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 #306902 Presentation
Title: Doubly Robust Inference Procedures for Analyzing the Cancer Registry Data
Author(s): Sho Komukai* and Satoshi Hattori
Companies: Osaka University Graduate School of Medicine and Graduate School of Medicine, Osaka University
Keywords: cancer registry; net survival; relative survival ratio; doubly robust estimator; robust score test

Cancer registries are useful to various research questions for cancer populations, such as the comparisons for the cancer prognosis among nations. Since the cancer registry data has no information about cause of death, one of the measures for cancer prognosis based on the cancer registry data is a net survival, defined as a survival probability if a patient would not die due to reasons other than the cancer. For net survival, a doubly robust estimator was proposed in the presence of the covariate-dependent censoring (Komukai and Hattori, 2017). A relative survival ratio is also used as alternative to the net survival, and it is defined as the ratio between the survival probability of the cancer population and that of the general population. In this research, we propose a doubly robust estimator for the relative survival ratio. Although the inference for the net survival by Komukai and Hattori (2017) relies on untestable assumptions on dependence among time-to-events and censoring, it is possible to verify the underlying assumption in the inference for the relative survival ratio. We also propose a doubly robust test to assess the underlying assumption for relative survival ratio.

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

Back to the full JSM 2019 program