Severity of illness within DRGs. Homogeneity study

Variations in the impact of the new case-based payment reform on medical costs, length of stay, and quality across different hospitals in China: an interrupted time series analysis

BACKGROUND

In 2018, an innovative case-based payment scheme called Diagnosis-Intervention Packet (DIP) was piloted in a large developed city in southern China. This study aimed to investigate the impact of the new payment method on total medical expenditure per case, length of stay (LOS), and in-hospital mortality rate across different hospitals.

METHODS

We used the de-identified patient-level discharge data of hospitalized patients from 2016 to 2019 in our study city. The interrupted time series model was used to examine the impact of the DIP payment reform on inflation-adjusted total expenditure per case, LOS, and in-hospital mortality rate across different hospitals, which were stratified into different hospital ownerships (public and private) and hospital levels (tertiary, secondary, and primary).

RESULTS

We included 2.08 million and 2.98 million discharge cases of insured patients before and after the DIP payment reform, respectively. The DIP payment reform resulted in a significant increase of the monthly trend of adjusted total expenditure per case in public (1.1%, P = 0.000), tertiary (0.6%, P = 0.000), secondary (0.4%, P = 0.047) and primary hospitals (0.9%, P = 0.039). The monthly trend of LOS increased significantly in public (0.022 days, P = 0.041) and primary (0.235 days, P = 0.032) hospitals. The monthly trend of in-hospital mortality rate decreased significantly in private (0.083 percentage points, P = 0.002) and secondary (0.037 percentage points, P = 0.002) hospitals.

CONCLUSIONS

We conclude that implementing the DIP payment reform yields inconsistent consequences across different hospitals. DIP reform encouraged public hospitals and high-level hospitals to treat patients with higher illness severities and requiring high treatment intensity, resulting in a significant increase in total expenditure per case. The inconsistencies between public and private hospitals may be attributed to their different baseline levels prior to the reform and their different responses to the incentives created by the reform.

Capturing patients' needs in casemix: a systematic literature review on the value of adding functioning information in reimbursement systems

BACKGROUND

Contemporary casemix systems for health services need to ensure that payment rates adequately account for actual resource consumption based on patients' needs for services. It has been argued that functioning information, as one important determinant of health service provision and resource use, should be taken into account when developing casemix systems. However, there has to date been little systematic collation of the evidence on the extent to which the addition of functioning information into existing casemix systems adds value to those systems with regard to the predictive power and resource variation explained by the groupings of these systems. Thus, the objective of this research was to examine the value of adding functioning information into casemix systems with respect to the prediction of resource use as measured by costs and length of stay.

METHODS

A systematic literature review was performed. Peer-reviewed studies, published before May 2014 were retrieved from CINAHL, EconLit, Embase, JSTOR, PubMed and Sociological Abstracts using keywords related to functioning ('Functioning', 'Functional status', 'Function*, 'ICF', 'International Classification of Functioning, Disability and Health', 'Activities of Daily Living' or 'ADL') and casemix systems ('Casemix', 'case mix', 'Diagnosis Related Groups', 'Function Related Groups', 'Resource Utilization Groups' or 'AN-SNAP'). In addition, a hand search of reference lists of included articles was conducted. Information about study aims, design, country, setting, methods, outcome variables, study results, and information regarding the authors' discussion of results, study limitations and implications was extracted.

RESULTS

Ten included studies provided evidence demonstrating that adding functioning information into casemix systems improves predictive ability and fosters homogeneity in casemix groups with regard to costs and length of stay. Collection and integration of functioning information varied across studies. Results suggest that, in particular, DRG casemix systems can be improved in predicting resource use and capturing outcomes for frail elderly or severely functioning-impaired patients.

CONCLUSION

Further exploration of the value of adding functioning information into casemix systems is one promising approach to improve casemix systems ability to adequately capture the differences in patient's needs for services and to better predict resource use.

Length of stay and costs for asthma patients by hospital characteristics--a five-year population-based analysis

This study sets out to explore the relationship between hospital characteristics, asthma length of stay (LOS), and costs per discharge. The study adopts hospitalization data from the Taiwan National Health Insurance Research Database covering the period from 1997 to 2001. Study subjects were identified from the database by principal diagnosis of asthma or asthmatic bronchitis, with a total of 139,630 cases being included in the study. Multiple-regression analyses were performed to explore the relationship between LOS, costs per discharge and hospital characteristics, adjusting for age, gender, and discharge status of patients, as well as complications or comorbidities. The regression analyses showed that, compared with district hospitals, medical centers and regional hospitals have longer and more statistically significant LOS, as well as higher costs. Hospitals operating on a for-profit basis have shorter LOS and lower costs than public and not-for-profit hospitals. This study shows the existence of wide variations in LOS and costs per discharge for asthma hospitalizations, between the various types of hospitals in Taiwan.

