Human papillomavirus detection in self-collected vaginal specimens and matched clinician-collected cervical specimens

Abstract

Human papillomavirus (HPV) detection is an integral part of cervical cancer screening, and a range of specimen collection procedures are being tested. Preliminary studies have found that the majority of women prefer self-collection of vaginal specimens instead of clinician-collected specimens of the cervix. The purposes of the current study were to explore the social and behavioral predictors of acceptance of self-collection of vaginal specimens among patients and to assess concordance in detection of HPV between clinician-collected cervical specimens and self-collected vaginal specimens. The study was conducted at a university family medicine clinic using a cross-sectional study design, and enrollment of women presenting for routine gynecological examination consecutively in a period of 1 year, self-administered questionnaires, collection of paired vaginal and cervical specimens for HPV DNA using Hybrid Capture 2, and cytologic analysis. Most women (79.8% [398/499]) agreed to collect vaginal specimens. In our study, 76.6% (216/282) African American women (AA), 88.1% (156/176) white non-Hispanic (WNH) women, and 63.4% (26/41) women of other races (P < 0.0001) agreed to self-collect vaginal specimens. HPV was detected in 16.0% (80/499) of clinician-collected cervical specimens and 26.1% (104/398) of self-collected vaginal specimens (P < 0.001). HPV detection was concordant in 13.4% (53/398) women in both cervical and vaginal specimens. Self-collection of vaginal specimens for HPV DNA detection is acceptable to most women presenting for routine gynecological examination. WNH women were more likely to obtain self-collected specimens than AA women. Vaginal specimens were more likely to be positive for HPV than were cervical specimens.