The proposed study will test hypotheses regarding microbiological, virological and behavioral risk factors for the development of bacterial vaginosis (BV), a common vaginal condition which is increasingly recognized as having serious health sequelae, including adverse pregnancy outcomes, pelvic inflammatory disease, and increased risk of HIV infection. The etiology and natural history of BV are poorly understood. The study will be conducted in Rakai District, Uganda, where approximately 50 percent in the general population of women of reproductive age have BV. We propose to conduct two complementary research activities: I, a BV natural history study in a cohort of 250 women (with and without BV, HIV and prior sexual experience), and II, a study in 50 polygamous family units which will enrol the husband, his wives and other women residing in the household. Repeated interview and sample collection in the two studies will be used to assess transition probabilities of BV onset, persistence, regression and recurrence in relation to: a) detailed sociodemographic, behavioral and health data; b) vaginal microflora, particularly Lactobacillus species (which will be characterized using DNA homology and assessed for H2O2 production) and c) the potential presence of lactobacillus bacteriophages, whose possible role in Lactobacillus depletion will be explored. In the polygamous household study, we will determine whether factors associated with normal vaginal flora or with BV (including lactobacilli, anaerobes and phages) may be transmitted sexually or via close household contact such as through the sharing of bathing utensils or water, by comparing women with a polygamous sexual network to other women within the household. The Rakai population offers a unique opportunity to assess BV. We previously enrolled and followed approximately 7,000 women in a population-based trial of STD control for AIDS prevention, have documented increased risk of HIV and adverse birth outcomes in women with BV, and have evidence of improved pregnancy outcomes with STD/BV treatment. The proposed study will provide unique epidemiological, microbiological and virological data regarding normal vaginal flora in this rural African population, the natural history of BV, and on potential causes of this prevalent condition. Such information will be critical for the design of future BV prevention, treatment and control trials, including selection of interventions to be tested, sample size requirements and definition of study end points.
