Akullian Adam, Ssempijja Victor, Breidenbecker Daniel, Nalugoda Fred, Nakigozi Gertrude, Santelli John, Kreniske Philip, Chang Larry W, Reynolds Steven J, Ssekubugu Robert, Gray Ronald H, Wawer Maria J, Quinn Thomas C, Galiwango Ronald M, Probert William J M, Imai-Eaton Jeffrey W, Ratmann Oliver, Fraser Christophe, Kagaayi Joseph, Kigozi Godfrey, Kate Grabowski M, Serwadda David
Institute for Disease Modeling, Bill and Measslinda Gates Foundation.
Clinical Monitoring Research Program Directorate, Frederick National Laboratory for Cancer Research.
medRxiv. 2025 Jan 12:2025.01.10.24319101. doi: 10.1101/2025.01.10.24319101.
Recent declines in HIV incidence among adolescent girls and young women (AGYW) in Africa are often attributed to the expansion of biomedical interventions such as antiretroviral therapy and voluntary medical male circumcision. However, changes in sexual behaviour may also play a critical role. Understanding the relative contributions of these factors is essential for developing strategies to sustain and further reduce HIV transmission.
We conducted a mathematical modelling study of data from the Rakai Community Cohort Study (RCCS), an open, population-based cohort of 15- to 49-year-olds in 30 communities in Rakai, Uganda, to investigate the biomedical and behavioural drivers of HIV incidence decline in AGYW (15-24 years of age). We estimated changes in the HIV incidence rate between 2000-2019 using retrospective cohort data to validate our modelled incidence estimates. We ran modelled counterfactual scenarios to quantify the independent and combined effects (cumulative infections averted and difference in incidence rates) of antiretroviral therapy (ART), voluntary medical male circumcision (VMMC), and delays in age of first sex (AFS) over historical (between 2000-2020) and projected (between 2000-2050) time horizons.
Incidence in women 15-24 years of age declined by 83% between 2000-2019 (from 1.72 per 100 person-years in 2000 to 0.30 per 100 person-years in 2019), the largest reduction in incidence of all age groups of women. Increasing AFS over the last two decades (by 3 years in women and 2 years in men) was the largest contributor to incidence decline in women 15-19 years of age, averting 17% of cumulative infections between 2000-2020 and 37% between 2000-2050. Incidence in 15-19-year-old women was 69% lower in 2020 and 75% lower in 2050 compared to counterfactual scenarios without changes in AFS. ART scale-up contributed the most to incidence declines among women 20-24 years of age, averting 13% of infections between 2000-2020 and 43% of infections between 2000-2050. VMMC averted < 5% of infections in 15-24-year-olds to-date, with larger reductions in incidence between 2000-2050 in both 15-19 year-olds (13% reduction in cumulative infections) and 20-24 year-olds (22% of cumulative infections). ART, VMMC, and increasing AFS acted additively to reduce HIV incidence in AGYW, with little redundancy when combined.
Our results provide strong support for maintaining both the protective changes in sexual behaviours and effective biomedical interventions to sustain continued reductions in HIV incidence among AGYW.
非洲青少年女性(AGYW)中艾滋病毒感染率近期的下降通常归因于生物医学干预措施的扩大,如抗逆转录病毒疗法和自愿医学男性包皮环切术。然而,性行为的改变也可能起关键作用。了解这些因素的相对贡献对于制定维持并进一步降低艾滋病毒传播的策略至关重要。
我们对来自拉凯社区队列研究(RCCS)的数据进行了数学建模研究,RCCS是乌干达拉凯30个社区中15至49岁人群的一个开放的基于人群的队列,以调查AGYW(15至24岁)中艾滋病毒感染率下降的生物医学和行为驱动因素。我们使用回顾性队列数据估计2000年至2019年期间艾滋病毒感染率的变化,以验证我们建模的感染率估计值。我们运行建模的反事实情景,以量化抗逆转录病毒疗法(ART)、自愿医学男性包皮环切术(VMMC)和首次性行为年龄(AFS)延迟在历史时期(2000年至2020年)和预测时期(2000年至2050年)的独立和综合影响(避免的累积感染数和感染率差异)。
2000年至2019年期间,15至24岁女性的感染率下降了83%(从2000年的每100人年1.72例降至2019年的每100人年0.30例),是所有年龄组女性中感染率下降幅度最大的。在过去二十年中,AFS增加(女性增加3岁,男性增加2岁)是15至19岁女性感染率下降的最大因素,在2000年至2020年期间避免了17%的累积感染,在2000年至2050年期间避免了37%的累积感染。与AFS无变化的反事实情景相比,2020年15至19岁女性的感染率低69%,2050年低75%。ART扩大规模对20至24岁女性感染率下降的贡献最大,在2000年至2020年期间避免了13%的感染,在2000年至2050年期间避免了43%的感染。迄今为止,VMMC在15至24岁人群中避免的感染不足5%,在2000年至2050年期间,15至19岁人群(累积感染减少13%)和20至24岁人群(累积感染的22%)的感染率下降幅度更大。ART、VMMC和AFS增加共同作用以降低AGYW中的艾滋病毒感染率,联合使用时几乎没有冗余。
我们的结果为维持性行为的保护性变化和有效的生物医学干预措施提供了有力支持,以持续降低AGYW中的艾滋病毒感染率。