A Time-Dependent Structural Model Between Latent Classes and Competing Risks Outcomes

• Activity Number: 177 - Statistical Modeling of Lifetime Data: LiDS Section Student Award Session
• Type: Contributed
• Date/Time: Tuesday, August 4, 2020 : 10:00 AM to 2:00 PM
• Sponsor: Lifetime Data Science Section
• Author(s): Teng Fei* and John Hanfelt and Limin Peng
• Companies: Emory University and Emory University and Emory University
• Keywords: Competing risks; Cumulative incidence function; Estimating equation; Latent class analysis; Structural Model

Abstract

Latent class analysis is an intuitive tool to characterize disease phenotype heterogeneity. With data more frequently collected on multiple phenotypes in chronic disease studies, it is of rising interest to investigate how the latent classes embedded in one phenotype are related to another phenotype. Motivated by a cohort with mild cognitive impairment (MCI) from the Uniform Data Set (UDS), we propose a time-dependent structural model between latent classes and competing risk outcomes. We develop a two-step estimation procedure which circumvents latent class assignment and is rigorously justified for accounting for the uncertainty in classifying latent classes. The new method also properly addresses the random censoring to the competing risks and the missing failure types of competing risks. The asymptotic properties of the resulting estimator are established. We develop sample-based inference procedures whereas standard bootstrapping inference is infeasible. Simulation studies demonstrate the advantages of the new method over benchmark tools. An application to the MCI data from UDS uncovers detailed pictures of the neuropathological relevance of the baseline MCI subgroups.