatically affect how and how well it operates. For example, the urban hospital has been stressed by AIDS, the drug epidemic with its related trauma, premature births, child abuse, lack of long-term care beds, abandonment of elderly dependents, increasing numbers of uninsured patients, and growing numbers of non-English-speaking patients. Hospital efforts are also complicated by the absence of an integrated insurance program for the elderly that covers both acute and long-term care beds. This encourages discontinuity of care and communication gaps across settings of care that affect both the quality and efficiency of care.
OCR for page 246
Medicare: A Strategy for Quality Assurance - Volume I Ambulatory Care Prevention As noted earlier, ambulatory care settings vary widely, and these variations can affect the systems used to prevent quality problems. For instance, credentialing systems in a health maintenance organization (HMO) may be quite organized and extensive; by contrast, the solo practitioner or small fee-for-service (FFS) group typically is subject only to state medical licensing requirements, and a practice partnership may require board certification in a medical or surgical specialty but no more organized assessment of partners’ performance. Each setting may have different resources devoted to developing patient care systems to prevent problems. For instance, systems for obtaining initial or full partnership privileges in an HMO may be extensive. By contrast, to practice as a solo practitioner the physician need only have a state medical license. Most ambulatory patient-care arrangements have not been designed to provide a systematic means of preventing or identifying quality problems. Follow-up and coordination are generally uneven; peer review is less intense, if it exists at all. In addition, the patient has a more central role. The health care system is limited in what it can do to prevent problems associated with a patient’s personal or work life. For ongoing review and management, a staff-model HMO may have teams of health professionals who engage in informal and formal peer-review programs and hold patient care conferences on a regular basis. A group-model HMO may contract with a multispecialty group practice that, in turn, has its own quality assurance programs. Both may have administrative and clinical professionals to design ambulatory data systems to track care and follow-up involving test results, specialty referrals, and preventive services. By contrast, independent practice associations (IPAs) or FFS solo and small group practice provide little opportunity for sharing observations and expertise of many practitioners, although some IPA-model HMOs have extensive internal claims systems that allow the monitoring of both ambulatory and inpatient services. Some large FFS multispecialty groups may maintain single patient records for their patients; others have separate records in each department in which the patient is seen. On the other hand, solo and small practices can have positive aspects. Long-term physician-patient relationships allow a practitioner to understand and appreciate the medical history and social needs of individual patients. Such relationships may compensate for the fragmentation of care and medical records that are common in systems where no single provider has primary responsibility for a patient’s care.
OCR for page 247
Medicare: A Strategy for Quality Assurance - Volume I Detection Quality evaluation must begin by examining either a process of care to which standards can be applied or an outcome that can be defined and measured such that the assessors can work back from a measurably unsatisfactory outcome to the relevant poor process of care. Standards of ambulatory practice are less well-defined than standards for inpatient care. For example, the proper use of insulin for diabetics, antibiotics for a sore throat, and diagnostic testing for sickle cell anemia are three common but controversial areas (Komaroff, 1983). Ambulatory quality assessment depends on monitoring the process of care for those patients who obtain it. This requires some system for identifying those who have received care for specific conditions or problems, which is referred to as case-finding. Case-finding based only on recorded diagnosis (when possible) will not detect missed or incorrect diagnoses, yet identifying early serious disease is a crucial aspect of ambulatory quality. Physician offices generally have neither manual nor computerized case registry systems for case-finding. Outcomes can also be used for case-finding. For example, recent efforts by Medicare Peer Review Organizations (PROs) and others to screen in HMOs for poor ambulatory care using the criterion of a hospital admission that might have been preventable may help to detect deficiencies in both diagnosis and treatment. Determining whether a poor outcome results from poor care is, however, very difficult. At least three factors (discussed below) are at work: (1) the difficulty of specifying and measuring outcomes, (2) the difficulty in defining an episode of care, and (3) the incomplete and fragmentary nature of data. Measuring outcomes. Outcomes of care are hard to specify in ambulatory care and even more difficult to measure. The natural history of conditions (for example, their tendency to be self-limiting in many cases) is seldom well defined, and often we have little solid information on how medical interventions should affect outcomes (Palmer, 1983, 1988; Palmer and Adams, 1988; Lohr, 1988). Furthermore, relying on patterns of outcomes for patients seen by specific physicians is difficult for three reasons. First, most physicians have low volumes of particular types of cases compared to hospitals. Second, case mix probably varies markedly across physicians. Third, patients may well choose physicians because of their interventionist or noninterventionist style. In addition, data systems in ambulatory care generally lack information about outcomes of care other than physiologic measures.
