Abstract
The Place Susceptibility Index (PSI) has the potential to be a critical tool in effectively managing infectious disease outbreaks or natural disasters in Africa. However, PSI availability for the African continent is limited and often, when available, is only at the national or regional level. Thus, lacking the required details to support locally relevant decision-making to support such activities. Here, outlined the method for modeling PSI at the 3rd-order administrative level for selected African Countries. This method combined Bayesian spatial statistical modeling with the utilization of the Population-based HIV Impact Assessments (PHIA) data and geospatial covariates. Using the Jenks Classification, substantial variations in PSI classes across the countries were observed. Across the 10 countries, about 45% of the spatial units were categorized as the low and relatively low susceptibility classes, while around 31% belonged to the high and very high classes. Botswana had 17% of the spatial units classified as high or very high, while Zambia had as many as 58% of its spatial units in these classes. The analysis showcased wide variations in susceptibility across countries, thus highlighting heterogeneity often missed in national datasets. This thereby provides insight into regions and areas within each country with the potential for severe negative outcomes from disease outbreaks and natural or man-made disasters. The datasets presented here are publicly available as part of the INFORM Africa Research project, and provide an evidence base to inform strategic decision-making.