Trials to assess equivalence: the importance of rigorous methods

Abstract

The aim of an equivalence trial is to show the therapeutic equivalence of two treatments, usually a new drug under development and an existing drug for the same disease used as a standard active comparator. Unfortunately the principles that govern the design, conduct, and analysis of equivalence trials are not as well understood as they should be. Consequently such trials often include too few patients or have intrinsic design biases which tend towards the conclusion of no difference. In addition the application of hypothesis testing in analysing and interpreting data from such trials sometimes compounds the drawing of inappropriate conclusions, and the inclusion and exclusion of patients from analysis may be poorly managed. The design of equivalence trials should mirror that of earlier successful trials of the active comparator as closely as possible. Patient losses and other deviations from the protocol should be minimised; analysis strategies to deal with unavoidable problems should not centre on an "intention to treat" analysis but should seek to show the similarity of results from a range of approaches. Analysis should be based on confidence intervals, and this also carries implications for the estimation of the required numbers of patients at the design stage.

A randomised clinical trial of two docetaxel regimens (weekly vs 3 week) in the second-line treatment of non-small-cell lung cancer. The DISTAL 01 study

“…The present study, powered to rule out a 0.67 hazard ratio of death with the weekly schedule, suggests survival be very similar with the two compared schedules (HR of death 1.04), although, of course, interpretation of data in terms of equivalence (Jones et al, 1996) is not allowed.…”

Genetic nurse counsellors can be an acceptable and cost-effective alternative to clinical geneticists for breast cancer risk genetic counselling. Evidence from two parallel randomised controlled equivalence trials

“…As this was an equivalence trial, a confidence interval (CI) approach was used (Jones et al, 1996). The difference in outcomes was calculated, adjusting for differences in baseline scores using multiple linear regression.…”

“…For each outcome, the 95% CIs around the difference were calculated. An outcome was considered 'equivalent' when the 95% CI for the difference between the intervention and control group fell completely within a predetermined equivalence limit (Jones et al, 1996). Where the 95% CI for the observed difference fell completely outside the equivalence limit, we considered the outcomes would be 'nonequivalent'.…”

The clinical effectiveness of concise cognitive behavioral therapy with or without pharmacotherapy for depressive and anxiety disorders; a pragmatic randomized controlled equivalence trial in clinical practice

“…The sample size calculation (see (Meuldijk et al, 2012) indicated that 500 patients (alpha=0·05, 1 -β=0·80, two-sided) were required to demonstrate equivalence (Jones et al, 1996). Concise care will be deemed equivalent to standard care, if the 95% confidence interval (CI) for the observed difference in the proportions of treatment success (defined as a 50% reduction in BSI score compared to baseline BSI score) do not cross these predefined clinical margins of equivalence (∆) of -15% and + 15% including an acceptable difference of 5% (Meuldijk et al, 2012;Wiens, 2002;Jones et al, 1996).…”

“…Concise care will be deemed equivalent to standard care, if the 95% confidence interval (CI) for the observed difference in the proportions of treatment success (defined as a 50% reduction in BSI score compared to baseline BSI score) do not cross these predefined clinical margins of equivalence (∆) of -15% and + 15% including an acceptable difference of 5% (Meuldijk et al, 2012;Wiens, 2002;Jones et al, 1996). The equivalence margins are specified a priori on the basis of a clinical notion of detection, a minimally clinically important difference in effect (see Meuldijk et al2012).…”