Trials to assess equivalence: the importance of rigorous methodsBMJ 1996; 313 doi: https://doi.org/10.1136/bmj.313.7048.36 (Published 06 July 1996) Cite this as: BMJ 1996;313:36
All rapid responses
This is to state/argue that to evaluate any indigenous
treatments/drugs, equivalence trials are more suitable. As well known, the
aim of 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.
Let me inform about "How this thinking started?" It was learned that
"requirement is to establish that traditional drugs are equally
effective" but they are to be preferred for other advantages (like easy
availability, less cost, less side-effects, easy administration, etc.).
Should not we plan "equivalence trials" and not classical superiority
trials in such situations?
HO : Effect of Treatment A = Effect of Treatment B i.e. HO : A = B
and H1 : A is not equal to B (two-sided H1) or A > B or A <B (one-sided H1)
This HO is tested (by appropriate test). If we "do not reject HO (i.e.
test statistic is not significant at given alpha level) we say that
"evidence is not enough to prove A=B". This is "lack of evidence" of
equivality. However, "absence of evidence is not evidence of absence".
Therefore, if you intend to prove "equivality" the answer is
In short, (and the fact is) : Indigenous treatment may be equivalent to
modern active comparator w.r.t. Efficacy. Nevertheless, superior (& so
preferable) w.r.t. Safety, Cost, Availability, Ease in administration,
etc. There could be more active comparators and or more new drug
formulations to be tested in one trial. In any case, analysis should be
based on "confidence intervals" and this also carries implications for
the estimation of the required number of patients at the design stage.
Best references are1,2, however, few important methodological details are
described in literature3,4.
1. Jones B., et al. (1996) Trials to assess equivalence : the importance
of rigorous methods. Br.Med.Jr., vol. 313, pp36-39.
2. Stefan Wellek (2002) Testing statistical hypotheses of equivalence. CRC
3. Sarmukaddam S.B. (2006), Fundamentals of Biostatistics. Jaypee Brothers
Medical Publishers Ltd., New Delhi, India.
4. Indrayan A. and Sarmukaddam S.B. (2001), Medical Biostatistics.
Marcel Dekker, Inc., New York, USA.
Sanjeev B. Sarmukaddam
Maharashtra Institute of Mental Health,
Sassoon Hospital Campus, Pune, India.
Competing interests: No competing interests