Avaliação da produtividade de hospitais brasileiros pela metodologiado diagnosis related group (DRG): 145.710 altas em 116 hospitais

Abstract

Worldwide resources spent on health vary between 5% to just over 20% of GNP (Gross National Product). In Brazil, the federal constitution mandates for health at least 15% of municipal revenues and 12% of the state. Managing these investments satisfactorily becomes a major challenge. Developed in the 80s for the US government, the Diagnosis Related Groups (DRG) is a patient classification system that seeks to relate the types of care to the resources consumed during hospitalization, enabling the comparison of costs and hospital productivity between institutions. This study measured the productivity of Brazilian hospitals beds compared to US hospitals. Methodology: 145.710 hospital discharge reports were analyzed in hospitals of Minas Gerais and Goiás, comparing the frequencies of stay by DRG product. DRG. The DRG categorization of the US government was used in its version 31.0 (MS-DRG) and Brasil® software was used to match the Brazilian coding system of procedures TUSS (Unified Terminology of Supplemental Health) and SUS (Unified Health System). Validation of correspondences made by the software occurred for each product by medical staff. Thus, it was established ratio between the number of days spent by product, where values above 1 represent Brazilian productivity lower than US, and below 1 indicate higher Brazilian productivity. Results: Females were predominant with 63.9% and the average age was 42.8 years. The median equal to or less than the 10th percentile was recorded at 31% of surgical and 5.5% of clinical. The proportion of DRG surgical type (65.8%) is higher than that found in clinical DRG (36.3%). Surgical type is 3.4 times likely to have hospital stay below or equal to the American 50th percentile. The permanence in Brazilian hospitals is higher for clinical DRG (63.7%) than surgical (34.2%). It can be inferred that the Brazilian productivity clinical hospital bed is smaller than the US, while surgical is equal to or greater (65.8%). The burden of care complexity of each product showed positive correlation with the median length of hospital stays. However, the difference between the correlation of clinical DRG (p = 0.001; R 2 = 28.2) compared with surgical (p = 0.001; R2 = 49.6) does not explain the findings. In the category of surgical DRG the correlation (p = 0.001; R 2 = 49.6), shows that the complexity explains about half of the factors which impact the time of bed occupancy. On the other hand, in clinical DRG, the part not correlated with the complexity of care, which represents about 70%, results from other no-biological factors inherent to the health system. Conclusions: The clinical beds are mainly responsible for the low productivity of the Brazilian hospitals studied in relation to the US. This indicates a window of opportunity for improving Brazils health systems through adjustments in the paradigm of management with the focus on hospital processes.