Simulating nationwide coupled disease and fear spread in an agent-based model.

Area of Science:

• Epidemiology
• Computational modeling
• Behavioral science

Background:

• Disease outbreaks involve complex interactions between pathogen spread, human behavior, and psychological responses like fear.
• Traditional epidemiological models often simplify or omit these crucial feedback loops.

Purpose of the Study:

• To develop and analyze a dynamic agent-based model (EpiCast) that couples disease transmission with the spread of fear.
• To investigate how fear influences protective behaviors and impacts epidemic trajectories.
• To compare agent-based modeling with compartmental models and explore various behavioral scenarios.

Main Methods:

• Implementation of the EpiCast simulation framework, a dynamic agent-based model capturing individual interactions and disease transmission.
• Coupling disease spread with fear propagation through both in-person contact and broadcast media.
• Comparison of agent-based results with compartmental models incorporating different disease states (asymptomatic, exposed, pre-symptomatic).
• Simulation of diverse behavioral scenarios by varying fear levels and response intensity.

Main Results:

• Compartmental models show that including various disease states (e.g., asymptomatic, exposed) affects outbreak progression and overall trajectory.
• In the EpiCast model, fear spread via broadcast media combined with strong behavioral responses typically results in multiple epidemic waves.
• Multiple waves in EpiCast occurred under a narrow parameter range when fear spread only through local contact.

Conclusions:

• The co-evolution of disease and fear dynamics significantly influences outbreak patterns.
• Agent-based modeling, like EpiCast, is essential for capturing the complex feedbacks between fear, behavior, and disease spread.
• Integrating fear dynamics into epidemiological models is critical for designing effective public health interventions and response strategies.