The association between psychiatrist numbers and hospitalization costs for schizophrenia patients: a population-based study

OBJECTIVE

This study explores the association between psychiatrist case volumes and costs for hospitalized schizophrenia patients.

METHODS

The study uses the Taiwan National Health Insurance Research Database for 2003, identifying the study subjects from the database by ICD-9-CM principal diagnosis code 295. Our study sample comprises of 135,621 admissions treated by 787 psychiatrists in 181 hospitals, with the sample being divided equally into three psychiatrist volume groups: <or=300 (low volume), 301-600 (medium volume) and >or=601 admissions (high volume). After adjusting for psychiatrist, patient and hospital characteristics, multiple regression analyses were performed to determine the association between psychiatrist case volume and hospitalization costs (total, drug, and non-drug).

RESULTS

The regression analyses showed that after adjusting for psychiatrist, patient and hospital characteristics, average treatment costs associated with hospitalized schizophrenia patients were inversely related to psychiatrist volume. The respective total costs, drug costs and non-drug costs of patients treated by high-volume psychiatrists were 369 US dollars (p<0.001), 26 US dollars (p<0.001) and 343 US dollars (p<0.001) lower than those of low-volume psychiatrists. The respective total costs, drug costs and non-drug costs for those treated by medium-volume psychiatrists were 248 US dollars (p<0.001), 22 US dollars (p<0.001) and 226 US dollars (p<0.001) lower than those of low-volume psychiatrists.

CONCLUSIONS

We find that after adjusting for patient, psychiatrist and hospital characteristics, an inverse volume-cost relationship exists for psychiatrists treating schizophrenia patients. Further studies should aim to investigate the volume-quality relationship to ensure that incremental cost savings associated with increased patient volume are not achieved at the expense of quality of patient care.

Effect of Disease Complications on Hospital Costs

Background: To test the hypothesis that complications increase the use of resources in managing patients in hospitals, we examined the costs of managing patients with the same disease with and without complications.

Methods: We used a database developed by the University HealthSystems Consortium that contains the costs of managing more than 1 million patients in 60 University hospitals. We created a simplified database of the costs of 457 445 patients in 111-paired diagnosis-related groups (DRGs) that were classified as either having or not having complications and/or comorbidities. Costs were calculated from the ratio of costs to charges within the individual hospitals.

Results: The median costs of managing patients with complications were higher than those for managing patients without complications, confirming the appropriateness of the dual classification. Notably, these extra costs were largely incurred through increased length of stay. Of note, the cost per day for DRGs with complications and/or comorbidities was most often less than that for the corresponding uncomplicated conditions. Although accommodation costs generally were the largest single component of total costs for both complicated and uncomplicated conditions, in only 31 DRGs (15 with complications, 16 without) did they account for more than one-half the total costs. Laboratory and drug costs were higher for complicated conditions, but as a proportion of total costs were comparable for complicated and uncomplicated conditions.

Conclusions: Complications in patients are associated with increased hospital costs, although the costs per day of hospitalization are often less than in patients without such complications.

Outcomes after lower-extremity reconstruction in DRGs 478 and 479

PURPOSE

To assess outcomes for 21,261 patients in DRGs 478 and 479 hospitalized in Veterans Affairs Medical Centers (VAMCs) during fiscal years 1991-1994. DRGs 478 and 479 contain patients undergoing a variety of vascular procedures including lower-extremity arterial reconstruction.

METHODS

VA Patient Treatment File (PTF) data were analyzed using Patient Management Category (PMC) software which defined illness severity, patient complexity as defined by PMC count, and calculated resource intensity scale (RIS), a measure of resource utilization, for each admission.

