Agreement between Observers for Pap smear and Visual Inspection Tests in Florianópolis, Southern Brazil

Aim: Despite well established validity and reliability of Pap test, a significant proportion of false negative test results still remains and is partially due to the lack of agreement between observers. Visual inspection tests (VIT) share this difficulty but their results are immediately available, thus making it easier to reduce disagreement between observers if these are well trained. The objective of this paper was to verify the agreement on a typical squamous cells of undetermined significance (ASC-US) or higher grade lesions between different examiners with Pap smear and visual inspection tests (VIT) in screening for cervical cancer. It is part of a wider study to evaluate the screening performance of Pap test and VIT in Brazil.

Study Design: A cross-sectional study was conducted among consenting clients attending cervical cancer screening services at a non-governmental medical facility in southern Brazil.

Place and Duration of Study: “Rede Feminina de Combate ao Câncer”, a female cancer prevention center in Florianópolis, southern Brazil, between June 2010 and July 2012.

Methodology: Two observers evaluated 353 Pap smear test specimen and 284 visual inspection tests (VIT) of the women screened for cervical cancer in female cancer prevention center in southern Brazil . VIT started with naked eye examination after visual inspection by applying acetic acid (VIA) and Lugol’s iodine (VILI) to cervix, followed by additional examination with 2x magnifying glass for both modalities .

Results: Concordance between the first and the second observer regarding Pap test was reasonably good (kappa=0.67) but was even better regarding all VIT modalities (kappa range 0.76-0.83 ). Magnifying glass did not significantly improve the agreement.

Conclusion: VIT showed very good agreement between observers . Adding VIT to Pap smear test may improve diagnostic accuracy in cervical cancer screening .