Bayesian Spatiotemporal Small-Area Estimation of HIV Testing Uptake in Ghana, 2008-2022: Integrating Machine-Learning-Derived Geospatial Covariates with District-Level BYM2-RW1 Modelling of the Ghana Demographic and Health Surveys

The study examined HIV testing expansion among women in Ghana using data from 2008, 2014, and 2022, including 4,769, 9,391, and 15,014 women respectively. National testing prevalence increased from 20.7% in 2008 to 46.7% in 2014 and 53.8% in 2022. Researchers used a Bayesian hierarchical model combining spatial and temporal factors with geospatial variables derived from machine learning. They found that higher urban population proportion was associated with increased testing, while greater distance to the nearest city was associated with lower testing. The analysis revealed a persistent north-to-south gradient in testing uptake that cannot be fully explained by accessibility and urbanization. Specific districts with low and high testing levels were identified as priorities for targeted testing scale-up.