Renshaw A A, Lee K R, Granter S R
Department of Pathology, Brigham & Women's Hospital and Harvard Medical School, Boston, Massachusetts 02115, USA.
Cancer. 1997 Aug 25;81(4):212-9.
Disagreements in cytologic interpretation can have several causes, including differences in diagnostic threshold and diagnostic accuracy. These can be distinguished by a combination of statistical analyses.
For demonstration purposes, a nonrandom collection of 80 cervicovaginal smears, the majority of which (74) were originally diagnosed as atypical cells of undetermined significance (ASCUS), were reviewed by 3 separate observers and classified as either negative, negative and reactive, ASCUS favor reactive, ASCUS not otherwise specified, ASCUS suggestive of a squamous intraepithelial lesion (SIL), low grade SIL, or high grade SIL. The results were compared with corresponding biopsies and analyzed with distribution analysis, the kappa statistic, threshold analysis, and receiver operating characteristic (ROC) curve analysis.
Distribution analysis of diagnoses from the three observers demonstrated statistically significant differences in how cases were classified and a low level of agreement. Kappa analysis confirmed a very poor interobserver agreement. Threshold analysis revealed that one observer used a threshold between negative and ASCUS that was statistically more specific but less sensitive than the other observers. ROC curve analysis showed that another observer was more accurate than this observer.
Variation in cytologic interpretation may have several causes. Distribution, threshold, and ROC analysis allow distinction between differences in diagnostic accuracy and diagnostic thresholds. This approach to analyzing cytologic interpretation may be useful for quality improvement efforts.
细胞学诊断意见不一致可能有多种原因，包括诊断阈值和诊断准确性的差异。这些可以通过多种统计分析方法加以区分。
为便于演示，收集了80份宫颈阴道涂片（非随机收集），其中大部分（74份）最初被诊断为意义不明确的非典型细胞（ASCUS），由3名独立观察者进行复查，并分类为阴性、阴性且有反应性、倾向反应性的ASCUS、未另行指定的ASCUS、提示鳞状上皮内病变（SIL）的ASCUS、低级别SIL或高级别SIL。将结果与相应活检结果进行比较，并采用分布分析、kappa统计、阈值分析和受试者工作特征（ROC）曲线分析。
对三名观察者的诊断结果进行分布分析，结果显示病例分类存在统计学上的显著差异，一致性较低。kappa分析证实观察者间的一致性非常差。阈值分析显示，一名观察者使用的阴性与ASCUS之间的阈值在统计学上比其他观察者更具特异性，但敏感性更低。ROC曲线分析表明，另一名观察者比该观察者更准确。
细胞学诊断的差异可能有多种原因。分布分析、阈值分析和ROC分析有助于区分诊断准确性和诊断阈值的差异。这种分析细胞学诊断的方法可能有助于质量改进工作。
