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 207
Medicare: A Strategy for Quality Assurance - Volume I 7 Quality Problems and the Burdens of Harm INTRODUCTION Good health care requires the technical proficiency and means to deliver services correctly and the cognitive and communication skills necessary to elicit and evaluate needed information and then decide which mix of available services is most likely to achieve desired health outcomes for particular patients. When these requirements are not met patients are at risk. Choices among methods to prevent, detect, and correct quality problems should be based on (1) how well they do at detecting different types of problems such as overuse versus underuse of care, and (2) what relative burden of harm is imposed by these different problems. This burden can be quantified in terms of incidence, distribution across populations, and degree of impact on patient outcomes (such as functional status and survival). Other things being equal, if we knew that problems of overuse caused twice as much harm as those of underuse, we would devote twice as much of our quality assurance resources to those specific techniques designed to discover and correct overuse problems. Importance of Poor Technical and Interpersonal Quality We use the term “technical quality” to refer generally to the ways health care is delivered by individuals and organizations, such as whether correct diagnoses are made, appropriate medications prescribed, or surgical procedures performed skillfully. It includes not only practitioner knowledge and manual or technical skills, but also interpersonal skills—listening, answering questions, giving information, and eliciting and including patient preferences in decision making. Poor interpersonal skills can deny patients the
OCR for page 208
Medicare: A Strategy for Quality Assurance - Volume I information they need to make informed choices, care for themselves, or adhere to treatment plans. Those problems, as well as poor organization and coordination of various aspects of patient care, can lead to harm ranging from discomfort or distrust to disfigurement or death. The classic approaches to quality assurance, including risk management and infection control, focus on poor technical quality, particularly problems of skill, performance, and system functioning. Within hospitals, such methods of problem detection as surgical review, morbidity and mortality conferences, incident reporting, and review for adverse occurrences have this focus. Problems with poor technical quality may involve one or more “outlier” practitioners whose skills are inadequate, or they may involve broader weaknesses in a system. Practitioners or institutions that are outliers in one area cannot be presumed to be outliers in all areas of practice. For instance, the hospital-specific mortality data released by the Health Care Financing Administration (HCFA) identified very few hospitals that were high-mortality outliers across several different diagnoses and very few that were outliers across several years (see Chassin et al., 1989). Moreover, “good” practitioners or providers cannot be assumed to be uniformly good (Palmer, 1988). In any case, methods for dealing with outliers may differ from those appropriate for responding to problems with average performance. Importance of Overuse in Quality of Care Some problems with quality of health care can be classified under the term overuse. Overuse is the provision of services whose likelihood of harm to the patient outweighs the likelihood of benefit. Benefits include increased life expectancy, relief of pain, reduction in anxiety, or improved functional capacity. Harms include the morbidity and mortality that accompany the provision of service (such as a surgical procedure). They also include other, less commonly measured adverse effects such as the anxiety of anticipating and undergoing a procedure, time lost from work, and time spent in rehabilitation. Excessive diagnostic services may have direct negative side effects and may also, if falsely positive, lead to other more invasive and hence more risky examinations and treatments. A case in point is the patient who has an unnecessary exercise stress test that is falsely positive and who as a result undergoes coronary angiography (Graboys, 1989). Excessive use of medications such as antibiotics (Foxman et al., 1987) exposes patients and the population in general to unjustified side effects, for example the proliferation of antibiotic-resistant organisms. When patients are hospitalized unnecessarily, they risk falls, medication errors, and hospital-acquired infections.
OCR for page 209
Medicare: A Strategy for Quality Assurance - Volume I In addition, resources are limited, and overuse of some services may preclude others from getting needed care. Use of the intensive care unit by a patient who does not need it, for instance, may delay or prevent care for another patient who does need it. Overuse of services adds to the cost of care for the individual and for society. The study committee’s concern with overuse as a quality issue focuses on whether the services provide a net expected benefit to a patient, not whether those resources might have produced more benefits if applied elsewhere. Importance of Underuse in Quality of Care Underuse is the lack of provision of services whose expected benefits outweigh their expected risks to the patient. In less technical terms, underuse is a missed opportunity, for instance, the omission of a preventive service such as mammography screening of older women that carries a small risk but greatly enhances the likelihood of survival if it speeds detection of an operable tumor. Harms can be a foregone cure, a condition not improved, or a symptom not ameliorated. Again, the study committee’s focus is on the net benefits and risks to the patient, not the relative costs and benefits for society. Differentiating Problems of Poor Technical Quality, Overuse, and Underuse Often it is quite apparent whether a problem stems from poor technical quality, overuse, or underuse. Problems of poor technical quality are most easily differentiated from overuse when a clear intervention, such as a surgical procedure or prescribing decision, is involved. The decision to intervene may itself be appropriate (not overuse), but the execution (technical quality) is improper. For instance, an outmoded technique is used, the wrong dose of medications prescribed, an instrument misread, or a test result ignored. In other situations, such as in long-term management of chronic conditions, distinguishing poor technical quality, overuse, or underuse may be much more difficult. A physician who suspects essential hypertension could, for example, explore the problem over a series of patient visits, laboratory tests, and trials of medication. Incorrect choices in the process of diagnosis and treatment could result in delayed diagnosis (underuse by insufficient testing), incorrect medication (poor technical quality), or unnecessary or potentially harmful tests (overuse). Iatrogenic illness or complications arising from medical treatment result from side effects of unnecessary medications or drug interactions (overuse), infection because hospital procedures are not followed (underuse), or preventable complications of a procedure
OCR for page 210
