KwabenaAsabere/MAPPING-MALARIA-PREVALENCE-IN-THE-GAMBIA

# **Malaria Prevalence Mapping in The Gambia** ## **Project Overview** This project investigates **malaria prevalence in The Gambia** using spatial statistical modeling. The analysis leverages geostatistical methods, particularly the **Stochastic Partial Differential Equation (SPDE) approach in INLA**, to produce high-resolution prevalence maps. The work is motivated by the need to better understand spatial heterogeneity in malaria burden and to inform targeted interventions. ### **Why This Project Is Needed** #### Malaria remains a major health burden in sub-Saharan Africa The Gambia, like many West African countries, has seen substantial progress in malaria control but transmission persists, often in pockets or hotspots. National-level statistics hide important local variation in prevalence. Resource allocation requires high-resolution maps Ministries of Health and NGOs (e.g., WHO, PMI, Global Fund) need spatially detailed maps to identify areas with higher malaria burden. Without spatial modeling, interventions (bed nets, indoor residual spraying, drug distribution) may be spread too thinly or miss high-risk communities. #### Surveys alone are insufficient Household surveys provide cluster-level estimates, but they are sparse (only 63 clusters in this dataset). We need statistical models (like INLA-SPDE) to “fill in the gaps” and predict prevalence continuously across the country. #### Modern Bayesian geostatistics (SPDE in INLA) The SPDE approach allows computationally efficient Bayesian modeling of large spatial datasets. It accounts for spatial correlation between nearby survey clusters, producing smoother, more realistic maps than ad hoc methods. #### Decision-making and equity By mapping where malaria is highest, policymakers can ensure equitable distribution of interventions. Helps move from “one-size-fits-all” to targeted strategies that save resources and maximize impact. --- ## Data The dataset contains **63 survey clusters** across T …