Prestige of training programs and experience of bypass surgeons as factors in adjusted patient mortality rates.

OBJECTIVES: The relation of physician performance to physician training and experience is not well understood. The aim of this study was to examine whether indicators of physician background and experience were associated with an objective measure of physician performance. METHODS: Physician background information obtained from the Directory of Board-Certified Medical Specialists was linked to physician risk-adjusted mortality rates obtained from three statewide data bases of coronary artery bypass surgeons. Subjects were 275 surgeons who performed CABG surgery on 83,547 patients during the years 1989 to 1992. Surgical performance was measured by the mortality ratio (MR), the ratio of the observed to the predicted patient mortality rate as determined by detailed clinical information. Training institutions and physicians were characterized as prestigious if they were listed as outstanding in published articles. RESULTS: Surgical performance was not associated with graduation from an American medical school; attendance at a prestigious medical school, residency, or fellowship program; or an academic appointment. Mortality ratios decreased with increased volume and increased with years of experience, age, and academic rank. Surgeons were more likely to be considered a "best doctor" if they had more years experience and trained at a prestigious residency or fellowship program. CONCLUSIONS: Training at a prestigious institution was associated with identification as a "best" doctor but not with lower mortality ratios.

Pre-transplant identification of risk factors that adversely affect length of stay and charges for renal transplantation.

BACKGROUND: In the current era of renal transplantation, increasing attention is being focused on resource utilization. The purpose of this study was to identify demographic, medical and immunologic risk factors that are associated with changes in length of stay (LOS) and charges for renal transplantation. METHOD: The study was a retrospective analysis of 311 consecutive renal transplants performed at a single institution. Univariate and multivariate analyses were used to examine relationships between risk factors, LOS, charges and post-operative complications. RESULTS: The following pre-transplant variables were found to be independently significant in predicting increased LOS and/or charges: African-American race, obesity for women, chronic obstructive pulmonary disease (COPD), presence of cardiac disease or previous stroke, pre-transplant dialysis time > or = 1 yr, a 10% increase in panel reactive antibody (PRA), cadaver donor and retransplantation. The analyses were performed with and without adjustment for key outcome variables such as delayed graft function (DGF) and use of induction antibody therapy. Increased LOS or charges for specific risk factors could be attributed to increased complication rates, including delayed graft function seen with various co-morbidities, or increased immunologic risk and more frequent use of induction antibody therapy. CONCLUSION: Analysis of linked financial and clinical databases can reveal demographic, medical and immunologic risk factors that correlate with LOS, charges and complications for renal transplantation. Efforts to establish quantitative relationships for various risk factors relative to resource utilization will become important in managed care and/or capitated healthcare delivery systems.

Hospitalization costs associated with homelessness in New York City.

BACKGROUND: Homelessness is believed to be a cause of health problems and high medical costs, but data supporting this association have been difficult to obtain. We compared lengths of stay and reasons for hospital admission among homeless and other low-income persons in New York City to estimate the hospitalization costs associated with homelessness. METHODS: We obtained hospital-discharge data on 18,864 admissions of homeless adults to New York City's public general hospitals (excluding admissions for childbirth) and 383,986 nonmaternity admissions of other low-income adults to all general hospitals in New York City during 1992 and 1993. The differences in length of stay were adjusted for diagnosis-related group, principal diagnosis, selected coexisting illnesses, and demographic characteristics. RESULTS: Of the admissions of homeless people, 51.5 percent were for treatment of substance abuse or mental illness, as compared with 22.8 percent for the other low-income patients, and another 19.7 percent of the admissions of homeless people were for trauma, respiratory disorders, skin disorders, and infectious diseases (excluding the acquired immunodeficiency syndrome [AIDS]), many of which are potentially preventable medical conditions. For the homeless, 80.6 percent of the admissions involved either a principal or a secondary diagnosis of substance abuse or mental illness -- roughly twice the rates for the other patients. The homeless patients stayed 4.1 days, or 36 percent, longer per admission on average than the other patients, even after adjustments were made for differences in the rates of substance abuse and mental illness and other clinical and demographic characteristics. The costs of the additional days per discharge averaged $4,094 for psychiatric patients, $3,370 for patients with AIDS, and $2,414 for all types of patients. CONCLUSIONS: Homelessness is associated with substantial excess costs per hospital stay in New York City. Decisions to fund housing and supportive services for the homeless should take into account the potential of these services to reduce the high costs of hospitalization in this population.

