Affordable Access

deepdyve-link
Publisher Website

Optimum design and sequential treatment allocation in an experiment in deep brain stimulation with sets of treatment combinations.

Authors
  • Atkinson, Anthony1
  • Pedrosa, David2
  • 1 Department of Statistics, London School of Economics, London WC2A 2AE, UK.
  • 2 Klinik und Poliklinik für Neurologie, Universitätsklinik Gießen und Marburg, Baldingerstrasse, D35043 Marburg.
Type
Published Article
Journal
Statistics in Medicine
Publisher
Wiley (John Wiley & Sons)
Publication Date
Dec 30, 2017
Volume
36
Issue
30
Pages
4804–4815
Identifiers
DOI: 10.1002/sim.7493
PMID: 28960373
Source
Medline
Keywords
Language
English
License
Unknown

Abstract

In an experiment including patients who underwent surgery for deep brain stimulation electrode placement, each patient responds to a set of 9 treatment combinations. There are 16 such sets, and the design problem is to choose which sets should be administered and in what proportions. Extensions to the methods of nonsequential optimum experimental design lead to identification of an unequally weighted optimum design involving 4 sets of treatment combinations. In the actual experiment, patients arrive sequentially and present with sets of prognostic factors. The idea of loss due to Burman is extended and used to assess designs with varying randomization structures. It is found that a simple sequential design using only 2 sets of treatments has surprisingly good properties for trials with the proposed number of patients. Copyright © 2017 John Wiley & Sons, Ltd.

Report this publication

Statistics

Seen <100 times