The 60-film set was developed by experts (expert group) for examining 8 indices (sensitivity and specificity for pneumoconiosis, sensitivity and specificity for large opacities, sensitivity and specificity for pleural plaque, profusion increment consistency for small opacities, and shape differentiation for small opacities) of physicians’ reading skills on pneumoconiosis X-ray according to ILO 2000 Classification. The aims were firstly to examine differences in 8-index data between groups, secondly to investigate the factor structure of 8-index data, and thirdly to examine differences in factor scores between groups.
Methods
53 physicians (A-Group) attended the “Asian Intensive Reader of Pneumoconioses” training course, and 22 physicians (B-Group) attended Brazilian course. After training, they took examination of reading 60-film exam set. The 8-index data were compared between the two examinee groups and the expert group by t-test. The 8-index data of the pooled group from two groups were subjected to exploratory factor analysis. Then scores of factors were examined in groups for comparison.
Results
There were significant differences in sensitivity for pneumoconiosis and sensitivity for large opacities at p<0.05 between A-Group and B-Group. There was a significant difference in sensitivity for large opacities at p<0.05 between pooled group and B-Group. The examinee groups were inferior to the expert group in all of indices. A 4-factor was analyzed to structure 8-index: The specificity for pneumoconiosis, specificity for large opacities, specificity for pleural plaque and shape differentiation for small opacities loaded on Factor 1; The sensitivity for pneumoconiosis and sensitivity for large opacities loaded on Factor 2; The sensitivity for pleural plaque loaded on Factor 3; The profusion increment consistency loaded on Factor 4. Factor 2 score in A-Group were significantly higher than in B-Group.
Discussion
Accordingly, the 60-film set providing 8-index and 4 factor scores structuring 8-index was suggested appropriately to assess physicians’ reading proficiency at different training settings.