Medicare: A Strategy for Quality Assurance - Volume I such as hemorrhage following careless surgical technique (poor technical quality). Similarly, failure to obtain informed consent for services might be viewed as underuse of the informed consent procedure, overuse of services that the informed patient would have foregone, or poor interpersonal care because the patient’s preferences were not elicited. These examples suggest that classifying problems may sometimes be rather arbitrary. Underuse of services can be conceptually linked to overuse in that both are concerned with the appropriate use of services as opposed to the technical competence in providing them. Underuse and overuse may also coexist. For instance, some admissions might be regarded as overuse of hospital care because of underuse of nonhospital nursing care. A quality assurance technique that detects the former may not detect the latter. Whether the hospital care or lack of nursing care is avoidable or governed by external factors such as geography or availability of nursing staff is an issue for any quality assurance program. To the extent that both overuse and underuse of services arise from lack of physician knowledge of the natural history of the patient’s condition, of the patient’s preferences, or of the probabilities of various outcomes of diagnostic tests and therapy, they may represent the same problem; regulatory and financial mechanisms are not likely to be effective. Other interventions, such as better dissemination and use of available medical knowledge or continuing medical education, may be more useful. Sources of Information About Burdens of Harm Ideally, before we design a quality assurance program for the elderly, we should understand the relative burdens of harm created by poor technical quality, overuse, and underuse. This knowledge would permit us to build specific mechanisms tailored to the efficient identification and effective amelioration of these quality problems. Unfortunately, our knowledge in this area is meager. Physicians and other practitioners frequently complain that current programs require much effort but often overlook or fail to address what they consider to be serious problems. We have attempted to assess the burden of harm imposed on the elderly by the three major categories of quality problems. To do this the committee used four sources of information. First, we commissioned two background papers—one from a medical perspective (Rubenstein et al., 1989) and one from a nursing perspective (Lang and Kraegel, 1989)—to examine the state of health care for the elderly as it has been documented in the research literature. A third commissioned paper concerned home health care (Hawes and Kane, 1989). Second, during public hearings, we asked respondents whether they could estimate how widespread quality problems were and
OCR for page 211
Medicare: A Strategy for Quality Assurance - Volume I whether those were problems primarily of overuse, underuse, or poor technical quality (see Chapter 2 in Volume II). Third, we raised these same questions in the focus groups held among elderly individuals and among practicing physicians and during our extensive site visits around the country (Chapters 3 and 4 in Volume II). Finally, we used published studies and available data on disciplinary and malpractice actions. The following examination of the burden of harm necessarily emphasizes problems. It should not imply that health care in this country is generally of dubious quality. Participants in the beneficiary focus group were generally satisfied with their primary physicians and the medical care they received. Many commented positively on the Medicare program itself, asserting that adequate health care would be a financial burden without the assistance of Medicare. They viewed medical care in the United States as very good and as one of the best medical systems in the world. Other positive aspects of medical care frequently mentioned were scientific advances, the level of medical technology, increased efficacy of medications and pharmaceuticals, and a high skill level among providers of care. EVIDENCE OF PROBLEMS IN TECHNICAL QUALITY OF CARE We reviewed both direct and inferential evidence to assess the frequency and severity of poor technical quality of care. Sources of such evidence include the literature on clinical and health services research, national data on malpractice compensation, and disciplinary actions by state boards of medical examiners. For evidence specific to the elderly, we considered sanctions recommended by Medicare Peer Review Organizations (PROs) and imposed by the Department of Health and Human Services (DHHS) (see Chapter 6) and the results of the committee’s public hearings, site visits, and focus groups. General Evidence of Poor Technical Quality Malpractice Data Malpractice data are often assumed to be a good source of information concerning poor technical performance of physicians or other providers. Although a review of these data by the Office of Technology Assessment (OTA, 1988) throws that assumption into considerable question, we report some of the better known work, because it provides a sense of the more egregious problems with technical quality of care. A Pennsylvania study, which was commissioned by the Pennsylvania Medical Society and Pennsylvania Trial Lawyers Association, showed that 1 percent of Pennsylvania physicians (228) were responsible for 25 percent
OCR for page 212
Medicare: A Strategy for Quality Assurance - Volume I of the malpractice loss payments by the Pennsylvania Medical Professional Liability Catastrophe Loss Fund (Wolfe, 1986). Of this 1 percent, nearly half had three or more loss claims over a 10-year period. Similarly, in Michigan, 2.5 percent of physicians accounted for almost 20 percent of all claims, and just under 20 percent of physicians accounted for over 70 percent of all claims (Wolfe, 1986). In Florida, the Orlando Sentinel reported that 3 percent of doctors were responsible for almost half of all malpractice claims paid in the state between 1975 and 1984 (cited in Wolfe, 1986). In 1987, there were 6.7 professional liability claims per 100 physicians (Slora and Gonzales, 1988). About 37 percent of physicians have had a malpractice claim filed against them at some time during their practicing career. The mere filing of a claim provides no information, however, about whether malpractice occurred. In the most comprehensive analysis to date, the General Accounting Office reported on claims closed in 1984 (GAO, 1987).1 Fifty-seven percent were dismissed without a verdict, settlement, or any compensation going to the claimant. Seventy-one percent of the providers sued were physicians; 21 percent were hospitals, and the remainder involved other facilities, nurses, dentists, and others. Forty-two percent of the physicians had previous malpractice claims filed against them. The three most frequent allegations, surgical error, failure to diagnose, and treatment errors, accounted for 69 percent of all malpractice claims reviewed. The data on principal allegations in closed claims files are not differentiated by overuse, underuse, and poor technical quality, although they might be so categorized by imputation. For instance, the nearly 18,700 allegations of surgical errors accounted for 25 percent of the claims, and almost 90 percent of these could be classified as poor technical quality (retained foreign bodies, improper positioning, or wrong body part) and 8 percent as overuse (unnecessary surgery and failure to obtain consent for surgery). For claims of diagnostic errors (about one quarter of all allegations), 38 percent could be linked to poor technical quality (misdiagnosis, improper performance of diagnostic test), 1 percent overuse (failure to obtain consent), and 60 percent to underuse (failure to diagnose, or delay in diagnosis). Allegations of treatment errors constituted nearly 20 percent of all allegations. Of these, 75 percent can be classified as poor technical quality (improper performance, improper choice of treatment) and 21 percent for underuse (failure to render treatment, delay in treatment). Data provided by one of the nation’s largest underwriters of malpractice liability insurance, St. Paul Fire and Marine Insurance Company, reflect similar experience. The St. Paul’s 1988 Annual Report to Policyholders (1987) cited 15.4 claims per 100 physician policyholders. Major allegations involved surgical errors (29 percent of claims, mainly related to complications), failure to diagnose
OCR for page 213
