Related Experiment Videos
A random-effects model for analysis of infectious disease final-state data
1Department of Mathematics and Computer Science, University of Salford, England.
Biometrics
|September 1, 1995
Summary
This study fits an extended General Epidemic Model to household infection data, incorporating individual heterogeneity in disease transmission. The methodology efficiently estimates transmission rates and assesses model fit using Shigella sonnei outbreak data.
Area of Science:
- Epidemiology
- Mathematical Biology
- Biostatistics
Background:
- The General Epidemic Model (GEM) is a foundational tool in mathematical epidemiology.
- Previous extensions allowed for specified distributions of infectious periods.
- Fitting such models to real-world household data presents statistical challenges.
Purpose of the Study:
- To fit an extended GEM to household infection data.
- To develop a likelihood function for analyzing final infection states.
- To incorporate and estimate random-effects heterogeneity in transmission rates.
Main Methods:
- Fitting Ball's extended GEM to household infection data.
- Derivation of a likelihood function for final infection states.
- Development of an algorithm for numerical computation of maximum likelihood estimators (MLEs).
- Assessment of goodness-of-model-fit.
Main Results:
- The methodology was successfully applied to Shigella sonnei outbreak data from 102 households.
- Estimated transmission rates showed consistency with age and sex-related infectiousness/susceptibility patterns.
- The extended model with random-effects heterogeneity provided a robust fit.
Conclusions:
- The extended GEM provides a flexible framework for analyzing household infectious disease dynamics.
- The developed methods allow for efficient estimation and model assessment.
- Findings support the influence of individual characteristics on disease transmission within households.