OCR for page 248
Medicare: A Strategy for Quality Assurance - Volume I Defining episodes of care. Measurements of the process of care hinge on the ability to define and measure an episode of care.6 Such definition is occasionally possible, for example, with a series of vaccinations or the diagnosis and complete healing of a fracture. Typically, reviewers must make arbitrary decisions about the beginning and end of an episode and about the relevance of intervening office visits, emergency room visits, or telephone calls involving one or more (known) practitioners. Such arbitrary decision making reflects our lack of understanding of the natural history of most ambulatory conditions. It is difficult to judge whether a series of visits by the same patient for similar problems represents (1) continued appropriate treatment (one episode of care), (2) an independent recurrence (two episodes), or (3) a failure of treatment (one episode). Many efforts at assessment avoid the issue of an episode of care and base the assessment on some other unit of analysis, such as a single-patient contact, completion of a health screening protocol, or a patient’s summary assessment of satisfaction with care. Incomplete and fragmented data systems. Case-by-case review of the process of care using ambulatory records is often seriously compromised by poor organization of data and lack of documentation of key aspects of care in the medical record. The degree of documentation in the medical record is said to vary inversely with the experience of the clinician—the most extensive notes reportedly are written by medical students. Where used the problem-oriented medical record (POMRs) might help a reviewer to evaluate the medical logic of care (such as the relationship between a set of patient complaints and their diagnostic evaluation), but the use of the POMR in its complete format is rare. In general, clinical data systems in ambulatory care are poorly developed, and billing systems rarely include clinical information. The Uniform Ambulatory Care Data Set, approved in June 1989 by the National Committee on Vital and Health Statistics, includes certain clinical data (DHHS, 1981; Felts, 1989). In addition to patient and provider identifiers, patient demographic descriptors, the date and place of encounter, and itemized charges, recommended encounter information would include: problem; diagnosis or assessment; laboratory, radiology, and surgical services; and the patient’s reason for encounter (optional). If this minimal data set is widely adopted, claims data might be suitable for initial screening in ambulatory care for those who have access to and the expertise to analyze these data. Even beyond the difficulties in assessing care from the review of one chart, patients may independently see multiple providers, with little or no communication or record linkage among them. In the hospital setting, practitioners, patients, and their records are gathered in one place; in the ambu-
OCR for page 254
Medicare: A Strategy for Quality Assurance - Volume I Correction Seventy-five percent of states report terminating contracts with home health agencies as their most common enforcement mechanism in the provision of home health services. Many withhold funds or recoup money already paid out. Some states impose intermediate sanctions for noncompliance, such as cutting off the intake of new clients or reassigning caseloads to other providers (Riley, 1988). A more appealing method of improving quality (in home care as in other settings) would be the feedback of timely information to providers on their patterns of care with enough specificity and precision to generate improved performance. Such feedback, although it may occur as informal evaluation by supervisors, has not been as fully developed in home health care agencies as in hospital care. PAYMENT SYSTEMS The organizational, financial, and philosophical differences between traditional fee-for-service (FFS) practice and prepaid group practice in this country have been the subject of very intense examination over the past two decades—yielding a literature far too extensive to recount here (Luft, 1981, 1988; Brown, 1983). As the prepaid capitated approach has matured, it has spawned several variants and hybrids, complicating the task of specifying what systems factors are likely to encourage or impede successful quality assurance. This section attempts to highlight some key variables that should be taken into consideration in designing a quality assurance strategy for Medicare. Today, three-quarters of nonfederal, nonresident physicians have a practice (including those who belong to preferred provider organizations, or PPOs9); that is, approximately 300,000 physicians practice through FFS rather than salaried or capitated arrangements. The percentage of FFS physicians is falling, however, and differences by age of physician are impressive. For example, excluding residents, 47 percent of physicians under age 36 are now (at least partly) salaried, compared to only 19 percent over age 55 (cited in Feinglass and Salmon, forthcoming). Still, 97 percent of Medicare beneficiaries remain in FFS arrangements, despite substantial growth in capitated and salaried payments systems and strong promotion by the Health Care Financing Administration (HCFA) of beneficiary enrollment in Medicare HMOs and competitive medical plans. Thus, for the purposes of this study, the factors that must be considered in designing a strategy for quality assurance for Medicare are overwhelmingly those of the FFS sector.