Medicare: A Strategy for Quality Assurance - Volume I (28 percent, mainly related to cancer), improper treatment (27 percent, mainly birth-related), and anesthesia errors (3.5 percent). Other types of allegations accounted for 12 percent of claims. Of the total, 67 percent referred to the hospital, 32 percent to the physician’s office or clinic, and 1 percent to a surgicenter. Data on claims filed or closed cannot be used to estimate the prevalence of quality problems in medical care in the United States for several reasons. Patients may not know that malpractice has occurred; patients may know but may choose not to make a claim; or they may act, but legitimate claims may be rejected. These reasons can be labelled “false negatives,” that is, the absence of a successful claim despite actual malpractice. Among successful claims, on the other hand, may be some false-positives. For example, a claim may be settled even though it lacks merit, or for strategic reasons a malpractice complaint may include practitioners who were only tangentially involved in the patient’s care. Such false-positives mean that claims closed with compensation cannot give even a minimum estimate of poor quality. Research Evidence Although more than a decade old, the California Medical Insurance Feasibility (CMIF) Study (Mills, 1977) remains the most comprehensive estimate of the incidence of “potentially compensable events” (an injury worthy of seeking legal recourse) occurring in short-stay acute hospitals. This study, which reviewed records of about 21,000 discharged patients, found that 4.65 percent of admissions resulted in some potentially compensable event (PCE), defined as a “disability (temporary or permanent) caused by health care management (including acts of commission and omission by health providers).” Seventeen percent of these were judged to have involved legal liability. By applying these figures to the 38.8 million patients hospitalized in the United States in 1983, Wolfe (1986) estimated that there were 310,400 people injured (or killed) as a result of negligent behavior. Comparing the frequency of PCEs in the CMIF data with data on insurance claims led investigators to believe that only about 1 valid claim in 10 was brought at that time. Since claims frequency has doubled since the 1974 data were gathered, Danzon (1985) estimates that today only one in five actionable claims are brought and that poor care resulting in temporary or permanent disability or death occurs in 1 percent of hospital admissions.2 A prospective study by Couch and his colleagues (1981) looked at the frequency of surgical mishaps (adverse outcomes because of error) in the field of general surgery. They identified 36 mishaps in a review of 5,612 admissions (0.6 percent) and described five sources of physician error: overestimation of surgical skills, unwarranted urgency in performing major sur-
OCR for page 214
Medicare: A Strategy for Quality Assurance - Volume I gical procedures, urge for perfection beyond the patient’s needs, uncritical performance of vogue therapies, and insufficient restraint and deliberation in patient care. The last four problems might be considered to result from overuse rather than poor skills. When Wolfe (1986) applied this rate of mishaps to all surgical admissions in the United States, he estimated 136,000 injuries to surgical patients caused by doctor error. Steele and his colleagues (1981) reviewed the care of 815 consecutively admitted elderly patients to a medical service of a university hospital and found that 36 percent suffered some form of adverse complications attributable to medical management. About 9 percent of the 815 had a major event that was life-threatening or disabling, and in 2 percent of the cases the patient died. The most frequent complications involved drug reactions, complications of cardiac catheterization, and falls. The authors cautioned that the teaching hospital where the study was done might have had a higher proportion of seriously ill patients who might have been more difficult to treat. Other studies reviewed by McPhee et al. (1982) point to avoidable and sometimes severe reactions resulting from polypharmacy, transfusion reactions, and nosocomial (hospital-acquired) infections. Hospitals using Medical Management Analysis (MMA), a system of standardized reporting of adverse occurrences, are reported to identify “adverse patient occurrences” (APOs) in about 18 to 20 percent of patient hospital records (Craddick and Bader, 1983). Adverse patient occurrences are defined as “any untoward patient event which, under optimal conditions, is not a natural consequence of the patient’s disease or procedure” (Craddick and Bader, 1983, p. 7). A study of surgeons who had been recommended by peers as “good” surgeons revealed that some 23 percent of charts showed one or more APOs. These records were then further reviewed by their peers to determine whether management had been acceptable, questionable, or a breach of the standard of care. Questionable management was defined as “those instances in which the peer reviewer might have managed the case differently, but the subject physician’s management was within the range of variation of standard practice” (Craddick and Bader, 1983). The reviewers identified questionable management for 12 percent of the APOs involving 4 percent of the patients and no cases that suggested that the standard of care was breached. The percentage of APOs per surgeon ranged from 18 percent to 31 percent and was correlated with the complexity of the surgery (Craddick and Bader, 1983). Dubois and Brook (1988) reported that among 182 deaths reviewed independently by at least three physicians, 14 percent were probably or definitely preventable. Reasons varied by type of admission and included errors in management and or diagnosis. Research studies in which medical practice as recorded in the medical
OCR for page 215
Medicare: A Strategy for Quality Assurance - Volume I chart is compared with standards set by peers and experts virtually always demonstrate substantial deficiencies. In fact, differences among providers and among practice settings are usually smaller than those between average and “ideal” practice, even when practitioners have set their own standards. For instance, a review of antibiotic use in a community hospital by Jogerst and Dippe (1981) found substantial misuse. Only 72 percent of therapeutic uses and only 36 percent of prophylactic uses were appropriate, according to standards developed by the physicians in the hospital under study. Disciplinary Actions by State Medical Boards Reports on disciplinary actions by state medical boards can provide information about the prevalence of medical incompetence. The Federation of State Licensing Boards (FSLB) reported that in 1986, 2,302 disciplinary actions were taken against the country’s approximately 500,000 practicing physicians, an overall rate of 4.6 per 1,000 physicians.3 The rate of disciplinary actions by medical boards varied by state and territory from 0 to 15 per 1,000 practicing physician. This variation probably reflects state differences in laws, attitudes toward regulation, and board willingness to engage in disciplinary activities (OTA, 1988). Disciplinary actions covered by the above figures are revocation of license, suspension, probation, and reprimand. The State boards may also take milder actions. One example is a letter agreement in which no harm is alleged and the physician acknowledges a problem (such as impairment) and agrees to enter a program for impaired physicians; other informal actions include letters of concern and recommendations for continuing education. The Public Citizen Health Research Group (Public Citizen) compared 1987 FSLB data on serious disciplinary actions to previous years (Wolfe, 1989). Disciplinary actions rose for three consecutive years and in 1987 reached 2.78 serious disciplinary actions for every 1,000 United States doctors.4 The lowest ranking state had 0.45 serious actions per 1,000 physician. The highest ranking state had 8.58, a 19-fold difference. Although there is no evidence that the proportion of poor practitioners is evenly distributed among the states, Public Citizen estimates that if all states had the highest rate the overall rate of disciplinary actions would have been three times higher. Wolfe (1989) also cited a Tufts University study that found that physician-owned insurance companies terminated coverage of 6.6 per thousand physician policyholders in 1985 because of “negligence-prone behavior.” They restricted the scope of covered practice or imposed other sanctions on an additional 7 per 1,000 policyholders because of substandard care. Public
OCR for page 216
