Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 442 - Disease Prediction, Statistical Methods for Genetic Epidemiology and Mis
Type: Contributed
Date/Time: Thursday, August 12, 2021 : 4:00 PM to 5:50 PM
Sponsor: Section on Statistics in Epidemiology
Abstract #318140
Title: An Augmented Likelihood Approach for the Discrete Proportional Hazards Model and a Complex Survey Design, with Application to HCHS/SOL Study
Author(s): Lillian Boe* and Pamela Shaw
Companies: University of Pennsylvania and Kaiser Permanente Washington Health Research Institute
Keywords: measurement error; augmented likelihood; misclassification; surrogate endpoint; proportional hazards

In certain epidemiologic settings, it is common to have infrequent follow-up for an expensive or otherwise burdensome gold standard outcome and more frequent follow-up for an error-prone or proxy outcome. Examples include a biopsy or intensive clinical exam that may be done infrequently and regular follow-up with a self-report instrument. We have developed a new augmented likelihood approach that incorporates error-prone, auxiliary outcomes into the analysis of gold standard time-to-event data in order to improve the efficiency of inference. This estimator is implemented for a complex survey design that includes both clustering and unequal probability sampling. We conduct a numerical study to show how we can improve statistical efficiency by using the proposed method instead of standard approaches for interval censored survival data that do not leverage the auxiliary data. We apply this method to data from the Hispanic Community Health Study/Study of Latinos in order to assess the association between energy and protein intake and the risk of incident diabetes mellitus when the dietary exposures of interest are also subject to measurement error.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2021 program