A multivalued agent-based model for the study of noncommunicable diseases

Area of Science:

• Epidemiology
• Computational modeling
• Health behavior science

Background:

• Noncommunicable diseases (NCDs) are significantly influenced by individual behavioral choices.
• Existing models like SimNCD simulate NCD dynamics but often lack detailed individual decision-making processes.
• Epidemiological questionnaires are valuable tools for understanding population health behaviors.

Purpose of the Study:

• To extend the SimNCD (Simulation of NonCommunicable Diseases) model by integrating agent-based reasoning mechanisms.
• To incorporate epidemiological questionnaires into agent decision-making to represent individual attitudes and choices.
• To apply the enhanced model to study childhood obesity and predict corpulence variations.

Main Methods:

• Agent-based modeling (ABM) was used as the core framework.
• A novel reasoning mechanism was developed for agents, incorporating preferences modeled via the linguistic 2-tuple method.
• Multi-attribute decision-making (MADM) methods were employed to simulate activity choices.
• The enhanced SimNCD model was applied to a childhood obesity scenario.

Main Results:

• The study demonstrates the utility and extensibility of the enhanced SimNCD model.
• The integration of questionnaire-based preferences allows for a more nuanced simulation of individual health-related choices.
• The model successfully predicted children's corpulence variations under different simulated scenarios.

Conclusions:

• The enhanced SimNCD model provides a powerful tool for simulating NCDs by capturing individual decision-making.
• Integrating behavioral factors through agent reasoning enhances the predictive power of epidemiological models.
• This approach offers valuable insights for public health interventions targeting NCDs, such as childhood obesity.