Medicare: A Strategy for Quality Assurance - Volume I Citizen noted that the combined rate of 13.6 terminations or other sanctions per 1,000 is almost five times the 1987 rate of serious disciplinary actions by state licensing boards. The most common violation, and the one accounting for one-half of disciplinary actions, is inappropriate prescription writing, often for controlled substances. The second major category of violation is substance abuse (drugs or alcohol, or both). A report from the Office of Inspector General (OIG, 1986) points out that this second category is expanding both in absolute and proportionate terms. Inappropriate prescription writing and substance abuse together account for three-quarters or more of all disciplinary actions. Other kinds of violations acted on by the state medical boards, such as professional misconduct, fraud, economic violations, or felony conviction, are not directly related to technical competence. Derbyshire (1984) has hypothesized that 10 percent of physicians are professionally “incompetent” to practice, based on his experience on a state board of medical examiners and as past President of the FSLB.5 Evidence of Poor Technical Quality for the Elderly PRO Sanctions and Corrective Actions Sanctions imposed by DHHS on the basis of recommendations by PROs are a potential source of data about the rate of poor quality care for Medicare beneficiaries. As of September, 1989, the OIG reported that since the start of the PRO program it had received 197 referrals for possible sanctions, had imposed 110 sanctions, rejected 79, and had 8 cases pending or moot because of the physicians’ death or retirement (unpublished data). Sanction data should be considered, at best, a very minimal measure of the rate of poor quality care because most of those sanctioned are cited for multiple violations, because PROs undertake many more corrective actions than sanction recommendations, and because of the extensive due process accorded physicians during the sanctioning process, including the “willing or able” requirement for physician exclusion (see Chapter 6). The number of corrective action plans together with sanction recommendations might provide a more reasonable estimate of the number of physicians with recognized quality-of-care problems. During the second PRO Scope of Work (a two-year period from 1986 to 1988), PROs identified more than 82,600 physicians with quality problems among the estimated 300,000 physicians who bill Medicare. It resolved 75,200 cases and instituted nearly 65,750 interventions (HCFA, 1989). When asked during the public hearing whether the number of sanctions could be used to estimate the prevalence of physicians that one “would not send a neighbor to,” a representative of the California PRO replied that this
OCR for page 217
Medicare: A Strategy for Quality Assurance - Volume I description would apply to perhaps 6 to 8 percent of the 50,000 physicians in the state. However, during a 19-month contract period the California PRO had sent first notices of sanction to only 137 physicians and forwarded just 14 recommendations for sanction to the OIG, a very small fraction of the almost 50,000 physicians practicing in that state. The medical director of another PRO estimated that about 5 percent (200) of 3,800 practicing physicians in his state account for 95 percent of the quality problems with perhaps 60 physicians accounting for approximately 80 percent of the quality problems. Quality problems were instances detected by PRO quality screens and confirmed after review by a PRO physician and possibly other specialists. To further hinder the estimation problem, more than one respondent during our site visits pointed out that PRO review includes hospital care only. They estimated that far more physicians with poor skills evade detection than are detected during PRO review, because they do not have hospital privileges or do not claim reimbursement from Medicare. SuperPRO data (relating to re-review of charts initially reviewed by Medicare PRO physicians) were discussed in Chapter 6, and will be only briefly recapped here. According to a GAO report (1988a), the SuperPRO found that between 2.0 and 8.5 percent of hospital admissions had a quality problem, but this figure is based on a nonrandom selection of 17 percent of claims paid.6 Clinical Research Evidence on Quality of Care for the Elderly In their reviews of the literature, both Rubenstein et al. (1989) and Lang and Kraegel (1989) examined the research evidence of poor technical quality of care for the elderly. Both papers cited extensive evidence of deficient treatment for conditions such as breast cancer, psychiatric disorders and confusion, pneumonia, urinary tract infection, incontinence, pressure sores, and malnutrition. They also reported patient management deficiencies in drug therapy, functional disability, and monitoring of fluids. Their review suggested the need for quality standards for ethical aspects of care and for rehabilitation and supported indications that higher nurse staffing ratios might reduce patient falls, improve the mental status of cognitively impaired elderly, strengthen hospital discharge planning, and provide more structured teaching for self-care. Beneficiary Complaints Data on beneficiary complaints about poor quality care are sketchy. The numbers of complaints and the percentages of confirmed quality problems vary greatly by PRO. During the second year of the second Scope of Work (roughly 1988), the rate of complaints per 100,000 beneficiaries ranged
OCR for page 227
Medicare: A Strategy for Quality Assurance - Volume I be detected by the usual methods of quality assessment. These tend to be based less on population-based evidence than on adverse occurrences or evidence from samples of patients. Underuse by specific subgroups can be identified by purposive sampling. Evidence of Underuse Among the Elderly Medicare Data Continued concern about the effects of Medicare’s PPS has prompted greater attention to the possibility of premature hospital discharge, which is a type of underuse among the elderly. An OIG review of over 7,000 hospital admissions by Medicare patients between October 1984 and March 1985 concluded that only 0.8 percent of discharges were premature (OIG, 1988b). Premature discharges were characterized by inadequate treatment and incomplete therapies; occurred most often in small, rural, nonteaching facilities; and were often associated with quality-of-care problems during the hospital stay. PROs have been directed to review hospital readmissions as a screen for possible premature discharge as well as other problems in quality that may have led to readmission within 31 days. The Medicare Prospective Payment Assessment Commission (ProPAC, 1989) reviewed data on patterns of readmission from Medicare claims files from 1984 to 1986 but found no evidence in readmission statistics to indicate significant quality problems related to premature discharge. In their review of generic screen failures based on a 3-percent random sample of Medicare discharges, PROs confirmed that patients were not stable at the time of discharge in less than 1 percent (0.87 per 100) of the records reviewed (ProPAC, 1989). One area where declining length of stay has been thought to provide evidence that quality of care has been jeopardized is in care of patients with hip fracture. Fitzgerald et al. (1988) found that since implementation of PPS and declining lengths of stay, hospitals have reduced the amount of rehabilitative care given and have discharged more patients to nursing homes (an increase from 38 to 60 percent); in addition, the number of patients remaining in nursing homes after one year rose from 9 to 33 percent. HMO patients in that study were also discharged to nursing homes at a higher rate than before PPS, but only 16 percent of HMO patients were still in nursing homes after a year, compared to 35 percent of other patients. Russell (1989) regards this finding as indicative of differences in nursing home care management rather than an underuse effect of PPS. If so, the data do not permit differentiating such patterns as underuse or poor technical quality. The OIG study cited earlier also sought to identify the incidence of poor care, defined as “medical care clearly failing to meet professionally recog-
OCR for page 228