OCR for page 255
Medicare: A Strategy for Quality Assurance - Volume I Fee-for-Service Practice Fee-for-service office practice has not generally been a focus of quality assurance programs.10 A strategy for quality review for FFS practitioners would require a system to choose review methods and topics, to establish review criteria, to gather data, to disseminate results, to monitor change, and to intervene when necessary to correct problems. No such systems are currently in place. Hence, we have little past history or experience to guide the design of such a strategy.11 Data In the FFS system (with billing to an insurance carrier), virtually all services paid by a health plan will be recorded and coded. In principle, this makes case-finding for quality assessment possible. In practice, many obstacles to the use of claims data must be overcome. They include incomplete or inaccurate recording or coding of diagnoses and services; noncovered (but key) services that are not recorded at all (e.g., prescription drugs for Medicare); covered services for which no claims are made; and noncovered services that are miscoded as covered ones (e.g., preventive services, such as screening mammography coded as diagnostic mammography). Most important, because billing systems are not created to provide any clinical data (e.g., results of laboratory tests, measures of severity of illness, or even accurate diagnoses), quality review that uses claims data as a screen must eventually look to other sources of information to assess the actual care rendered. As mentioned earlier with respect to ambulatory care generally, medical records are scattered in individual physician offices and use nonstandardized recording systems. Practitioners may develop their own terminology, check-off lists, and flow sheets to record progress. They may use the briefest of notations knowing that no other practitioner (other than, perhaps, a practice partner) is likely to need or read them (the main exceptions to this statement are when a patient transfers to another physician or an attorney requests the record pursuant to malpractice action). Thus, the office medical record may be a very poor or uneven source of information nationally about either the process or the outcome of patient care, especially if that care involves a considerable amount of listening and counseling (as contrasted with provision of diagnostic tests and procedures or prescriptions). HCFA is now requiring diagnostic coding on ambulatory Part B claims. Both the reliability and validity of that information will have to be assessed before it can be considered a useful data source for quality assurance. For
OCR for page 256
Medicare: A Strategy for Quality Assurance - Volume I instance, because most of the elderly have multiple diagnoses and the diagnostic codes have no impact on payment or treatment, many physicians may merely note a legitimate diagnosis, such as hypertension, even though another condition was the primary reason for a visit. If reliability of diagnostic data is established, it may be possible to apply automated logic programs to assess the appropriateness of ambulatory services based on acute or recurring diagnoses. Also the possibility exists that services can be linked across settings (such as hospital to outpatient), but, again, this deserves careful investigation. Types of Problems Because of the fundamental financial incentives for overuse of services, FFS quality assurance systems need to look carefully for overuse. When the FFS coverage has large deductibles and copayments or major coverage exclusions, however, then underuse also becomes a concern. The underuse problem may especially manifest itself among people never “entering the system,” whereas overuse may occur through additional services even within a single visit. Procedure-oriented practices may make overuse more likely because of a reimbursement structure that rewards procedures at the expense of counseling, cognitive services, or lengthy personal interaction between patient and physician; in this sense, the payment scheme may underwrite the cost of other services such as patient education and preventive care. In all cases, however, attention must be given to finding poor technical quality (PPRC, 1989). Prepaid Capitated Systems In some ways the prepaid group practice sector, although small, has advanced on the quality assurance front more than the FFS sector has. Prepaid plans were developing quality assurance programs long before the HMO Act of 1973 (P.L. 93–222) required