The relationship of the value of outcome comparisons to the number of patients per provider.

PURPOSE: Monte Carlo methods were used to assess how the value of outcome comparisons depends on the number of patients per provider. METHODS: We simulated two patient data sets that have been used for well-known studies of outcome comparisons: mortality rates for coronary artery bypass surgeons from New York and Pennsylvania, and 30-day hospital mortality rates of Medicare patient from a national data set. In the simulated data sets, each surgeon or hospital provider was assigned a true or underlying probability of mortality. RESULTS: For the simulated CABG surgery data set, the underlying probability of mortality explained 30% of the variation in the observed mortality rate when there were 100 patients per physician, and 63% when there were 400 patients. The positive predictive value of using an observed mortality rate in the bottom 10% to identify a surgeon whose underlying probability of mortality was in the bottom 10% was 31% for 100 patients and 59% for 400 patients. The relationship between underlying and observed rates was weaker in the simulated Medicare data set with the same number of patients per provider. For a given data set, the amount of random variation in the observed rates of adverse outcomes among providers can be estimated with a simple equation. CONCLUSION: The results show that the assessment of provider outcomes may be greatly affected by random variation. An indication of the amount of random variation in a given data set can be obtained from the examples in this study and an equation for estimating random variation.

The evaluation of screening methods to identify medically unnecessary hospital stay for patients with pneumonia.

OBJECTIVES: This study compared three methods to screen charts of pneumonia patients for excess days. METHODS: A derivation data set was used to statistically derive a severity measure to predict length of stay for pneumonia patients and to refine a clinical algorithm for identifying excess stay. A validation data set was used to compare three computerized methods to screen for unnecessary hospital days: (1) an observed length of stay greater than a target value; (2) an observed stay greater than predicted for the specific patient; and (3) an algorithm that tested whether there were clinical justifications for the entire hospital stay. RESULTS: The sensitivity and specificity for detecting excess stay for the three methods were (1) 0.48 and 0.85 for the observed stay greater than the target value; (2) 0.56 and 0.73 for observed stay greater than predicted; and (3) 0.83 and 0.85 for the algorithm. CONCLUSIONS: These results suggest that computerized clinical algorithms may provide a useful method to detect unnecessary hospital stay.

Correlation of rates of coronary artery bypass surgery, angioplasty, and cardiac catheterization in 305 large communities for persons age 65 and older.

OBJECTIVE: The rate of coronary artery bypass surgery (CABG) has been shown to vary greatly across geographic regions. This study examined whether these rates were associated with the rate of coronary artery angioplasty (PTCA) and with other community characteristics. DATA SOURCES/STUDY SETTING: The health care financing administration provided the number of Medicare hospitalizations in 1988 for conditions and procedures related to coronary artery disease. Information on physicians and hospitals was obtained from the Area Resource File, and the number of persons in each age, sex, and race category was obtained from US. census data. STATISTICAL METHODS: Age-and sex-adjusted hospitalization rates based on the patient's zip code of residence were calculated at the level of the Metropolitan Statistical Area (MSA) for white patients age 65 or older. Rates were obtained for 305 MSAs for CABG, PTCA, cardiac catheterization, angina, and myocardial infarction. PRINCIPAL FINDINGS: The rate of cardiac catheterization had a correlation of .72 with the CABG rate and .64 with the PTCA rate. The correlation of the PTCA and CABG rates with each other was .49. This correlation was not charged by adjusting for the rates of hospitalization for angina or myocardial infection, but it was reduced to only .05 (ns) after adjusting for the rate of cardiac catheterization. The rates of all three procedures had rank correlations of about .15 with the density of thoracic surgeons and about .30 with the density of hospitals with cardiac catheterization and open heart surgery units. CONCLUSIONS: Community CABG and PTCA rates tend to move in the same direction due to community factors that also affect the rates of cardiac catheterization. These community factors do not appear to include the rate of coronary artery disease, but may include resources or attitudes toward aggressive treatment of coronary artery disease.