Medicare: A Strategy for Quality Assurance - Volume I nized standards under any circumstances in any locale” (OIG, 1989b, p. 1). After data were adjusted for hospital size to reflect the population of Medicare patients in all hospitals, the proportion receiving poor quality care was estimated to be 5.5 percent. Eighty percent of the reasons for poor quality care involved the omission of necessary services, for instance, failure to order or provide appropriate tests and services either at all or in a timely manner. Many hospitals with high rates of poor quality care also had high rates of unnecessary admissions and premature discharges. Six DRGs were frequently associated with poor quality care: DRG 14 (strokes, except transient ischemic attack); DRG 15 (transient ischemic attacks); DRG 87 (pulmonary edema and respiratory failure); DRG 89 (simple pneumonia and pleurisy, patients over age 69); DRG 141 (fainting, patients over age 69); and DRG 320 (kidney and urinary tract infections patients over age 69) (OIG, 1989b). In response to a legislative emphasis on quality, the PROs have been reviewing cases failing generic screens. ProPAC reported that a generic screen-based study of 3,250 records found 162 quality problems (5.0 percent). Indication of underuse might be found in these researchers’ estimate that an additional 5.3 percent of records had problems (not detectable in the generic screens) that involved “insufficient attention to medical problems, rather than problems with the care that was provided” (ProPAC, 1989, p. 34). PRO review of “sentinel admissions” by patients enrolled in risk-contract prepaid group practices is oriented to possibly insufficient ambulatory care (see Chapter 6), but no findings have yet been reported. Other Research Evidence of Underuse of Services for the Elderly With respect to underuse of services, Rubenstein et al. (1989) cited two articles indicating underuse of emergency services and intensive care units for the elderly. They also documented many different forms of underdiagnosis and undertreatment in their review of the clinical literature. Physicians spend less time with the elderly and diagnose a smaller proportion of older patients’ medical problems. When diagnostic assessments made by geriatricians are compared with those by patients’ primary care or hospital physicians, all 11 studies they reviewed showed less complete diagnostic services for the elderly. Diagnoses often missed by primary care physicians include: treatable incontinence, curable infections, gait disorders, and metabolic problems. Similar findings were reported in six studies of nursing home patients, in six studies of elderly patients in general medical clinics, and in five of six studies of elderly patients later diagnosed with cancer. In all these settings, it appeared that underdiagnosis was prevalent among elderly patients, and, in those studies that made the comparison, less common among comparable groups of younger patients.
OCR for page 229
Medicare: A Strategy for Quality Assurance - Volume I Rubenstein and his colleagues also reviewed studies of undertreatment of older patients in hospital settings, nursing homes, and general outpatient settings. Studies in all settings found a high prevalence of treatable conditions for which treatment was less adequate for older compared to younger patients. Problems included underprovision of rehabilitation in both hospital and nursing home settings, a need for more acute care in nursing homes, deficiencies in disease-specific care including underuse of mental health services for depression, delayed and less aggressive treatment for cancer patients, and underprovision of acute care and rehabilitation for older versus younger stroke patients. In short, Rubenstein and his colleagues saw major gaps in the health care delivered to the elderly. They also pointed out that because they reviewed only published studies of patients already receiving care, their estimates probably understated problems of underuse in the community. In their review of the nursing literature for areas of underuse of those nursing interventions that maintain or restore function, Lang and Kraegel (1989) emphasized evaluation of the functional status of the elderly and the impact of loss of function and increased dependency for those with chronic disease. They reported evidence of underuse of home-care nursing services because some elderly do not know that services are available or because services are prematurely terminated. Older females have differentially fewer home visits than males despite greater functional impairment and fewer available caretakers. Public Hearings, Focus Groups, and Site Visits During the public hearings, focus groups, and site visits, the study committee heard frequently about lack of Medicare coverage for long-term care, preventive services, and primary care. Cost barriers were reported to result in implicit rationing based on class, sex, and ability to pay. Underuse was also linked to isolation (economic, social, transportation, and housing). Nearly one-half of all public testimony respondents expressed concern that the quality of health care would decrease as a result of cost-containment measures. Premature discharges, utilization review, financial factors for underuse, and health care decisions being made by the “wrong people” (e.g., staff of FIs, PROs, and third-party payers) are examples of the cost-containment concerns mentioned. More than one-third of public hearing respondents believed that the quality of care for the elderly is less than optimal because of deficiencies in Medicare coverage of health-related services that encourage independent living for those with certain chronic conditions. These include custodial, homemaker, and other services. Other respondents stressed that the Medicare reimbursement system does not take into account the health needs of the elderly that relate to quality of life. The public hearing testimony did
OCR for page 230
Medicare: A Strategy for Quality Assurance - Volume I not elicit specific information about underdiagnosis or undertreatment of patients who were already using the system. Elderly participants in focus group mentioned interpersonal issues in care more than problems of access or undertreatment. Issues mentioned included the amount of time and interest devoted to them by providers (especially in contrast to younger patients), the amount of nursing care and the quality of attention in the nursing home. They did express concern about PPS leading to premature discharge. During the site visits, our contacts raised numerous concerns about underuse. These included failure to diagnose (particularly acute pneumonia, myocardial infarction, and meningitis) and inadequate follow-up of positive diagnostic findings. Other points raised were the lack of preventive services (influenza vaccination and cancer screening), inadequate mental health care, and lack of home care services. Findings to date do not indicate that Medicare PPS has had demonstrable negative effects on quality of care or the health of the elderly population (e.g., in terms of higher mortality rates). Nevertheless, the hearings, focus groups, and site visits show that patient and provider groups remain apprehensive about the future. SUMMARY Evidence of overuse of health care services is substantial; virtually all factors of a fee-for-service system promote it. Information on underuse is more sketchy and inferential; incentives of capitation are a source of concern. Underuse is hard to detect through available surveillance systems. One background paper for this study concluded that, for the elderly, underuse is more prevalent than overuse, but no estimate could be made of the actual burden of harm (frequency and severity). Finally, setting- and disease-specific examples of poor technical quality can be found in many arenas, although they do not translate easily into national estimates of levels of quality of care. Little information is available on national estimates of prevalence of quality problems, and no data document the burden of harm by problem, by disease category, by setting, or by population group such as the elderly. This lack of data is not surprising, for several reasons. Compiling data on national quality problems has not been seen as Medicare’s responsibility; no current monitoring system can provide such data; and provider groups do not view their internal quality assurance programs as needing to contribute to national estimates of quality problems. Nevertheless, there is good reason to believe that problems exist in all three areas, of different magnitudes in different settings. For this reason,
OCR for page 231
