MODELLING THE TRANSMISSION DYNAMICS OF SARS-CoV-2

Abstract

Infectious disease modelling helps cultivate a quantitative understanding of the transmission patterns of a pathogen. The inference of key model parameters is facilitated by Bayesian methods, which update initial knowledge of these factors with empirical observations. In this thesis, taking the coronavirus disease 2019 (COVID-19) as the disease of interest, we developed hierarchical models to characterize the transmission patterns and estimate impacts of various non-pharmaceutical interventions on institutional, national and international outbreaks. For estimation of model parameters, we utilized various numerical algorithms, such as Markov chain Monte Carlo and Integrated Nested Laplace Approximation, efficiently pooling information from the partially observed disease surveillance data. Overall, this thesis informs the transmission dynamics and control of COVID-19 through the epidemic models constructed at different scales.