Exploring the Impact of Human and Pathogen Behavior on Infectious Disease Dynamics

Abstract

The COVID-19 pandemic has spurred recognition of the many ways humans interact to influence infectious disease dynamics. The success of non-pharmaceutical interventions revealed the importance of contact rates between individuals in driving outbreaks. Travel between locations, meanwhile, was shown to be responsible for introducing local outbreaks and sustaining transmission. The pandemic also prompted interest in the way pathogens interact to cause infection and disease, as the co-circulation of multiple respiratory pathogens was highly anticipated following disruption to their typical seasonality patterns. In this context, mathematical models and epidemiological studies have been useful in lending insight into how human and pathogen behavior influence population-level disease dynamics. In the first two chapters, we explore the impact of human behavior on two different scales. In the first, we analyze cross-sectional survey data on age-structured contact rates from six cities in the US during the COVID-19 pandemic. We find highly disrupted contact rates relative to baseline rates, but in general, limited differences between the six cities. Second, we assess the impact of differences in mobile phone operators’ geographic coverage on the implied population mobility patterns estimated from call detail records (CDRs). We find that even in relatively simple metapopulation models, the chosen mobility source substantially influences both descriptive travel patterns between localities and implied transmission dynamics from mathematical models. Lastly, in Chapter 3, we explore the use of cross-sectional, co-detection prevalence studies in ascertaining virus-virus interactions using samples from clinical settings. We show that selection bias is typically expected in this study design, except under strict assumptions regarding the probability of having symptoms given different viral states, and we propose alternative, unbiased study designs to study virus-virus interactions.