Abstract
INTRODUCTION: The National Early Warning Score (NEWS) is commonly used to identify patients at high mortality risk. However, it has notable limitations. In this study, to enhance the accuracy, we revised it and evaluated the performance of modified NEWS (MNEWS) in predicting the outcomes of suspected sepsis patients. METHODS: This single-center, prospective cohort study was conducted on patients with suspected sepsis to evaluate the accuracy of MNEWS in predicting mortality, survival to discharge, vasopressor requirements, and the need for mechanical ventilation. The MNEWS comprises the NEWS variables plus age, chronic major organ dysfunction, malignancy, functional status, and specific infected organ involvement. Sensitivity, specificity, likelihood ratio (LR), and area under the receiver operating characteristic curve (AUROC) were used to evaluate the performance of the MNEWS in predicting the studied outcomes. RESULTS: Of the 1,393 patients included in this study, 209 died. Mean MNEWS was significantly higher in non-survivors than survivors (19.8 vs. 14.9, p<0.001). The AUROC of MNEWS in predicting 30-day mortality was 0.82 (95% CI: 0.79-0.85). MNEWS ≥ 18 had the highest accuracy for 30-day mortality prediction with 76.1% sensitivity, 75% specificity, positive LR of 3.13, and AUROC of 0.76 (95% CI: 0.73-0.79). The AUROC of MNEWS ≥18 for predicting survival until discharge, need for vasopressors, and need for mechanical ventilation were 0.75 (95% CI: 0.72-0.78), 0.72 (95% CI: 0.69-0.75), and 0.76 (95% CI: 0.73-0.79), respectively. Additionally, MNEWS ≥18 demonstrated superior predictive performance, compared with NEWS ≥7 and qSOFA ≥2 for various clinical outcomes. CONCLUSIONS: The MNEWS was similar to the NEWS in overall predictive accuracy for 30-day mortality but exhibited a higher predictive accuracy than did the qSOFA score. Notably, MNEWS ≥18 was a significant indicator of 30-day mortality risk, as well as the likelihood of requiring vasopressors, survival to discharge, and 7-day mortality.