Competing Risk Survival Analysis of Time to In‐Hospital Recovery or Death Among COVID‐19 Patients: A Hospital‐Based Multicenter Retrospective Cohort Study in Southeast Ethiopia

ABSTRACT Background and Aims In‐hospital death and recovery are competing risks in COVID‐19 patients, complicating prognosis. While international prognostic scores exist, their reliance on complex biomarkers limits their utility in resource‐constrained settings. This study aimed to estimate the duration and identify determinants of in‐hospital outcomes in southeast Ethiopia and to develop a clinically accessible risk stratification tool. Methods Data from 827 confirmed COVID‐19 patients (October 2022–May 2023) across six treatment centers were analyzed using the Fine‐Gray Competing Risk Survival Analysis (CRSA). Additionally, a novel “Asella COVID‐19 Risk Score” was developed using readily available bedside clinical parameters (age, comorbidity, and antibiotic use) and validated using receiver operating characteristic analysis. Results Overall, 139/827 (17%) died, and 516/827 (62%) recovered. Risk of death was significantly higher for patients aged ≥ 50 years (acsHR = 2.62; 95%CI: 1.29, 5.29; p < 0.001) and those with an immunocompromised state (acsHR= 1.46; 95%CI: 1.08, 1.98; p = 0.014). Median time to death was 5 days. The “Asella COVID‐19 Risk Score” demonstrated an area under the curve of 0.65. At an optimal cutoff of ≥ 4, the score achieved a sensitivity of 47.2%, specificity of 73.2%, and a high negative predictive value (NPV) of 83.5%. Conclusion Advanced age and immunocompromised status significantly increase mortality risk. The Asella COVID‐19 Risk Score provides a clinically validated, high‐NPV triage tool suitable for resource‐limited settings, facilitating the early identification of high‐risk patients and more efficient resource allocation.