Subgroup inference for multiple treatments and multiple endpoints in an Alzheimer’s disease treatment trial

Patrick Schnell, Qi Tang, Peter Müller, Brad Carlin

Research output: Contribution to journalArticlepeer-review

13 Scopus citations

Abstract

Many new experimental treatments outperform the current standard only for a subset of the population. Subgroup identification methods provide estimates for the population subset which benefits most from treatment. However, when more than two treatments and multiple endpoints are under consideration, there are many possible requirements for a particular treatment to be beneficial. In this paper, we adapt notions of decision-theoretic admissibility to the context of evaluating treatments in such trials. As an explicit demonstration of admissibility concepts, we combine our approach with the method of credible subgroups, which in the case of a single outcome and treatment comparison provides Bayesian bounds on the benefiting subpopu-lation. We investigate our methods’ performance via simulation, and apply them to a recent dataset from an Alzheimer’s disease treatment trial. Our results account for multiplicity while showing patient covariate profiles that are (or are not) likely to be associated with treatment benefit, and are thus useful in their own right or as a guide to patient enrollment in a second stage study.

Original languageEnglish (US)
Pages (from-to)949-966
Number of pages18
JournalAnnals of Applied Statistics
Volume11
Issue number2
DOIs
StatePublished - Jun 2017

Bibliographical note

Publisher Copyright:
© Institute of Mathematical Statistics, 2017.

Keywords

  • Bayesian inference
  • Clinical trials
  • Heterogeneous treatment effect
  • Linear model
  • Simultaneous inference
  • Subgroup identification

Fingerprint

Dive into the research topics of 'Subgroup inference for multiple treatments and multiple endpoints in an Alzheimer’s disease treatment trial'. Together they form a unique fingerprint.

Cite this