Retrospective analysis of hospitalization costs using two payment systems: the diagnosis related groups (DRG) and the Queralt system, a newly developed case-mix tool for hospitalized patients

Júlia Folguera,Elisabet Buj,David Monterde,Gerard Carot-Sans,Isaac Cano,Jordi Piera-Jiménez,Miquel Arrufat
DOI: https://doi.org/10.1186/s13561-024-00522-6
2024-06-27
Health Economics Review
Abstract:Hospital services are typically reimbursed using case-mix tools that group patients according to diagnoses and procedures. We recently developed a case-mix tool (i.e., the Queralt system) aimed at supporting clinicians in patient management. In this study, we compared the performance of a broadly used tool (i.e., the APR-DRG) with the Queralt system.
economics,health policy & services
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to compare the performance of two case - mix tools (i.e., the widely - used APR - DRG and the newly - developed Queralt system) in predicting hospitalization costs in the hospital service payment system. Specifically, the study aims to evaluate the predictive ability, efficiency, and within - group homogeneity of these two tools to determine which tool can more effectively support hospital management and cost control. ### Research Background - **Traditional Payment Models**: The traditional fee - for - service model and the fixed - daily - fee payment model have problems of inefficiency and poor quality, such as unnecessary procedures and a longer average length of stay. - **Case - mix Tools**: To overcome these problems, many healthcare systems have turned to case - mix tools, which simplify patient complexity and procedures into a limited number of groups in order to classify all hospitalization situations. - **Limitations of Existing Tools**: Although most evaluations focus on clinical or hard endpoints, such as mortality during and after hospitalization, there are fewer studies on the performance of these tools in cost estimation. ### Research Objectives - **Primary Objective**: Compare the performance of APR - DRG and Queralt system in predicting hospitalization costs. - **Secondary Objective**: Evaluate the efficiency and within - group homogeneity of the two tools. ### Methods - **Data Sources**: A retrospective analysis of all hospitalization records in eight hospitals managed by the Catalan Institute of Health (ICS) in 2019, accounting for approximately 30% of all hospitalizations in Catalonia. - **Cost Calculation**: Both direct and indirect costs are obtained from complete cost accounting. - **Grouping Methods**: - **APR - DRG**: Grouping is based on diagnosis - related groups (DRG) and severity levels. - **Queralt System**: Grouping is based on the principal diagnosis (Queralt DxP), comorbidities at admission (Queralt DxS), complications during hospitalization (Queralt DxC), and major and minor procedures (Queralt Px). ### Results - **Overall Costs**: The analysis includes 166,837 hospitalization events with an average cost of 4,935 euros. - **Efficiency and Heterogeneity**: The components of the Queralt system perform better in terms of efficiency (i.e., the percentage of covered costs and the number of hospitalizations) and within - group homogeneity. - **Predictive Models**: - **Logistic Regression**: The DP_strata of the Queralt system performs best in predicting high costs (80th, 90th, and 95th percentiles). - **Linear Regression**: The DP_strata of the Queralt system has a higher fit in predicting actual costs. ### Conclusions - **Main Findings**: The Queralt system not only performs well in predicting hospital outcomes but also outperforms APR - DRG in predicting hospitalization costs. - **Future Research**: It is necessary to further validate the performance of the Queralt system in other countries and regions to assess its universal applicability. Through this study, the authors hope to provide medical managers with a free and flexible tool that can accurately reflect hospital costs and can be adjusted according to specific needs.