Premium
Proffered Papers
10.30–11.15 Monday 15 September 2003 4 An audit of the positive predictive value of high‐grade dyskaryosis for CIN 2 or worse on histology
Author(s) -
Wadehra V.,
Johnson S. J.
Publication year - 2003
Publication title -
cytopathology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.512
H-Index - 48
eISSN - 1365-2303
pISSN - 0956-5507
DOI - 10.1046/j.1365-2303.14.s1.1_9.x
Subject(s) - colposcopy , medicine , cytology , predictive value , cervical screening , biopsy , gynecology , obstetrics , radiology , pathology , cervical cancer , cancer
The positive predictive value (PPV) of high‐grade dyskaryosis for CIN2 or worse on histology is published annually for the laboratories in the UK National Health Service Cervical Screening Programme (NHSCSP). The laboratories collect this data every year for a 3‐month period from April to June as part of the KC61 return. The laboratory PPV is expected to fall between 65% to 85% and provides an inter‐laboratory comparison figure for the specificity of the laboratory reporting. The PPV of individual consultants reporting the smears is monitored internally. The RVI the PPV fell from 84.9% in 2000–2001 to 78.71% in 2001–2002. PPV for individual consultants showed a drop for four of the five consultants, including a fall from 91.6% to 77.94%. The reasons for the fall were investigated by pursuing data rechecking, accuracy of reporting, cyto‐histological correlation, histology and colposcopy factors and follow‐up cytology. The main cause of the fall lay outside the laboratory in the type of biopsy taken at colposcopy. We conclude that the type of biopsy taken at colposcopy influences the PPV calculation. Informal discussions with others suggest comparison between laboratories is not reliable as departments vary in how the KC61 data is collected. As the type of data effects the correlation results, we conclude that caution must be exercised in inter‐laboratory comparisons until more standardization is introduced in histology data collection for PPV for the KC61. It is suggested that other laboratories view their PPV data with our findings in mind.