Temporal and probabilistic comparisons of epidemic interventions

Area of Science:

• Epidemiology
• Mathematical Modeling
• Public Health

Background:

• Disease forecasting is vital for public health interventions but challenged by stochasticity, heterogeneous contact patterns, and behavioral changes.
• Predicting epidemic trajectories is complex due to inherent randomness and dynamic population behaviors.

Approach:

• Utilizes time-dependent probability generating functions (PGFs) to model stochastic branching processes of disease spread over contact networks.
• Introduces a general transmissibility equation incorporating time-varying rates for transmission, recovery, contact patterns, and immunization.
• Develops a framework for temporal and probabilistic analysis of intervention impacts, validated against computationally intensive simulations.

Key Points:

• Enables probabilistic forecasting of disease spread and intervention effectiveness.
• Provides metrics for comparing intervention strategies, including expected cases, worst-case scenarios, and probability of critical case levels.
• Offers a computationally efficient alternative to continuous-time stochastic simulations.

Conclusions:

• Facilitates more accurate short-term forecasts of disease spread.
• Supports informed comparison and selection of public health intervention strategies.
• Addresses the dynamic and unpredictable nature of epidemics for improved public health planning.