Diagnostic accuracy of self collected vaginal specimens for human papillomavirus compared to clinician collected human papillomavirus specimens: a meta-analysis.

BACKGROUND/OBJECTIVES Providing summary recommendations regarding self collection of vaginal specimens for human papillomavirus (HPV) testing is difficult owing to the wide range of published estimates for the diagnostic accuracy of this approach. To determine summary estimates from analyses of reported findings of the sensitivity, specificity and summary receiver operating characteristic curves (SROC) for self collected vaginal specimens for HPV testing compared to the reference standard, clinician collected HPV specimens. METHODS Standard search criteria for a diagnostic systematic review were employed. Eligible studies were combined using a random effects model and summary ROC curves were derived for overall and for specific subgroups. RESULTS Summary measures were determined from 12 studies. Six studies where patients used Dacron or cotton swabs or cytobrushes to obtain samples were pooled and had an overall sensitivity of 0.74 (95% CI 0.61 to 0.84) and specificity of 0.88 (95% CI 0.83 to 0.92), with diagnostic odds ratio of 22.3 and an area under the curve of 0.91. Self specimens using Dacron or cotton swabs or cytobrushes collected by women enrolled at referral clinics had an overall sensitivity of 0.81 (95% CI 0.65 to 0.91) and specificity of 0.90 (95% CI 0.80 to 0.95). Sensitivity and specificity of tampons ranged from 0.67-0.94 and 0.80-0.85 respectively. CONCLUSIONS Our findings indicate that the combined sensitivity for HPV-DNA is more than 70% when patients use Dacron swabs, cotton swabs, or cytobrushes to obtain their own vaginal specimens for HPV-DNA evaluation. Self collected HPV-DNA swabs may be an appropriate alternative for low resource settings or in patients reluctant to undergo pelvic examinations.