Medicare: A Strategy for Quality Assurance - Volume I the study committee believes it prudent, indeed necessary, that a Medicare quality assurance program develop and use appropriate measures to track all three types of problems and to find ways to estimate the burden of harm better, so that differential emphasis on improvement can be placed where most needed. NOTES 1. The investigators sampled 25 insurers from a pool of 102 insurers that in 1984 closed an estimated 73,500 claims involving about 103,300 health care providers. They then analyzed a sample of 2,781 claims for types of allegations, which include claims closed with and without compensation. 2. The final report of the Harvard Medical Practice Study, which is expected to be published early in 1990, has been designed to investigate the incidence during hospitalization of injuries resulting from medical intervention and the proportion of these injuries that are due to substandard care (Hiatt et al., 1989). 3. The 1986 figure was based on numbers of all duplicated disciplinary actions reported to the Federation of State Licensing Boards for 1986, divided by numbers of practicing physicians reported by the American Medical Association (AMA, 1986) using the sum of nonfederal office-based practitioners and full-time hospital-based (nonresident) physician staff. 4. The figure is lower than the 1986 data quoted above because it includes only “serious actions”—revocations, suspensions, and probations—and does not include reprimands (or remedial education). 5. The terms professional incompetence, professional malfeasance, moral turpitude, and repeated malpractice are not used with consistency or precision in state legislation. Derbyshire (1984) defines professional incompetence as the “inability of a physician to care for patients satisfactorily because of such failings as faulty judgment, unreliability, unavailability, and professional obsolescence” (p. 136B). He intends this definition to include impairment from substance abuse, but it is not clear whether inappropriate prescription writing or moral turpitude are intended to be or are, in practice, widely included in the term incompetence. Nor is it clear whether incompetence can be established by one incident or requires a pattern of behavior. Thus, comparisons of various estimates of “rates of incompetence” are probably not valid. 6. One might expect the SuperPRO sample to have a higher rate of quality problems than a true random sample of claims. The SuperPRO reviews randomly selected cases reviewed by each PRO, but the PRO pool of cases is not a random selection of all hospital admissions. In addition to the “3 percent random sample” nearly five times as many cases are selected either because there is reason to believe a quality problem exists for that case or for reasons unrelated to poor technical quality per se, but perhaps related to overuse of services or to coding problems. Generally speaking, however, because of the nonrandom nature of PRO cases, the SuperPRO estimates cannot be used to estimate either the incidence of poor quality care in the 83 percent of claims not
OCR for page 232
Medicare: A Strategy for Quality Assurance - Volume I reviewed or the number of inappropriate admissions. Of the 1,187 cases identified (and re-reviewed) by the SuperPRO, they found 58 cases where the potential for patient risk was of a serious nature (4.9 percent); 3 cases where actual reversible or minor harm was done to the patient (0.25 percent); and 9 cases where irreversible or significant harm was done to the patient (0.76 percent). 7. Retrospective review of appropriateness depends on the information that is recorded in the medical record. Factors relevant to the treatment decision might not have been documented. 8. Many of these studies were done in the early 1980s; rates of inappropriate use may have changed since then. Furthermore, such rates may vary among different populations. For instance, Chassin et al. (1987) found one-third of carotid endarterectomies to be inappropriate in each of three large geographic areas in 1981. Merrick et al. (1986) found that only 13 percent of procedures among 95 patients in California Veterans Administration Medical Centers were “clearly inappropriate.” Physicians in the United Kingdom and in the United States reviewed the same cases of coronary artery bypass surgery. U.K. physicians judged 35 percent of cases and U.S. physicians 13 percent of cases to be inappropriate, a difference that was interpreted as indicating substantial cultural differences (Brook et al., 1988). 9. When asked about the distribution of overuse, underuse, or poor technical quality, most study respondents found it difficult to make estimates. Issues of underuse were most frequently expressed as access and benefits problems. Concerns expressed about continuity, fragmentation, and shared decision making, which occurred with some frequency, did not fall easily into any of our three categories; we generally viewed them as reflecting poor technical quality. 10. The most extensive evidence of the effect of cost-sharing on use of services comes from The RAND Corporation’s decade-long Health Insurance Experiment (HIE). Although the HIE did not include the elderly, its strong results across all other age groups make clear that even relatively mild levels of copayments and deductibles have marked impacts on use of services (Newhouse et al., 1981, 1987). The evidence concerning the effect of cost-sharing on adult health status is less powerful—the main findings indicate that having completely free care significantly improved the health of adults in two areas of particular importance for the elderly: hypertension and visual acuity (Brook et al., 1983; Keeler et al., 1985). Cost-sharing may also have had a differential effect on persons of low income (particularly children). Finally, cost-sharing was found to be at best only a blunt instrument for curtailing the use of inappropriate services; except for persons of higher income and education, cost-sharing reduced the use of appropriate and inappropriate ambulatory services to about the same degree (Lohr et al., 1986).
OCR for page 233
Medicare: A Strategy for Quality Assurance - Volume I REFERENCES American Bar Association. The Black Box of Home Care Quality (a report prepared for the Chairman of the Select Committee on Aging, U.S. House of Representatives). Comm. Publ. No. 99–573, Washington, D.C.: Government Printing Office, August 1986. American Medical Association. Physician Characteristics and Distribution in the U.S., 1986 Edition. Chicago, Ill.: American Medical Association, 1986. Banta, H.D. and Kemp, K.B. Background Paper No. 4: The Management of Health Care Technology in Ten Countries. Washington, D.C.: Office of Technology Assessment, 1980. Brook, R.H. and Lohr, K.N. Will We Need To Ration Effective Health Care? Issues in Science and Technology 3:68–77, Fall 1986. Brook, R.H., Ware, J.E., Jr., Rogers, W.H., et al. Does Free Care Improve Adults’ Health? Results from a Randomized Controlled Trial. New England Journal of Medicine 309:1426–1434, 1983. Brook, R.H., Kosecoff, J.B., Park, R.E., et al. Diagnosis and Treatment of Coronary Disease: Comparison of Doctors’ Attitudes in the U.S.A. and the U.K. Lancet i:750–753, 1988. Brown, R.E., Sheingold, S.H., and Luce, B.R. Options of Using Practice Guidelines in Reducing the Volume of Medically Unnecessary Services. BHARC-013/89/ 027. Washington, D.C.: Battelle Human Affairs Research Centers, 1989. Bunker, J.P. Surgical Manpower. A Comparison of Operations and Surgeons in the United States and in England and Wales. New England Journal of Medicine 282:135–144, 1983. Chassin, M.R., Brook, R.H., Park, R.E., et al. Variations in the Use of Medical and Surgical Services by the Medicare Populations. New England Journal of Medicine 314:285–290, 1986. 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. Chassin, M.R., Park, R.E., Lohr, K.N., et al. Differences among Hospitals in Medicare Mortality. Health Services Research 24:1–31, 1989. Couch, N.P., Tilney, N.L., Rayer, A.A., et al. The High Cost of Low-Frequency Events—The Anatomy and Economics of Surgical Mishaps. New England Journal of Medicine 304:634–637, 1981. Craddick, J.W. and Bader, B.S. Medical Management Analysis, A Systematic Approach to Quality Assurance and Risk Management, Vol. I. Auburn, Calif.: J.W.Craddick, 1983. Danzon, P.M. Medical Malpractice: Theory, Evidence, and Public Policy. Cambridge, Mass.: Harvard University Press, 1985. Derbyshire, R.C. Medical Discipline in Disarray: Retrospective and Prospective. Hospital Practice 19:136A-136P, June 1984. Dubois, R.W. and Brook, R H. Preventable Deaths: Who, How Often, and Why? Annals of Internal Medicine 109:582–589, 1988.
