National Early Warning Score (NEWS) independently predicts death or unexpected ICU admission in general surgical or orthopaedic patient.

Abstract

Background: National Early Warning Score (NEWS) has been shown accurate in predicting severe adverse event (SAE; defined as death or unexpected ICU admission) for in-ward patients. Unfortunately, its trigger level varies among hospitals and population studied. The main objective of the study is to determine the best in-patient trigger level in predicting occurrence of SAE. Methods: All patients aged >12 years old admitted to surgical or orthopaedic ward from 1st December 2016 to 31st January 2017 were included. They were prospectively followed up until discharged or developed SAE. NEWS was calculated on admission and during every nursing shift. Other than demographic data, parameters analysed were diagnosis at presentation, specialty involved, surgical procedure done, type of surgery (elective or emergency) and Charlson’s co-morbidity index. Results: Among 226 patients, eight of them developed SAE (3.5%; 4 deaths and 4 unexpected ICU admissions). Mean NEWS was significantly different between non-SAE and SAE groups (1.77±1.642 vs 6.5±3.162; p=0.0001 respectively). SAE was independently associated with NEWS score (OR=2.828; 95% CI: 1.632-4.902; p=0.0001) on multivariate stepwise logistic regression with AUROC of 0.905 (p=0.0001). Sensitivity, specificity, positive and predictive value for NEWS ≥ 5 in predicting SAE was 87.5%, 91.3%, 26.9% and 99.5% respectively with number needed to screen was 3.6. Conclusion: NEWS is an excellent model to predict death or unexpected ICU admission in general surgical and orthopaedic patients. A score of ≥ 5 is an ideal trigger level to further escalate clinical care.