Interobserver agreement in assessment of chest radiographs for the diagnosis of pneumonia between residents and radiologists in a pediatric emergency department: The role of patient and resident characteristics

Abstract

Abstract Background Chest radiograph is often considered to be a standard procedure for diagnosis of pneumonia; however, its interpretation shows considerable variability among observers.Aims This study goal is to assess the extent of agreement between pediatric residents and board-certified radiologists in interpretation of chest films for detection of pneumonia and to evaluate the impact of resident’s experience, patient’s age and signs of infection on this phenomenon.Methods 935 patients with suspected pneumonia admitted to the pediatric emergency department at a non-tertiary medical hospital in Israel during 2019–2021 were included in this study. All patients had chest radiographs interpreted by a resident and a radiologist. Interobserver agreement was assessed using Kappa and prevalence-adjusted bias-adjusted kappa (PABAK) with 95% confidence intervals (CI). Results were stratified by resident experience (junior or senior), patient’s age (≤ 3 or > 3 years), white blood cells (≤ 15,000 or > 15,000 cells/mL), C-reactive protein (≤ 5 or > 5.0 mg/dL) and temperature (< 38.0°C or ≥ 38.0°C).Results Moderate agreement between pediatric residents and radiologists was demonstrated for diagnosis of pneumonia (kappa = 0.45). After adjustment for disease prevalence, the extent of agreement increased to near-substantial (PABAK = .59, 95% CI 0.54–0.64). The extent of agreement was higher for children over three years and in patients without clinical or biochemical features of pneumonia, especially when diagnosis of pneumonia was ruled out.Conclusions Second reading of chest radiographs by an experienced radiologist should be considered particularly for patients below age of three years and those with signs of infection and an initial diagnosis of pneumonia.