Adequacy of national mortality data for internal hospital use: a case study.

OBJECTIVE: The study investigated the value of using national or regional data bases to examine care in a specific hospital. DATA SOURCES: The following data sources were included: (1) the results of the 1992 HCFA analysis of the index hospital for patients hospitalized in fiscal year 1990; (2) the 1989 Medicare Provider Analysis and Review (MEDPAR) file; and (3) clinical information from bypass surgery patients in Wisconsin and from the index hospital. PRINCIPAL FINDINGS: The assessment of the mortality rates in the index hospital for all conditions combined and for CABG patients differed depending on what data base was used and how the data were analysed. The national data were most useful in establishing that the coding practices for all patients and the mortality rate for intra-aortic balloon patients differed between the index hospital and other hospitals. The regional clinical data base for bypass surgery patients was used to establish that the high mortality rates for intra-aortic balloon patients were due to patient selection. CONCLUSIONS: National claims data must be analysed carefully before applying results to an individual hospital. Even a careful analysis is more for raising questions about care at a specific hospital rather than for reaching definitive conclusions.

The relationship of hospital ownership and teaching status to 30- and 180-day adjusted mortality rates.

Hospital characteristics have been shown previously to be associated with variations in the probability of death within 30 days of admission. In the current study, the authors extend the examination of the relationship between hospital type to both short-term and long-term adjusted mortality. Observed and predicted 1988 hospital mortality rates were obtained from the Health Care Financing Administration (HCFA). A total of 3,782 acute care hospitals were divided into six mutually exclusive groups on the basis of their status as osteopathic, private for-profit, public teaching, public nonteaching, private teaching, and private nonteaching hospitals. After adjusting for the HCFA predicted mortality, Medicaid admissions, and emergency visits, 30-day and 30-to-180-day patient mortality rates were compared for these hospital types. Separate comparisons also were performed after stratifying hospitals into three groups defined by community size. The risk-adjusted 30-day mortality per 1,000 patients was 91.5, ranging from 85.4 for private teaching hospitals to 95.3 for nonteaching public hospitals, and 97.4 for osteopathic hospitals. The adjusted 30-to-180-day mortality was 84.7, ranging from 82.6 for nonteaching public hospitals to 87.4 and 88.2, respectively for public teaching and osteopathic hospitals. Differences among hospital types were minimal for small communities and increased with community size. In the large communities, the types of hospitals with high 30-day mortality also had higher mortality after 30 days. There was a strong association of hospital type with adjusted 30-day mortality, which should depend on the quality of hospital care, and a much weaker association with post-30-day mortality, which may be more dependent on patient risk. There was no evidence that types of hospitals with low 30-day mortality were postponing rather than preventing mortality.

Severity of illness measures derived from the Uniform Clinical Data Set (UCDSS).

The Health Care Financing Administration (HCFA) plans to use the Uniform Clinical Data Set System (UCDSS) to collect data on hospitalized Medicare patients. This study examined the value of UCDSS data for creating severity of illness measures. UCDSS data were obtained from a study hospital and from a national data set for patients with pneumonia (n = 528) and stroke (n = 565). Models to predict length of stay or an adverse event were derived for each condition using HCFA claims data alone, UCDSS data alone, and UCDSS data supplemented with additional information also abstracted from charts. The models were derived from one set of patients and validated on another. The R2 for predicting length of stay in the validation data for the UCDSS model was 0.29 for pneumonia and 0.19 for stroke compared to R2 values from the claims model of 0.09 for stroke and 0.06 for pneumonia. UCDSS models also were better than claims models for predicting adverse events. The best UCDSS models included International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes and other information requiring clinical judgment, and were improved by adding more information on patient functional status. Some findings were more strongly associated with outcome for the study hospital than for the national data. These results suggest that UCDSS models will predict outcome much better than the claims based models currently used by HCFA for the analysis of hospitalization-related mortality; more functional status information should be added to UCDSS; and despite an extensive objective database, the most predictive UCDSS models require clinician-assigned diagnostic codes.

