|
Activity Number:
|
135
- Multiplicity, Missing Data and Other Topics
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 9, 2021 : 1:30 PM to 3:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
|
Abstract #318395
|
|
|
Title:
|
Simulating Treatment Discontinuation in Clinical Trials
|
|
Author(s):
|
Ahmad Hakeem Abdul Wahab* and Hege Michiels and Arman Sabbaghi and Stephen J Ruberg
|
|
Companies:
|
Purdue University, Department of Statistics and Department of Applied Mathematics, Computer Science and Statistics, Ghent University and Purdue University, Department of Statistics and Analytix Thinking, LLC
|
|
Keywords:
|
causal;
potential outcomes;
compliance;
Rubin Causal Model;
missing;
principal stratification
|
|
Abstract:
|
Randomized clinical trials will inevitably have some patients whose outcomes are missing due to disruptions such as excess efficacy, lack of efficacy and adverse events from treatment. These complications or intercurrent events to the planned trial protocol can confound the effect of both control and experimental treatments under investigation. Specifically, such complications yield latent subpopulations or strata of patients characterized by their treatment adherence or compliance behaviors. These strata must be taken into account to obtain valid causal inferences on the effect of the receipt of treatment, and not simply the assignment of treatment. To that end, we developed a Data Generating Model embedded in an Rshiny app that simulates clinical trials under the Rubin Causal Model that captures the treatment causal effect while accounting for intercurrent events. This app enables users to control patient adherence through different sources of discontinuity with varying functional behaviors, and understand operating characteristics of treatment effect estimators obtained by different models for various estimands.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2021 program
|
