Novel data and approaches for dynamic modeling of human behavior and infectious disease ecology

PROJECT SUMMARY Demographic and social forces that influence human behavior fundamentally shape the epidemiology of infectious diseases, but are often left out from mathematical models of disease transmission. Our recent work has added to the growing body of evidence that patterns of how humans come into contact with one another,

and the adoption of behaviors to reduce transmission, are highly heterogeneous across individuals and groups in a population; these behaviors are strongly influenced by how societies are structured and the socio- economic, environmental, and political contexts in which epidemics occur. These same forces influence

patterns of exposure, susceptibility, and health outcomes. Human behavior also varies temporally and spatially, leading to heterogeneities in transmission across space and time and the possibility of dynamical feedback between the disease environment and the socio-demographic processes that influence human

behavior. Driven by the public health imperative to understand and address disparities in infection exposure and risk, I propose to develop new models and methods for understanding how demographic and social processes – from individual heterogeneities to large-scale patterns of population behavior – drive infectious

disease risk across scales. Specifically, my research group will develop a generalized framework for jointly modeling human behavioral changes and disease dynamics, with bi-directional feedback between behavior and the disease environment. We will also build and validate models for incorporating data on human mobility

and aggregation at the population level, while incorporating other extrinsic drivers of transmission such as climate. We will parameterize these models by leveraging detailed human behavioral data from an ongoing survey platform as well as novel sources of mobile phone and digital trace data, and validate model results by

fitting to epidemiological geo-located time series data. Our modeling approach will be flexible and dynamic, allowing us to adapt it for specific pathogens and regions of interest. The project results will provide a better understanding of the complex interactions between human behavior and the fundamental biological processes

underlying disease transmission, with the ultimate aim of guiding control efforts, designing equity-focused interventions, and improving our ability to predict and prepare for outbreaks. 1