|
Activity Number:
|
548
- Innovative Methods in Precision Medicine
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Thursday, August 11, 2022 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biometrics Section
|
|
Abstract #322676
|
|
|
Title:
|
Estimation Following Two-Stage Adaptive Enrichment Design
|
|
Author(s):
|
Peter Kimani*
|
|
Companies:
|
Warwick Medical School
|
|
Keywords:
|
Subpopulation selection;
Two-stage designs;
Estimation;
Enrichment designs
|
|
Abstract:
|
Patients with a medical condition may be heterogenous in how they respond to a new treatment. The two-stage adaptive enrichment design is an efficient clinical trial design to test whether a new treatment is beneficial and for which subpopulation. Recruitment in stage 1 is from the full population whilst testing in stage 2 is on the subpopulation that is likely to benefit based on stage 1 data. This is efficient since resources in stage 2 focus on the promising subpopulation. Estimators that include stage 1 data and stage 2 data have smaller mean squared errors than estimators based on the unbiased stage 2 data. However, including stage 1 data introduces complexity because of selection bias that arise from using stage 1 data to select the subpopulation that is likely to benefit from the new treatment. In this presentation, we will describe bias adjusted estimators for the treatment effect in the selected subpopulation that we have developed.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2022 program
|
