Incorporating Cross-Sectional Information into a Joint Model of Longitudinal and Survival Data Using a Power Prior

• Activity Number: 320 - Innovative Approaches for Modeling Time-to-Event Data in the Presence of Competing Risks and/or Time-Varying Covariates
• Type: Topic Contributed
• Date/Time: Tuesday, August 9, 2022 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Juned Siddique* and Michael Daniels and Hongyan Ning and Norrina Allen and John Wilkins and Donald Lloyd-Jones
• Companies: Northwestern University Feinberg School of Medicine and University of Florida and Northwestern University Feinberg School of Medicine and Northwestern University Feinberg School of Medicine and Northwestern University Feinberg School of Medicine and Northwestern University Feinberg School of Medicine
• Keywords: joint model; power prior; cohort; CVD

Abstract

Joint modeling of longitudinal data and survival outcomes is a useful approach for understanding how cardiovascular risk factor trajectories over time affect the development of cardiovascular disease (CVD) later in life. Prospective cardiovascular cohort studies provide an appropriate source of information to address these questions but a limitation is that many of these cohort studies are relatively small and/or have a low number of events. Conversely, there exist large representative cross-sectional surveys that provide an abundant source of information on the relationship between cardiovascular risk factors and CVD, but do not contain information on risk factor trajectories. In this talk, we describe a flexible Bayesian approach for obtaining more precise inferences by incorporating cross-sectional risk factor data and its association with outcomes into a joint model through the use of a power prior. We use longitudinal data from the Coronary Artery Risk Development in Young Adults (CARDIA) cohort study and cross-sectional data from the Third National Health and Nutrition Examination Survey (NHANES) Linked Mortality File.