Abstract
Epidemics of infectious diseases, such as the one caused by the rapid spread of the coronavirus disease 2019 (COVID-19), have tested the world's more advanced health systems and have caused an enormous societal and economic damage. The mechanism of contagion is well understood. As people move around, over time, they regularly engage in social interactions. The spatiotemporal network representing these interactions constitutes the backbone on which an epidemic spreads, causing outbreaks. At the same time, advanced technological responses have claimed some success in controlling the epidemic based on digital contact tracing technologies. Motivated by these observations, we design, develop and evaluate a stochastic agent-basedSEIRmodel of epidemic spreading in spatiotemporal networks informed by mobility data of individuals (trajectories). The model focuses on individual variation in mobility patterns that affects the degree of exposure to the disease. Understanding the role that individual nodes play in the process of disease spreading through network effects is fundamental as it allows to (i) assess the risk of infection of individuals, (ii) assess the size of a disease outbreak due to specific individuals, and (iii) assess targeted intervention strategies that aim to control the epidemic spreading. We perform a comprehensive analysis of the model employing COVID-19 as a use case. The results indicate that simple individual-based intervention strategies that exhibit significant network effects can effectively control the spread of an epidemic. We have also demonstrated that targeted interventions can outperform generic intervention strategies. Overall, our work provides an evidence-based data-driven model to support decision making and inform public policy regarding intervention strategies for containing or mitigating the epidemic spread.
Collapse