Michelle Li & Quinn Lewis - Apr 28 2021
How Can Data Improve HIV Prevention Programs for Adolescent Girls and Young Women?

Source: Global Fund for Children

In sub-Saharan Africa, the number of new HIV infections among adolescent girls and young women remains exceptionally high. In 2019, adolescent girls and young women ages 15 to 24 years accounted for about one in four new HIV infections, despite making up only 10 per cent of the total population.

Compared with their male peers, adolescent girls and young women ages 10 to 24 years old are up to 14 times more likely to be newly infected with HIV. This disproportionate impact is due to a myriad of overlapping cultural, social, and economic factors, such as gender-based violence, discrimination, and exclusion from economic opportunities, which make it difficult for girls to negotiate safe sex, stay in school, seek health services, and make empowered decisions about their sexual and reproductive health and lives.

As the world’s youth population continues to grow, it’s clear that preventing new infections among girls and young women is critical to ending the HIV epidemic.

In response to this, the United States President’s Emergency Plan for AIDS Relief (PEPFAR) and partners initiated the DREAMS (Determined, Resilient, Empowered, AIDS-free, Mentored, and Safe) program to address the range of factors that make girls and young women particularly vulnerable to HIV in the highest HIV burden countries globally. The program aims to address the root cause of girls’ and young women’s vulnerability to HIV through the implementation of a broad evidence-based package of interventions aimed at adolescent girls and young women, their families, communities, and male partners that address HIV risk behaviours, socioeconomic vulnerabilities, and gender-based violence.

Through the program, girls receive multiple services, such as school and community-based HIV and violence prevention interventions, access to youth-friendly sexual and reproductive health services, inclusion in social networks with peer mentorship and support, educational subsidies, and socioeconomic strengthening interventions such as savings groups, to empower girls and reduce their risk of HIV. The program has been shown to reduce new HIV diagnoses by at least 25 per cent among young women in the majority of districts that have implemented the program.

So, where does data come in?

In partnership with DREAMS, the Data.FI project, funding by the United States Agency for International Development (USAID) conducts high-impact analyses on adolescent girls and young women to inform programming decisions. The team leverages existing gender data from traditional data sources, such as representative household surveys, in combination with novel data sources such as satellite imagery, with machine-learning software to fill gaps in these data sources.

Specifically, this data-informed approach is designed to:

Characterise Risk Among Girls and Young Women

Data.FI uses traditional population-based household survey data, for instance from the Demographic and Health Surveys and the Population-based HIV Impact Assessments, to understand risk among adolescent girls and young women based on known factors including early sexual debut, having multiple sexual partners, alcohol use, a history of violence, being orphaned, and not being in school — among others.

The team applies that data to characterise risk, such as identifying the most salient risk factors for HIV and understanding patterns of risky behaviour. This information can be used by DREAMS program implementers to inform the design of HIV prevention activities, target most at-risk girls for enrolment, or tailor the delivery of services for specific profiles of adolescent girls and young women.

Estimate the Population Size of Vulnerable Girls

Typically, traditional surveys only identify representative populations at the national or regional levels. Which can limit the ability of programs to estimate the number of adolescent girls and young women in need of services at the district level. To address this, Data.FI uses machine learning and artificial intelligence to combine geotagged survey data with satellite imagery data to predict survey data values at a one-kilometre square level.

This technique gives the team the ability to estimate the number of vulnerable adolescent girls and young women in need of HIV prevention services. This information can be used by DREAMS program planners to monitor the proportion of girls in need of services who are being reached with the full package of prevention interventions and advocate for program expansion to other high-priority districts.

Locate Most At-Risk Adolescent Girls and Young Women

From this process, Data.FI can generate a modelled surface that depicts, at a 1-square-kilometre area, where vulnerable adolescent girls and young women are likely to live and plan where to target HIV prevention activities and how to implement interventions based on existing infrastructure, access, and other geographic characteristics.

Reaching HIV epidemic control requires protecting the most at-risk adolescent girls and young women and other populations most likely to acquire HIV before they are infected. Using all the tools and data available, including artificial intelligence, can help to better locate the most vulnerable girls at heightened risk for HIV and more equitably inform prevention services.

Palladium leads the USAID-funded Data.FI project. Learn more about Data.FI on its website and for more information contact info@thepalladiumgroup.com