Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 323 - Estimating Treatment Effects: Applications in Health Policy
Type: Contributed
Date/Time: Wednesday, August 5, 2020 : 10:00 AM to 2:00 PM
Sponsor: Health Policy Statistics Section
Abstract #312394
Title: An Approach to the Analysis of Composite Endpoints with No Prioritized Components
Author(s): Abdullah Masud* and John Schoenfelder
Companies: AbbVie and AbbVie
Keywords: prioritized clinical endpoints; composite endpoints; generalized Wilcoxon test; significance test; bootstrap procedure

In data analysis, when a variety of clinical endpoints are combined into a composite endpoint for evaluating the effect of treatment, several approaches for analysis are mainly based on prioritized clinical endpoints. By standard analysis framework of a composite endpoint, a subject is characterized as a winner or loser based on the outcome of the first prioritized endpoint. If that outcome is unknown, then the subject is characterized as a winner or loser depending on the outcome of the second prioritized endpoint, and so on. Yet, a subject can be categorized as winner or loser without ordering the endpoints. In this work, we construct a composite score using the generalized Wilcoxon testing procedure by combining all endpoints together and making comparisons within all available between-treatment pairs. Within each pair, the subject who performs better on every single component of the composite endpoint is assigned a better score. We propose a parametric significance and bootstrap testing procedures to evaluate the effect of treatment. We perform an extensive simulation study to assess the operating characteristics of the proposed methods.

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

Back to the full JSM 2020 program