Predictive accuracy of Early Warning Score Systems for Detecting Critically Ill Patients in an Outpatient Setting

Background No systematic methods exist for triaging outpatients with severe conditions. Our previous pilot study suggested that the National Early Warning Score (NEWS) predicts admission and unexpected intensive care unit (ICU) transfer in outpatients. The Visensia Score Index (VSI) is another artificial intelligence-based Early Warning Score System that automatically analyzes data from bedside monitors. This study aimed to compare the predictive accuracy of NEWS and VSI in identifying potentially critically ill patients in outpatient settings. Methods This single-center retrospective cohort study was conducted at a 344-bed tertiary-care hospital in Japan. The study population comprised first-visit patients at an internal medicine clinic from June to November 2018. Age, sex, physiological measurements, NEWS, VSI, disposition, and outcomes were collected. Results 3,301 patients were included. A total of 108 (3.3%), 16 (0.5%), and 5 (0.2%) patients were admitted to the general ward, high-dependency unit (HDU), and ICU, respectively. The area under the curve (AUC) of NEWS for hospital admission, HDU or ICU admission, and ICU admission were 0.71 (95% confidence interval (CI, 0.66 - 0.76); 0.88 (95% CI, 0.80 - 0.97); and 1.00 (95% CI, 0.996-1.0), respectively, while those of VSI were 0.66 (95% CI, 0.60-0.71), 0.82 (95% CI, 0.71-0.93), and 0.97 (95% CI, 0.96-0.98] ). The AUC of NEWS was significantly superior to that of VSI for both hospital (p = 0.03) and ICU admissions (p