Volume 21 Supplement 2
Prediction of in-hospital mortality and admission to ICU using vital signs - a study of Early Warning Score as an alternative to traditional triage
© Illum Vendler et al; licensee BioMed Central Ltd. 2013
Published: 9 September 2013
Triage of patients in the Emergency Department includes scoring of vital parameters. The objective of this study was to compare two such triage systems for assessing vital parameters - a single-parameter system, T-vital, as used in Danish Emergency Process Triage, and a multiple-parameter system, T-EWS, which we based on Early Warning Score (EWS) - and correlate the triage scores to in-hospital mortality and admission to ICU. Studies examining EWS in triage are currently limited in number.
Using data from the Acute Admission Database of Nordsjællands Hospital (n = 6164 admissions), we calculated and stratified EWS into four T-EWS colour codes (red, orange, yellow, and green), testing different stratifications' correlation to in-hospital mortality and admission to ICU. Afterwards, we compared the ability of the chosen T-EWS and T-vital to predict patients at risk (red and orange category) of in-hospital mortality or admission to ICU. The data were analysed using area under the receiver operating curve (AUROC), sensitivity, specificity, overtriage, undertriage, and diagnostic rates.
T-vital allocated 10.6% of patients to the orange or red category, whereas T-EWS allocated 5.8% to these categories. There was no significant difference in the ability of T-EWS to predict in-hospital mortality compared to T-vital (AUROC (95% CI): T-EWS = 0.74 (0.70-0.79); T-vital = 0.76 (0.72-0.80)). Likewise, there was no significant difference in prediction of ICU admission (AUROC (95% CI): T-EWS = 0.76 (0.70-0.81); T-vital = 0.73 (0.67-0.79)). The specificity (95% CI) of T-EWS compared to T-vital was higher for both in-hospital mortality (0.95 (0.94-0.95) and 0.90 (0.90-0.91), respectively) and for admission to ICU (0.95 (0.94-0.95) and 0.90 (0.89-0.91), respectively). There was a trend of higher sensitivity of T-vital, and no difference in overtriage, undertriage or diagnostic rates.
The two triage systems are largely similar in their ability to discriminate patients at high risk of in-hospital mortality or admission to ICU. However, T-vital's larger proportion of orange and red patients might yield a larger workload in the Emergency Department. Replacement of T-vital with T-EWS could be considered, as EWS is already in use as a monitoring tool after triage, but more studies are needed for further clarification.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.