Determinants of Incidence and Clearance of High-Risk Human Papillomavirus Infections in Rural Rakai, Uganda

Abstract

BACKGROUND: We used self-administered vaginal swabs to assess the incidence and clearance of carcinogenic human papillomavirus (HPV) infections in rural Rakai, Uganda. METHODS: Women provided self-administered vaginal swab at annual home-based visits. Type-specific carcinogenic HPV incidence and clearance and risk factors were assessed. RESULTS: Carcinogenic HPV incidence was 17.3 per 100 person-years among HIV-positive women compared with 7.0 per 100 person-years among HIV-negative women (P < 0.001). HPV-51 had the highest incidence followed by HPV-16 (1.8 per 100 and 1.5 per 100 person-years, respectively). In multivariate model, HIV-positive women were twice as likely to have incident infection compared with HIV-negative women. Younger women were at higher risk for incident infection, as were women with higher lifetime and recent sexual partners, and high perception of AIDS. Married women were less likely to have incident infection. Approximately half of all carcinogenic HPV infections cleared over the study follow-up of 3 years. HPV-31, HPV-35, and HPV-16 had the lowest clearance (16.7%, 27.9%, and 38.3%, respectively). In multivariate model, HIV-positive, women over 30 years with higher HPV viral, burden and more lifetime sex partners were less likely to clear infections. CONCLUSIONS: Self-collected vaginal swabs provide accurate HPV exposure assessment for studying HPV exposure and epidemiology and can be an important tool for research in populations unwilling to undergo pelvic exam.