An open-data-driven agent-based model to simulate infectious disease outbreaks.

Area of Science:

• Epidemiology
• Computational modeling
• Public health

Background:

• Infectious disease outbreaks are complex, influenced by vaccination, population density, and age structure.
• Understanding outbreak dynamics requires models that capture individual behaviors and factor interactions.
• Agent-based models (ABMs) can reveal emergent patterns not explicitly programmed.

Purpose of the Study:

• To develop and validate a data-driven agent-based model for simulating airborne infectious disease spread.
• To investigate how town characteristics and agent interactions influence outbreak size and severity.
• To test the hypothesis that complex interactions, not just town features, dictate outbreak outcomes.

Main Methods:

• Created a data-driven agent-based model using open data to simulate airborne disease spread.
• Simulated a 2012 measles outbreak in Schull, Ireland, across 33 towns.
• Correlated model results with town characteristics (population, area, vaccination rates, age structure).

Main Results:

• Simulated outbreaks showed minimal correlation with individual town characteristics.
• Model results indicated that interactions between factors and agent decisions significantly influenced outbreak dynamics.
• The agent-based model successfully captured variations in outbreak patterns.

Conclusions:

• Agent-based models are effective tools for understanding infectious disease outbreak complexity.
• Outbreak severity is driven by intricate interactions and individual behaviors, not solely by demographic factors.
• This modeling approach provides valuable insights for public health preparedness and intervention strategies.