Equivalence test

An equivalence test is a test that allows to conclude, with a specified confidence level, equivalence between observations.

When using equivalence tests, you must specify how large of a difference between group averages would represent a clinically or practically important (significant) difference, or how large a difference can be to be still considered insignificant. These limits must be decided beforehand and are based on scientific and clinical judgment, or industry-specific standards.

Equivalence may be concluded when the observed differences are smaller than the prespecified maximum insignificant differences. When differences are larger than the prespecified limits, equivalence is rejected.

In MedCalc, Schuirmann’s (1987) two one-sided tests (TOST) approach can be used to test equivalence. To do this, you

• define the range of differences that are not clinically or practically important (i.e. not significant)
• depending on the experimental design, use the Independent samples t-test or Paired samples t-test
• select the option for a 90% (!) Confidence Interval
• if the entire range of the 90% Confidence Interval of the mean difference lies within prespecified range of indifference, then conclude with 95% confidence that the two treatments are equivalent.

Literature

• Schuirmann DJ (1987) A comparison of the Two One-Sided Tests Procedure and the Power Approach for assessing the equivalence of average bioavailability. Journal of Pharmacokinetics and Biopharmaceutics 15:657-680.

See also

• Independent samples t-test
• Paired samples t-test
• Bland-Altman plot

External links

• Equivalence test on Wikipedia