Concordance of Self- and Clinician-Collected Anal Swabs to Detect Human Papillomavirus in a Sample of HIV-Negative Men

Objective The aim of this study was to determine the concordance of self- and clinician-collected anorectal swabs for the detection of human papillomavirus (HPV) DNA in a population of HIV-negative men who have sex with men (MSM). Methods This cross-sectional study involved recruitment of HIV-negative MSM in a Midwestern US metropolitan area to collect paired sequential self- and clinician-collected anorectal swabs using illustrated instructions. Swabs were tested for type-specific HPV DNA with a comparison of type-specific HPV categories detected by each method. The sensitivity and specificity of self-collection were calculated assuming clinician collection as the criterion standard. McNemar's test and κ statistics were used to determine percent agreement and concordance of self- and clinician-collected swab results. Results Seventy-eight participants had paired anorectal swab samples of adequate quality for analyses. The sensitivity and specificity of self-collected swabs for detection of all high-risk HPV DNA types were 69.8% and 91.4%, respectively. Similar degrees of sensitivity and specificity of self-collection were seen for other groups of high-risk HPV types. Percent agreement and κ statistic for self- and clinician-collected swabs for all high-risk HPV types were 80.8% and 0.53, respectively. Conclusions Self-collected anorectal swab samples showed lower sensitivity but moderate to high specificity for detection of high-risk and vaccine-preventable HPV types compared with clinician-collected swab samples. Self-collection instructional details and the thoroughness of clinician collection of samples may have impacted sensitivity and specificity, suggesting a need to optimize and standardize instructions.