Objectives: To analyze the effect of age on chest computed tomography (CT) features and disease severity in COVID-19.Materials and Methods: The study included the chest CT images of 367 patients with COVID-19 between April 1, 2020, and June 1, 2020. The patients were divided into four age groups: Group A (≤19 years), group B (20-39 years), Group C (40-59 years), and Group D (≥60 years). On chest CT, laterality, involved lobes, number of lesions, lesion types, and lesion distribution were evaluated and compared.Results: The study population consisted of 367 patients (210 male and 157 female, mean age 47.50 years, range 12-92 years). There was no significant gender difference between the age groups. The most common CT finding in all age groups was ground-glass opacity. Consolidations, crazy-paving patterns, and air bronchograms were more common in elderly patients, in which the number of lesions was also higher. The upper lobes and the right middle lobe were affected more frequently in elderly patients. While no mortality was found in groups A and B, the rates of intensive care admission and mortality were higher in groups C and D. The ROC curve analysis revealed that 55 years of age was the optimal cut-off value to predict a worse outcome.Conclusion: The CT findings of COVID-19 patients vary in different age groups. Bilateral, multiple and diffuse infiltrations, consolidation, air bronchogram, and crazy-paving patterns, upper lobe involvement, and older age should be considered as an indicator for worse outcomes.