The relationship between adjusted hospital mortality and the results of peer review.

This study assessed the relationship between the Health Care Financing Administration adjusted mortality rate for a hospital and the errors in care found by the peer review process. The three data sets used were: (1) the 1987-1988 completed reviews from 38 peer review organizations (PROs) of 4,132 hospitals and 2,035,128 patients; (2) all 1987 hospital mortality rates for Medicare patients as adjusted by HCFA for patient mix; and (3) the 1986 American Hospital Association Survey. The PRO data were used to compute the percentage of cases reviewed from each hospital confirmed by a reviewing physician to have a quality problem. The average percentage of confirmed problems was 3.73 percent with state rates ranging from 0.03 percent to 38.5 percent. The average within-state correlation between the problem rate and the adjusted mortality rate for all PROs was .19 (p < .0001), but the correlations were much higher for relatively homogeneous groups of hospitals, .42 for public hospitals and .36 for hospitals in large metropolitan statistical areas (MSAs). These results suggest that the HCFA adjusted hospital mortality rate and the PRO-confirmed problem rate are related methods to compare hospitals on the basis of quality of care. Both methods may compare quality better if used within a group of homogenous hospitals.

Assessing providers of coronary revascularization: a method for peer review organizations.

OBJECTIVES: Current methods to evaluate quality of care are usually limited to reviews of individual cases or comparisons of hospital mortality rates. We present an alternative method that compares complication rates adjusted for patient characteristics. METHODS: Detailed clinical data that were specifically designed for quality comparisons of providers of revascularization procedures were abstracted from the medical records of 1998 Medicare patients, in 16 hospitals, who had coronary artery bypass surgery and 2091 patients, in 16 hospitals, who had angioplasty. Providers were ranked on the basis of an unadjusted risk, a risk adjusted for detailed clinical information, and a risk adjusted only for patient comorbidities. RESULTS: Complication rates differed significantly and substantially among the hospitals. Clinical adjustment changed the hospital rankings for the bypass surgery hospitals, but not for the angioplasty hospitals. Adjustment for comorbidities did not affect hospital rankings for either procedure. CONCLUSIONS: When sample sizes are limited, adverse outcome rates may be a more sensitive measure of quality of care than mortality rates. Rates that are unadjusted or adjusted only for comorbidities may be inadequate for evaluating some providers of bypass surgery.

Outcome and cost analysis of scheduled versus emergency scleral buckling surgery.

BACKGROUND: Retinal detachments are usually considered to be a surgical emergency. However, there are additional risks and costs for unnecessary emergency surgeries. The purpose of this study is to evaluate whether the conventional wisdom for treating all retinal detachments as emergencies needs to be re-examined. METHODS: Forty-eight patients who had an emergency scleral buckle and 89 patients who had a scheduled procedure were randomly selected from 884 consecutive patients who had a primary scleral buckling procedure during a 4 1/2-year period. The medical records of each patient were used to obtain detailed information related to prognosis. The visual acuity measurements of each patient, taken 6 months after the procedure, were obtained from the records of the ophthalmologist following the patient. Linear regression analysis was used to compare the final visual outcome for patients who had emergency surgery with patients who had scheduled surgery after taking into account patient factors related to prognosis. RESULTS: Patients selected for emergency surgery had better visual prognoses than scheduled patients but had the same risk of systemic complications and the same extent of detachment if the macula was not involved. None of the 18 patients with an attached macula experienced macular involvement while awaiting scheduled surgery. There were no differences between emergency and scheduled patients in ocular or systemic complications, rate of reattachment, rate of decreased visual acuity after surgery, visual outcome adjusted for prognosis, or, since 1985, length of hospital stay. A greater cost was incurred for the patients having emergency surgery due to difference in pay scales for support personnel. CONCLUSIONS: Because the study is not large and the patients were not randomized to treatment, the results are not definitive. However, they suggest that emergency surgery is unnecessary for many patients with a detached retina.