federally qualified HMOs to have ongoing programs that “stress health outcomes” (see Weiner and Densen, 1958; Shapiro et al., 1960, 1967, 1976; Morehead et al., 1964). Federally qualified HMOs have to meet certain financial and structural requirements, to have ongoing quality assessment programs, and to be able to demonstrate this to outside reviewers. Even nonfederally qualified HMOs and HMOs serving Medicaid patients are often called on by state departments of health to demonstrate their quality assurance programs. For instance, in California the Knox-Keene Health Care Service Plan Act of 1975 stipulated that California HMOs assess the acceptability and accessibility of care and the adequacy with which the HMO met the health needs of the served population. In Kansas, a new law requires independent, on-site quality-of-care inspec-
OCR for page 257
Medicare: A Strategy for Quality Assurance - Volume I tions at least once every three years. Despite these regulatory requirements, in practice HMO quality assurance programs have ranged from “paper plans” with little substance to extensive and well-supported internal programs with research components. More recently, Medicare PROs have begun to review the ambulatory care for the one million Medicare beneficiaries belonging to HMOs and competitive medical plans (CMPs) with Medicare risk contracts (see Chapter 6 and Volume II, Chapter 8). PRO review, which is fairly complex and exceeds what is required for the FFS sector, includes review of ambulatory care in addition to inpatient and other settings and calls for an emphasis on “episodes of care.” For instance, the records of outpatient care for patients admitted for any of 13 specified diagnoses that might indicate substandard ambulatory management can be subjected to review. Moreover, some HMOs and CMPs that opt for so-called limited review are subject to evaluation of both their existing quality assurance program and a periodic sample of the cases that their program itself is investigating. Data Data systems in the prepaid practice sector can be both better and worse than those in the FFS sector. The major difference, of course, is that there are generally no claims forms—a source of appreciable problems in implementing Medicare PRO review of HMO and CMP care. In principle, an advantage of prepaid over FFS systems is that the former can describe and document their enrolled populations (and can correlate use of services to this base). In practice, problems arise because people are enrolled in multiple systems, for example when each employed spouse covers all family members under his or her employer’s health plan. When this occurs, an apparent “nonuser” may actually be receiving services in another system. Management information systems (MIS) in HMOs typically have minimal clinical information. They may, for instance, use CPT-412 codes for classifying procedures or ICD-9-CM codes for hospital diagnoses, but the quality of these data might be suspect because they are for information management use, not for payment per se. MIS data are typically recorded in modules, such as membership-information files, specialty referrals, and encounter (usually a patient visit) information. A hospitalization module may include dates of admission and discharge, summary charges and diagnoses, but no more detailed information. Encounter forms may record aspects of the clinical encounter (e.g., date, practitioner, type of visit, disposition, and primary diagnoses), but there are no strong incentives or disincentives (other than the time required of the practitioner to complete the form) to record services or aspects of the visit accurately. Other information may be
OCR for page 258
Medicare: A Strategy for Quality Assurance - Volume I retrievable from the MIS (e.g., registration and demographic data on members and authorized specialty referrals), but integration of the data may be cumbersome, not especially timely, and expensive. The MIS may be developed internally or be commercial with major modification by software vendors. As a result, the definitions of basic terms, coding, and recording systems are idiosyncratic and vary widely from one HMO to another. Clinical data for review of care are typically available only in the medical record and supplemented by other sources such as laboratory or pharmacy logs. The medical record system within any one HMO may have a standard format, and a unit record may be designed so that information about care in many settings (e.g., emergency room visits, hospital discharge