Predictors of HIV seroconversion in Botswana: machine learning analysis in a representative, population-based HIV incidence cohort.

While HIV prevalence and incidence in Botswana have been well-characterized at the population level, the individual-level predictors of HIV seroconversion using machine learning approaches had not been identified in a population-representative dataset.

Target RWE's HIV team and Botswana Harvard AIDS Institute Partnership collaborators applied machine learning methods to a large representative population-based Botswana dataset to identify predictors of HIV seroconversion.

Identifying machine learning-derived predictors of HIV seroconversion provides HIV prevention program designers with a more accurate risk stratification tool for targeting prevention interventions to the highest-risk individuals.

Machine learning HIV seroconversion predictor identification for targeted prevention intervention in Botswana, Predictive analytics evidence supporting targeted HIV prevention program design and resource allocation