Abstract
The outbreak and continuing impact of COVID-19 have significantly increased the rates of hospitalization and admissions to intensive care units (ICU). This study evaluates clinical outcomes in critically ill patients and investigates variables tied to poor prognosis. A secondary database analysis was conducted to investigate the predictors of poor outcome among critically ill COVID-19 patients in Saudi Arabia. Multivariable logistic regression analysis was used to assess the association between various demographic characteristics, comorbidities, and COVID-19 symptoms and patients' poor prognosis, as a composite outcome. A total of 2257 critically ill patients were identified (male (71.8%), and elderly (37.3%)). The mortality rate was 50.0%, and the composite poor outcome was 68.4%. The predictors of poor outcome were being elderly (OR = 4.79, 95%CI 3.19-7.18), obesity (OR = 1.43, 95%CI 1.1-1.87), having a severe or critical case at admission (OR = 6.46, 95%CI 2.34-17.8; OR = 22.3, 95%CI 11.0-45, respectively), and some signs and symptoms of COVID-19 such as shortness of breath, feeling fatigued or headache, respiratory rate >= 30/min, PaO2/FiO(2) ratio < 300, and altered consciousness. In conclusion, identifying high-risk populations that are expected to have a poor prognosis based on their criteria upon admission helps policymakers and practitioners better triage patients when faced with limited healthcare resources.