Accelerated Failure Time Model for Case-Cohort Design with Longitudinal Covariates Subject to Measurement Error and Detection Limits

• Activity Number: 411
• Type: Contributed
• Date/Time: Tuesday, July 31, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Xinxin Dong*+ and Lan Kong and Abdus S. Wahed
• Companies: and Penn State College of Medicine and University of Pittsburgh
• Keywords: case cohort ; longitudinal biomarker ; limit of detection (LOD) ; joint analysis ; mixed effects model ; accelerated failure time model

Abstract

Biomarkers are often measured over time in epidemiological and clinical studies to better understand the mechanism of diseases. Case cohort sampling provides a cost effective method to collect expensive biomarker data for revealing the relationship between biomarker trajectories and time to event. However, biomarker measurements are often limited by the sensitivity and precision of a given assay, resulting in data that are censored at detection limits and prone to measurement errors. Additionally, occurrence of an event of interest may preclude biomarkers from being further evaluated. Under a classical case cohort design, we propose a modified likelihood-based approach to accommodate special features of longitudinal biomarker measurements in the accelerated failure time (AFT) models. The maximum likelihood estimators are obtained by Gaussian quadrature method. We evaluate the performance of our case-cohort estimator and compare its relative efficiency to the full cohort estimator through simulation studies. The proposed method is further illustrated using the data from a biomarker study of sepsis among patients with community acquired pneumonia.