A spatial statistical approach to malaria mapping

Background Good maps of malaria risk have long been recognized as an important tool for malaria control. The production of such maps relies on modelling to predict the risk for most of the map, with actual observations of malaria prevalence usually only known at a limited number of specific locations. Estimation is complicated by the fact that there is often local variation of risk that cannot be accounted for by the known covariates and because data points of measured malaria prevalence are not evenly or randomly spread across the area to be mapped. Methods We describe, by way of an example, a simple two-stage procedure for producing maps of predicted risk: we use logistic regression modelling to determine approximate risk on a larger scale and we employ gee-statistical ('kriging') approaches to improve prediction at a local level. Malaria prevalence in children under 10 was modelled using climatic, population and topographic variables as potential predictors. After the regression analysis, spatial dependence of the model residuals was investigated. Kriging on the residuals was used to model local variation in malaria risk over and above that which is predicted by the regression model. Results The method is illustrated by a map showing the improvement of risk prediction brought about by the second stage. The advantages and shortcomings of this approach are discussed in the context of the need for further development of methodology and software.