Online Program Home
My Program

Abstract Details

Activity Number: 163 - SPEED: Longitudinal/Correlated Data
Type: Contributed
Date/Time: Monday, July 30, 2018 : 10:30 AM to 12:20 PM
Sponsor: Biometrics Section
Abstract #329653 Presentation
Title: Mixed Latent Markov Models for Longitudinal Multiple Diagnostics Data with an Application to Salmonella in Malawi
Author(s): Marc Henrion* and Angeziwa Chirambo and Tonney C. Nyirenda and Melita Gordon
Companies: Malawi Liverpool Wellcome Trust Clinical Research Programme and Malawi Liverpool Wellcome Trust Clinical Research Programme and College of Medicine and Malawi Liverpool Wellcome Trust Clinical Research Programme
Keywords: latent Markov model; mixed model; latent variable model; Bayesian modelling; random effects; Salmonella

Latent Markov models (LMMs) are commonly used to analyze longitudinal data from multiple diagnostic tests. LMMs consist of a structural model for the latent infection state, defining probabilities for the initial state and transmission between states, and a measurement model for the observed test results, defining the item response probabilities and thus test sensitivities and specificities. LMMs typically assume that tests are independent conditional on the latent infection state. This is likely to be violated for tests using similar technologies. We introduce random effects to relax the conditional independence assumption and we derive a generalization of the basic LMM for an application to Salmonella infection data. We analyze longitudinal data from four molecular PCR tests and a stool culture test from patients in Blantyre, Malawi. To assess the tests' performances, we consider basic and mixed LMMs, both with time homogeneous and heterogeneous transition probabilities. We compare the different models and discuss technical considerations. A PCR assay using primers from the TTR gene achieves the best sensitivity / specificity trade-off.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program