Interobserver Variability in CT Severity Scoring System in COVID-19 Positive Patients

Background: Chest CT scans are done in cases of coronavirus disease 2019 (COVID-19)-positive patients to understand the severity of the disease and plan treatment accordingly. Severity is determined according to a 25-point scoring system, however, there could be interobserver variability in using this scoring system thus leading to the different categorization of patients. We tried to look for this interobserver variability and thus find out its reliability. Methods: The study was retrospective and was done in a designated COVID center. Some 100 patients were involved in the study who tested positive for COVID-19 disease. The research was conducted over six months (January 2021 to June 2021). Images were given to three radiologists with a minimum of 10 years of experience in thoracic imaging working in different setups at different places for interpretation and scoring further and their scores were compared. Before the study, the local ethics committee granted its approval. Results: There was no significant variability in the interobserver scoring system thus proving its reliability. The standard deviation between different observers was less than three. There was almost perfect agreement amongst all the observers (Fleiss’ K=0.99 [95% confidence interval, CI: 0.995-0.998]). Maximum variations were observed in the moderate class. Conclusion: There was minimum inter-observer variability in the 25-point scoring system thus proving its reliability in categorizing patients according to severity. There was no change in the class of the patient according to its severity. A 25-point scoring system hence can be used by clinicians to plan treatment and thus improve a patient's prognosis.