RESULTS

In-hospital mortality rate was 3.16% (671/21,261) for all patients. Mortality did not differ between the 14,155 patients who underwent extremity arterial reconstruction (3.22%) and the remaining patients (3.03%). The incidence of ICD-9-CM-coded complications was 20.4% after limb revascularization versus 12.8% for remaining patients (P < 0.001). Length of stay (LOS) was 18.6 +/- 17.6 days with versus 10.3 +/- 14. 5 days without limb revascularization (P < 0.001). As defined in this study, patients who underwent limb revascularization were older (64.1 +/- 9.6 vs 62.2 +/- 11.0, P < 0.001); had higher illness severity scores (3.63 +/- 1.60 vs 2.72 +/- 1.72, P < 0.001); were more complex (had higher PMC count: 2.59 +/- 1.35 vs 2.54 +/- 1.34, P = 0.016); and required utilization of more resources (had higher RIS: 2.16 +/- 0.81 vs 1.68 +/- 0.76, P < 0.001) than remaining patients. Logistic regression analysis limited to those undergoing extremity revascularization revealed that age, presence of complications, patient complexity, illness severity, and acute arterial thromboembolism were increasingly and independently associated with greater in-hospital mortality. The logistic regression model also showed that the type of arterial reconstruction was related to in-hospital mortality: arterial bypass (ICD-9-CM 39.29) was associated with lower mortality. Outcomes were defined for the subgroup (n = 7,728) undergoing arterial bypass (ICD-9-CM 39.29) who were assigned to Patient Management Category 4101, 4113, or 4141: Mortality rates were 2.26, 2.19, and 5.03% for those undergoing elective bypass (n = 3003), urgent bypass (n = 3,513), and bypass for gangrene (n = 1212), respectively. Octogenarians did not experience higher mortality rates after elective bypass ¿1.4% (1/73) vs 2.3% (67/2,930), n.s., but experienced higher mortality rates after urgent bypass ¿8.6% (8/93) vs 2.0% (69/3,420), P < 0.001 and after bypass for gangrene ¿11.6% (5/43) vs 4.8% (56/1,169), P < 0.045.

CONCLUSIONS

Outcomes for patients in DRGs 478 and 479 who underwent extremity revascularization differed from those who did not. Outcomes varied by the type of arterial reconstruction and its urgency and indication and within selected subpopulations (i.e., octogenarians). DRG-based reimbursement would not be sensitive to these clinically important factors which have a major impact on outcomes and resource utilization.

Modeling of high-cost patient distribution within renal failure diagnosis related group

Modeling by mixed-distribution was proposed in order to analyze heterogeneity of costs and length of stays within Diagnosis Related Groups (DRGs). A mixed-distribution model based on Weibull distributions was applied to 791 discharge abstracts of French DRG no. 450 (Health Care Financing Administration 3 DRG no. 316 "Renal failure") from a national database. Three subgroups of cost and length of stay were identified. Except for age, clinical criteria significantly linked with the long-stay subgroup were the same as those associated with the high-cost subgroup: acute renal failure, intensive care, infectious complications, and vascular investigations. The identification of factors associated with high costs, based on the proposed model, will allow physicians to understand more accurately how their choice of specific procedures influences hospital costs.

High cost factors for leukaemia and lymphoma patients: a new analysis of costs within these diagnosis related groups

STUDY OBJECTIVE

To determine high cost factors to help managers and clinicians to analyse the reasons of adverse costs and provide indications for financial negotiation.

DESIGN

To locate high cost or long stay patients, the analysis was designed on the basis of a mixture of Weibull distributions. In this new model, the proportion of high cost patients was expressed according to the multinomial logistic regression, permitting the determination of high cost factors.

SETTING

The 1993 French reference database, constituted in the framework of the national study of DRG costs, conducted by the French Ministry of Health. The database of discharge abstracts recorded in 1993 in the Dijon public teaching hospital.

PARTICIPANTS

The analyses were based on 1352 abstracts from the French reference database and 368 from the Dijon database concerning patients, aged 18 and over, suffering from leukaemia and lymphoma.

MAIN RESULTS

High cost and long stay factors were the same: number of stays, death, transfer, acute leukaemia, neutropenia, septicaemia, high dose aplastic chemotherapy, central venous catheterisation, parenteral nutrition, protected or laminar airflow room, blood transfusion, and intravenous antibiotherapy.

CONCLUSIONS

Taking into account high cost predictive factors, as shown in the case of leukaemia and lymphoma patients, would help to reduce the adverse effects of a prospective payment system.

Adjusting outcome measurements for case-mix in a vascular surgical register--is it possible and desirable?

OBJECTIVE

We analysed the variation in the outcome of infrainguinal bypass surgery between departments in a register for clinical audit to see if variation in case-mix influenced the results.

