Contents
Academy · Research Writing · Non-inferiority vs superiority vs equivalence
Non-Inferiority, Superiority and Equivalence Trials: Hypotheses, Margins and Design
These three trial types are sometimes treated as variations of a single design. Each, however, addresses a structurally different question, and the chosen margin determines what a result can legitimately claim.
A superiority trial examines whether a new treatment is better than a comparator. A non-inferiority trial examines whether it is not unacceptably worse, within a pre-specified margin. An equivalence trial examines whether it falls within a margin of similarity in either direction. Each requires a different hypothesis, margin and analytic approach.
Learning objectives
- How the hypothesis and margin differ across the three trial types.
- Why the non-inferiority margin has to be justified clinically and statistically, not chosen arbitrarily.
- Why a failed superiority trial cannot simply be re-labelled as a non-inferiority result after the fact.
Common misconceptions
- “A trial that fails to show superiority has effectively shown non-inferiority instead.” Christensen and Lesaffre both note that a non-inferiority conclusion requires a pre-specified margin and a trial designed and powered to test against it; a superiority trial that fails to reach statistical significance has not demonstrated non-inferiority but has produced an inconclusive result, because the margin and analysis required to support a non-inferiority claim were never defined or tested.
- “The non-inferiority margin can be set wherever makes the trial easiest to pass.” The margin must be justified clinically (the largest difference considered acceptable) and statistically (sufficiently narrow that, where the comparator’s own effect over placebo is known, non-inferiority to the comparator still implies meaningful superiority over no treatment); an arbitrarily wide margin undermines the entire basis of the comparison.
Hypothesis tested in superiority trials
The null hypothesis of a superiority trial is that there is no difference between treatments, and the trial aims to reject that null hypothesis in favour of the new treatment being better. This is the default and most familiar trial design, and it is appropriate when there is genuine reason to expect, and clinical value in demonstrating, an improvement over the existing standard1.
Hypothesis tested in non-inferiority trials
A non-inferiority trial is appropriate when the new treatment is not expected to be more effective but may offer other advantages (fewer adverse effects, lower cost, simpler administration), and the aim is to demonstrate that it is not worse than the comparator by more than a pre-specified margin. Christensen emphasises that this margin must be set before the trial and justified both clinically (the largest loss of effect still considered acceptable) and statistically, and that the per-protocol analysis is typically reported alongside the intention-to-treat (ITT) analysis, because the conservative bias of ITT towards no difference can impede the detection of genuine inferiority2.
Hypothesis tested in equivalence trials
An equivalence trial, which is less common than the other two designs, tests whether a treatment falls within a specified margin of the comparator in either direction, being neither meaningfully better nor meaningfully worse. Lesaffre notes that this design is typically used for generic or biosimilar comparisons, in which the aim is to demonstrate therapeutic equivalence rather than a directional difference, and that it requires a two-sided margin rather than the one-sided margin used in non-inferiority testing1.
Key points
Key points
Superiority, non-inferiority and equivalence trials test structurally different hypotheses with different pre-specified margins, and a trial designed for one cannot be reinterpreted as having answered another after the fact — especially a failed superiority trial, which does not become a non-inferiority finding by default.
Practical application
Practical application
- The trial type should be decided and stated before the trial begins. Superiority, non-inferiority or equivalence should be pre-specified in the protocol, together with the corresponding margin where applicable, rather than inferred after the results are known.
- The non-inferiority or equivalence margin should be justified explicitly. The clinical and statistical reasoning underlying the chosen margin should be stated, with reference to the comparator’s established effect size where relevant.
- Both intention-to-treat (ITT) and per-protocol analyses should be reported for non-inferiority trials. Both analyses should be presented and their agreement discussed, since either alone can be misleading for this trial type.
Frequently asked questions
Yes. With appropriate statistical adjustment, hierarchical testing from non-inferiority to superiority is an accepted approach that can be pre-specified; the reverse, inferring non-inferiority from a failed superiority trial not designed for that purpose, is not.
The margin should be justified with input from clinical experts on the loss of effect that remains acceptable and from statisticians on its relationship to the comparator’s established effect, and it is often informed by regulatory guidance for the specific disease area.
