A Framework for the Joint Modeling of Longitudinal Diagnostic Outcome Data and Latent Infection Status

• Activity Number: 243
• Type: Contributed
• Date/Time: Monday, July 30, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Geoffrey Jones*+ and Wesley Johnson and Daan Vink and Nigel French
• Companies: Massey University and University of California at Irvine and Massey University and Massey University
• Keywords: Bayesian statistics ; Biomarkers ; Diagnostic Test ; Sensitivity ; Specificity

Abstract

For many diseases the infection status of individuals cannot be observed directly, but can only be inferred from biomarkers that are subject to measurement error. Diagnosis based on observed symptoms can itself be regarded as an imperfect test of infection status. The temporal relationship between infection and disease may be complex, especially for recurrent diseases where individuals can experience multiple bouts of infection. Given repeated measures of a biomarker for infection and apparent disease status of a number of individuals at multiple time points, together with relevant covariates, we propose and fit a model in which the unobserved infection status is a correlated latent process. This model can be used to investigate the temporal dynamics of infection, and to evaluate the usefulness of the biomarker for monitoring purposes. Our work is motivated and illustrated by a longitudinal study of Bovine Digital Dermatitis on commercial dairy farms, where the biomarkers of interest are a continuous ELISA test outcome and a dichotomous outcome, lesion status in feet.