Post date: Dec 28, 2014 9:55:36 PM
" Nested Partially-Latent Class Models (npLCM) for Estimating Disease Etiology from Case-Control Data"
Zhenke Wu, Department of Biostatistics, JHU
Abstract:
The Pneumonia Etiology Research for Child Health (PERCH) study attempts to use modern technology to infer the distribution of pneumonia-causing bacterial or viral pathogens from multiple measurements with different precisions outside of the lung and to predict the cause for an individual child with pneumonia. The paper describes a latent variable model to address these two analytic goals using data from a case-control design. We assume each observation is a draw from a mixture model for which each component represents one pathogen. Conditional dependence among multivariate binary measurements on a single subject is induced by nesting latent subclasses within each disease class. Measurement precision can be estimated using the control sample for whom the etiologic class is known. We use stick-breaking priors on the subclass weights to estimate the population and individual etiologic distributions that are averaged across models indexed by different numbers of subclasses. Assessment of model fit and individual diagnosis is done using posterior samples drawn by Gibbs Sampling. We demonstrate the method's operating characteristics via a simulation study tailored to the motivating scientific problem and illustrate the model with a detailed analysis of PERCH study data.