ABSTRACT
PURPOSE: To identify factors associated with human immunodeficiency virus (HIV) infection among adolescent females in Zimbabwe and appropriate prevention strategies for this vulnerable population. METHODS: A total of 1807 females aged 15-19 years completed a questionnaire and provided a blood sample for HIV testing as part of a nationally representative survey. Associations between HIV infection and factors operating at the individual, household, partner and community levels, as well as sexual behavior, were explored through bivariate and multivariate logistic regression analyses. Two multivariate models were fitted: the first model considered sexual risk behaviors and contextual variables, whereas the second model considered only contextual variables. RESULTS: Of 1807 adolescent females, 192 (10.6%) were HIV positive, and 41% of HIV-positive adolescent females reported no sexual risk behaviors. In the first multivariate model, the risk associated with number of lifetime sexual partners was increased for 1 partner (odds ratio [OR] = 2.4, 95% confidence interval [CI] = 1.57-3.6), 2 partners (OR = 4.4, 95% CI = 2.22-8.55), and 3 or more partners (OR = 6.3, 95% CI = 2.56-15.7) as compared with having 0 partners. Believing that people with HIV have many sexual partners (OR = 1.71, 95% CI = 1.14-2.57) and that the man should take the initiative to have sex (OR = 1.55, 95% CI = 1.03-2.32) were also risk factors. In the second model, increased risk was associated with having ever married or lived with a man (OR = 1.99, 95% CI = 1.18-3.35) as well as the attitudes above. Decreased risk of HIV infection was associated with having a job (OR = .39, 95% CI = .18-.88), main activity in past 12 months was as a student (OR = .39, 95% CI = .19-.80), participation in school-based lectures on sexual health (OR = .49, 95% CI = .27-.87), and perceiving that AIDS is a somewhat serious problem in the community (OR = .55, 95% CI = .33-.92). CONCLUSIONS: Adolescent females in Zimbabwe who are married, not attending school and/or are unemployed, are at heightened risk for HIV infection. Interventions that improve their educational and employment opportunities, strengthen school-based prevention services, foster more equitable gender attitudes, and make marriage safer by, for example, promoting knowledge of partners' serostatus before marriage, may reduce their risk. Future research priorities are proposed.
