Interobserver variability in the radiographic diagnosis of adult outpatient pneumonia

Acta Radiol. 1992 Jan;33(1):79-81.


Acute chest radiographs were obtained from 319 adult patients with acute respiratory infections. Where a lower respiratory infection was diagnosed, follow-up chest radiographs were obtained in most patients. A radiologic panel diagnosed pneumonia in 21 patients. The agreements between the panel and 3 independent interpreters, 2 residents in radiology, and one senior chest physician, were assessed. Also the reports given by the specialist in radiology at the Department of Radiology were compared with the panel's evaluation. While the kappa-agreements between the panel's interpretations and those by the Department of Radiology and the consultant in chest medicine was 0.71 and 0.72, respectively, the corresponding kappa-values between the residents and the panel was only 0.50. The proportion of agreement when pneumonia was diagnosed was 0.56 between the panel and the Department of Radiology, and 0.59 between the panel and the chest consultant, compared to 0.36 between the panel and the residents. The study demonstrates the difficulty of diagnosing outpatient pneumonia and the importance of experience.

MeSH terms

  • Adolescent
  • Adult
  • Aged
  • Ambulatory Care
  • Data Interpretation, Statistical
  • Female
  • Humans
  • Male
  • Middle Aged
  • Observer Variation
  • Pneumonia / diagnostic imaging*
  • Radiography