Evidence-appraisal glossary

Non-inferiority trial

A non-inferiority trial tests whether a new treatment is not meaningfully worse than an existing one, rather than better. Researchers set a pre-specified "non-inferiority margin," the largest acceptable loss of benefit, and conclude non-inferiority only if the new treatment stays within that margin.

Also called: Noninferiority trial, Non-inferiority study, Non-inferiority design.

What it is

Most trials ask "is the new treatment better?" A non-inferiority trial asks a different question: "is it not unacceptably worse?" This design suits a new option that offers other advantages (cheaper, safer, easier to take) if it roughly matches a proven comparator on effectiveness.

How to read one

  • Find the margin. The non-inferiority margin is the largest drop in benefit the researchers agreed to tolerate before calling the new treatment worse. It should be pre-specified and clinically justified, not chosen to make results look good. A wide margin makes non-inferiority easy to "prove."
  • Check the confidence interval. Non-inferiority is shown when the whole confidence interval for the difference stays on the acceptable side of the margin. Compare the CI to the margin, not just the point estimate.
  • Watch the analysis set. In superiority trials, intention-to-treat is conservative; in non-inferiority trials, sloppiness and dropout can falsely blur two treatments together and favor "non-inferior," so both ITT and per-protocol analyses should agree.
  • Confirm the comparator truly works, or "non-inferior" can mean equally ineffective.

This is a plain-language methodology definition for reading research. It is general education, not medical advice.

Back to the glossary