Analysis of Statistical Tests to Compare Cumulative Proportion of Responders for Pain Data in Clinical Trials

• Activity Number: 256
• Type: Contributed
• Date/Time: Monday, August 2, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Song Wang*+
• Companies: PPD, Inc.
• Keywords: cumulative ; proportion ; responders ; pain ; power

Abstract

The use of cumulative proportion of responders analysis (CPRA) to present pain data has gained popularity in recent years. The goal of this study was to evaluate the effectiveness of several analysis techniques (K-S test, Wilcoxon rank sum test, and cumulative odds ordinal logistic regression test) to compare the cumulative proportion of responders. Several datasets were generated containing acute or chronicle pain data using different pain measurement scales. Multiple samples were then drawn from these data and used to assess the type I error rate and the power of these tests. In general, all 3 tests controlled type I error rate well. The ordinal logistic regression test had the highest Type I error rate but was also the most powerful test; while the K-S test was least powerful with lowest type I error rate. Applications of these tests under various assumptions are discussed.