Efficient Real-Time Monitoring of an Emerging Influenza Pandemic: How Feasible?

Area of Science:

• Epidemiology
• Computational Biology
• Public Health

Background:

• Real-time epidemic monitoring is crucial for prompt public health response.
• The 2009 A/H1N1 outbreak highlighted challenges with noisy, multi-source pandemic data.
• Accurate real-time inference is difficult with imperfect, accumulating data.

Purpose of the Study:

• To assess the feasibility of real-time epidemic inference using imperfect data.
• To develop an analytic tool combining transmission models with observation models.
• To improve the timeliness and accuracy of epidemic assessments during outbreaks.

Main Methods:

• Developed an age-stratified SEIR transmission model integrated with observation models.
• Employed a sequential Monte Carlo (SMC) algorithm for synthesizing multiple data streams.
• Utilized SMC for iterative inference, model adequacy assessment, and reduced computation time compared to MCMC.

Main Results:

• SMC effectively synthesizes noisy, multi-source data for real-time inference.
• The method significantly reduces computation time for epidemic assessments.
• Demonstrated benefits in assessing predictive performance and handling parameter nonidentifiability.

Conclusions:

• SMC provides a feasible and efficient approach for real-time epidemic monitoring.
• The developed tool enhances the capacity for timely public health decision-making during epidemics.
• This methodology is valuable for managing future pandemic data challenges.