Ross-Macdonald models: Which one should we use?

Area of Science:

• Mathematical modeling
• Epidemiology
• Vector-borne diseases

Background:

• Ross-Macdonald models are foundational for vector-borne disease dynamics.
• Existing models often use varied formulations without exploring dynamical consequences.
• A comprehensive comparison of different Ross-Macdonald model assumptions is lacking.

Purpose of the Study:

• To present and analyze diverse Ross-Macdonald model formulations.
• To investigate the dynamical impact of varying assumptions on disease periods.
• To compare model outputs with empirical data and highlight key drivers.

Main Methods:

• Developed an agent-based model with arbitrary latency/infectious period distributions.
• Created a deterministic Volterra integral equations model with arbitrary waiting times.
• Compared model solutions using epidemic statistics (peak, final size) and basic reproduction number (R0).

Main Results:

• Different distributions for latent and infectious periods significantly alter epidemic curves.
• Agent-based models provided empirical estimations for R0.
• Seasonality was identified as a critical factor influencing epidemic dynamics and duration.

Conclusions:

• Realistic distributions for latent and infectious periods are crucial for accurate modeling.
• Model formulation choices profoundly affect vector-borne disease dynamics.
• Seasonality plays a key role in shaping epidemic patterns.