Evaluation of an interdisciplinary data set for national intensive care unit assessment

OBJECTIVES: To evaluate the ability of an interdisciplinary data set (recently defined by the Austrian Working Group for the Standardization of a Documentation System for Intensive Care [ASDI]) to assess intensive care units (ICUs) by means of the Simplified Acute Physiology Score II (SAPS II) for the severity of illness and the simplified Therapeutic Intervention Scoring System (TISS-28) for the level of provided care. DESIGN: A prospective, multicentric study. SETTING: Nine adult medical, surgical, and mixed ICUs in Austria. PATIENTS: A total of 1234 patients consecutively admitted to the ICUs. INTERVENTIONS: Collection of data for the ASDI data set. MEASUREMENTS AND MAIN RESULTS: The overall mean SAPS II score was 33.1+/-2.1 points. SAPS II overestimated hospital mortality by predicting mortality of 22.2%+/-2.9%, whereas observed mortality was only 16.8%+/-2.2%. The Hosmer-Lemeshow goodness-of-fit test for SAPS II scores showed lacking uniformity of fit (H = 53.78, 8 degrees of freedom; p < .0001). TISS-28 scores were recorded on 8616 days (30.6+/-1.5 points). TISS-28 scores were higher in nonsurvivors than in survivors (30.4+/-0.9 vs. 25.7+/-0.4, respectively; p < .05). No significant correlation between mean TISS-28 per patient per unit on the day of admission and mean predicted hospital mortality (r2 = .23; p < .54) or standardized mortality ratio per unit (r2 = -.22; p < .56) was found. CONCLUSIONS: Implementation of an interdisciplinary data set for ICUs provided data with which to evaluate performance in terms of severity of illness and provided care. The SAPS II did not accurately predict outcomes in Austrian ICUs and must, therefore, be customized for this population. A combination of indicators for both severity of illness and amount of provided care is necessary to evaluate ICU performance. Further data acquisition is needed to customize the SAPS II and to validate the TISS-28.