Estimating Infectious Disease Parameters for the Transmission of Malaria in Ugandan Children

Abstract

In this thesis, we used malaria data from Ugandan children to estimate malaria-related mortality and determinants. We derived a link between mathematical and statistical models enabling the estimation of the malaria prevalence and force of infection (FOI) by assuming an SIS compartmental model and the generalised linear mixed model (GLMM). Finally we developed a methodology to account for outcome-dependent sampling (ODS) using a joint model. The results indicated that malaria is the leading cause of death among children who died when aged 29 days to 14 years in some parts in Uganda. The mortality hazard was higher among children who had fever or those that were admitted in the health centres or those who died on the way to seek care. The malaria prevalence and FOI were highest among children aged 5-10 years in areas with high transmission intensities and were highest among children aged about 1 year in areas of low transmission intensities. Children who were previously asymptomatic had higher hazards of being re-infected with malaria compared to their counterparts who were symptomatic. Heterogeneity in malaria infection was quantified and was highest between households compared to between household members. The simulation study indicated that ignoring ODS could lead to biased estimates which would later lead to incorrect assessment of intervention strategies.