A stochastic agent-based model to evaluate COVID-19 transmission influenced by human mobility

Area of Science:

• Epidemiology and Public Health
• Computational Modeling
• Mathematical Biology

Background:

• The COVID-19 pandemic highlighted the need for accurate epidemic forecasting models.
• Understanding multiscale human mobility is crucial for predicting disease transmission and evaluating interventions.
• Existing models often struggle to incorporate complex travel behaviors and their impact on infection dynamics.

Purpose of the Study:

• To develop a novel agent-based model, Mob-Cov, integrating human mobility patterns and individual health conditions.
• To investigate the influence of local and global travel on COVID-19 outbreak dynamics and the feasibility of achieving zero-COVID-19.
• To provide a flexible and accurate tool for researchers and policymakers to analyze pandemic scenarios.

Main Methods:

• Utilized a stochastic agent-based modeling strategy with hierarchical spatial containers.
• Modeled individual movements using power-law local movements and global transport between container levels.
• Incorporated dynamic infection and recovery processes within the population.

Main Results:

• Frequent local, long-distance travel and smaller population sizes reduce disease transmission.
• Increased travel between large geographical areas accelerates global outbreaks.
• Achieving zero-COVID-19 is feasible with reduced population size (<400) and restricted global travel (<20% highly mobile individuals).

Conclusions:

• The Mob-Cov model offers a realistic simulation of human mobility across various spatial scales.
• Mobility patterns, population density, and health conditions significantly influence epidemic trajectories.
• The model serves as a valuable tool for understanding pandemic dynamics and informing public health strategies.