Individual-level Modeling of COVID-19 Epidemic Risk Using GAMA simulation data
Résumé
The COVID-19 pandemic highlights the need for a multi-faceted response comprising a range of public health interventions including quarantining and targeted lockdowns, in conjunction other measures such as vaccination campaigns and genomic disease surveillance. Many of these interventions need to be informed by epidemic risk predictions given the available data, including clinical symptoms, contact patterns, and environmental factors. Here we propose a novel probabilistic formalism based on Individual-Level Models (ILMs) that offers rigorous formulas for the probability of infection of individuals, which can be parameterized via Maximum Likelihood Estimation (MLE) applied on compartmental models defined at the population level. We integrate individual data collected in real-time with overall case counts to update a predictor of the susceptibility of infection of a single person as a function of their individual risk factors (e.g.: age, immune status, etc.) In order to generate realistic synthetic data for the purpose validating models that depend on such individual-level covariates, we used an agentbased model (ABM) in the GAMA simulation platform using the COMOKIT parameters for COVID-19. The initial simulation experiments are promising and suggest that is possible to: (1) obtain good estimates for the individual-level parameters by applying MLE on the population level data, (2) predict what individuals in the population are at higher risk of infection, and (3) inform effective public health interventions, such as quarantine, based on the predicted individual risks.
Origine | Fichiers produits par l'(les) auteur(s) |
---|