Evaluation of the HCFA model for the analysis of mortality following hospitalization.

From 1987 through 1990, the Health Care Financing Administration (HCFA) evaluated variations in the mortality rates experienced by patients admitted to hospitals participating in the Medicare program. This study was conducted to evaluate the adequacy of the model used for that purpose. Detailed clinical data were gathered on 42,773 patients admitted to 84 statistically selected hospitals. The effect of risk adjustment using the HCFA model, which is based on claims data, was compared to a risk-adjustment model based on physiologic and clinical data. Models that include claims data were markedly superior to those containing only demographic characteristics in predicting the probability of patient death, and the addition of clinical data resulted in further improvement. The correlation of ranks of hospitals based on a model that uses only the claims data and on one that uses, in addition, clinical data, was .91. As a screen for the identification of "high (mortality) outlier" hospitals, the claims model had moderate sensitivity (81 percent) and specificity (79 percent), a high negative predictive value (90 percent), and a low positive predictive value (64 percent) when compared to the clinical model. The two mortality models gave similar results when used to determine which structural characteristics of hospitals were related to mortality rates: hospitals with a higher proportion of registered nurses or board-certified physician specialists, or with a greater level of access to high-technology equipment had lower risk-adjusted mortality rates. These data suggest that the current claims-based risk-adjustment procedure may satisfactorily be used to characterize variations in mortality rates associated with hospitalization. The procedure could also be used as a basis for further epidemiological analyses of factors that affect the probability of patient death. However, it does not positively identify outlier hospitals as providers of problematic care.

Mortality after coronary angioplasty and coronary artery bypass surgery (the national Medicare experience).

Mortality rates for Medicare patients who underwent coronary artery bypass surgery were compared with those who had angioplasty or angioplasty and bypass surgery. Two data sets were used for this study: The first contained information on demographic factors, co-morbidities and subsequent mortality on all 96,666 Medicare patients who had bypass surgery or angioplasty in 1985; the second contained additional detailed clinical data collected using the MedisGroups method on a random sample of 2,931 revascularization patients from 6 states. From the national data set 30-day and 1-year mortality rates were 3.8 and 8.2% for 25,423 angioplasty patients and 6.4 and 11.8% for 71,243 bypass surgery patients (p less than 0.001 for both time periods). Mortality rates for the MedisGroups data were 4.4 and 8.5% for the angioplasty patients and 6.5 and 11.9% for the bypass surgery patients. After eliminating patients admitted with a myocardial infarction, mortality rates were 1.9 and 6.0% for 632 angioplasty patients and 5.1 and 10.8% for 1,730 bypass surgery patients. The risk-adjusted relative risk of mortality for bypass surgery versus angioplasty was 1.72 (p = 0.001) for all patients, 2.15 (p less than 0.001) for low-risk patients and 0.90 (p = not significant) for high-risk patients. Results suggest that low-risk patients have better survival with angioplasty because of lower short-term mortality.

Racial and community factors influencing coronary artery bypass graft surgery rates for all 1986 Medicare patients.

