Online Program Home
My Program

Abstract Details

Activity Number: 476 - SPEED: Clinical Trial Design, Longitudinal Analysis, and Other Topics in Biopharmaceutical Statistics
Type: Contributed
Date/Time: Wednesday, August 1, 2018 : 8:30 AM to 10:20 AM
Sponsor: Biopharmaceutical Section
Abstract #330467 Presentation
Title: Analysis of Multiple Thresholds in a Responder Analysis of Patient-Reported Outcome Measures
Author(s): Lysbeth Floden* and Melanie L Bell and Stacie Hudgens
Companies: Clinical Outcome Solutions and University of Arizona and Clinical Outcome Solutions
Keywords: Patient Reported Outcomes; Responder Analysis; MCT; meaningful change threshold; empirical cumulative distribution function
Abstract:

Background: The desire to interpret randomized controlled trial (RCT) results in a meaningful way for various stakeholders has led to the responder analysis (RA). Responders are generally defined as patients whose change from baseline meet or exceed the meaningful change threshold (MCT). Criticisms include the potentially arbitrary responder definition. RA evaluates responder proportions, usually by a chi-square test. The empirical cumulative distribution function (eCDF) is an informative presentation of all responder definitions and group differences can be evaluated with Kolmogorov-Smirnov (KS) test. Methods: We simulated a continuous outcome from a 2-arm RCT and compared power of the KS test with multiple responder definitions to the chi-square test of a single definition. We identified distribution characteristics of the change score that are more likely to show group differences. Finally, we present a case-study from an open-label RCT. Results: The KS test had an advantage in power over the chi-square test when the variance of the change score differed between groups. The shape of the eCDF determined when the KS finds differences in proportions of those improving.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program