How Generalized Pairwise Comparisons can help reducing sample size in clinical trials

Listen to the latest IDDI podcast on Generalized Pairwise Comparisons to learn how this statistical method helped reduce sample sizes in clinical trials 

Today’s podcast focuses on the nouveau statistical methodology called Generalized Pairwise Comparisons IDDI Podcast Generalized Pairwise Comparisons (GPC) and how it can help reduce sample sizes in clinical trials. 

The GPC method fulfills a recent literature trend that attempts addressing a common concern for the analysis of randomized clinical trial data.

While multiple outcomes of interest are typically measured on patients, traditional statistical investigations only concentrate on a ‘’primary’’ outcome. All other outcomes are then either only considered as of secondary importance, or not considered at all in the analysis.

The core idea of GPC is to allow patients and clinicians to elicit among the outcomes of the trial, an order of importance that is most relevant to them. The outcomes can be of any type (categorical, continuous, or time-to-event), and can include efficacy, toxicity, quality of life, or cost-related information.

KEY LEARNING OBJECIVES:

  • What is GPC and how this can help reducing the sample size in clinical trials
  • How was the method applied to the specific study designed by Institut Jules Bordet

SPEAKERS:

  • Francesco Sclafani, Head of Clinic, Department of Gastrointestinal Oncology, Institut Jules Bordet
  • Mickaël De Backer, Senior Research Biostatistician, IDDI