The Interplay between Human Behavior and Infectious Disease Dynamics

Abstract

Human behavior and infectious disease are related in a dynamic feedback loop. This integral and ubiquitous relationship is often ignored in epidemiological modeling, leading to findings that overlook a crucial element in determining infectious disease transmission and thus have limited utility for public health purposes. Significant past work focuses on one side of the behavior-disease relationship: how contact behavior determines disease transmission. Far less attention has been paid to how disease changes social behavior and the dynamic effects of these behavioral changes on future disease spread. The work that does consider the entire behavior-disease feedback loop is largely theoretical and lacks support from empirical data.

In this dissertation, I move this problem forward by characterizing the effects of disease on behavior through empirical data, exemplifying the downstream effects of the identified behavior changes on infectious disease dynamics through epidemiological models, and considering the impacts of population heterogeneity in these behaviors. In chapter 1, I consider how disease physiologically drives behavior change through sickness behaviors. In chapter 2, I characterize the ways that disease modifies behavior socio-politically, through public health policy and information-driven risk perception. In chapter 3, I demonstrate the effects of health inequities that drive population heterogeneity in infectious disease-related behaviors. I achieve this through integration of large-scale empirical data with complex epidemiological and statistical models to address case studies of respiratory-transmitted infectious diseases: influenza and COVID-19. I find that 1) sickness behaviors and absenteeism reduce transmission, and only a small portion of the population must engage in sickness behaviors in order to alter epidemic outcomes; 2) subjective risk perception and state-level policies result in the largest reductions in social contact behavior during the COVID-19 pandemic, and typical behavior-disease models that account for objective risk will overestimate the impacts of behavior change on transmission; and 3) health inequities result in increased influenza among vulnerable populations that are often overlooked by epidemiological surveillance.

The COVID-19 pandemic has exemplified that behavior is our first defense against transmission, thus understanding how behavior changes due to disease is a crucial step to understanding epidemiological dynamics and improving public health.