A0245
Title: The mean residual life model for the survival data with covariate measurement errors: Application to stage 5 CKD data
Authors: Chyong-Mei Chen - Institute of Public Health (Taiwan) [presenting]
Abstract: The mean residual life regression model with covariate measurement errors is considered. In the whole cohort, the surrogate variable of the error-prone covariate is available for each subject. In contrast, the instrumental variable (IV), which is related to the underlying true covariates, is measured only for some subjects, the calibration sample. Without specifying distributions of measurement errors and assuming that the IV is missing at random, the calibration and cohort estimators of the regression parameters are proposed by solving estimation equations (EEs) developed from the calibration and cohort samples, respectively. A synthetic estimator is derived by synthesizing the EEs based on the generalized method of moments to improve estimation efficiency. The large sample properties of the three proposed estimators are derived, and their finite sample performance is evaluated via simulation studies. Simulation results show that the cohort and synthetic estimators outperform the IV calibration estimator, and the relative efficiency of the cohort and synthetic estimators depends on the missing rate of IV. The proposed method is illustrated by application to data from patients with stage 5 chronic kidney disease in Taiwan.
