Online Program Home
My Program

Abstract Details

Activity Number: 341 - Random Effects and Mixed Models
Type: Contributed
Date/Time: Tuesday, July 30, 2019 : 10:30 AM to 12:20 PM
Sponsor: Biometrics Section
Abstract #304102
Title: A Bayesian Joint Model for Longitudinal Frequency and Duration Outcomes in a Migraine Study
Author(s): Gul Inan*
Companies: Istanbul Technical University
Keywords: Bayesian inference; count data; joint modeling; migraine; multiple outcomes; random-effects

Motivated by a migraine study, we are interested in jointly modeling migraine frequency and duration outcomes collected repeatedly from patients by specialists to investigate the mechanisms of migraine over time. In this motivating longitudinal study, at each hospital visit, the migraineurs were asked to report number of days with migraine attacks they have had within past 30 days and average duration of the attacks. While the frequency outcome is a count variable bounded by 30 days, the duration outcome is completely reported in discrete hours, including 0 for non-migraine days, resulting in a count variable. Furthermore, this bivariate longitudinal count data has an unbalanced structure with irregularly spaced time points due to that each patient visited the hospital with varying number of times within irregular intervals. In this study, we propose a bivariate generalized linear mixed-effects model to jointly analyze the two longitudinal count outcomes within Bayesian framework. We investigate performance of the proposed joint model via a Monte Carlo simulation study and compare its performance with separate models, where each longitudinal count outcome is modeled individually.

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

Back to the full JSM 2019 program