Online Program Home
My Program

Abstract Details

Activity Number: 69
Type: Contributed
Date/Time: Sunday, July 31, 2016 : 4:00 PM to 5:50 PM
Sponsor: Biometrics Section
Abstract #320847 View Presentation
Title: Multiple Imputation Framework to Estimate Causal Effect of Testing on Treatment Decision
Author(s): Irina Bondarenko* and Yun Li
Companies: University of Michigan and University of Michigan
Keywords: causal inference ; multiple imputation ; principal strata ; test utility ; diagnistic test ; breast cancer

Newly diagnosed cancer patients and physicians face complex decisions regarding diagnostic tests and treatment options. Due to advances in medical science a number of genomic, genetic and imaging tests became available during the last decade. However, utility of these tests is still debated. Utility of the test can be expressed as causal effect of testing on a treatment decision. We propose to use imputation framework to estimate causal effect of testing on treatment decision. Such framework allows to impute potential outcomes (treatments) conditional on the results of the test, and other relevant covariates. Using multiply-imputed data, we assess average causal effect, causal effect by test result and estimate how key covariates predict principal strata membership. The later would allow to identify the subset of patients who may benefit from the test. We conducted simulation study to study properties of the proposed method. We applied the proposed method to examine the impact of a genomic test on chemotherapy using data from ICanCare Study, a survey of newly diagnosed with breast cancer patients.

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

Back to the full JSM 2016 program

Copyright © American Statistical Association