Incorporation of Clinical Meaningfulness into the Analysis of a Continuous Variable to Improve Efficiency

• Activity Number: 591
• Type: Topic Contributed
• Date/Time: Thursday, August 7, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Qi Jiang*+ and Steven Snapinn
• Companies: Amgen and Amgen
• Keywords: responder ; dichotomization ; continuous ; clinically meaningful ; weighting

Abstract

The responder analysis, which involves dichotomization of a continuous outcome variable to classify subjects as responders or non-responders, is an attempt to evaluate whether the effect of a treatment on the analysis variable is clinically meaningful. Due to a large cost in power, critics have argued that the analysis should be based on the continuous variable. However, a flaw in the analysis of the continuous outcome variable is that it treats changes in that variable as equally clinically meaningful across the entire scale, when in fact changes in some parts of the scale may be more meaningful than in others. In this presentation I will discuss the concept of a clinical meaningfulness weighting function, and show how it can be used with the continuous outcome variable to produce an analysis that retains much of the power of the analysis of continuous variable, and evaluates whether the treatment effect is clinically meaningful in a way that the responder analysis does not.