Background: Adolescent girls and young women (AGYW) in East and Southern Africa experience a disproportionate HIV burden, accounting for more than a quarter of new infections in the regions. Their elevated risk reflects intertwined biological, behavioural, and structural factors. Although advanced statistical and computational models are increasingly applied to estimate HIV risk, the existing evidence remains fragmented. This review will synthesise the methodological evidence on HIV risk prediction among AGYW, identifying key strengths, limitations, and evidence gaps across existing modelling approaches.
Methods: Studies published during the period 1 January 2000 to 31 October 2025 will be eligible, as this period captures the emergence and evolution of Bayesian, machine learning, and traditional statistical models applied to HIV risk estimation among AGYW or comparable populations in East and Southern Africa. Searches will span PubMed, Scopus, ScienceDirect, and relevant institutional repositories. Two reviewers will independently screen titles/abstracts and full texts and extract data on study characteristics, predictors, validation methods, and model performance; disagreements will be resolved by discussion or, if unresolved, adjudicated by a third reviewer. Methodological quality will be assessed in duplicate using PROBAST+AI, with the third reviewer arbitrating any discrepancies in domain ratings or applicability judgments. Data synthesis will utilize a qualitative analysis approach, and this review will be reported using the PRISMA-ScR checklist.
Discussion: The findings of this study will guide the refinement of predictive frameworks and support the design of more robust, context-specific models for HIV risk estimation. Ultimately, this synthesis will contribute to strengthening data-driven approaches for HIV prevention and improving the precision of interventions targeting AGYW in East and Southern Africa. To the best of our knowledge, this study represents the first scoping review in this area and has the potential to offer important insights and guidance for improving HIV prevention interventions targeting adolescents and young women.