Accuracy of Pap tests reported as CINI

Papanicolaou (Pap) tests reported as CIN I (cervical intraepithelial neopla sia, grade 1) may be subject to laboratory misclassification because of scr eening and interpretative errors. A peer-group consensus review was conduct ed to measure the misclassification rate of Pap rests reported as CIN I and to analyze the undercalled and overcalled tests for due cause. Four hundre d and forty-nine Pap rests originally reported as CIN I were independently reviewed twice by a panel of four pathologists, and disagreements were reso lved by consensus review. Results were based on the original screening for the first review and following the removal of those markings, were based on a second independent rescreening for the second review. A review result of low-grade squamous intraepithelial lesion (LSIL) and atypical squamous cel ls of undetermined significance (ASCUS)Savoring LSIL was equated with the o riginal CIN I result. Final classification was based on the second consensu s review. Misclassified tests were categorized as screening or interpretati ve errors, based on a comparison of the review classifications. LSIL and AS CUS Savoring LSIL were repel-led in 85.1% and 73.9% of the first and second reviews, respectively. In the final classification there were 362 (80.6%) LSIL and ASCUS-LSIL and 87 (19.4%) misclassifications: 31 (6.9%) undercalls and 56 (12.5%) overcalls. Screening error accounted for 35.5% of undercall ed tests, and the remainder were interpretative errors, as were all those o vercalled. In this study Pap tests reported as CIN I were subject to miscla ssification because of a laboratory error in 19.4% of tests. Reductions in screening and interpretative errors were identified as mechanisms for impro ving accuracy. Diagn. Cytopathol, 1999;21:129-136. (C) 1999 Wiley-Liss, Inc .