Integrating opportunistic data into respiratory disease models to enhance surveillance, explain seasonality, and reveal spatial transmission landscapes - Project Summary My research program addresses fundamental questions about population-scale infectious disease dynamics and brings to light the role of social factors in these dynamics. The impact that carefully formulated and extensively validated models can have in predicting spatio-temporal disease dynamics and providing a rational assessment of alternative intervention strategies is understood by policymakers and clinical practitioners alike. However, as the COVID-19 pandemic has demonstrated, continuing to neglect socio-behavioral processes presents a critical barrier to future model development: behavioral surveillance gaps jeopardize our ability to predict pathogen emergence; a poor understanding of the feedback loops between behavior and disease hampers the forecasting of disease dynamics; and a limited appreciation of the nonlinear impacts of anti-mitigation behavior reduces hope of eliminating diseases before they take hold. At the same time, public health inequities fueled by income inequality and systemic racism pose a dire and urgent threat. To address these pressing gaps, my research team uses a multi-scale socio-behavioral disease modeling approach to integrate interacting elements of health, physical and socially-constructed environments, and community and individual behavior to predict social and spatial heterogeneities in respiratory disease burden. We develop generative and inferential models for a systematic understanding of the constant, compounding socio-behavioral processes that give rise to disease heterogeneities across individuals, communities, and systems. We also leverage opportunistic datasets to characterize behavior and disease across geography and time to resolve questions that have eluded explanation without socio-behavioral data. Our future work will advance the theory of respiratory disease dynamics with a focus on two case studies, SARS- CoV-2 and influenza, in the United States. The work will make significant contributions to our understanding of respiratory virus epidemiology, seasonality, spatial epidemiology, health inequities, and public health policy and will spur innovation for integrating large data streams into infectious disease models. These advances will generalize to improve our understanding of other respiratory, partially immunizing viruses that cause epidemics or pandemics. Our focus on the US public health system also serves as a crucial case study to characterize the consequences of intense variation across social, environmental, economic, and demographic dimensions and inform the impact of heterogeneity on data collection, model complexity, disease outcomes, and management strategies. Our work has broad implications at a time when heterogeneities will be amplified by future perturbations, including emerging diseases, climate change, and sociopolitical unrest. Understanding the mechanisms underlying the causes of epidemiological heterogeneity across space, time, and the landscape of vulnerability will help inform resource allocation, design outbreak intervention, optimize disease surveillance, strengthen health systems, and improve access to healthcare.
