GENERALIZED ACCELERATED RECURRENCE TIME MODEL IN THE PRESENCE OF A DEPENDENT TERMINAL EVENT

• Type of Material: Article
• Authors: By Bo Wei, Emory University; Zhumin Zhang, University of Wisconsin–Madison; HuiChuan J Lai, University of Wisconsin–Madison; Limin Peng, Emory University
• Language: English
• Date: 2020-06-01
• Publisher: INST MATHEMATICAL STATISTICS
• Publication Version: Author Accepted Manuscript (After Peer Review)
• Copyright Statement: © 2020 Institute of Mathematical Statistics
• Final Published Version (URL): http://dx.doi.org/10.1214/20-AOAS1335
• Title of Journal or Parent Work: ANNALS OF APPLIED STATISTICS
• Volume: 14
• Issue: 2
• Start Page: 956
• End Page: 976
• Grant/Funding Information: Supported in part by NIH R01 grant HL113548 and NIH R01 grant AG055634.
• Keywords: Statistics & Probability; CYSTIC-FIBROSIS; PSEUDOMONAS-AERUGINOSA; REGRESSION; terminal event; Inverse probability censoring weighting; MECONIUM ILEUS; Science & Technology; Physical Sciences; Recurrent events; Mathematics; counting process; SEMIPARAMETRIC ANALYSIS; RISKS; FAILURE TIMES

Abstract

Recurrent events are commonly encountered in longitudinal studies. The observation of recurrent events is often stopped by a dependent terminal event in practice. For this data scenario, we propose two sensible adaptations of the generalized accelerated recurrence time (GART) model (J. Amer. Statist. As-soc. 111 (2016) 145–156) to provide useful alternative analyses that can offer physical interpretations while rendering extra flexibility beyond the existing work based on the accelerated failure time model. Our modeling strategies align with the rationale underlying the use of the survivors’ rate function or the adjusted rate function to account for the presence of the dependent terminal event. For the proposed models, we identify and develop estimation and inference procedures which can be readily implemented based on existing software. We establish the asymptotic properties of the new estimator. Simulation studies demonstrate good finite-sample performance of the proposed methods. An application to a dataset from the Cystic Fibrosis Foundation Patient Registry (CFFPR) illustrates the practical utility of the new methods.