Discrimination of the National Early Warning Score versus the 2018 Chinese Emergency Triage Standard for in-hospital mortality in emergency department patients with confirmed high-risk cardiovascular and aortic emergencies: a single-centre retrospective cohort study

Abstract

The National Early Warning Score (NEWS) is a physiological severity score, whereas the 2018 Chinese Emergency Triage Standard is a categorical triage system used to assign treatment priority. Both use information available at first emergency department (ED) assessment, but their relative ability to discriminate mortality risk in high-risk cardiovascular and aortic emergencies has not been directly compared. We compared their discrimination for all-cause in-hospital mortality and explored a possible adjunctive role for NEWS in routine triage. This single-centre retrospective cohort study included adults presenting to the ED of a tertiary teaching hospital between January 2019 and December 2023 with a confirmed high-risk acute cardiovascular or aortic emergency, including acute myocardial infarction, acute pulmonary embolism, acute aortic syndrome, or aortic aneurysm. A retrospectively reconstructed NEWS (supplemental-oxygen status was unavailable and scored 0 for all patients) was calculated from data recorded at first ED triage. Triage grade, assigned by the triage nurse according to the 2018 Chinese Emergency Triage Standard, was extracted as a recorded variable. The primary outcome was all-cause in-hospital mortality. Discrimination was assessed using the area under the receiver-operating-characteristic curve (AUROC) and compared using DeLong’s test. Internal validation was performed by bootstrap resampling, and associations between vital signs and mortality were explored using restricted cubic splines. Among 3,168 included patients, in-hospital mortality was 4.1% (129/3,168). NEWS showed higher discrimination for in-hospital mortality than the 2018 Chinese Emergency Triage grade (AUROC 0.750 [95% CI 0.706–0.794] vs. 0.584 [95% CI 0.538–0.629]; DeLong’s test P <  0.001). At the Youden-optimal threshold of NEWS ≥ 3, sensitivity was 68.2% and specificity was 71.6%. Discrimination was similar when NEWS was recalculated without the respiratory-rate component (AUROC 0.758 [95% CI 0.715–0.800]), and bootstrap internal validation showed negligible optimism. Higher NEWS was associated with increased mortality, and several vital signs showed non-linear associations with mortality. In this retrospective single-centre cohort of patients with confirmed high-risk acute cardiovascular and aortic emergencies, a retrospectively reconstructed, incomplete NEWS discriminated all-cause in-hospital mortality better than the 2018 Chinese Emergency Triage Standard. Whether NEWS adds prognostic value beyond triage grade requires prospective evaluation. Improved physiological data recording, prospective multicentre validation, and development and external validation of an appropriate NEWS-to-risk mapping are needed before clinical implementation.