summary, and specialty referrals) are recorded in the primary care chart. As in the FFS system, however, medical records may differ dramatically across different HMOs. Group and staff models have the benefit of one or several practice locations in which records are gathered. Independent practice association (IPA) and network models, on the other hand, can resemble FFS practice. An IPA would find it very difficult to require participating physicians to use common data systems because physicians often participate in many different IPAs with only a small proportion of their practice enrolled in any given IPA. Types of Problems Because of the financial incentives for restricting access to expensive services such as specialty care or elective hospitalization, HMOs need to look carefully for underuse. The structure of financial incentives in HMOs may lead fewer people to be nonusers, but it may also reduce services among users. HMOs are also especially prone to “procedural” barriers to access (e.g., lengthy queues for appointments) and to barriers to physician-initiated or authorized services where “gatekeeper” physicians are at financial risk for patient use of these services. HMOs tend not to have incentives that favor overuse in general or procedures over cognitive services, but they can nevertheless be subject to problems of overuse of services. Problems of poor technical care must be sought as vigorously as in FFS system. OTHER SETTING AND SYSTEM FACTORS We emphasized in Chapter 1 that our definition of quality of care—and by extension, the quality assurance system that should be designed to reflect that definition—was intended to include both individuals and popula-
OCR for page 259
Medicare: A Strategy for Quality Assurance - Volume I tions. Thus, a feature of singular importance to a quality assurance system that crosses both setting and payment systems is its ability to define the population intended to be served and the population actually served. For the HMO and the Medicare program, the populations of users as well as nonusers can be defined (subject to the caveats noted above). In the hospital setting, in ambulatory FFS practice, and in home care, however, only the population served is defined. For these settings and payment systems, quality assessment systems can develop such measures as rates of compliance with standards, rates of adverse occurrences, and rates of patient satisfaction for those who receive services, but they cannot evaluate the status of those who do not obtain care. Furthermore, we also emphasized “desired patient outcomes” as the central focus of our definition. Leaving aside the appreciable (but not insurmountable) difficulties with defining and measuring outcomes, it is important to note that outcome measures that are applied only to those who receive care may be misleading. They may point to ostensible differences in outcome that are in fact caused by differences in access. A quality assurance program must be able to correct for this problem in one of two ways (or both). It must be able to adjust for case mix before comparing care rendered in different institutions, settings, and payment systems. It should also be able to assess population-wide outcomes through mechanisms that will reach both users and nonusers. SUMMARY This chapter has introduced issues for quality assurance that are related to the setting of care (hospital, physician office-based care, and home health care) and to the payment mechanism (FFS and prepaid capitated systems). The more highly organized and integrated the provider setting, the more likely it is to have in place the informal and formal mechanisms (for example, specifications for granting practice privileges; good data systems; and formal peer review activities) that help to assure the quality of care delivered. Thus, the barriers to successful internal quality assurance are lower for hospitals than for ambulatory and home care. With respect to payment systems, FFS and HMO provider groups probably have offsetting advantages and disadvantages relating to availability of data, types of problems, and focus on individuals versus populations. In general, the design of a quality assurance strategy for Medicare faces more difficulties for the ambulatory and home health settings than for hospitals, regardless of payment systems. For that reason and because of the overwhelming preference of Medicare beneficiaries for the FFS system, the challenge to designing such a strategy for that system is greater than for the HMO sector.
