A Spatio-Temporal Modeling Framework for Surveillance Data of Multiple Infectious Pathogens with Small Laboratory...

Area of Science:

• Epidemiology
• Infectious Disease Surveillance
• Biostatistics

Background:

• Syndrome-based surveillance systems capture infectious disease cases by clinical signs, with limited laboratory confirmation.
• Understanding pathogen-specific transmission dynamics and risk factors is crucial for effective disease control, especially for diseases like hand, foot, and mouth disease (HFMD).

Purpose of the Study:

• To develop and validate a Bayesian spatio-temporal modeling framework for infectious disease surveillance data with small validation sets.
• To improve the understanding of transmission dynamics and risk factors for enteroviruses causing HFMD in China.
• To compare novel and existing sampling approaches for unobserved pathogen-specific patient counts.

Main Methods:

• Developed a Bayesian spatio-temporal modeling framework to estimate pathogen-specific case counts from syndrome-based surveillance data.
• Proposed a novel approach for sampling unobserved pathogen-specific patient counts, compared with an existing method.
• Assessed the framework's utility through simulations with varying validation set sizes and sampling designs.
• Applied the methodology to the 2009 HFMD epidemic in southern China, focusing on enterovirus 71 and Coxsackie A16.

Main Results:

• The developed framework effectively samples unobserved pathogen-specific patient counts.
• Simulations demonstrated the framework's practical utility in identifying key epidemiological parameters across different validation set sizes.
• Comparison of sampling designs highlighted the impact of validation data aggregation on parameter estimation.
• The application to the 2009 HFMD epidemic provided insights into transmissibility and climatic effects of key enteroviruses.

Conclusions:

• The Bayesian spatio-temporal modeling framework offers a robust method for analyzing infectious disease surveillance data with limited laboratory validation.
• This approach enhances the understanding of disease transmission dynamics and risk factors, crucial for public health interventions.
• The methodology is applicable to real-world epidemics, aiding in the identification of key pathogens and their associated environmental factors.