Under-five mortality (U5M) rates are among the health indicators of utmost importance globally. It is the goal 3 target 2.1 of the Sustainable Development Goals that is expected to be reduced to at least 25 per 1000 livebirths by 2030. Although a considerable reduction in U5M was observed globally, several countries especially those in sub-Saharan Africa such as Ghana are struggling to meet this critical target. To provide opportunities for efficient U5M surveillance and targeted control and elimination efforts amidst limited public health resources in poor-resource settings like Ghana where universal intervention is effectively impossible, there is the need to produce robust small-area (more localised) estimates of U5M risk for both sampled and unsampled locations continuously over Ghana to support policymakers and other stakeholders responsible for the survival of children. We aimed to estimate and map U5M risk, with the goal of identifying communities at high risk where urgent interventions can be targeted. To achieve this, we demonstrate how geostatistical modelling under the novel Integrated Nested Laplace Approximation (INLA) can be implemented to analyse U5M risk in Ghana using the most recent 2014 Ghana Demographic and Health Survey data. Our modelling approach also allowed clusters with small counts to borrow information from their neighbouring clusters thereby reducing the risk of inflated prevalence attributable to possible presence of small counts at a cluster location. We will also showcase how visualization of the predicted U5M risk can be improved by developing interactive web-based predictive maps. Our approach and maps facilitate the development of targeted public health interventions and further research in a resource-limited setting where universal intervention is practically impossible. To better support policymakers and program managers in their efforts to reduce U5M and to improve child survival, we developed exceedance probabilities so map users can easily identify communities where the predicted U5M risk is likely to exceed allowable thresholds.
