Age-disparate sexual relationships with older men may drive high rates of HIV acquisition in young women in sub-Saharan Africa, but evidence is limited. We investigate the association between age-disparate relationships and HIV incidence in Manicaland, Zimbabwe.Design:
A general-population open-cohort study (six surveys) (1998–2013).Methods:
A total of 3746 young women aged 15–24 years participated in consecutive surveys and were HIV-negative at the beginning of intersurvey periods. Last sexual partner age difference and age-disparate relationships [intergenerational (≥10 years age difference) and intragenerational (5–9 years) versus age-homogeneous (0–4 years)] were tested for associations with HIV incidence in Cox regressions. A proximate determinants framework was used to explore factors possibly explaining variations in the contribution of age-disparate relationships to HIV incidence between populations and over time.Results:
About 126 HIV infections occurred over 8777 person-years (1.43 per 100 person-years; 95% confidence interval = 1.17–1.68). Sixty-five percent of women reported partner age differences of at least 5 years. Increasing partner age differences were associated with higher HIV incidence [adjusted hazard ratio (aHR) = 1.05 (1.01–1.09)]. Intergenerational relationships tended to increase HIV incidence [aHR = 1.78 (0.96–3.29)] but not intragenerational relationships [aHR = 0.91 (0.47–1.76)]. Secondary education was associated with reductions in intergenerational relationships [adjusted odds ratio (aOR) = 0.49 (0.36–0.68)]. Intergenerational relationships were associated with partners having concurrent relationships [aOR = 2.59 (1.81–3.70)], which tended to increase HIV incidence [aHR = 1.74 (0.96–3.17)]. Associations between age disparity and HIV incidence did not change over time.Conclusion:
Sexual relationships with older men expose young women to increased risk of HIV acquisition in Manicaland, which did not change over time, even with introduction of antiretroviral therapy.