Rossiter C E, Browne K, Gilson J C
TUC Centenary Institute of Occupational Health, London School of Hygiene and Tropical Medicine, UK.
Br J Ind Med. 1988 Aug;45(8):538-43. doi: 10.1136/oem.45.8.538.
A series of 100 full size radiographs illustrating the range of asbestos related radiographic changes was collected by the Asbestos International Association to provide a demonstration and teaching supplement to the ILO 1980 International Classification of the Radiographic Appearances of Pneumoconioses. Each film was read by 12 experienced readers from ten countries; the readings have been summarised by a median reading, with the range covered by two thirds of the readers. The occupational histories and some relevant clinical information are also available. It is proposed that, in the use of this set, readers should classify each film using the ILO classification rules, particularly without reference to the summary reading or the additional information. Comparing the individual readings by the 12 readers, most of the variation in reading the profusion of small opacities was ascribable only to random variation, with little consistent bias between readers. By contrast, two readers recorded pleural changes much more frequently than their colleagues. This illustrated a major problem with the ILO 1980 Classification. Several others which occurred in this reading trial are also discussed.
石棉国际协会收集了一系列100张全尺寸X光片,展示与石棉相关的X光变化范围,以作为国际劳工组织1980年《尘肺病放射影像国际分类》的演示和教学补充资料。每张片子由来自十个国家的12位经验丰富的阅片者阅读;阅片结果以中位数阅片总结,范围涵盖三分之二阅片者的判断。职业病史及一些相关临床信息也可供参考。建议在使用这套资料时,阅片者应依据国际劳工组织的分类规则对每张片子进行分类,尤其不要参考总结阅片结果或附加信息。比较12位阅片者的个人阅片结果,阅读小阴影密集度时的多数差异仅归因于随机变化,阅片者之间几乎没有一致的偏差。相比之下,两位阅片者记录胸膜变化的频率比其他同事高得多。这说明了国际劳工组织1980年分类法存在的一个主要问题。本文还讨论了此次阅片试验中出现的其他几个问题。