Medicare’s DRG-weights in a European environment: the Spanish experience

Abstract

Hospital payment systems are being changed to mixed systems, composed of case-mix categories and structure indicators. The Health Care Financing Administration’s Diagnosis-Related-Groups (HCFA’s DRGs) and DRG-weights are used in Catalonia as Prospective Payment System (PPS)-instruments for hospital inpatient reimbursement. The Catalonian and Spanish health systems, however, are very different from the US health environment. The aim of this study is to determine whether the HCFA’s DRGs and DRG-weights fit the special characteristics of a European environment. To do this, cost-based weights, determined from information from the cost accounting system of two public hospitals in Barcelona, are compared with Medicare-weights. A total of 35 262 discharges representing 12 794 million pesetas are analyzed. Medicare-weights do not differ globally from cost-based-weights and the adjusted correlation weighted least squares regression between the two weight-scales is 95%. There are, however, systematic deviations in six DRG-groupings. The most important deviations are concentrated in Ambulatory Surgery categories, in DRGs in which prostheses are used, and in specialties excluded from several PPSs because of extreme variability in treatment intensity. In conclusion, Medicare-weights can be used to pay hospital output in a European environment but they should be adjusted to avoid systematic deviations.

Introduction

Following the tendency of other European countries, prospective payment systems (PPS) capable of defining and valuing hospital output have recently been introduced in Spain.

At the beginning of the 1980s, Catalonia (Spain) began a process of decentralization which gave this autonomous region control over its own health system within the general Spanish context. In the last few years, this system has served as an example for reforms, mainly regarding hospital care, in other regions of Spain as well as in several South American countries. The Catalan public health sector comprises around 70 general hospitals of which only ten are owned by the regional health service authority, the Catalan Health Service (CHS). The public health sector hospitals, which serve a population with public health insurance (100% of the population), are owned by the municipalities, by non-profit making organizations and by institutions composed of a mixture of these entities and the CHS.

Until recently, a PPS was used which measured and valued inpatient and outpatient hospital activity but which did not distinguish among case-mix categories [1]. Since 1997, hospital output measuring systems have been introduced to allocate resources [2]. Patients have been classified according to the Diagnosis-Related-Groups (DRG)-system used by the US Health Care Financing Administration (HCFA). The HCFA’s DRG-weights have been used to weight hospital-wide discharges and to determine how much each hospital should be reimbursed. Unlike a true PPS, the model introduced allocates a fixed budget among hospital-wide activity which is the total of the forecasted activity detailed by each hospital. Thus, priority is given to criteria which tend to reduce the providers’ and purchasers’ financial risk, as has been explained by Newhause [3], Magnusen [4] and Ellis [5].

The usefulness of DRGs and their positive results for clinical analysis in Spain and other European countries have been partially validated [6], [7]. They have been used on several occasions as a patient classification system for hospital cost analysis [8], [9]. Internationally, this system has been validated and adapted to reimburse hospital output [10] in several countries, such as Portugal [11], Norway [4], United Kingdom [12], [13], [14], [15], Australia [16], [17] and Ontario in Canada [18] among others. These countries have modified the patient classification system and adjusted the DRG-weights to various degrees. As yet, there has been no study of the workings of DRGs and their associated weights in Spain.

There are clear differences between European National Health Services and the US Medicare system. These differences affect the structure of hospital costs and consequently make the transfer of financial tools between systems difficult. As a primary issue, the following differences must be considered:

• Although the information used to create DRG was the widest possible [19] and based not solely on Medicare patients, the recalibration of weights was based on Medicare’s case-mix [20], [21]which is very different from the universal health coverage provided by European National Health Services. Within the US system itself, it is questioned whether DRG-weights can be applied to such different hospitals and forms of recalibration.

• US weights do not incorporate the cost of physicians’ fees for which there is a different form of payment [22], whereas in Europe, physicians are considered integral to the centers’ staff and no distinctions are made among the financing of operating costs, whether of staff or of any other type.

• Because of differences between European systems and the Medicare system, and because output-based payment methods divide up health care without considering it as a whole, the moment in which a patient is admitted to hospital or is discharged may also be different in each system [23]. The reasons for this difference may be due to the structure of the public sector in terms of primary care, long-stay hospitals or nursing homes, to the financial incentives of the alternatives to inpatient care, or to the incentives of the hospital payment system [24].

• Ambulatory surgery is performed much more in the US than in Europe [25], [26], [27]; in Catalonia it is performed in only a few centers and specialties [28], [29]. In the US, ambulatory surgery is financed separately from the general PPS [30], [31] while the new model implemented in Catalonia does not distinguish between inpatient and outpatient surgery, reimbursing both through the weight of the corresponding DRG [2].

• The cost of prostheses forms part of Medicare’s DRG-weights and also of the actual costs of the hospitals analyzed. The price of prostheses is much higher in Spain than in the US.

• Several US health care purchasers who use a DRG-based payment system have decided to apply different forms of payment to certain case-mix categories. This is because of the excessive variability in the types of treatment provided by certain specialties, namely: rehabilitation, psychiatry, drug and alcohol abuse, and transplants [22], [32], [33], [34].

• The deficient classification of secondary diagnoses leads to the undervaluation of the DRG to which the patient is assigned [35], [36], [37]. This most likely leads to actual costs being underestimated in European health services where until now there has been no financial incentive to improve the quality of classification.

The aim of this study is to determine whether a positive and sufficient relationship can be established between Medicare’s DRG-weights (MW)-structure and the cost-based DRG-weights (CW)-structure determined for the purposes of this study. Our hypothesis is that the seven above-mentioned differences will invalidate the direct importation of MW to Spain. A further aim is to quantify the intensity of these differences and to propose possible adjustments.

The opportunity to carry out this study was provided by exhaustive information on the cost of 35 262 discharges from the hospitals belonging to the Municipal Institute of Health (MIH) in Barcelona from 1995 to 1996. Per discharge cost information in Spanish hospitals is not normally available.