Tehran University of Medical Sciences Frontiers in Emergency Medicine 2717-3593 5 4 2021 03 28 e40 e40 Alireza Jalali Prehospital and Hospital Emergency Research Center, Tehran University of Medical Sciences, Tehran, Iran Ehsan Karimialavijeh Prehospital and Hospital Emergency Research Center, Tehran University of Medical Sciences, Tehran, Iran Parto Babaniamansour Department of Biomedical Engineering, University of Kentucky, Lexington, Kentucky, USA Ehsan Aliniagerdroudbari School of Medicine, Shahid Beheshti University of Medical Sciences, Tehran, Iran Sepideh Babaniamansour School of Medicine, Islamic Azad University Tehran Faculty of Medicine, Tehran, Iran. 2021 02 03 2021 03 07 Introduction: Coronavirus Disease (COVID‐19) has become the most important global health issue, and chest computed tomography (CT) scan can help determine the severity of the infection. Objectives: This study aimed to provide an emergency scoring tool for predicting 30-day adverse outcomes in non-critical new-onset COVID-19 patients. Methods: This derivation study was conducted on new-onset COVID-19 patients presenting to the emergency department of an urban teaching hospital in Tehran, Iran, between 20 February and 20 March 2020. The total lobe severity score (TSS), age, history of comorbidities, and 30-day adverse outcomes (death, ICU admission or intubation) were taken into account to produce three prediction models. Results: Overall, 137 patients were included in the study. Their mean age was 59.9±16.8 years and 62% were male. The ground glass nodule, patch B/punctate ground-glass opacity, fibrous stripes, and air bronchogram sign with perihilar distribution, bilateral and ≥ 2 affected lobes were the most common findings. The mean TSS (model 1) was significantly higher in patients with an adverse outcome (9.4±3.2) compared to the discharged patients (7.2±3.3) (p<0.001, AUC: 0.703, sensitivity: 64.4% and specificity: 74.1%). The optimal cut-off point of model 2 (TSS and age) had the following parameters: AUC: 0.721, sensitivity: 71.2% and specificity: 67.2%. The optimal cut-off point of model 3 (TSS, age, comorbidities) had: AUC: 0.755, sensitivity: 79.7% and specificity: 65.5%. The discrimination achieved with model 3 based on Bonferroni’s test was significantly better than that achieved with TSS (p<0.001). Conclusion: TSS combined with age and history of at least one comorbidity had a better predictive value for adverse outcomes with a cut-off point above 8. https://fem.tums.ac.ir/index.php/fem/article/view/662 https://fem.tums.ac.ir/index.php/fem/article/download/662/351