Noninferiority Clinical Trials With Binary Outcome: Statistical Methods Used in Practice

Abstract

The use of noninferiority (NI) clinical trials continues to increase. At the same time, this study design remains complex with numerous issues related to the planning, implementation, and analysis of such trials. In this systematic review, we expand previous reviews, and evaluate reporting of the statistical methods used for design and analysis of NI trials with binomial proportions. We assessed 70 NI clinical trials, published between June 2017 and May 2018. Based on these trials, we outline the gaps in statistical reporting in the published articles, and provide recommendations to minimize the gaps. The recommendations, and the awareness they raise are a helpful addition to the efforts the research community is facing today in combating reproducibility and replicability issues.

KEYWORDS:

• Noninferiority design
• Replicability
• Reproducibility
• Review