Online Program Home
My Program

Abstract Details

Activity Number: 404 - Quantile, Semiparametric and Nonparametric Methods in Survival Analysis
Type: Contributed
Date/Time: Tuesday, July 30, 2019 : 2:00 PM to 3:50 PM
Sponsor: Biometrics Section
Abstract #307183 Presentation
Title: Methods for Survival Analysis Leveraging Data from Randomized Clinical Trials and Observational Studies
Author(s): Jean De Dieu Tapsoba* and Ying Qing Chen
Companies: Fred Hutchinson Cancer Research Center and Fred Hutchinson Cancer Research Center
Keywords: Big data; Clinical trial; Cox proportional hazards model; Observational data; Risk prediction; Shrinkage

In many biomedical studies, researchers are often interested in investigating the association between some covariates (e.g. exposure) and the time to an event of interest. Such an investigation is usually conducted under a randomized clinical trial design with a sample of limited size taken from the targeted population. Large datasets from external observational studies are increasingly available nowadays and may contain important (although possibly biased) information on the association under investigation from a population that shares some similarities with the one targeted by the clinical trial. In this paper, we develop methods combining the clinical trial and observational datasets to improve the estimation of the covariate effects as well as the prediction of hazard based on the Cox regression model. The proposed methods use weighting and penalization techniques to strike a balance between the possible bias increase and variance reduction originating from the use of the observational data. Their finite sample size performances are assessed through extensive simulation studies. We illustrate our methods with an application to HIV- prevention clinical trial and program data.

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

Back to the full JSM 2019 program