|
Activity Number:
|
153
- Contributed Poster Presentations: Biopharmaceutical Section
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 8, 2022 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
|
Abstract #322333
|
|
|
Title:
|
Doubly-Robust Estimation for Interim Monitoring in Sequential Multiple Assignment Randomized Trials
|
|
Author(s):
|
Cole Manschot* and Marie Davidian and Eric Laber
|
|
Companies:
|
North Carolina State University and North Carolina State University and Duke University
|
|
Keywords:
|
Sequential multiple assignment randomized trials;
Group sequential analysis;
Interim monitoring;
Double robustness;
Dynamic treatment regimes;
Clinical trials
|
|
Abstract:
|
Sequential multiple assignment randomized trials (SMARTs) are the gold standard for generating data for evaluation of multi-stage treatment regimes. Interim monitoring allows the possibility of early stopping; however, few methodological developments for principled interim analysis of SMARTs exist. A challenge for interim analysis of SMARTs is that not all enrolled participants will have progressed through all treatment stages at the time of the analysis. Prior procedures estimate mean outcome for a regime using data only from participants who have completed all stages and have an ascertained outcome of interest. We propose an efficient doubly-robust estimator for mean outcome for interim analysis that exploits partial information from enrolled participants who have not completed all stages and develop associated Pocock and O'Brien-Fleming-type testing procedures for early stopping. In simulation experiments, the estimator controls type I error and achieves nominal power while reducing expected sample size relative to previous estimators. The procedure is illustrated via a case study based on a recent SMART evaluating behavioral pain interventions for breast cancer patients.
|
Authors who are presenting talks have a * after their name.
