A model structure for estimating malaria risk

Malaria is one of the leading causes of death in the developing world today. While prevention and treatment methods are available, their large-scale usage is a major drain on governmental budgets, and not applied whenever necessary. For this reason, understanding the endemicity of a region will allow the efficient implementation of suitable prevention methods. More importantly, the prediction of extraordinary malaria outbreaks will allow the recruitment of emergency facilities before transmission becomes widespread. In addition, climate change may influence the endemicity pattern of a region, causing malaria incidence to rise in areas in which it was non-existent or controlled. For all these purposes, a seasonal to decadal malaria forecast is needed. A novel approach has been attempted, using dynamic mathematical biological modelling. There has been initial work on the prediction of malaria epidemic based on seasonal climate forecasts, in areas of unstable transmission, which may be used to provide early warning. Here we describe a mathematical biological model of the weatherdependent parasite transmission dynamics, within-host and within-vector. The biological structure and the mathematic formulation permit computer simulation of infection patterns under various climatic and control conditions. Here we present the model structure and results at a local scale using reanalysis weather data. We then discuss the role of different aspects of the impact of unusual climatological effects and their potential implications, as well as further developments in the simulation structure and outline pathways for future progress. We also suggest further aspects of biological research, required for model improvement.

Keywords: malaria transmission; dynamic modelling; mathematical model; endemicity; Africa