ABSTRACT
Background: The number of visits and mortality rates of emergency patients at Emergency Department (ED) have been increasing from time to time. Those patients have a wide spectrum of conditions. Appropriate identification of patients with high mortality risk is crucial. The Hypotension, Oxygen Saturation, Low Temperature, ECG Changes, and Loss of Independence (HOTEL) score is an easy and important tool to be applied in the ED. However, the score has not been validated in Indonesia.
Objective: To evaluate the performance of HOTEL score in predicting the 24-hour mortality of nonsurgical patients in ED of Cipto Mangunkusumo hospital.
Methods: The design is a prospective cohort study. The research subjects were the nonsurgical patients who were admitted to ED of Cipto Mangunkusumo hospital between October-November 2012. We collected systolic blood pressure, peripheral oxygen saturation, body temperature, ECG changes, and loss of independence data. Those data were evaluated based on the HOTEL scoring system. The outcome were evaluated in 24-hour after admission (alive or dead). The calibration was evaluated with the Hosmer-Lemeshow test. The discrimination performance was measured with area under the curve (AUC).
Results: There are 815 non-surgical patients who were admitted to the ED between October until November 2012, but only 804 (98.7%) subjects were included in this study. The 24-hour mortality rate is 3.7% (30 subjects). The calibration performance with the Hosmer-Lemeshow test showed p value of 0.753. The discrimination performance is shown with the AUC score of 0.86 (95% CI 0.781; 0,93).
Conclusion: The HOTEL score has a good calibration and discrimination performance in predicting the 24-hour mortality of the nonsurgical patients in ED of Cipto Mangunkusumo hospital.
Key words: Validation, HOTEL score, mortality, nonsurgical patients, emergency department