OCR for page 234
Medicare: A Strategy for Quality Assurance - Volume I Eddy, D.M. Variations in Physician Practice: The Role of Uncertainty. Health Affairs 3:74–89, 1984. Eddy, D.M. and Billings, J. The Quality of the Medical Evidence. Health Affairs 7:20–32, 1988. Eisenberg, J.M. Sociologic Influences on Decision-Making by Clinicians. Annals of Internal Medicine 90:957–964, 1979. Eisenberg, J.M. Doctors’ Decisions and the Cost of Medical Care. Ann Arbor, Mich.: Health Administration Press Perspectives, 1986. Fitzgerald, J.F., Moore, P.S., and Dittus, R.S. The Care of Elderly Patients with Hip Fracture. New England Journal of Medicine 319:1393–1397, 1988. Foxman, B., Valdez, R.B., Lohr, K.N., et al. The Effect of Cost Sharing on the Use of Antibiotics in Ambulatory Care: Results from a Population-Based Randomized Controlled Trial. Journal of Chronic Diseases 40:429–437, 1987. Gertman, P.M. and Restuccia, J. The Appropriateness Evaluation Protocol: A Technique for Assessing Unnecessary Days of Hospital Care. Medical Care 19:855–871, 1981. GAO (General Accounting Office). Medical Malpractice: Characteristics of Claims Closed in 1984. HRD-87–55. Washington, D.C.: General Accounting Office, April 1987. GAO. Internal Controls: Need to Strengthen Controls Over Payments by Medicare Intermediaries. HRD-89–8. Washington, D.C.: General Accounting Office, November, 1988a. GAO. Medicare PROs. Extreme Variation in Organizational Structure and Activities. GAO/PEMD-89–7FS. Washington, D.C.: General Accounting Office, November, 1988b. Graboys, T.B. Conflicts of Interest in the Management of Silent Ischemia. Journal of the American Medical Association 261:2116–2117, 1989. Greenspan, A.M., Kay, H.R., Berger, B.C., et al. Incidence of Unwarranted Implantation of Permanent Cardiac Pacemakers in a Large Medical Population. New England Journal of Medicine 318:158–163, 1988. Hawes, C. and Kane, R.L. Issues Related to Quality Review and Assurance in Health Care. Paper prepared for the Institute of Medicine Study to Design a Strategy for Quality Review and Assurance in Medicare, 1989. HCFA (Health Care Financing Administration). Utilization and Quality Control Peer Review Organizations Second Scope of Work. Executive Data Summary. Report through February 1989. Report dated 5/30/89. Baltimore, Md.: Health Care Financing Administration, 1989. Hermanson, B., Omenn, G.S., Kronmal, R.A., et al. Beneficial Six-Year Outcome of Smoking Cessation in Older Men and Women with Coronary Artery Disease. New England Journal of Medicine 319:1365–1369, 1988. Hiatt, H.H., Barnes, B.A., Brennan, T.A., et al. Special Report: A Study of Medical Injury and Medical Malpractice. New England Journal of Medicine 320:480–483, 1989. Jogerst, G.L. and Dippe, S.E. Antibiotic Use Among Medical Specialties. Journal of the American Medical Association 245:842–846, 1981. Kassirer, J.P. Sounding Board. Our Stubborn Quest for Diagnostic Certainty—A Case of Excessive Testing. New England Journal of Medicine 320:1489–1491, 1989.
OCR for page 235
Medicare: A Strategy for Quality Assurance - Volume I Keeler, E.B., Brook, R.H., Goldberg, G.A., et al. How Free Care Reduced Hypertension in the Health Insurance Experiment. Journal of the American Medical Association 254:1926–1931, 1985. Lang, N. and Kraegel, J.M. Quality of Health Care for Older People in America. Paper prepared for the Institute of Medicine Study to Design a Strategy for Quality Review and Assurance in Medicare, 1989. Lohr, K.N., Brook, R.H., Kamberg, C.J., et al. Use of Medical Care in the Rand Health Insurance Experiment. Diagnosis- and Service-Specific Analyses in a Randomized Controlled Trial. Medical Care 24:S1-S87, September Supplement, 1986. McPhee, S.J., Myers, L.P., and Schroeder, S.A. The Cost and Risks of Medical Care—An Annotated Bibliography for Clinicians and Educators. Western Journal of Medicine 137:145–161, 1982. McPhee, S.J., Garnick, D.W., and Schroeder, S.A. Cardiac Imaging and Cost-Containment: On a Collision Course. American Journal of Cardiac Imaging 1:204–206, 1987. McPherson, K., Wennberg, J.E., Hovind, O.B., et al. Small Area Variations in the Use of Common Surgical Procedures: An International Comparison of New England, England, and Norway. New England Journal of Medicine 307:1310–1314, 1982. Merrick, N.L., Brook, R.H., Fink, A., et al. Use of Carotid Endarterectomy in Five California Veterans Administration Medical Centers. Journal of American Medical Association 258:2531–2535, 1986. Mills, D.H., ed. California Medical Association and California Hospital Association Report on the Medical Insurance Feasibility Study. San Francisco, Calif.: Sutter, 1977. Moskowitz, A.J., Benjamin, J.K., and Kassirer, J.P. Dealing with Uncertainty, Risks, and Tradeoffs in Clinical Decisions. Annals of Internal Medicine 108:435–449, 1988. Moloney, T.W. and Rogers, D.E. Medical Technology—A Different View of the Contentious Debate over Costs. New England Journal of Medicine 301:1413–1419, 1979. Newhouse, J.E., Manning, W.G., Morris, C.N., et al. Some Interim Results from a Controlled Trial of Cost Sharing in Health Insurance. New England Journal of Medicine 305:1501–1507, 1981. Newhouse, J.P., Manning, W.G., Duan, N., et al. The Findings of the Rand Health Insurance Experiment—A Response to Welch et al. Medical Care 25:157–179, 1987. OIG (Office of Inspector General, Department of Health and Human Services). Medical Licensure and Discipline: An Overview, Department of Health and Human Services. P-01–86–0064. Washington, D.C.: Office of Inspector General, Office of Analysis and Inspections, June 1986. OIG. National DRG Validation Study: Special Report on Premature Discharges. OAI-05–88–00740. Washington, D.C.: Office of Inspector General, Office of Analysis and Inspections, February 1988a. OIG. National DRG Validation Study: Unnecessary Admissions to Hospitals. Draft. OAI-09–88–00880. Washington, D.C.: Office of Inspector General, Office of Analysis and Inspections, April 1988b.
