SEARCH

SEARCH BY CITATION

Keywords:

  • cervical screening program;
  • human papillomavirus;
  • self-sampling;
  • nonresponders;
  • cervical cytology

Abstract

Women not attending cervical screening programs are at increased risk of cervical cancer. We investigated in these nonresponders to what extent offering self-sampling devices for cervicovaginal brushes for high-risk human papillomavirus (hrHPV) testing would induce participation and, if so, what the yield of precursor (i.e. CIN2 or worse) lesions following self-sampling would be. In addition, we assessed screening history of participants and costs per detected high-grade CIN2 or worse (“CIN2+”) lesion in comparison to the regular program in the Netherlands. Nonresponders received a device for hrHPV testing (self-sampling group, n = 2,546) or an extra recall for conventional cytology (control group, n = 284). The percentage of self-sampling responders were compared with responders in the recall group. hrHPV positive self-sampling responders were invited for cytology and colposcopy. CIN2+ yield and costs per detected CIN2+ were evaluated. Active response was higher in the self-sampling than in the control group (34.2 vs. 17.6%; p < 0.001). hrHPV positive self-sampling responders were less likely to have a prior screening history than screening participants (p < 0.001), indicating that they are regular nonresponders. hrHPV prevalence was similar (8.0 vs. 6.8%; p = 0.11), but CIN2+ yield was higher in self-sampling responders compared to screening participants (1.67 vs. 0.97%; OR = 2.93, 95% CI 1.48–5.80; p = 0.0013). Costs per CIN2+ lesion detected via self-sampling were in the same range as those calculated for conventional cytological screening (€8,836 vs. €7,599). Offering self-sampling for hrHPV testing in nonresponders is an attractive adjunct to effectively increase population coverage of screening without the adverse effect of markedly increased costs per detected CIN2+ lesion. © 2006 Wiley-Liss, Inc.