Equivalence trial
An equivalence trial tests whether two treatments differ by less than clinically justified limits in both directions.
Why this level?
Participants or clusters are randomly allocated to concurrent comparison groups. NHMRC 2009 sourceRandom allocation supports a treatment comparison, but equivalence also depends on justified limits, sensitive measurement, adherence and credible handling of missing outcomes.
Fit
When to use it, and when not to
Use it when
- You need to exclude important differences in either direction.
- A new treatment may offer practical benefits even if its clinical effect is similar.
Not when
- You need to exclude unacceptable loss in only one direction. Use a noninferiority trial. Read: Noninferiority trial
- Your aim is to show benefit. Use a superiority trial. Read: Randomized controlled trial
Before you start
What you need
- The outcome, time point and treatment contrastDefine the population, scale and handling of treatment changes.
- Two clinical equivalence limitsJustify the limits with external evidence before enrollment or outcome review.
- A margin-based sample-size planUse the expected difference, variance, limits and chosen error rates.
- Analysis populations and missing outcomesPrespecify assignment-based and adherence-based questions with their assumptions.
- The trial could detect an important difference.How to check: Protect treatment contrast, adherence and outcome measurement quality.
- The limits are clinically defensible.How to check: Explain why differences inside them are acceptable for the intended use.
- Missing outcomes have a stated model.How to check: Justify missing at random if used; assess departures that favor apparent similarity.
Plain language
How it works
Randomize and conceal allocation as in other trials. Define the clinical limits before outcomes are known. Preserve study conduct that could reveal a real treatment difference. Plan sample size using the margin-based test.
Piaggio and colleagues explain reporting for equivalence trials; Schuirmann develops the two one-sided testing rule. The trial design includes its margin and conduct, while the linked analysis page explains the test.
Worked example
A worked example
- Setting
- A large randomized trial compares two ways of delivering rehabilitation.
- Question
- Can the mean outcome difference be kept within prespecified limits of -5 and 5 points?
- Approach
- Assume a difference of 1 point and standard error of 2. Use normal Wald tests with one-sided alpha 0.05.
- Result
- The matching 90% interval is -2.29 to 4.29 points. The larger one-sided p-value is 0.023.
- What it means
- Both bounds lie inside the limits, supporting equivalence for this outcome and margin. The result does not prove identical effects.
In practice
Recent studies PubMed search · not vetted
Watch out
Common mistakes
The one to remember
A nonsignificant superiority test does not establish equivalence; poor adherence or missing outcomes can hide important treatment differences.
- Selecting limits after seeing results.Instead: Choose limits from clinical needs and external evidence first.
- Assuming intention-to-treat always gives a conservative equivalence test.Instead: Examine adherence, missingness and the assumptions behind each analysis population.
Inside Mallard
How Mallard uses this
Plans do not link to this design yet. Learn search and the Atlas offer it as a reference.
design.equivalence_trial, the id a plan cue records.How to report it
Use the CONSORT extension for noninferiority and equivalence trials. Report both prespecified limits and their rationale, allocation, sample-size assumptions, analysis populations, adherence, missing outcomes and confidence interval. A reporting guideline is not a design manual: use it to make the study transparent, not as a substitute for careful design. Check the target journal's own requirements as well, which this page cannot know.
Related Learn pages
Related designs and methods
Evidence
Sources & further reading
- Reporting of noninferiority and equivalence randomized trials: extension of the CONSORT 2010 statement (opens in a new tab) · JAMA · doi 10.1001/jama.2012.87802
- A comparison of the two one-sided tests procedure and the power approach for assessing the equivalence of average bioavailability (opens in a new tab) · J Pharmacokinet Biopharm · doi 10.1007/BF01068419