MATERIALS AND METHODS

The study was a retrospective analysis of 764 infrainguinal bypass operations performed from 1988 to 1990 at six Swedish surgical departments. Results were assessed at 30 days and at 1 year postoperatively.

RESULTS

There was a significant variation (p < 0.01) in mortality and amputation rates both at 30 days and at 1 year and in patency rate at 30 days. There were also differences in case-mix. Differences were found in indication, location of distal anastomosis and graft type. Regression analysis found that mortality was influenced by age, diabetes and heart disease and patency rate by location of the distal anastomosis and graft type. When 'hospital' was added as a variable in the regression analysis it was also found to be a significant indicator.

CONCLUSION

Location of the distal anastomosis was the main factor in adjusting patency for case-mix.

Linking cost control measures to health care services by using activity-based information

Canada's health care institutions are under pressure to limit expenditures, maintain or increase productivity, and assimilate new technology. Even though more than 75% of hospital operating expenditures are controllable, according to a study by the Economic Council of Canada, cost systems are needed to provided essential management information. The new Canadian Management Information System (MIS) Guidelines for health care are designed to provide accurate cost measurement of patient treatment and to help managers evaluate the impact of planned program changes on areas of operational responsibility. Other potential benefits of implementing the MIS guidelines include correcting dysfunctional funding of health care units with benchmarking and setting high reporting standards for resource use at the patient level (MIS, 1991). This paper focuses on one important aspect of bringing these costs under control by examining the relation between cost deviations (variances) and underlying cost drivers. Our discussion will lead to the conclusion that incompatibility of DRG methodology and traditional cost accounting models may be an important source of cost variability within diagnostically-related disease groupings.

Relationship between severity, costs and claims of hospitalized patients using the Severity of Illness Index

The results of the prospective application of Horn's 'Severity of Illness Index' in a teaching hospital during 1987, 1989, and 1990 constitute the basis of the present report. The average overall severity of illness scores for the three years were 1.42 in 1987, 1.65 in 1989, and 1.46 in 1990. Most of the processes evaluated in the three periods showed an overall distribution among severity levels 1 and 2, both overall and when the seven dimensions of the severity of illness index were analyzed. A statistically significant correlation between the overall severity of illness and average length of stay was found for patients in 1989 and 1990. The length of stay differed significantly in the different severity levels. When the four levels of the seven dimensions of the severity of illness index for 1987, 1989, and 1990 were compared, it was observed that figures were not uniformly distributed. There was a statistically significant association between severity of illness for hospital service and pharmacy charges per hospital stay for both 1989 and 1990, as well as a statistically significant inverse relationship between severity of illness and the number of claims per hospital service in both periods of time. Case-mix methods that account for the severity of patients constitute a useful indicator of quality for the management of different hospital services and of the hospital as a whole.

Explaining cost variations in DRGs 'Acute Myocardial Infarction' by severity of illness

The empirical relationship is analyzed between the severity of illness and costs of medical care for 464 patients classified into DRGs 121-123, Acute Myocardial Infarction (AMI), in the University Hospital, Maastricht. Severity of cardiac and cardiovascular disorders characteristic of acute myocardial infarction is defined and operationalized in a sense that closely resembles the clinical practice of cardiologists. The effect of the severity of illness on DRG cost variations is studied separately for the costs of acute care (such as thrombolytic therapy, cardiac catheterization and percutaneous transluminal coronary angioplasty (PTCA)), length of hospital stay, costs of intensive nursing care at the coronary care unit (CCU) and the costs of ECGs, laboratory tests, echocardiography, exercise tests and drugs. For AMI patients, severity of illness measured by specific clinical criteria is found to give better predictions (higher R2) for costs of medical care than the DRG classification.

A description and clinical assessment of the Computerized Severity Index

Recent initiatives expanding health care data networks have increasingly emphasized severity of illness information, both to improve fairness of hospital payment and to assist in widespread assessment of hospital and physician quality. The Computerized Severity Index (CSI), one of the newest severity tools to generate interest, is disease specific and produces scores from 1 to 4 at both the disease and overall patient levels. Severity is defined as "the treatment difficulty presented to physicians due to the extent and interactions of patient's diseases." The clinical logic of the severity rating system is readily available through the "severity matrices" associated with over 820 disease groups. Questions exist about the CSI's dependence on diagnostic coding and the qualitative nature of some of the clinical criteria. More study is required to assess the utility of the CSI for various health policy purposes.

