Rossiter C E
Br J Ind Med. 1972 Oct;29(4):407-19. doi: 10.1136/oem.29.4.407.
407-419. The UICC/Cincinnati classification was developed in 1967 as an extension of the ILO 1958 classification, but it was decided not to publish it until evidence on the practicability and on the repeatability of readings using this classification was available. The scheme proved to be very practicable to use, with each reader able to read 300 to 500 films per day. There were only 0·3% omissions of any part of the complete reading in 42 000 readings. The results of early trials of the classification show that both inter- and intra-observer variability are very similar to those for the ILO 1958 scheme and also to those reported from many other studies of variability in medical fields. A representative selection of the results of these early trials is presented. Familiarity with the classification and the standard films improved repeatability considerably, and the differences between practised readers tended to be consistent.
407 - 419。国际抗癌联盟/辛辛那提分类法于1967年作为国际劳工组织1958年分类法的扩展而制定，但在获得关于使用该分类法进行读数的实用性和可重复性的证据之前，决定暂不公布。事实证明，该方案使用起来非常实用，每个读者每天能够阅读300至500张胸片。在42000次读数中，完整读数的任何部分遗漏率仅为0.3%。该分类法早期试验的结果表明，观察者间和观察者内的变异性与国际劳工组织1958年方案的变异性非常相似，也与许多其他医学领域变异性研究报告的结果相似。本文展示了这些早期试验结果的代表性样本。熟悉该分类法和标准胸片可显著提高可重复性，且经验丰富的读者之间的差异往往较为一致。
