B0696
Title: Inference for individual-level models of infectious diseases with an application to the COVID-19 pandemic
Authors: Leila Amiri - University of Manitoba (Canada) [presenting]
Abstract: When individual-level data are available, more complex types of models may also be applied, then is otherwise the case. We initially formulate this in the context of the progression of an epidemic, for a disease where there is removal and no re-infection (i.e., a susceptible-exposed-infected-removed (SEIR) framework), as it is an appropriate model for the COVID-19. In the SEIR model, it is assumed that the infection rate will stay constant over time. We propose an individual-level statistical model (ILM) to rigorously predict the infection rate at each given time by incorporating the corresponding covariates and characteristics of infected people at the individual and area levels. We will then incorporate the infection rates at each given time into the SEIR model to accurately predict the outbreak over time that help policymakers for possible interventions. The performance of the proposed approach is evaluated through simulations. We also apply our proposed model to Manitoba COVID-19 datasets.