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

Area of Science:

• Radiology
• Infectious Diseases
• Medical Imaging Analysis

Background:

• Chest CT scans are crucial for assessing coronavirus disease 2019 (COVID-19) severity and guiding treatment.
• A 25-point scoring system is used, but interobserver variability may affect patient categorization.
• This study evaluated the reliability of the 25-point scoring system by assessing interobserver variability.

Purpose of the Study:

• To determine the interobserver variability of the 25-point scoring system for COVID-19 chest CT scans.
• To assess the reliability of the scoring system in categorizing disease severity.
• To confirm the utility of the scoring system for clinical decision-making.

Main Methods:

• Retrospective analysis of chest CT scans from 100 COVID-19 positive patients.
• Three experienced thoracic radiologists independently scored the CT images using the 25-point system.
• Radiologists' scores were compared to evaluate interobserver agreement and variability.

Main Results:

• No significant interobserver variability was found, indicating high reliability of the scoring system.
• The standard deviation between observers was less than three, with near-perfect agreement (Fleiss' Kappa = 0.99).
• Minor variations were noted primarily within the moderate severity class.

Conclusions:

• The 25-point scoring system demonstrates minimal interobserver variability, confirming its reliability for COVID-19 severity assessment.
• Consistent categorization of patient severity is achievable, supporting its use in clinical practice.
• This reliable scoring system aids clinicians in treatment planning and improving patient prognosis.