Probabilistic modelling of malaria prevalence in regions with high transmission rates – Nigeria as a case study

Abstract

Malaria continues to pose significant health threat in many parts of the world, particularly in regions where transmission rates are high like Nigeria. Early detection and prediction of its prevalence are essential for effective public health interventions and resource allocation. This study focuses on modelling the prevalence of malaria based on some key demographic and environmental factors for effective decision making. It also explores the interplay among the predictors as it affects the outcome in terms of direction and strength. The generalized linear modelling approach with logit link was adopted in this study. The parameters of the model were estimated via Maximum Likelihood technique. Likelihood ratio test was used to assess the model fit.  The classification accuracy of the model was also examined using the confusion matrix, Receiver Operating Characteristic and Area Under the Curve. The results revealed that mothers, level of education and children ages have more significant impacts on malaria risk than other factors examined (maternal education levels, gender, region, and household wealth). The model has a good fit with high level of discriminatory or classification power based on the evaluation metrics and test. By identifying high-risk factors and groups, health authorities can prioritize interventions/medical resources directed to areas where they are mostly needed. This study emphasizes the importance of data-driven approaches in the ongoing fight against malaria and provides room for future research and model refinement.