A regression-like approach to developing a severity index for EMS patients

An approach to constructing a severity index for emergency medical services patients with cardiac-related problems is developed. The procedure is based on two linear programming models and produces a set of weights which can be added to estimate the severity of a patient's condition. A set of patients independent of the set used to derive the weights was ranked with respect to severity by a set of physicians who were also independent of the model development process. The average value for Spearman's rank order correlation coefficient (rho) between a ranking based on the severity weights and the physicians was 0.6897. The average value of rho calculated over all possible pairs of the physician rankings was 0.6859. Thus, the ranking based on the severity weights correlated as well with the physicians ranking as did the physician rankings among themselves.

Practice databases and their uses in clinical research

A few large clinical information databases have been established within larger medical information systems. Although they are smaller than claims databases, these clinical databases offer several advantages: accurate and timely data, rich clinical detail, and continuous parameters (for example, vital signs and laboratory results). However, the nature of the data vary considerably, which affects the kinds of secondary analyses that can be performed. These databases have been used to investigate clinical epidemiology, risk assessment, post-marketing surveillance of drugs, practice variation, resource use, quality assurance, and decision analysis. In addition, practice databases can be used to identify subjects for prospective studies. Further methodologic developments are necessary to deal with the prevalent problems of missing data and various forms of bias if such databases are to grow and contribute valuable clinical information.

The "corporatization" of U.S. hospitals: what can we learn from the nineteenth century industrial experience?

While broad parallels have been noted between the current "corporatization" of health care and developments in U.S. manufacturing in the late 19th century, there has been little in-depth analysis of these parallels. This article explores trends in the industrial organization of the hospital industry from the perspective of the manufacturing experience. Efforts to use corporate managerial techniques to rationalize hospitals have played an important role in the development of the modern structure of the hospital industry since the 1920s. But the emergence of multihospital systems is a new phenomenon. Some significant similarities exist between current conditions in the hospital industry and conditions in manufacturing at the time of the great industrial merger boom at the turn of the century. The subsequent experience of multiplant manufacturing firms created during the great industrial merger boom varied considerably. The characteristics of successful industrial consolidations are not present in the hospital industry; but motives for consolidation exist that were not present in manufacturing, while changes in the organization of production loom in the future.

Improving the homogeneity of diagnosis-related groups (DRGs) by using clinical laboratory, demographic, and discharge data

For 48 of the most common diagnosis-related groups (DRGs) at our hospital, we examined the ability of clinical laboratory tests, demographic data, and ICD-9-CM codes, which provide a measure of severity of illness, to predict patients' length of stay (LOS) more accurately than DRGs alone. For 10 of 20 medical DRGs and 13 of 23 surgical DRGs examined, we were able to increase the ability to predict LOS by at least 10 per cent. The laboratory tests that proved most predictive of LOS over all DRGs were the mean serum sodium, potassium, bicarbonate, and albumin. The system is data driven, objective, and flexible, thus ensuring its utility for the purpose of equitable reimbursement.

The effect of resident involvement on community hospital charges

Attempts to explain the high cost of care in teaching hospitals have yielded conflicting results. This study was conducted to compare hospital charges and lengths of stay for two groups of patients: one cared for by a resident team and the other cared for by attending physicians. The study was conducted at a university-affiliated hospital in Portland, Oregon. An initial group of 5,451 admissions was examined, considering type of doctor (resident or attending), severity of illness, and patient demographic characteristics. A regression analysis revealed that total charges were similar in the two groups, but only 14% of the variance in log total charges was explained. A subgroup of 1,058 admissions in the eight most common diagnosis-related groups (DRGs) was further evaluated. In this analysis total charges for the resident patient group were 52% higher than charges for the patient group cared for by attending physicians. Forty-one per cent of the variance was explained, with type of doctor and severity of illness accounting for 5% and 10%, respectively. Further examination of one DRG indicated that additional factors not included in previous studies, such as extent of preadmission evaluation, ethical factors influencing treatment options, and patient expectations for care, may be important determinants of hospital charges. This study demonstrates that the high cost of resident care is not fully explained by currently available measures.

DRGs and severity of illness measures: an analysis of patient classification systems

Concerns about inefficiency and excessive expenditures has stimulated interest in the measurement of the output of the hospital, an organization that produces a wide variety of products. Diagnosis Related Groups and a number of severity measures including Disease Staging, the Severity of Illness Index, and Patient Management Categories are described and evaluated. Using criteria such as reliability, clinical validity, homogeneity, number of categories, and data acquisition costs, the analysis suggests that while none of the hospital patient classification systems is perfect, great strides have been made in reflecting the multiproduct nature of the hospital.

