Agent-based modelling of complex factors impacting malaria prevalence.

Area of Science:

• Epidemiology
• Mathematical Biology
• Computational Science

Background:

• Malaria transmission models are becoming increasingly complex, incorporating factors like control measures, vector behavior, and socio-economic variables.
• The complexity of these models makes parameter calibration difficult, especially with limited directly usable field data.
• Household size is a known factor in malaria elimination, but its integration into models presents challenges.

Purpose of the Study:

• To present a novel approach for combining in situ field data with malaria transmission model parameters.
• To demonstrate a method for calibrating classical malaria models using agent-based stochastic simulations.
• To enable the incorporation of complex transmission factors into malaria models without developing entirely new dynamical systems.

Main Methods:

• Utilized agent-based stochastic simulations initially calibrated with hut-level experimental data.
• Generated synthetic data from simulations for regression analysis to calibrate key model parameters (e.g., biting rates, vector mortality).
• Applied the approach to classical malaria models, calibrating parameters to account for complex factors and tested against Entomological Inoculation Rate (EIR) field data.

Main Results:

• Demonstrated that transmission characteristics can be estimated by incorporating factors influencing EIR and malaria incidence.
• Showcased how reducing mosquito-human contact rates and increasing vector mortality (via control measures or socio-economic factors) impacts transmission.
• Validated the approach against diverse field data for EIR values.

Conclusions:

• Complex phenomena like long-lasting insecticidal net (LLIN) coverage, vector behavior changes, and socio-economic impacts can be integrated into continuous-level modeling.
• The computational approach is generic and applicable to other diseases or systems with available in situ data.
• While a proof of concept, the study offers valuable insights into enhancing malaria transmission modeling accuracy.