Safe thresholds for hybrid capture 2 test in primary cervical screening

BMJ 2011; 342 doi: https://doi.org/10.1136/bmj.d2941 (Published 23 May 2011) Cite this as: BMJ 2011;342:d2941
  1. Peter Sasieni, professor of biostatistics and cancer epidemiology,
  2. Alejandra Castanon, epidemiologist
  1. 1Centre for Cancer Prevention, Wolfson Institute of Preventive Medicine, Queen Mary University of London, London EC1M 6BQ, UK
  1. p.sasieni{at}qmul.ac.uk

Higher cut offs could be used in some circumstances, but equivocal values should not be ignored

Science Photo Library

Substantial evidence shows that testing for human papillomavirus (HPV) DNA with hybrid capture 2 is more sensitive but less specific than cytology at detecting high grade cervical intraepithelial neoplasia.1 The sensitivity of hybrid capture 2 relative to cytology depends on the quality of the cytology, but its absolute sensitivity is uniformly high (about 96% overall and >90% in most studies).2 Its specificity depends on the level of HPV infection in the population. Typically 6-10% of women aged 30-64 will test positive.

For many years studies have looked at whether increasing the threshold for positivity would make the test more useful by trading a small loss in sensitivity for a large reduction in the clinical false positive rate. In the linked systematic review (doi:10.1136/bmj.d2757), Rebolj and colleagues assess randomised controlled trials that investigated the effect of different cut-off points on the sensitivity and specificity of the hybrid capture 2 test.3 The authors suggest that the standard cut off of 1 relative light units/cut off (rlu/co) should be changed to 10 rlu/co.

Although the data show that the loss of sensitivity associated with increasing the cut-off point is small …

View Full Text

Log in

Log in through your institution


* For online subscription