Modelling the Epidemiological Dynamics of Seasonal Influenza Viruses at Local Scales

Seasonal influenza viruses are a substantial source of disease burden globally, causing epidemics across all climatic regions. Through error-prone RNA replication, influenza viruses can escape pre-existing humoral immunity and reinfect humans, resulting in recurrent epidemics within populations. From year to year, individual epidemics differ substantially in timing, duration and size. Despite intensive study, characterising the spatiotemporal patterns of virus circulation and identifying the underlying sources of this variability at global, regional and local scales remain as ongoing challenges. There is a need to reconcile environmental, virological and host drivers of virus epidemiological dynamics across diverse contexts. Such insights can only be generated through a holistic approach that integrates observational, ecological, experimental and modelling studies: this would enable more accurate and timely epidemiological forecasts and more efficient allocation of public health resources. In this thesis, I investigate the phylodynamical interactions between the seasonal influenza virus, environment and human host population, integrating analyses from observational study and theoretical modelling approaches. The current knowledge gap on the drivers of local city-level epidemics is identified in Chapter 2 and subsequently addressed over 4 research chapters. In Chapter 3, I review existing epidemic detection algorithms and present a novel statistical model that I developed for use with noisy disease surveillance data and is optimised for the context of seasonal influenza. In Chapter 4, I apply this novel algorithm and analyse a 15-year dataset of 18,250 typed, subtyped, and antigenically characterised seasonal influenza viruses from the five most populous cities in Australia. With the necessary geographical and virus resolution, I quantify the effects of previously hypothesised environmental and virological factors. Most surprisingly, despite an apparent lack of marked change in virus antigenicity, individual antigenic variants are capable of reinvading the same population over consecutive seasons, which runs contrary to predictions made by existing mathematical models. In Chapters 5 and 6, I investigate how antigenic variants are capable of causing recurrent epidemics at local scales by building upon previous theoretical modelling studies and developing a modelling framework to investigate the interactions between and joint effects exerted by the topology of cross-immunity and host contact structure within a population. In Chapter 5, I investigate the effects of correlations between network structure and individual susceptibility. In Chapter 6, I examine the population-level significance of age-specific changes to an individual's immune response. In Chapter 7, I review my findings and discuss how these new insights into virus ecology can open new avenues for better influenza control and future research.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2020

Author dc:creator

• Lam, Edward Kong Seng

Advisors dc:contributor.advisor

• Russell, Colin
• Restif, olivier

• Influenza
• Epidemiology
• Phylodynamics
• Modelling
• Dynamics
• Infectious Diseases

Licence

• https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/9a20ced4-a324-4b3d-b137-081265a6a0fb/download
• https://www.rioxx.net/licenses/all-rights-reserved/

Language dc:language

eng

DOI dc:identifier.doi

https://doi.org/10.17863/CAM.64742

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/317626