Detection of lung cancer on the chest radiograph: a study on observer performance

Study objectives: to study the validity and observers consistency in the de tection of lung cancer on the chest radiograph. Materials and methods: the chest radiographs of 100 clinical cases were interpreted by 14 observers. T he radiographs were obtained from 30 patients with initially missed but his to pathologically proven non-small cell lung cancer (NSCLC), 35 patients wi th other cardiopulmonary diseases and 35 patients with no abnormalities. Th e observers consisted of ten experienced radiologists, two-experienced ches t physicians and two residents in radiology. All observers were unaware of the study design. The validity and observer consistency was determined for each observer. Results: the mean sensitivity and specificity of the ten rad iologists were 0.36 and 0.90. For the two chest physicians, the mean sensit ivity and specificity were 0.29 and 0.96. For the two residents in radiolog y, mean sensitivity and specificity were 0.25 and 0.94. The mean interobser ver kappa and mean intraobserver kappa for the radiologists were 0.38 and 0 .54. For the two chest physicians, the mean interobserver kappa was 0.43, w hile the intraobserver kappa was 0.59. For the two residents in radiology, mean interobserver kappa was 0.35 and the intraobserver kappa was 0.42. The re was no significant relation between the consistency parameters and valid ity parameters. The interobserver and intraobserver kappa values showed goo d correlation. Conclusion: the validity of the chest radiograph and observe rs consistency in the detection of nodular lung cancer varies widely. The l evel of experience is likely to influence the diagnostic performance. (C) 2 001 Elsevier Science Ireland Ltd. All rights reserved.