An individual-based model of Plasmodium falciparum malaria transmission on the coast of Kenya.

Individual-based models provide powerful tools to model complex interactions characterized by individual variability. This paper presents an object-oriented design for individual-based modelling of Plasmodium falciparum malaria transmission. Two kinds of objects, human and mosquito, that exhibit variability among individuals for parameters such as recovery and survival rates are defined. The model tracks the dynamics of human hosts and adult female mosquitoes individually. Immunity, modelled as a function of exposure history, is represented by reduced susceptibility and increased recovery rate. The model was calibrated using epidemiological data collected at 30 sites along the coast of Kenya. The sites were grouped into low, intermediate and high transmission based on mean daily human-biting rates. Simulation results show that malaria transmission was stable even in low transmission areas where the human-biting rate is approximately 0.5 bite per day. The model was used to examine the effect of infection control programmes that aim at interrupting transmission by reducing human-vector contact rates and implementing active case detection and drug treatment of infections. With this intervention, local elimination of malaria is likely with a probability of extinction of approximately 0.8 in low transmission areas. However, a small amount of immigration (> 0.3%) by infected people into the community could prevent local extinction of the parasite. In intermediate and high transmission areas, reduction in prevalence is short-lived and the probability of local elimination is low, even at high coverage levels of the intervention.