Abstract
BACKGROUND: Early identification of clinical deterioration in respiratory patients is essential for timely intervention and ICU admission in emergency departments. The National Early Warning Score (NEWS) and the Modified Early Warning Score (MEWS) are widely used and well-established tools for this purpose. However, there is limited comparative evidence regarding their predictive power in respiratory patients. Therefore, this study aimed to compare the predictive power of NEWS and MEWS for ICU admission in respiratory patients presenting to the emergency departments of four tertiary hospitals in Tehran during the COVID-19 pandemic. METHODS: A Multicenter descriptive cross-sectional study was conducted using 755 medical records of respiratory patients admitted to the emergency departments of four tertiary hospitals in Tehran (Amir A’lam, Sina, Baharloo, and Shariati) between March 2020 and February 2024. Medical records were selected using a consecutive and convenience sampling method. Data were collected using a demographic information form, the National Early Warning Score (NEWS) form, and the Modified Early Warning Score (MEWS) form. Statistical analyses were performed using SPSS (version 16) and MedCalc (version 3.5.19), employing descriptive and inferential methods. RESULTS: Median NEWS and MEWS scores were significantly higher in ICU-admitted patients than in non-ICU patients (P < 0.001). Specifically, the optimal cutoff value for NEWS was determined to be 9, which resulted in a sensitivity of 60.9% and specificity of 82.2%. Similarly, for MEWS, the optimal cutoff point was identified as 6, yielding a sensitivity of 61.4% and specificity of 76.5%. Notably, the AUC for NEWS (0.761; 95% CI: 0.722–0.799) was significantly higher than that for MEWS (0.729; 95% CI: 0.688–0.770; P = 0.002). CONCLUSION: Both NEWS and MEWS appeared to be effective in predicting ICU admission among respiratory patients. However, NEWS demonstrated slightly superior predictive power. These findings indicate that the integration of NEWS in the emergency triage process may result in early detection of high-risk respiratory patients, leading to improved critical care resource allocation and patient outcomes. Future research is warranted to confirm the generalizability and predictive applicability of NEWS for ICU admission across different clinical contexts. ETHICAL CONSIDERATIONS: The code of ethics (IR.TUMS.FNM.REC.1403.041) was obtained from the joint ethics committee of the Faculty of Nursing and Midwifery and the Faculty of Rehabilitation, Tehran University of Medical Sciences. Confidentiality of patient information and anonymity were also observed throughout the research process. CLINICAL TRIAL NUMBER: Not applicable.