巴氏涂片解读的实验室内部可重复性：一项实验的结果

Intralaboratory reproducibility of interpretation of Pap smears: results of an experiment.

作者信息

Duca P, Braga M, Chiappa L, Piffer R, Turolla E

机构信息

Istituto di Statistica Medica e Biometria, University di Milano, Italia.

出版信息

Tumori. 1988 Dec 31;74(6):737-44. doi: 10.1177/030089168807400620.

DOI:10.1177/030089168807400620

PMID:3232218

Abstract

One hundred and twenty Pap-smears were examined by 3 cytotechnologists on two occasions in the Legnano Hospital Cytologic Center. A 10-category classification system was used. Chance-adjusted overall and category-specific agreement was estimated (Cohen's unweighted kappa-statistics) within and between cytotechnologists. Mean K for the Center, weighted for precision, was 0.432, S.E. 0.029 (between cytotechnologists). The most reproducible categories (between cytotechnologists) were "9: malignant cells" (K = 0.683) and "1: normal" (K = 0.533); the least reproducible categories were "4: endocervical metaplastic cells" (K = 0.024) and "8: severe squamous dysplasia CIN III" (K = 0.227).

摘要

3名细胞技术专家在莱尼亚诺医院细胞学中心对120份巴氏涂片进行了两次检查。采用了一种10分类系统。在细胞技术专家内部和之间估计了机会调整后的总体和类别特异性一致性（科恩未加权kappa统计量）。该中心的平均K值（按精度加权）为0.432，标准误为0.029（细胞技术专家之间）。（细胞技术专家之间）最具可重复性的类别是“9：恶性细胞”（K = 0.683）和“1：正常”（K = 0.533）；最不可重复的类别是“4：宫颈内膜化生细胞”（K = 0.024）和“8：重度鳞状上皮发育异常CIN III”（K = 0.227）。