Objective: To examine the concordance between self-collected and clinician-collected samples for human papillomavirus (HPV) DNA.
Methods: Sexually active adolescent and young adult women aged 14-21 years (N = 101) were enrolled in a prospective cohort study of HPV testing. Participants self-collected vaginal samples for HPV DNA, and clinicians collected cervicovaginal samples for HPV DNA and a cervical cytology specimen. We determined concordance between the results of self- and clinician-collected specimens using a kappa statistic and McNemar's test.
Results: Of the 51% of participants who were HPV positive, 53% had 1 type, 25% had 2 types, and 22% had 3 types or more; 25 different HPV types were identified. Self-collected samples detected more participants with HPV than clinician-collected samples (45% versus 42%, P =.65). When results were categorized into presence or absence of high-risk HPV types, agreement between self- and clinician-collected specimens was high (kappa 0.72) and the difference between test results was not significant (McNemar's P =.41). However, when all HPV types detected were considered, agreement was perfect in only 51% of those with 1 or more types of high-risk HPV type. There was no association between agreement and age or HPV type.
Conclusion: Self testing for HPV DNA may be sufficiently sensitive for the detection of high-risk HPV DNA among adolescent and young adult women in clinical settings.