P A de Ruíz, E C Lazcano Ponce, R Duarte Torres, I Ruíz Juárez, I Martínez Cortez
{"title":"Diagnostic reproducibility of Pap testing in two regions of Mexico: the need for quality control mechanisms.","authors":"P A de Ruíz, E C Lazcano Ponce, R Duarte Torres, I Ruíz Juárez, I Martínez Cortez","doi":"","DOIUrl":null,"url":null,"abstract":"<p><p>To assess the reproducibility of diagnostic results obtained by examining Pap smears for cervical neoplasia, a study was conducted using a single group of 20 Pap smears, 3 negative and 17 from patients with varying degrees of neoplasia. These smears were examined by 14 volunteer readers (13 cytotechnologists and 1 cytopathologist) from the Mexican states of Oaxaca and Veracruz, and also by a highly experienced cytopathologist certified by the Mexican Board of Pathological Anatomy whose work provided a reference standard. Individual variability, as assessed by the Kappa coefficient of concordance, showed considerable difference in the diagnostic results obtained by different readers-the degree of agreement depending on the type of cervical lesion involved and the number of specimens from patients with that type of lesion. There was little diagnostic agreement when the specimens were assessed for particular classes of cervical neoplasia-mild, moderate, or severe neoplasia, carcinoma in situ, or invasive cervical cancer. (The greatest concordance was found in diagnosing specimens from subjects with invasive cervical cancer.) However, when the diagnosis was assessed continuously, using Kappa weighted in accordance with the five possible diagnoses of cervical neoplasia, the apparent reproducibility of the diagnoses improved greatly, Kappa coefficients for the 14 readers ranging from 0.31 to 0.72. In general, these data support the view that there is a need in Mexico and other parts of the Americas to establish quality control mechanisms monitoring cytologic diagnosis of cervical neoplasia, to standardize diagnostic nomenclature using a system such as the Bethesda System, to institute periodic certification, and to provide continuing training. As this suggests, it is necessary not only to evaluate but also to bring about organizational changes in order to expeditiously prevent or correct the problems that currently constrain achievement of efficient and effective cytologic diagnosis.</p>","PeriodicalId":75654,"journal":{"name":"Bulletin of the Pan American Health Organization","volume":"30 4","pages":"330-8"},"PeriodicalIF":0.0000,"publicationDate":"1996-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Bulletin of the Pan American Health Organization","FirstCategoryId":"1085","ListUrlMain":"","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
To assess the reproducibility of diagnostic results obtained by examining Pap smears for cervical neoplasia, a study was conducted using a single group of 20 Pap smears, 3 negative and 17 from patients with varying degrees of neoplasia. These smears were examined by 14 volunteer readers (13 cytotechnologists and 1 cytopathologist) from the Mexican states of Oaxaca and Veracruz, and also by a highly experienced cytopathologist certified by the Mexican Board of Pathological Anatomy whose work provided a reference standard. Individual variability, as assessed by the Kappa coefficient of concordance, showed considerable difference in the diagnostic results obtained by different readers-the degree of agreement depending on the type of cervical lesion involved and the number of specimens from patients with that type of lesion. There was little diagnostic agreement when the specimens were assessed for particular classes of cervical neoplasia-mild, moderate, or severe neoplasia, carcinoma in situ, or invasive cervical cancer. (The greatest concordance was found in diagnosing specimens from subjects with invasive cervical cancer.) However, when the diagnosis was assessed continuously, using Kappa weighted in accordance with the five possible diagnoses of cervical neoplasia, the apparent reproducibility of the diagnoses improved greatly, Kappa coefficients for the 14 readers ranging from 0.31 to 0.72. In general, these data support the view that there is a need in Mexico and other parts of the Americas to establish quality control mechanisms monitoring cytologic diagnosis of cervical neoplasia, to standardize diagnostic nomenclature using a system such as the Bethesda System, to institute periodic certification, and to provide continuing training. As this suggests, it is necessary not only to evaluate but also to bring about organizational changes in order to expeditiously prevent or correct the problems that currently constrain achievement of efficient and effective cytologic diagnosis.