|
Activity Number:
|
375
- Modern Statistical Methods for Comparative Effectiveness Research
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, July 30, 2019 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
|
Abstract #300333
|
Presentation
|
|
Title:
|
Incorporating Information from a Network of Personalized Trials to Facilitate Individualized Treatment Choice
|
|
Author(s):
|
Christopher Schmid*
|
|
Companies:
|
Brown University
|
|
Keywords:
|
network meta-analysis;
N-of-1 trials;
personalized medicine;
multilevel models;
pain studies;
shrinkage estimates
|
|
Abstract:
|
Single-case experimental designs can be used to create personalized protocols to make personalized treatment decisions. The N-of-1 trial uses a multi-crossover randomized design to measure individual treatment efficacy. Combining a series of N-of-1 trials in a multilevel structure enables assessment of average treatment effects in the population and treatment effect heterogeneity in subgroups. In a randomized trial comparing the use of N-of-1 trials vs usual care for patients with chronic pain, participants in the N-of-1 arm had few restrictions on the two treatments they could compare. This led to a wide variety of comparisons that can be considered a treatment network. This talk describes a meta-analytic approach to analyzing this network and compares the estimates of individual treatment effects from the network with those that come from analyzing each participant’s data separately. The gains in precision from analyzing the network are weighed against the model assumptions required for the network model.
|
Authors who are presenting talks have a * after their name.