Economic outcome under Medicare prospective payment at a tertiary-care institution: the effects of demographic, clinical, and logistic factors on duration of hospital stay and part A charges for medical back problems (DRG 243)

We investigated the effects of prospectively identified factors on the duration of hospital stay and part A charges in 240 hospitalizations (of 230 patients) for the diagnosis-related group "medical back problems" (DRG 243) at a tertiary-care institution in 1985 to determine whether heterogeneity existed within this reimbursement category. We confirmed our initial postulates that nonosteoporotic fractures and neck problems, as well as hospitalizations primarily for myelography after outpatient neurologic evaluation, had considerably different economic outcomes and thus excluded these categories from further analysis. Statistical analysis (forward stepwise regression) of the remaining 132 patients who had "general medical back problems" showed that increasing age, associated osteoporosis, and therapeutic injections best explained variation in the natural logarithm of duration of stay (R2 = 0.16). Total number of diagnoses, spondylosis, associated osteoporosis, age, therapeutic injections, and performance of special procedures best explained the variation in the logarithm of part A charges (R2 = 0.29). The ability to identify factors within a specified category that affect the duration of hospitalization and part A charges jeopardizes the fairness of prospective payment, and we believe that DRG 243 should be adjusted for age, comorbidity, and readily identifiable clinical syndromes that have disparate economic consequences. Because of poorly substantiated efficacy and a significant association with longer hospital stays and higher part A charges, clinicians should review the use of therapeutic injections for medical back problems. Analysis of case-mix such as ours should be helpful in promoting efficient practice and ensuring the fairness of any reimbursement system.

Hospital length of stay of frail elderly patients. Primary care by general internists versus geriatricians

Hospitalized patients in a medical diagnosis-related group (DRG) who were cared for by physicians in a division of Geriatric Medicine (geriatric group) were compared with a control group drawn from a stratified random sample of patients cared for by general internists (internist group) in the same hospital. Despite an older age, longer predicted length of stay, and higher DRG reimbursement, the geriatric group patients had a significantly shorter length of stay (8.8 vs 15.8 days; P less than 0.05) than the internist group. A shorter length of stay for the geriatric group was noted in each of five subgroups, sorted by admission and discharge status. Comparison to national data reveals that, despite the shorter length of stay in the geriatric group, length of stay data used by hospitals for management purposes would still classify the patients of the geriatrics group as "revenue losers" under the Medicare hospital reimbursement system.

The Computerized Severity Index. A new tool for case-mix management

We describe the new Computerized Severity Index (CSI) that is obtained from an expanded discharge abstract data set, based on a 6th-digit severity addition to the ICD-9-CM coding system. The new 6-digit code book (called ICD-9-CMSA) is used to label existence and severity of each principal and secondary diagnosis. It can be used to produce an overall severity of illness level for each hospital inpatient. The impact of severity-adjusted DRGs on prospective payment and uses of the CSI for assessing quality of care, efficiency, physician practice profiles, and prediction of posthospital resource needs are discussed.

Severity of illness within DRGs: impact on prospective payment

This study compares the financial impact of a Diagnosis Related Group (DRG) prospective payment system with that of a Severity of Illness-adjusted DRG prospective payment system. The data base of about 106,000 discharges is from 15 hospitals, all of which had a Health Care Financing Administration (HCFA) DRG case mix index greater than 1. In order to pool the data over the 15 hospitals, all charges were converted to costs, normalized to Fiscal Year 1983, and adjusted for medical education and wage levels. The findings showed that, for the study population as a whole, DRGs explained 28 per cent of the variability in resource use per case while Severity of Illness-adjusted DRGs explained 61 per cent of the variability in resource use per case. When we simulated prospective payment systems based on DRGs and on Severity-adjusted DRGs, we found that the financial impact of the two systems differed by very little in some hospitals and by as much as 35 per cent of total operating costs in other hospitals. Thus, even with a data set that is relatively homogeneous (with respect to the HCFA DRG case mix index definition of hospitals), we found substantial inequities in payment when DRGs were not adjusted for Severity of Illness. These findings suggest that, with a more representative set of hospitals, the difference between unadjusted and Severity-adjusted DRG-based prospective payment could be greater than 35 per cent of a hospital's total operating costs.