Chimbatata, Nathaniel Weluzani Banda2010-01-292010-01-292010-01-29http://hdl.handle.net/10539/7487Thesis (M.Sc.(Med.)(Epidemiology and Biostatistics)), Faculty of Health Sciences, University of the Witwatersrand,2009BACKGROUND Studies suggest an association between Bacterial Vaginosis (BV) and HIV infection; however, its temporal effect has not been greatly investigated. METHODS This is a secondary data analysis of a cohort study: set out to describe the association between BV infection and HIV acquisition. There were 750 participants enrolled in the primary cohort study. The main exposure, BV, was measured from a gram stain slide prepared from a vaginal swab. The slide was read in a laboratory qualitatively and scored by Nugents scoring. A score of 7 or above was considered positive for BV. The outcome variable (HIV) was determined by dual rapid tests and confirmed in the laboratory by a third generation ELISA. Descriptive statistics was done to describe demographic characteristics and the prevalence of BV and STIs. HIV incidence rate was calculated. Kaplan Meier survival time analysis and log rank test for significance were performed. Cox regression (univariate and multivariate) was done to determine association of BV with HIV infection. RESULTS The baseline prevalence of BV was 52 %, 95 % CI; 45 – 59. There were 21 HIV seroconversions experienced of which 7 had BV results missing and were excluded in the analysis. The remaining 14 seroconversions were followed for a mean time of 0.40 of a year and accumulated follow up time at risk of 286 person years, this represented an HIV incidence rate of 4.9 per 100 person years of follow up, 95 % CI: 2.9 – 8.27. Kaplan Meier curves revealed a higher risk of HIV-1 acquisition among women who were BV positive than the women who were BV negative. A log rank test showed that the v probability of seroconversion was different among the women depending on BV status, chi-square value 3.8, p 0.05. Controlling for confounding variables, seroconversion was high, but not significant, among BV positive women, adjusted hazard ratio 3.21; 95 % CI; 0.85-12.12, p value 0.08. CONCLUSION This study suggests that BV increases HIV seroconversion risk though statistical significance was not achieved. Vaginal cleansing education, screening and treating women with BV could maintain normal vaginal flora and reduce their susceptibility to HIV.enbacterial vaginosisHIV-1infectionThesis