OCR for page 260
Medicare: A Strategy for Quality Assurance - Volume I NOTES 1. Conditions of Participation for hospitals to be reimbursed by the Medicare program, and the role of Joint Commission accreditation by which accredited hospitals are “deemed” to meet those conditions, are discussed more fully in Chapter 5 and in Volume II, Chapter 7. 2. This aspect of quality assurance is emphasized by the Joint Commission and is receiving increasing attention as a result of passage of the Health Care Quality Improvement Act (1986). For more discussion, see Chapter 9 and Volume II, Chapter 6. 3. International Classification of Diseases, 9th Revision, Clinical Modification. 4. Cox and Zeelenberg, 1987; DeTore, 1988; Kaplan, 1988. 5. At the National Institutes of Health Clinical Center, for example, orders for antibiotics entered on the clinical management information system are screened for appropriateness. 6. The need for defining an episode of illness is sometimes voiced. Defining an episode of illness is even more difficult than defining an episode of care because of the complexities of establishing either a beginning or an ending to the illness, both of which occur outside the physician’s office and are unlikely to be recorded systematically, if at all, in the office records. 7. Mammography guidelines, for instance, are the subject of considerable controversy among various medical groups (McIlrath, 1989). 8. Much of this section is based on a paper prepared for this study by Catherine Hawes and Robert Kane on measuring outcomes in noninstitutional long-term care, hereafter referred to as Hawes and Kane, 1989. 9. PPOs are organizations or insurers that contract with practitioners to provide services to enrollees on a discounted FFS basis. The attraction to physicians is mainly a more stable, “guaranteed” patient population without the constraints or requirements of belonging to a traditional group or staff model HMO. 10. The most prominent exception may have been the efforts of the Experimental Medical Care Review Organizations (EMCROs) and Professional Standards Review Organizations (PSROs) to carry out ambulatory care review efforts in the 1970s. Efforts of earlier foundations of medical care were also carried out by physicians in “traditional” practices. The Association for Accreditation of Ambulatory Health Care also promotes ambulatory care review. One major research effort concerned the California Medicaid program, in particular, the Prepaid Health Research and Evaluation Demonstration (PHRED) project in which reviewers took microcomputers to physicians’ offices to evaluate process of care. None of these efforts, however, has been translated into a widely accepted, national consensus or activity for review of care rendered in physicians’ offices. 11. The Medical Outcomes Study (Tarlov et al., 1989) is measuring the process and outcomes of care in both capitated and fee-for-service settings. See Volume II, Chapter 6. 12. Current Procedural Terminology, fourth edition.
OCR for page 261
Medicare: A Strategy for Quality Assurance - Volume I REFERENCES Brown, L.D. Politics and Health Care Organization: HMOs as Federal Policy. Washington, D.C.: The Brookings Institution, 1983. Buehler, J.W., Smith, L.F., Wallace, E.M., et al. Unexplained Deaths in a Children’s Hospital: An Epidemiologic Assessment. New England Journal of Medicine 313:211–216, 1986. Burnum, J.F. The Misinformation Era: The Fall of the Medical Record. Annals of Internal Medicine 110:482–484, 1989. Chassin, M.R., Kosecoff, J., Park, R.E., et al. Does Inappropriate Use Explain Geographic Variations in the Use of Health Care Services? A Study of Three Procedures. Journal of the American Medical Association 258:2533–2537, 1987. Cox, J.R. and Zeelenberg, C. Computer Technology: State of the Art and Future Trends. JAAC 9:204–214, 1987. Craddick, J.W. and Bader, B.S. Medical Management Analysis: A Systematic Approach to Quality Assurance and Risk Management. Volume I. Auburn, Calif.: J.W.Craddick, 1983. Darling v. Charleston Community Memorial Hospital, 33 ILL. 2D 326, 211 N.E.2d 253 (1965), cert denied, 383 U.S. 946 (1966). DHHS (Department of Health and Human Services). Uniform Ambulatory Medical Care. Minimum Data Set. A Report of the United States National Committee on Vital and Health Statistics. DHHS Publ. No. (PHS) 81–1161. Washington, D.C.: U.S. Government Printing Office, 1981. Dubois, R.W., Rogers, W.H., Moxley, J.H., et al. Hospital Inpatient Mortality: Is it a Predictor of Quality? New England Journal of Medicine 317:1674–1680, 1987. DeTore, A.W. Medical Informatics: An Introduction to Computer Technology in Medicine. American Journal of Medicine 85:399–403, 1988. Eddy, D.M. and Billings, J. The Quality of Medical Evidence. Health Affairs 7:19–32, Spring 1988. Feinglass, J. and Salmon, J.W. Corporatization and the Use of Medical Management Information Systems: Increasing The Clinical Productivity of Physicians. International Journal of Health Services, forthcoming. Felts, W.R. Need for Standardized Ambulatory Care Data: National Provider Perspective. Paper presented at the American Public Health Association annual meeting, Chicago, Ill., October, 1989. Flood, A.B. and Scott, W.R. Professional Power and Professional Effectiveness: The Power of the Surgical Staff and the Quality of Surgical Care in Hospitals. Journal of Health and Social Behavior 19:240–254, 1978. GAO (General Accounting Office). VA Health Care: VA’s Patient Injury Control Program Not Effective. HRD-87–49, Washington, D.C.: General Accounting Office, May 1987. GAO. Improving Quality of Care Assessment and Assurance. PEMD-88–10, Washington, D.C.: General Accounting Office, May 1988.