OBJECTIVE: This study examines the differences in the rates of coronary artery bypass grafting (CABG) between white and black Medicare patients. DESIGN: This is a cross-sectional study with data from the 1986 Health Care Financing Administration hospital claims records on all Medicare patients, the 1988 update of the Bureau of Health Professions area resource file, and the 1985 Census Bureau's county population estimates file. SETTING: Data are from all Medicare patients in the United States in 1986. MAIN OUTCOME MEASURES: Sex- and age-adjusted CABG rates for whites and blacks over the age of 65 years were computed for each of 50 states and 305 Standard Metropolitan Statistical Areas (SMSAs). RESULTS: Nationally the CABG rate was 27.1 per 10,000 for whites (40.4 for white men and 16.2 for white women), but only 7.6 for blacks (9.3 for black men and 6.4 for black women). Racial differences were greater in the Southeast, particularly in nonmetropolitan areas, than in other regions. Neither white nor black SMSA rates were associated with the rate of admission for acute myocardial infarction (an indication of the amount of coronary artery disease). White rates, but not black rates, were associated with the number of thoracic surgeons per 100,000 people. CONCLUSIONS: For patients insured by Medicare, race is strongly associated with CABG rates, and this association is greater for men than for women and greater in the Southeast than in other parts of the country. Physician supply may relate to the CABG rates for whites.

The relationship of hospital characteristics and the results of peer review in six large states.

To determine which characteristics of hospitals may be related to a higher quality of care, the association of hospital characteristics with the outcomes of medical record review by state Peer Review Organizations (PROs) was studied. The two data sources were: 1) the AHA 1986 Annual Survey of Hospitals and 2) reviews completed between July 1987 through June 1988 from six large PROs. For each hospital the percentage of cases that failed physician review (the confirmed problem rate) was computed. Hospital characteristics evaluated included financial status, ownership, medical training, technological sophistication, and size. The following characteristics were significantly associated with a lower confirmed problem rate: a higher occupancy rate, greater payroll expenses per bed, a higher proportion of physicians who were board-certified specialists, greater technological sophistication, a higher number of beds, a higher proportion of nurses who were registered, and membership in the Council of Teaching Hospitals. Public hospitals had higher problems rates than private not-for-profit hospitals. All characteristics significantly related to higher confirmed problem rates were also related to higher adjusted mortality rates in a previous study of 3,100 U.S. hospitals. The results suggest that hospital resources, including financial status, training of medical personnel, and availability of sophisticated equipment, are related to the quality of care provided by the hospital.

Outcome and cost analysis of scheduled versus emergency scleral buckling surgery.

Conventional wisdom holds that a retinal detachment of recent onset should be regarded as a surgical emergency. A delay in surgery may result in an extension of detachment for patients with an attached macula and a worse visual outcome for patients with a detached macula. However, the potential disadvantages of performing surgery on an emergency basis must be weighed against the risks of delaying surgery; disadvantages include a greater frequency of operative complications resulting from fatigue factors among the operating personnel, an increased anesthetic risk due to inadequate time to assess and stabilize coexisting medical problems, and higher hospital costs. In this retrospective study covering 4 1/2 years, we compared the risks, benefits, length of hospitalization, and costs of scleral buckling surgery for retinal detachments performed as an emergency procedure or on the day following admission. After a 15% random selection from 884 consecutive operations, 48 emergency procedures were compared with 89 scheduled procedures. Patients selected for emergency surgery had better visual prognoses than scheduled patients. The potential for risk of systemic complications was not a reason for postponing surgery. None of the 18 patients with an attached macula experienced macular involvement while awaiting scheduled surgery. There were no differences between emergency and scheduled patients in ocular or systemic complications, rate of reattachment, rate of decreased visual acuity following surgery, visual outcome adjusted for prognosis, or, since 1985, length of hospital stay. Cost was greater for patients having emergency surgery, because of a difference in pay scales for support personnel.

Unmet needs of persons with cancer in Pennsylvania during the period of terminal care.

A stratified random sample of recent cancer deaths was drawn from the Pennsylvania death registry, and 433 family members or close friends were interviewed concerning unmet needs during the last month of life. It was estimated that 72% of persons who died of cancer in Pennsylvania experienced at least one unmet service need during this period. The most frequently reported was help with activities of daily living, estimated at 42% of cancer deaths, involving over 11,000 persons each year in the state. There were significantly more unmet needs during the terminal period, compared with just after diagnosis, in activities of daily living, obtaining health care, transportation, and problems with medical staff. Our findings indicate a need to increase a broad range of support programs during the terminal period, especially of home-care services.