Modified Early Warning Score Performance in Predicting the Outcome of Head Trauma Patients in Emergency Department

Modified Early Warning Score (MEWS) is a useful scoring system that detects physiological changes in identifying adult medical surgical patients at risk for deterioration and death. Although it is a good predictor of deterioration and mortality in hospitalized patients, the performance of MEWS in the ED and specifically in head trauma patients is still untested. The purpose of this study was to assess the performance of MEWS in predicting the outcome of head trauma patients in ED. We conducted a retrospective observational study at RSUD dr. Soedono Madiun, East Java Province, Indonesia. The sample using the medical record of patients with moderate-severe head trauma adjusted to the inclusion and exclusion criteria and used purposive sampling technique as many as 181 respondents. Multivariate analysis of ordinal logistic regression to determine the MEWS parameters that most influenced the outcome of head trauma patients, while the ability to predict outcomes was assessed using The Area Under Receiver Operating Characteristic (AUROC) analysis. The ordinal logistic regression results show that AVPU parameters show an Odd Ratio (OR) value of =2.89, meaning that the poor AVPU parameter measurement results have a risk of obtaining a death outcome of 2.89 greater than the outcome of moderate disability, severe disability, vegetative state persistence. The AUC MEWS value to predict the outcome of head trauma patients was 0.777 (95% CI; 0710-0,836) so that based on the AUC value the level of accuracy of MEWS’s scoring predictions was in good classification. In conclusion, MEWS’s ability shows good performance in predicting the outcome of head trauma patients.