Equivalence Clinical Trial

From GM-RKB
(Redirected from equivalence design)
Jump to navigation Jump to search

An Equivalence Clinical Trial is a Clinical Trial which primary objective is to show that participants' response to two or more treatment are essentially equivalent (i.e. clinically similar).



References

2021

  • (Wikipedia, 2021) ⇒ https://en.wikipedia.org/wiki/Glossary_of_clinical_research#E Retrieved:2021-12-26.
    • QUOTE: Equivalence trial
      • A trial with the primary objective of showing that the response to two or more treatments differs by an amount which is clinically unimportant. This is usually demonstrated by showing that the true treatment difference is likely to lie between a lower and an upper equivalence margin of clinically acceptable differences. (ICH E9)

2017

$H_0:\mu_1-\mu_0\leq-\delta$ vs. $H_1:\mu_1-\mu_0>-\delta$ (5)
where $\delta \geq 0$ and is also called the margin of clinical significance which is usually small(...)

"Equivalence” does not mean “equal” or "same" as in practice. When we say the treatment and the control are equivalent, we mean that they are "similar". By quantifying “Similarity” using a tolerance range, the hypotheses for an equivalence trial are specified as

$H_0:|\mu_1-\mu_0|\geq\delta$ vs. $H_1:|\mu_1-\mu_0|<\delta$

(6)
where $\delta > 0$ is a pre-specified tolerance margin. If the null hypothesis is rejected, then the mean difference of two groups is within the tolerance range and the treatment and control are equivalent.

A closer look at (6) shows the hypotheses in an equivalence trial are the same as

$H_0:\mu_1-\mu_0\leq-\delta$ and $\mu_0-\mu_1\leq -\delta$ vs. $H_1:\mu_1-\mu_0>\delta$ and $\mu_0-mu_1\geq>-\delta$ (6)
Comparing (5) and (6) we can see that the equivalence trial is the intersection of two non-inferiority trials. Intuitively, the treatment and control are equivalent, if and only if neither one is inferior to the other.

2015

2011

  1. Garret AD. Therapeutic equivalence: fallacies and falsification. Statist Med. 2003;22:741–762. doi: 10.1002/sim.1360.

2009