OCR for page 262
Medicare: A Strategy for Quality Assurance - Volume I Hawes, C. and Kane, R.L. Issues Related to Quality Review and Assurance in Home Health Care. Paper prepared for Institute of Medicine Study to Design a Strategy for Quality Review and Assurance in Medicare, 1989. Hetherington, R.W. Quality Assurance and Organizational Effectiveness in Hospitals. Health Services Research 17:185–201, 1982. Hsia, D.C., Krushat, W.M., Pagan, A.B., et al. Accuracy of Diagnostic Coding for Medicare Patients Under the Prospective-Payment System. New England Journal of Medicine 318:325–355, 1988. Istre, G.R., Gustafson, T.L., Baron, R.C., et al. A Mysterious Cluster of Deaths and Cardiopulmonary Arrests in a Pediatric Intensive Care Unit. New England Journal of Medicine 313:205–211, 1985. Joint Commission (Joint Commission on Accreditation of Healthcare Organizations). Agenda for Change Update 1:1, 1987. Joint Commission. Agenda for Change Update 2:1, 5, 1988. Kaplan, B. Development and Acceptance of Medical Information Systems: An Historical Overview. Journal of Health and Human Resources Administration Summer:9–29, 1988. Komaroff, A.L. The Doctor, the Hospital, and the Definition of Proper Medical Practice. Appendix E. in President’s Commission. Summing Up. The Ethical and Legal Problems in Medicine and Biomedical and Behavioral Research. Washington, D.C.: President’s Commission for the Study of Ethical and Legal Problems in Medicine and Biomedical and Behavioral Research, 1983. LaLonde, B. Quality Assurance Manual of the Home Care Association of Washington, 2nd Edition. Edmonds, Wash.: The Home Care Association of Washington, 1988. Lohr, K.N. Outcome Measurement: Concepts and Questions. Inquiry 25:37–50, 1988. Luft, H.S. Health Maintenance Organizations: Dimensions of Performance. New York, N.Y.: Wiley, 1981. Luft, H.S. HMOs and the Quality of Care. Inquiry 25:147–156, 1988. Luke, R.D. and Modrow, R.E. Professionalism, Accountability, and Peer Review. Health Services Research 19:113–123, 1982. Lynch, P. and Jackson, M.M. Monitoring: Surveillance for Nosocomial Infections and Uses for Assessing Quality of Care. American Journal of Infection Control 13:161–173, 1985. McDonald, C.J., Hui, S.L., Smith, D.M., et al. Reminders to Physicians from an Introspective Computer Medical Record. Annals of Internal Medicine 100:130–138, 1984. McIlrath, S. 11 Medical Groups Endorse Mammogram Guidelines. AMA News, July 14, 1989, pp. 3, 48. Meyer, H. Peer Review’s Limits Visible Once Again. AMA News, May 5, 1989, p. 1, 9–12. Morehead, M.A., Donaldson, R.S., Sanderson S., et al. A Study of the Quality of Hospital Care Secured by a Sample of Teamster Family Members in New York City. New York, N.Y.: Columbia University School of Public Health and Administrative Medicine, 1964. O’Leary, D.S. The Need for Clinical Standards of Care. Editorial. Quality Review Bulletin 14:31–32, 1988.
