The recent COVID-19 pandemic has led to the reliance on epidemiological models to forecast the potential benefits associated with various mitigation strategies to control the spread of this novel disease. The mathematical models focused on COVID-19 vary in their complexity from the relatively simple to more elaborate models that follow individuals and their movement over the course of the epidemic. Model performance depends both on model structure and the underlying data available. During the beginning of an epidemic when limited data are available, simple models should work best. As the epidemic progresses, more complex models may then perform better. Yet, most research focuses on a single model and does not consider comparing across multiple models. This research will address immediate societal concerns over the course of the epidemic by using a multi-model approach to ask questions regarding epidemic intensity (for example, the number of expected COVID-19 cases) and the impacts of mitigation strategies on these dynamics. While the research will develop models focused on COVID-19, the models along with the statistical approaches to fit models to data will be applicable for future outbreaks of novel diseases. Additionally, this project provides training opportunities for a graduate student and a post-doctoral researcher.
In general, this research will use a multi-model approach that considers a range of models from relatively simple compartmental models to more complex models that incorporate age- and/or network-structure based on social contacts. The models considered will provide estimates of important epidemiological parameters (for example, R0) along with their associated uncertainty. Understanding this uncertainty and establishing if and when epidemic model complexity is a hindrance or utility for estimating epidemic parameters and epidemic trajectories is a key issue. This research will combine data with multiple epidemic models on the COVID-19 pandemic to: 1. understand the relative performance of traditional compartmental approaches to network-based age-structured models during the course of an on-going epidemic; 2. quantify the most crucial sources of variation for estimating threshold vaccination criteria, epidemic trajectories, and potential mitigation strategies; and, 3. identify if and when more complex network-based models outperform more traditional approaches as the epidemic progress over time and data availability changes. To meet the above goals, the analyses conducted will use a Bayesian approach of fitting models to data. A Bayesian approach will allow for the quantification of model and parameter uncertainty as well as assess the uncertainty of forecasted epidemic dynamics and mitigation strategies.
This project is jointly funded by the Ecology and Evolution of Infectious Diseases Program (EEID), Division of Environmental Biology, Directorate of Biological Sciences and the Established Program to Stimulate Competitive Research (EPSCoR).
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
