|
Activity Number:
|
208
- Personalized and Precision Medicine
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 10, 2021 : 1:30 PM to 3:20 PM
|
|
Sponsor:
|
Biometrics Section
|
|
Abstract #317924
|
|
|
Title:
|
Investigating Stability in Subgroup Identification for Stratified Medicine
|
|
Author(s):
|
Gleicy Macedo Hair* and Thomas Jemielita and Shahrul Mt-Isa and Patrick Schnell and Richard Baumgartner
|
|
Companies:
|
Merck & Co., Inc. and Merck & Co., Inc. and MSD and The Ohio State University College of Public Health and Merck Research Laboratories
|
|
Keywords:
|
subgroup analysis;
survival data;
Bayesian;
risk benefit
|
|
Abstract:
|
Subgroup analysis methods investigate treatment effect heterogeneity among subsets of the study population defined by baseline characteristics. Several methodologies have been proposed in recent years and with these, some statistical issues such as multiplicity have been widely discussed. Bayesian credible subgroups is an approach that addresses this issue by constructing simultaneous credible bands for treatment effects. It results in a pair of credible subgroups (D, S) from which practitioners may conclude that all patient types in D have beneficial treatment effect and that those in Sc (S complement) have no or harmful treatment effect, while deferring conclusions about patient in the uncertainty region S\D (S remove D). Here, we applied bootstrap resampling to assess the stability of both subgroup-level and patient-level assignments. Our findings show that some patients move between S\D and other subgroups, but rarely between D and Sc. We propose exploring the stability of subgroups as a sensitivity analysis step to further assess the robustness of the identified benefitting subgroup. This result can be extended to other approaches and endpoints besides time-to-event data.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2021 program
|