OCR for page 263
Medicare: A Strategy for Quality Assurance - Volume I Palmer, R.H. and Reilly, M.C. Individual and Institutional Variables Which May Serve As Indicators of Quality of Medical Care. Medical Care 17:693–714, 1979. Palmer, R.H. Ambulatory Health Care: Evaluation Principles and Practice. Chicago, Ill.: American Hospital Association, 1983. Palmer, R.H. Beyond Hospital Inpatient Care: The Challenges and Prospects for Quality Assessment and Assurance in Ambulatory Care. Inquiry: 119–131, 1988. Palmer, R.H. and Adams, M.E. Considerations in Defining Quality of Health Care. Paper prepared for Institute of Medicine Study to Design a Strategy for Quality Review and Assurance in Medicare, 1988. PPRC (Physician Payment Review Commission). Annual Report to Congress. 1989. Washington, D.C.: Physician Payment Review Commission, 1989. Rhee, S.O. Organizational Determinants of Medical Care Quality: A Review of the Literature. Pp. 127–146 in Organization and Change in Health Care Quality Assurance, Luke, R.D., Krueger, J.C. and Modrow, R.E., eds. Rockville, Md.: Aspen Systems Corporation, 1983. Riley, P.A. Quality Assurance in Home Care. Report prepared for the National Academy for State Health Policy, an affiliate of the Center for Health Policy Development. Washington, D.C., December 1988. Sacks, J.F., Stroup, D.F., Will, M.L., et al. A Nurse-Associated Epidemic of Cardiac Arrests in an Intensive Care Unit. Journal of the American Medical Association 259:689–695, 1988. Scott, W.R. Managing Professional Work: Three Models of Control for Health Organizations. Health Services Research 17:213–240, 1982. Shapiro, S., Jacobziner, H., Densen, P., et al. Further Observations on Prematurity and Perinatal Mortality in a General Population and in the Population of a Prepaid Group Practice Medical Care Plan. American Journal of Public Health 50:1304–1317, 1960. Shapiro, S., Steinwachs, D.M., Skinner, E.A., et al. Survey of Quality Assurance and Utilization Review Mechanisms in Prepaid Group Practice Plans and Medical Care Foundations. Baltimore, Md.: Johns Hopkins Health Services Research and Development Center, 1976. (HEW contract HSM 110–73–435). Shapiro, S., Williams, L, Yerby, A., et al. Patterns of Medical Use by the Indigent Aged under Two Systems of Medical Care. American Journal of Public Health 57:784–790, 1967. Shortell, S.M. and LoGerfo, J.P. Hospital Medical Staff Organization and Quality of Care: Results for Myocardial Infarction and Appendectomy. Medical Care 19:1041–1055, 1981. Steinwachs, D.M., Weiner, J.P., and Shapiro, S. Management Information Systems and Quality. Pp. 160–182 in Providing Quality Care: The Challenge to Clinicians. Goldfield, N. and Nash, D.B., eds. Philadelphia, Pa.: American College of Physicians, 1989. Tarlov, A.R., Ware, J.E., Greenfield, S., et al. The Medical Outcomes Study. Journal of the American Medical Association 262:925–930, 1989. Tierney, W.M., Hui, S.L., and McDonald, C.J. Delayed Feedback of Physician Performance Versus Immediate Reminders To Perform Preventive Care. Medical Care 24:659–666, 1986.
OCR for page 264
Medicare: A Strategy for Quality Assurance - Volume I Tierney, W.M., McDonald, C.J., Hui, S.L., et al. Computer Predictions of Abnormal Test Results: Effects On Outpatient Testing. Journal of the American Medical Association 259:1194–1198, 1988. Weiner, L. and Densen, P. Comparison of Prematurity and Perinatal Mortality in a General Population and in the Population of a Prepaid Group Practice Medical Care Plan. American Journal of Public Health 48:170–185, 1958. Wennberg, J.E., Freeman, J.L., and Culp, W.J. Are Hospital Services Rationed in New Haven or Over-Utilised in Boston? The Lancet i:1185–1188, May 23, 1987.
Representative terms from entire chapter: