High-Fidelity Agent-Based Modeling to Support Prevention Decision-Making: an Open Science Approach

Area of Science:

• Public Health
• Systems Science
• Computational Modeling

Background:

• Preventing adverse health outcomes requires understanding multi-level contexts and social systems.
• System science methods, particularly agent-based models (ABMs), are increasingly called for to address this complexity.
• The utility of ABMs hinges on their accuracy and acceptability to stakeholders.

Purpose of the Study:

• To provide recommendations for adopting and using agent-based models (ABMs) in health outcome prevention.
• To ensure the creation of high-fidelity models and reliable decision-making based on model outcomes.
• To promote open science principles for enhanced model accuracy and acceptability.

Main Methods:

• Adopting system science methods, specifically agent-based models (ABMs).
• Incorporating stakeholder engagement throughout the modeling lifecycle.
• Implementing rigorous model verification, validation, and replication processes.
• Illustrating application through case studies in HIV and overdose prevention.

Main Results:

• Agent-based models (ABMs) offer unique insights into prevention strategies by simulating complex interactions.
• Recommendations address enhancing model fidelity and ensuring reliability for evidence-based decision-making.
• Stakeholder involvement and open science practices are crucial for model utility and acceptance.

Conclusions:

• Adoption of agent-based models (ABMs) requires a focus on accuracy, acceptability, and open science principles.
• Systematic recommendations facilitate the effective use of ABMs in public health prevention.
• Successful implementation, exemplified by HIV and overdose prevention, can improve health outcomes.