Article Text

other Versions

PDF
Superiority trials: statistical trickery or mass blindness?
  1. Nusrat Shafiq,
  2. Samir Malhotra
  1. Department of Pharmacology, Postgraduate Institute of Medical Education & Research, Chandigarh, India
  1. Correspondence to Professor Samir Malhotra, Department of Pharmacology, Postgraduate Institute of Medical Education & Research, Sector 12, Chandigarh 160012, India; smal.pgi{at}gmail.com

Statistics from Altmetric.com

Non-inferiority trials are conducted to demonstrate whether a new treatment is not inferior to an existing treatment by more than a prespecified margin, known as the non-inferiority margin (δ). δ is thus a critical element in these studies and must be justifiable both statistically and clinically.1 We have recently argued that there should be a superiority margin similar to the non-inferiority margin in superiority trials.2 The current practice is that if 95% CI lies within the inferiority margin (δ) of the no-difference line, the drug is considered to be non-inferior. The drug is considered inferior only if the 95% CI lie outside the δ (figure 1).

Figure 1

Superiority margin similar to the non-inferiority margin showing the erroneousness of claiming superiority. From the results of the primary composite outcome (death from cardiovascular causes, non-fatal myocardial infarction, or non-fatal stroke) in the Empagliflozin, Cardiovascular Outcomes, and Mortality in Type 2 Diabetes study. The primary hypothesis was to show non-inferiority with a margin of 1.3. If superiority is to be claimed, the results …

View Full Text

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.