Insights into the dynamics of malaria infection using mathematical modelling of cohort studies

Both P. vivax and P. falciparum continue to be the leading causes of malaria morbidity and mortality. The work in this thesis seeks to use mathematical modelling to gain insights into the within-host dynamics of these infections and with the aim to infer better control methods for reducing their burdens.P. vivax infection involves hypnozoite formation. These hypnozoites remain dormant in the liver before they reactivate at varying time intervals. The relative contribution of a new P. vivax infection and hypnozoite reactivation to initiation of blood stage infection is unclear. Also, the mechanisms governing the duration of hypnozoite latency and frequency of reactivation of hypnozoites are not fully understood. This thesis investigates the contribution of new inoculations of P. vivax sporozoites to primary infection versus hypnozoite reactivation to total infection by combining both mathematical modelling and data from two clinical cohorts. I found that the majority of the blood stage infections arise from hypnozoite reactivation.The work on P. vivax infection is extended to determine how hypnozoites reactivate from latency using conceptual models of different hypnozoite reactivation mechanisms from the literature. Some of these mechanisms are complex explanations of the reactivation events including re-infection or without re-infection. Applying these models to published cohort data, I found that heterogeneity in exposure to infection can explain many of the features of hypnozoite reactivation without the need for inclusion of additional factors. In P. falciparum endemic regions the incidence of both infection and clinical malaria risk decreases with age. The mechanisms of how an individual controls parasites are not completely understood. Also, patients with parasites before the malaria season have a reduced risk of clinical malaria. Whether the presence of parasites contributes to this observation is unknown. By analysing cohort data from a seasonal transmission setting, I found that both infants and adults have a reduced risk of infection and clinical malaria and that the clearance of parasites before the malaria season does not decrease the risk of clinical malaria.Overall, the work in this thesis demonstrates how mathematical modelling can be used to develop intervention strategies for controlling and eliminating malaria