Semiparametric Inference for a Two-Phase Failure-Time-Auxiliary-Dependent Sampling Design

Large cohort studies under simple random sampling could be prohibitive to conduct for epidemiological studies with a limited budget, especially when exposure variables are expensive or hard to obtain. Failure-time-dependent sampling (FDS) is a commonly used cost-effective sampling strategy for studies with failure times as outcomes. To further enhance study efficiency upon FDS, we propose a two-phase failure-time-auxiliary-dependent sampling (FADS) design that allows the probability of obtaining the expensive exposures to depend on both the failure time and some cheaply available auxiliary variables to the main exposure of interest. To account for the sampling bias, we develop a semiparametric maximum pseudo-likelihood approach for inference and a nonparametric bootstrap procedure for variance estimation. The proposed estimator of regression coefficients is shown to be consistent and asymptotically normally distributed. The simulation studies indicate that our proposed method works well in practical settings and is more efficient than other competing sampling schemes or methods. We illustrate our method with the analysis of two real data sets, the ARIC Study and the National Wilms' Tumor Study.