OCR for page 236
Medicare: A Strategy for Quality Assurance - Volume I OIG. Financial Arrangement Between Physicians and Health Care Businesses: State Laws and Regulation. OAI-12–88–0140. Washington, D.C.: Office of Inspector General, Office of Analysis and Inspections, April 1989a. OIG. National DRG Validation Study: Quality of Patient Care in Hospitals, OAI-09–88–00870. Washington, D.C.: Office of Inspector General, Office of Analysis and Inspections, July 1989b. OTA (Office of Technology Assessment). The Quality of Medical Care. Information for Consumers. OTA-H-386. Washington, D.C.: U.S. Government Printing Office, 1988. Palmer, R.H., Hall, J.A., Hargraves, J.L., et al. Performance on Other Tasks. Paper presented at the Annual Meeting of the American Medical Review Research Center, October 31, 1988. Park, R.E., Fink, A., Brook, R.H., et al. Physician Ratings of Appropriate Indications for Six Medical and Surgical Procedures. R-3280-CWF/HF/PMT/RWJ. Santa Monica, Calif.: The RAND Corporation, 1986. Paul-Shaheen, P., Clark, J.D., and Williams, D. Small Area Analysis: A Review and Analysis of the North American Literature. Journal of Health Politics, Policy and Law 12:741–809, 1987. Phillips, B.R., Schneider, B.W., Steele, K., et al. A Pilot Study of the Adequacy of Post-Hospital Community Care. Paper presented at the 6th Annual Meeting of the Association for Health Services Research and the Foundation for Health Services Research, Chicago, Ill., June 1989. ProPAC (Prospective Payment Assessment Commission). Medicare Prospective Payment and The American Health Care System. Report to the Congress. Washington, D.C.: Prospective Payment Assessment Commission, June 1989. Rubenstein, L.Z., Rubenstein, L.V., and Josephson, K.R. Quality of Health Care for Older People in America. Paper prepared for the Institute of Medicine Study to Design a Strategy for Quality Review and Assurance in Medicare, 1989. Russell, L.B. Effects on the Health of the Elderly. Pp. 47–68 in Medicare’s New Hospital Payment System. Is it Working? Washington, D.C.: The Brookings Institution, 1989. Schroeder, S.A. Strategies for Reducing Medical Costs by Changing Physicians’ Behavior. International Journal of Technology Assessment in Health Care 3:39–50, 1987. Schroeder, S.A. and Showstack, J.A. Financial Incentives to Perform Medical Procedures and Laboratory Tests: Illustrative Models of Office Practice. Medical Care 16:289–298, 1978. Siu, A.L., Sonnenberg, F.A., Manning, W.G., et al. Inappropriate Use of Hospitals in a Randomized Trial of Health Insurance Plans. New England Journal of Medicine 315:1259–1266, 1986. Slora, E.J. and Gonzalez, M.L. Medical Professional Liability Claims and Premiums, 1985–1987. Pp. 18–22 in Socioeconomic Characteristics of Medical Practice 1988. Gonzalez, M.L. and Emmonds, D.W., eds. Chicago, Ill.: American Medical Association Center for Health Policy Research, 1988. Sox, H.C., ed. Common Diagnostic Tests: Use and Interpretation. Philadelphia, Pa.: American College of Physicians, 1987.
OCR for page 237
Medicare: A Strategy for Quality Assurance - Volume I Steele, K., Gertman, P.M., Crescenzi, C., et al. Iatrogenic Illness on a General Medical Service at a University Hospital. New England Journal of Medicine 304:638–641, 1981. The St. Paul’s 1988 Annual Report to Policyholders. St. Paul, Minn.: The St. Paul Companies, 1987. Weisman, C.S., Morlock, L.L., Teitelbaum, M.A., et al. Practice Changes in Response to the Malpractice Litigation Climate. Results of a Maryland Physician Survey. Medical Care 27:16–24, 1989. Wennberg, J.E. Dealing with Medical Practice Variations: A Proposal for Action. Health Affairs 3:6–32, 1984. Wennberg, J.E. The Paradox of Appropriate Care. Journal of the American Medical Association 258:2568–2569, 1987. Wennberg, J.E. and Gittelsohn, A. Small Area Variations in Health Care Delivery. Science 142:1102–1108, 1973. Wennberg, J.E. and Gittelsohn, A. Health Care Delivery in Maine I: Patterns of Use of Common Surgical Procedures. Journal of the Maine Medical Association 66:123–149, 1975. Wennberg, J.E. and Gittelsohn, A. Variations in Medical Care among Small Areas. Scientific American 246:120–134, 1982. Wennberg, J.E., Barnes, B.A., and Zubkoff, M. Professional Uncertainty and the Problem of Supplier-Induced Demand. Social Science and Medicine 16:811–842, 1982. 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. Winslow, C.M., Kosecoff, J.B., Chassin, M., et al. The Appropriateness of Performing Coronary Artery Bypass Surgery. Journal of the American Medical Association 260:505–509, 1988a. Winslow, C.M., Solomon, D.H., Chassin, M.R., et al. The Appropriateness of Carotid Endarterectomy. New England Journal of Medicine 318:721–727, 1988b. Wolfe, S.M. Testimony of Sidney M.Wolfe, M.D. before Civil and Constitutional Rights Subcommittee, House Judiciary Committee, October 8, 1986. Wolfe, S.M., ed. State Medical Board Doctor Disciplinary Actions: 1987. The Public Citizen Health Research Group Health Letter 5:1–4, 1989.
Representative terms from entire chapter: