Conflicting aims. Voluntary health insurance and contemporary medical practice.

American medicine is financed today by a patchwork of systems formed around a concept of competitive voluntary health insurance that evolved over the past 50 years. This article reviews the theory of health insurance from a clinical perspective to examine whether changes in medical science and practice have made such an insurance system obsolete. As it is currently applied, a system of competitive voluntary health insurance conflicts with the goals of modern medical practice due to advances in screening and in treatment and the need to deal more effectively with paying for care of unknown efficacy. Proposals to reform health insurance must deal with the medical failings of competitive voluntary health insurance and should do more than simply extend the current system to cover more Americans.

The importance of physician communication on breast cancer screening of older women.

BACKGROUND: Breast cancer screening rates, especially for mammography, continue to lag for older women, particularly for women older than 65 years. METHODS: We investigated the associations of key variables with reported rates of mammography and clinical breast examination in a sample of 972 women older than 50 years; 724 of them were older than 65 years. They were surveyed in late 1990 through 30-minute bilingual telephone interviews. RESULTS: Although it was hypothesized that race, age, health status, and physician-patient communication variables would influence utilization rates, only the communication variables (and two access variables) significantly predicted a recent mammogram or clinical breast examination. In particular, the style of the communication--the patient's report of the physician's enthusiasm for mammography when it was discussed with women at the office visit--influenced the women's screening behavior significantly. Women who perceived that their physicians had some enthusiasm for mammography were more than four and a half times more likely than women whose physicians had no or little enthusiasm for mammography to have had one within the previous year. Other findings were that about half of the Los Angeles, Calif, women in this study reported a recent mammogram, an increase from the one third who reported one in the previous survey of 1988; a decline in screening was not reported until after age 75 years. About a quarter of the study women, on the other hand, had never been screened despite the long-standing recommendation for regular screening of women older than 50 years and the risk of breast cancer increasing with age. Surprisingly, women at higher risk of breast cancer were not being screened any more systematically than women at lower risk. CONCLUSION: We conclude that improved physician-patient communication skills could be a highly effective and easy-to-learn strategy to increase overall screening rates.

The cause of delirium in patients with hip fracture.

OBJECTIVES: To ascertain the most common causes of delirium, to establish the initiation and timing of delirium, and to determine the duration of delirium in patients with hip fracture. METHODS: Five hundred seventy-one (88%) of 650 patients with hip fracture admitted to 4 New York City hospitals were prospectively interviewed on a daily basis, 5 days a week, with the Confusion Assessment Method for the presence of delirium. The patients were enrolled within 48 hours of admission. Their medical charts and the data collected by the study staff were reviewed and summarized. Two of us (R.S.M. and A.L.S.) reviewed the case summaries independently and assigned a cause based on a previously developed classification system, estimated the onset of the delirious episode, and determined whether the delirium had cleared, improved, or persisted at discharge. Subsequently, discrepancies in cause, timing of initiation, and mental status on discharge between the 2 physicians reviewers were discussed until consensus was reached. RESULTS: The prevalence of delirium was 9.5% (54/ 571; 95% confidence interval, 7.0-11.9). Seven percent of episodes were assigned a definite cause, 20% a probable cause, 11% a possible cause, and 61% were attributable to 1 or more comorbid conditions. Twenty-eight (53%) of 54 subjects developed delirium after surgery. The delirium had cleared or improved in 40 (74%) of 54 subjects at the time of discharge. CONCLUSIONS: Delirium in patients with hip fracture appears to be a different syndrome from that observed in patients who are otherwise medically ill; it also appears to follow a different clinical course. These results have important implications for the management of delirium in patients with hip fracture.

Postdischarge geriatric assessment of hospitalized frail elderly patients.

BACKGROUND: The diffusion of comprehensive geriatric assessment services has been rather limited in North America partly because of reimbursement and organizational constraints. OBJECTIVE: To evaluate the impact of a comprehensive geriatric assessment intervention for frail older patients that is started before hospital discharge and is continued at home. METHODS: Patients older than 65 years were selected who had either unstable medical problems, recent functional limitations, or potentially reversible geriatric clinical problems. Patients (n = 354) were randomly assigned to either the intervention group or a control group. Information on survival, readmissions, nursing home placement, medication use, and health status was collected at 30 and 60 days after hospital discharge. RESULTS: No differences were observed between the two treatment groups in survival, hospital readmission, or nursing home placement by 60 days. After adjustment for baseline characteristics, no significant differences were observed between the two groups on measures of physical functioning, social functioning, role limitations, health perceptions, pain, mental health, energy and/or fatigue, health change, or overall well-being. CONCLUSIONS: Although efficacy has been demonstrated for some forms of comprehensive geriatric assessment, the types of services that are easier to establish (inpatient consultation services and ambulatory assessment) have not been shown to improve outcomes. Our results indicate that outcomes are unaffected by a limited form of comprehensive geriatric assessment begun in the hospital and completed at home. Further efforts are needed to develop and to evaluate realistic approaches to comprehensive geriatric assessment.

Screening for common problems in ambulatory elderly: clinical confirmation of a screening instrument.

PURPOSE: To develop a test that will be administered by nonphysician office staff to screen elderly persons seen in ambulatory settings for problems commonly contributing to functional disability. METHODS: We reviewed the literature to identify problems that reduce function and screening measures appropriate for use in office settings. Using this information, we developed an instrument including screening items for malnutrition/weight loss, visual impairment, hearing loss, cognitive impairment, urinary incontinence, depression, physical disability, and reduced leg mobility. This instrument was tested on 109 new patients at a university-based ambulatory geriatrics medicine clinic and validated using two standards: blinded and unblinded geriatricians' assessments. For each of the individual items on the screening instrument, we calculated sensitivity and specificity using both the blinded and unblinded geriatricians' evaluations as the reference standards, prevalence of the disorders, positive and negative predictive values, inter-rater reliability, and the direct annual costs of administering the test for an individual physician's office. RESULTS: The screen was administered in 8 to 12 minutes. Inter-rater agreement varied by item from 77% to 100%. The sensitivities of the items varied between 0.65 to 0.93 (blinded) and 0.70 to 0.95 (unblinded). Specificities ranged between 0.50 to 0.95 (blinded) and 0.64 to 0.95 (unblinded). Problem prevalences varied from 21% to 72%. Positive and negative predictive values were 0.60 to 0.91 and 0.77 to 0.96, respectively. Direct annual costs for a clinical practice include a one-time $530 fee for equipment and, depending on the screening administrator's salary, between $1 to $7 per patient screened. CONCLUSIONS: The screening instrument is relatively inexpensive and brief and easy to use in the ambulatory setting. It is good validity and reliability when compared to the assessment of a geriatrician. We are currently conducting a randomized trial to assess the effectiveness of the screen among older persons seen in community physicians' offices.

A randomized trial of office-based screening for common problems in older persons.

PURPOSE: To test the effectiveness of a 10-minute office-staff administered screen to evaluate malnutrition/weight loss, visual impairment, hearing loss, cognitive impairment, urinary incontinence, depression, physical limitations, and reduced leg mobility among older persons seen in office practice. This screen was coupled with clinical summaries to assist the physician in further evaluating and managing the screen-included problems. PATIENTS AND METHODS: Twenty-six community-based office practices of internists and family physicians in Los Angeles were randomized to intervention or control groups. Two hundred and sixty-one patients aged > or = 70 years and seeing these physicians for a new visit or a physical examination participated in the study. At the enrollment visit intervention group patients were administered the screening measure and their physicians were given the pertinent clinical summaries. Outcome measures were detection of, and intervention for conditions screened, and health status 6 months after the intervention. RESULTS: Hearing loss was both more commonly detected (40% intervention versus 28% control) and further evaluated (29% versus 16%) by physicians in the intervention group (P < 0.05). No other differences in the frequency of problem detection or intervention were noted between groups. Six months after the intervention no differences were noted in health status between groups. CONCLUSIONS: A brief measure to screen for common conditions in older persons was associated with more frequent detection and follow-up assessment of hearing loss. Although the measure was well accepted by physicians and their staffs, it did not appear to affect detection and intervention in regard to the other screen-included conditions, or health status at 6 months.

Physician implementation of and patient adherence to recommendations from comprehensive geriatric assessment.

PURPOSE: The goals of this study were to develop and determine the feasibility of interventions designed to increase both primary care physician implementation of and patient adherence to recommendations from ambulatory-based consultative comprehensive geriatric assessment (CGA), and to identify sociodemographic and intervention-related predictors of physician and patient adherence. PATIENTS AND METHODS: One hundred thirty-nine community-dwelling older persons who failed a screen for functional impairment, depressive symptoms, falls, or urinary incontinence received outpatient CGA consultation. These patients and the 115 physicians who provided primary care for them received one of three adherence interventions, each of which had a physician education component and a patient education and empowerment component. Recommendations were classified as physician-initiated or self-care and as "major" or "minor"; one was deemed "most important". Adherence rates were determined on the basis of face-to-face interviews with patients. RESULTS: Based on 528 recommendations for 139 subjects, physician implementation of "most important" recommendations was 83% and of major recommendations was 78.5%. Patient adherence with physician-initiated "most important" and "major" recommendations were 81.8% and 78.8% respectively. In multivariate models, only the status of the recommendation of "most important" (odds ratio 2.4, 95% CI [confidence interval] 1.3 to 4.5) and health maintenance organization (HMO) status of the patient (odds ratio 2.1, 95% CI 1.3 to 3.6) remained significant in predicting physician implementation. The logistic model predicting patient adherence to physician-initiated recommendations included male patient gender (odds ratio 3.1, 95% CI 1.3 to 7.0), the status of the recommendation of "most important" (odds ratio 1.9, 95% CI 1.0 to 3.8), total number of recommendations (odds ratio 0.7, 95% CI 0.5 to 0.9), and total number of problems identified by CGA (odds ratio 1.8, 95% CI 1.2 to 2.7). CONCLUSIONS: These findings indicate that relatively modest interventions strategies are feasible and lead to high levels of physician implementation of and patient adherence to physician-initiated CGA recommendations. These interventions appear to be particularly effective in HMO patients and for recommendations that were deemed to be "most important".

Change in self-reported functioning in older persons entering a residential care facility.

We evaluated the responsiveness of measures of function in admissions to a long-term care facility. Between baseline and follow-up assessment, one-fifth or more of the subjects either worsened or improved in most aspects of reported function. We compared two measures of self-reported function (COOP charts and a short-form survey). Convergent validity was observed for changes in pain, social health, and mental health (r = 0.39-0.74), but not for physical functioning. Although the short-form physical function measure discriminated worsening on several performance-based external criteria of physical functioning (area under ROC curves up to 0.82), the COOP and other measures of physical functioning were less likely to do so. All physical function measures were less responsive for detecting improvement. Clinicians and investigators intending to monitor change in function must consider the responsiveness of their measures.

The quality of medical care received by older persons.

This paper reviews studies published since 1970 on the quality of medical services received by older persons. Although many of the studies were flawed in design or limited in scope, they suggest that there are problems in the quality of care received by older persons in ambulatory, hospital, and nursing home settings. Changes in health care delivery and financing should attempt to improve, rather than merely maintain, the quality of medical and health services received by this vulnerable population.

An objective measure of physical function of elderly outpatients. The Physical Performance Test.

Direct observation of physical function has the advantage of providing an objective, quantifiable measure of functional capabilities. We have developed the Physical Performance Test (PPT), which assesses multiple domains of physical function using observed performance of tasks that simulate activities of daily living of various degrees of difficulty. Two versions are presented: a nine-item scale that includes writing a sentence, simulated eating, turning 360 degrees, putting on and removing a jacket, lifting a book and putting it on a shelf, picking up a penny from the floor, a 50-foot walk test, and climbing stairs (scored as two items); and a seven-item scale that does not include stairs. The PPT can be completed in less than 10 minutes and requires only a few simple props. We then tested the validity of PPT using 183 subjects (mean age, 79 years) in six settings including four clinical practices (one of Parkinson's disease patients), a board-and-care home, and a senior citizens' apartment. The PPT was reliable (Cronbach's alpha = 0.87 and 0.79, interrater reliability = 0.99 and 0.93 for the nine-item and seven-item tests, respectively) and demonstrated concurrent validity with self-reported measures of physical function. Scores on the PPT for both scales were highly correlated (.50 to .80) with modified Rosow-Breslau, Instrumental and Basic Activities of Daily Living scales, and Tinetti gait score. Scores on the PPT were more moderately correlated with self-reported health status, cognitive status, and mental health (.24 to .47), and negatively with age (-.24 and -.18). Thus, the PPT also demonstrated construct validity. The PPT is a promising objective measurement of physical function, but its clinical and research value for screening, monitoring, and prediction will have to be determined.

The geriatric "medical and public health" imperative revisited.

Attempts to reduce the future demand for institutional care through community services are likely to have limited success. For this reason, health professionals must focus on preventing or ameliorating functional decline in older persons. To focus attention on this aspect of the geriatric imperative, we use an epidemiologic model to estimate the potential impact of existing or potential medical and public health interventions that might decrease the incidence of functional decline. For at least three major causes (stroke, hip fracture, and incontinence) of disability, the potential exists for reducing the incidence and burden of functional disability by a number of mechanisms. For example, treating just half of adults age 65-74 with currently untreated diastolic or isolated systolic hypertension would reduce the incidence of stroke by 2.77% in this age group (or 1,500 fewer cases of stroke annually). The estimates presented indicate the need (1) to better implement those interventions that are known to be efficacious, and (2) to identify and to test new interventions for conditions contributing to functional impairment in the elderly.

Comprehensive geriatric assessment in a day hospital.

OBJECTIVE: To assess the effectiveness of comprehensive geriatric assessment conducted in day hospitals. DESIGN: Retrospective cohort comparison study with restricted inclusionary criteria and adjustment for baseline characteristics. SETTING: A hospital-based geriatric day hospital and geriatric clinic sites (both in the university and in the community). PATIENTS: Four hundred sixty-eight patients referred for comprehensive geriatric assessment during a 12-month period. INTERVENTION: Comprehensive geriatric assessment in a geriatric day hospital compared with assessment received in clinic sites without a day hospital. MAIN OUTCOME MEASURES: Services received in the first 2 weeks; hospitalization, emergency room visits, placement, death, and change in selected health status measures. Follow-up data was obtained from medical records, a telephone survey, and death certificates. RESULTS: Except in the case of rehabilitative services, day hospital patients were more likely to receive interdisciplinary services. The population seen in the day hospital was more functionally impaired and had significantly more dementia and depression. After adjusting for subjects' baseline characteristics and limiting the analyses to subjects meeting specific inclusionary criteria, the day hospital had no significant effect on mortality, use of emergency or hospital services, placement, or change on selected measures of health status. For example, compared with the clinic patients, receiving care in the day hospital was associated with an adjusted odds ratio of 1.01 (95% confidence interval: 0.53, 1.91) of being at a higher level of care at 6 months. The results were not sensitive to the choice of inclusionary criteria. CONCLUSION: Given their cost and uncertain effectiveness, day hospitals need additional evaluation before their further diffusion occurs.

Impacts of geriatric evaluation and management programs on defined outcomes: overview of the evidence.

Comprehensive geriatric assessment is a technique for multidimensional diagnosis of frail elderly people with the purpose of planning and/or delivering medical, psychosocial, and rehabilitative care. When comprehensive geriatric assessment is coupled with some therapy, then the term geriatric evaluation and management (GEM) will be used. Following a brief history of comprehensive geriatric assessment, we describe the varied patterns of GEM program organization and review the literature of studies examining GEM effectiveness. Program diversity complicates drawing firm conclusions about GEM effects; however, the vast majority of studies report positive, if not uniformly significant, results. Our analysis suggests that much of the variability in findings is due to sample size limitations. In order to reach conclusions of program effects across studies and to avoid problems of small sample sizes, we undertook a formal meta-analysis. In this initial meta-analysis, we sought to evaluate the effect of GEM programs on a single outcome: mortality. We pooled all published GEM controlled trials into four major groups: inpatient consultation services, inpatient GEM units, home assessment services, and outpatient GEM programs. Meta-analysis of 6-month mortality demonstrates a 39% reduction of mortality for inpatient consultation services (odds ratio 0.61, 95% confidence interval 0.46-0.81, P = 0.0008) and a 37% reduction of mortality for inpatient GEM units (odds ratio 0.63, 95% CI 0.42-0.93, P = 0.02). Home assessment services reduced mortality by 29% (odds ratio 0.71, 95% CI 0.55-0.90, P = 0.005). On the other hand, no significant survival effect was found for outpatient GEM programs (odds ratio 0.96, 95% confidence interval 0.61-1.49).(ABSTRACT TRUNCATED AT 250 WORDS)

Hierarchical measures of physical function in ambulatory geriatrics.

Brief and uncomplicated methods for obtaining information on functional status would facilitate the assessment of older patients. We evaluated the potential usefulness, reliability, and validity of four hierarchical measures of physical function in 123 elderly subjects seen in four ambulatory geriatrics settings. Although the vast majority (83.2%) of subjects were fully independent on the Katz Activities of Daily Living Scale, a broader scope of functional difficulty was reported on the Spector-Katz, five-item OARS, and Rosow-Breslau scales. The three scales all had either borderline or more acceptable coefficients of scalability (0.57-0.77); the hierarchical order of items was not observed in 5.3% to 13.6% of subjects. Combining items from these established measures resulted in two new scales with acceptable scalability and construct validity; however, some errors in item order persisted. Although their ease of administration is clearly advantageous, clinicians using short hierarchical scales to assess functional status of older patients should be aware of their limitations.

Geriatric medicine in the United States. The current activities of former trainees.

To improve the health care received by frail older persons, an effort has been made in the United States to increase the number of physicians trained in geriatric medicine and geropsychiatry. The goal of training has been to create leaders in education, research, and patient care. To assess the progress of this effort, we surveyed physicians (284 in geriatric medicine and 91 in geropsychiatry) who graduated from U.S. geriatrics fellowship programs. Responses were obtained from 224 medicine (79% response) and 59 psychiatry fellows (65% response). Sixty-five percent of former geriatric medicine fellows report spending 10% or less time on teaching; 44% report doing no research, and 44% report spending more than half their time in patient care. Compared to other primary care specialties, the geriatricians reported caring for larger proportions of older patients and spending more time per patient visit. However, their role in teaching, research, and long-term care is minimal.

Measuring physical function in community-dwelling older persons: a comparison of self-administered, interviewer-administered, and performance-based measures.

PURPOSE: To compare two self-administered, one interviewer-administered, and one performance-based measure of physical function in community-based older persons. METHODS: Eighty-three subjects were recruited from meal sites, senior recreation centers, and senior housing units for a comprehensive geriatric assessment program. At the time of screening, study participants self-administered the Functional Status Questionnaire (FSQ) and were administered the Katz Activities of Daily Living (ADL) and the Older Americans Resources and Services Instrumental Activities of Daily Living (OARS-IADL) instruments by interview. Participants also completed the Physical Performance Test (PPT) and were given the Medical Outcomes Study SF-36 to self-administer on site or at home and return by mail. RESULTS: All 83 subjects completed FSQ, Katz ADL, OARS-IADL, and PPT; 72 returned SF-36 forms. Correlations between the two self-administered physical function measures (FSQ and SF-36) were higher than between self-administered and interviewer-assessed (ADL and OARS-IADL) or performance-based (PPT) measures. When assessed for construct validity, the self-administered, OARS, and PPT measures had comparable correlations with role limitations as a result of physical health problems, but relationships between physical functional status measures and other SF-36 measures of health were inconsistent. CONCLUSION: The relationships between commonly used self-administered, interviewer-administered, and performance-based measures of physical function were inconsistent and weak, suggesting that these instruments are not measuring the same construct.

A cost and value analysis of two interventions with incontinent nursing home residents.

OBJECTIVE: More than half of nursing home residents suffer from urinary incontinence. These residents typically have long stays and, because of comorbid cognitive and physical impairments, have little hope of living again in a noninstitutional environment The value of interventions to change functional status of this chronically institutionalized population is often questioned. This paper explores this value issue in the context of two incontinence management interventions that have been shown to improve functional status: (1) Functional Incidental Training (FIT), and (2) Prompted Voiding (PV). The relative value of the different interventions for the nursing home population was estimated using paired preferences. DESIGN: The cost of two interventions (FIT and PV) that target incontinent nursing home residents was related to the value of these interventions as perceived by consumers of nursing home services. Both interventions decrease incontinence frequency, and one intervention also improves mobility endurance. PARTICIPANTS: Ninety incontinent nursing home residents received the intervention; 37 older nondemented board and care residents and 31 family members of the nursing home residents provided estimates of the intervention's value. MEASUREMENT: The staff-time allocations involved in implementing both interventions were documented in more than 85 resident care episodes. These time data were converted to labor cost based on the cost of nursing aides who would actually implement the intervention. The value of each intervention was assessed by asking consumers to make choices between the intervention and its associated outcomes (such as increased dryness) and other nursing home services of known cost (e.g., moving to a private room). RESULTS: Both interventions had labor costs that were greater than "usual care" costs. The additional cost was estimated to be $4.31 per resident per day for PV and $6.42 per resident per day for FIT if these programs were implemented from 7 AM to 7 AM. Consumer preference data indicated that consumers preferred the FIT and PV outcomes to more expensive alternative services, calculated to cost $10.00 per day, often marketed to consumers, CONCLUSION: Consumers may prefer the FIT and PV interventions relative to the typical services often marketed to the nursing home consumer. The analysis completed in this paper suggests that both interventions have value for frail residents likely to live out their lives in a nursing home.

Geriatric home assessment after hospital discharge.

OBJECTIVES: To evaluate the feasibility of an intervention involving post-discharge geriatric home assessment and follow-up and to describe the spectrum of significant clinical problems identified during the home assessment. DESIGN: Prospective observational study nested within a randomized controlled trial. SETTING: Inpatient service of a large academic medical center in Southern California. PATIENTS: There were 152 adults aged 65 or greater who had one or more specific risk factors for functional decline or increased mortality, who were awaiting discharge from the hospital, and who were assigned to the intervention arm of a randomized controlled trial of post-discharge comprehensive geriatric home assessment. MAIN RESULTS: During the home assessment, the gerontologic nurse practitioner (GNP) identified new or worsening problems in 150 patients (99%); 61 problems (eg, serum sodium 125 mg/dL; severe orthostatic hypotension) were considered by a reviewing physician to require urgent medical attention. Older age, non-white race, and new incontinence were associated independently with a greater number of findings (P < 0.05). Based on the findings, an interdisciplinary team made an average of 3.4 recommendations per patient; only two of 111 requests for written approval of recommendations were rejected. CONCLUSIONS: Post-discharge visitation by a GNP to patients at high risk is capable of detecting a high yield of important and potentially reversible clinical problems. This multidisciplinary approach is acceptable to physicians. Research is needed to identify additional links between short hospital stays, impairment or instability at discharge, and adverse outcomes.

Cost-effectiveness of low-air-loss beds for treatment of pressure ulcers.

BACKGROUND: Cost-effectiveness of low-air-loss beds for the healing of pressure ulcers was analyzed in the nursing home setting. A statistical model of pressure ulcer healing was used to estimate cost-effectiveness based on patient and ulcer characteristics. METHODS: Results of a previous randomized trial (84 patients from three nursing homes in Los Angeles) were reanalyzed and combined with estimates of costs to calculate the cost-effectiveness in dollars per added day free of pressure ulcers achieved by the use of low-air-loss beds compared to conventional foam mattresses. RESULTS: The cost-effectiveness of the low-air-loss bed was $26 per added day free of ulcers for our standard patient. Results were sensitive to low-air-loss bed lease costs and patient and wound healing characteristics. Results were less sensitive to expected mortality, daily wound care costs, and time-frame of consideration. Low-air-loss beds were more cost-effective for patients with good healing characteristics and mild ulcers. CONCLUSIONS: Findings support the expanded use of this technology for patients with mild pressure ulcers and good healing characteristics. For these patients, the cost-effectiveness of low-air-loss beds is comparable to other accepted health treatments. For patients with severe ulcers and poor healing characteristics, low-air-loss bed cost-effectiveness compares poorly with other accepted health treatments unless the lease cost can be substantially reduced, or unless life with a pressure ulcer is valued close to death.

New York State's Cardiac Surgery Reporting System: four years later.

This study examined changes in the risk-adjusted mortality associated with coronary artery bypass grafting procedures performed in New York State during the first 4 years of New York's Cardiac Surgery Reporting System (1989 to 1992). To track performance over time, surgeons and hospitals were subdivided into three groups on the basis of their performance in 1989. The risk-adjusted mortality for each of the three groups was computed for 1992 and compared with their 1989 mortality. The results indicate that all groups of providers exhibited large reductions in the risk-adjusted mortalities, with the groups that showed the highest initial mortalities manifesting the most improvement. However, the group rankings remained the same in 1992 as they were in 1989. For example, when the hospital groups were based on the terciles of risk-adjusted mortality observed in 1989, the risk-adjusted mortality decreased from 2.72% to 2.19% for group 1, from 4.24% to 2.51% for group 2, and from 7.12% to 2.77% for group 3. Notably, the risk-adjusted mortalities of the three groups were all significantly different from one another in 1989, but were not significantly different from one another in 1992. Another interesting finding was that the volume of operations performed by the various provider groups did not change substantially in the 4-year period.