Abstract Details
Activity Number:
|
645
|
Type:
|
Contributed
|
Date/Time:
|
Thursday, August 7, 2014 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Biometrics Section
|
Abstract #313729
|
View Presentation
|
Title:
|
Flexible Multivariate Methods for Binary-Event or Ordinal Composite Endpoints
|
Author(s):
|
Edward Mascha*+
|
Companies:
|
Cleveland Clinic
|
Keywords:
|
average relative effect ;
composite outcome ;
multivariate analysis ;
generalized estimating equations ;
treatment effect heterogeneity ;
empirical power
|
Abstract:
|
Clinical research studies comparing 2 or more treatments often include a composite primary endpoint comprising several binary components, such as distinct postoperative complications. Ideally, components have similar clinical severity and frequency and are similarly affected by treatment. When not true this leads to serious interpretation issues. Although not a substitute for prudent planning, multivariate (i.e., multiple outcomes per subject) methods are generally more powerful, flexible, and easier to interpret than standard methods which collapse the components into a single number. The multivariate framework facilitates severity weighting of components, assessment of heterogeneity of effects across components, and assuming either a common or distinct treatment effects. We highlight a multivariate distinct effects test of the average of the relative treatment effects (ARTE) across components. This improves existing multivariate methods by not allowing the overall effect estimate to be driven by highest frequency component(s), a key issue with composite endpoints. We assess the relative powers of the methods through simulations. Methods are demonstrated using 2 clinical trials.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2014 program
|
2014 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Professional Development program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.