They are less common than in generic or biosimilar development, in which demonstrating equivalence, rather than superiority, to the reference product is often the specific regulatory requirement.
References
- Lesaffre E. Superiority, equivalence, and non-inferiority trials. Bulletin of the NYU Hospital for Joint Diseases. 2008;66(2):150-154. read 10 October 2026. View on PubMed
- Christensen E. Methodology of superiority vs. equivalence trials and non-inferiority trials. Journal of Hepatology. 2007;46(5):947-954. read 10 October 2026. View on PubMed
Every source above is a peer-reviewed article, a publisher’s record of one, or a standard issued by the body that owns it. Each carries the date it was last read. Where a standard moves, this article is revised rather than left up with a note.
How to cite this article
Pick your referencing style and copy it. Your university or target journal may vary the punctuation — check its guide if it has one.
Pubrica Academy. (2026, October 10). Non-Inferiority, Superiority and Equivalence Trials: Hypotheses, Margins and Design. Pubrica. https://pubrica.com/academy/concepts-definitions/non-inferiority-superiority-equivalence/
Pubrica Academy. “Non-Inferiority, Superiority and Equivalence Trials: Hypotheses, Margins and Design.” Pubrica, 10 Oct. 2026, pubrica.com/academy/concepts-definitions/non-inferiority-superiority-equivalence/.
Pubrica Academy (2026) Non-Inferiority, Superiority and Equivalence Trials: Hypotheses, Margins and Design. Pubrica. Available at: https://pubrica.com/academy/concepts-definitions/non-inferiority-superiority-equivalence/ (Accessed: 15 October 2026).
Pubrica Academy. “Non-Inferiority, Superiority and Equivalence Trials: Hypotheses, Margins and Design.” Pubrica. October 10, 2026. https://pubrica.com/academy/concepts-definitions/non-inferiority-superiority-equivalence/
Pubrica Academy, ‘Non-Inferiority, Superiority and Equivalence Trials: Hypotheses, Margins and Design’ (Pubrica, 10 October 2026) <https://pubrica.com/academy/concepts-definitions/non-inferiority-superiority-equivalence/> accessed 15 October 2026.
Pubrica Academy. Non-Inferiority, Superiority and Equivalence Trials: Hypotheses, Margins and Design. [Internet]. Pubrica; 2026 Oct 10 [cited 2026 Oct 15]. Available from: https://pubrica.com/academy/concepts-definitions/non-inferiority-superiority-equivalence/
Pubrica Academy, “Non-Inferiority, Superiority and Equivalence Trials: Hypotheses, Margins and Design,” Pubrica, Oct. 10, 2026. [Online]. Available: https://pubrica.com/academy/concepts-definitions/non-inferiority-superiority-equivalence/
@misc{non_inferiority_superiority_equivalence_2026,
author = {{Pubrica Academy}},
title = {Non-Inferiority, Superiority and Equivalence Trials: Hypotheses, Margins and Design},
howpublished = {Pubrica},
year = {2026},
month = {oct},
url = {https://pubrica.com/academy/concepts-definitions/non-inferiority-superiority-equivalence/},
note = {Accessed: <span data-cite-acc="iso">2026-10-15</span>}
}This article is written to be learned from. Cite it if you are referring to what it says; if you are using the evidence behind it, cite the sources listed above instead.
Trial-type design check
For use when selecting and documenting the comparative design of a trial.
[ ] Trial type (superiority/non-inferiority/equivalence) pre-specified in the protocol [ ] Margin stated and justified clinically and statistically, where applicable [ ] Hierarchical testing approach pre-specified, if switching from non-inferiority to superiority is intended [ ] Both ITT and per-protocol analyses planned for non-inferiority trials
Pubrica AcademyWritten and maintained by Pubrica’s subject matter experts. Articles are revised when the underlying standards change, and each article states the date on which it was last reviewed.
How Pubrica can help
Pubrica assists in selecting and justifying the trial design and margin appropriate to a specific comparative question before the protocol is finalised.