Background HIV prevalence is decreasing in large parts of sub-Saharan Africa (SSA), but the impact of this decline on the geographical heterogeneity of HIV infectious burden is not well understood. We explored and described aspects of the spatial and temporal heterogeneity of the epidemic in three countries in SSA.
Methods Data were obtained from Demographic and Health Surveys conducted at different times in Malawi, Tanzania, and Kenya. We identified and compared spatial clusters with high numbers of HIV infections at two different times from each country using Kulldorff spatial scan test. The test locates areas with higher numbers of HIV infections than expected under spatial randomness. For each identified cluster, a likelihood ratio test was computed. A P-value was then determined through Monte Carlo simulations to evaluate the statistical significance of each cluster.
Results The table summarises the main results. We found no evidence of decline in HIV prevalence within clusters with high HIV prevalence despite the statistically significant decline in the national HIV prevalence in Malawi and Tanzania. National HIV prevalence decreased by 19% in Malawi, and 17% in Tanzania; meanwhile, HIV prevalence in areas outside of the clusters declined by 33% and 30%, respectively. There was no statistically significant decline in the national HIV prevalence in Kenya, but HIV prevalence within clusters increased by 27%.
Conclusions We found marked spatial variability in the decline of HIV prevalence in the three studied SSA countries. Even in the presence of declining national HIV prevalence, HIV prevalence in the high HIV prevalence clusters either did not decline or even increased. Most of the gains in reducing HIV disease burden did not occur in the areas of most intense HIV transmission, but in areas outside of the clusters. Our findings provide insights for resource allocation and HIV prevention interventions in these countries.
- Spatiotemporal dynamics
- Sub-Saharan Africa