LETTERS TO THE EDITORS
Letter: all or nothing—placebo effects in a non-drug clinical
trial in IBS
I note with interest and surprise that the study reported by Schumann
is missing elementary issues that usually are required when pre-
senting clinical trial data, especially in IBS. This is a mono-centric study
conducted in a medical unit known for and specialised in complemen-
tary and alternative medicine (CAM) approaches, thus likely to attract
a sub-selection of IBS patients interested in CAM treatment. This is
neither mentioned nor discussed in the paper. Furthermore, this is a
head-to-head trial comparing two interventions of completely differ-
ent kind (yoga, diet), but the power calculation is not based on non-
(eg, yoga is not inferior to low FODMAP) but
on conventional superiority (of which trial arm?). The study cited to
calculate statistical power is yet another non-controlled three-arm
cross-over trial of another CAM IBS intervention conducted by the
same group, which is highly questionable. Furthermore, the total num-
ber of statistical tests (for psychometric scales and subscales) per-
formed (see Supplementary Material) is higher than the total number
of patients included. While statistical adjustment for multiple compar-
isons may not be needed for secondary endpoints, beta errors can still
account for many significant differences reported. In addition, the pri-
mary endpoint chosen is certainly not in agreement with what is
required in IBS drug trials by approval authorities (eg, FDA and EMA),
especially with mixed IBS-type patients.
It is, therefore, quite unusual
to not report efficacy data in relation to IBS subtypes (especially IBD-
D and IBS-C), even if non-significant.
Although this is a non-blinded study without a placebo control
arm, the report is not mentioning the possibility that a placebo
response may exist; it even ignores the term placebo. If CAM studies
included a control arm, the placebo response was usually in the range
of conventional drug trials or even higher (42% in a meta-analysis
The same holds true for most drug and dietary interventions in IBS.
As the authors should know, placebo responses in head-to-head trials
are usually higher than in placebo-controlled trials (add +15%,
because patients are randomised to two or more effective treatments
and not to placebo control (which is the ethical reason why such
designs are often chosen). In head-to-head drug trials, it may be
allowed to assume similar placebo response rates in both arms; how-
ever, comparing yoga exercise to a dietary intervention (or apple to
pears) may need further adjustments for which data are not readily
In summary, given the size of the (similarly) high responses in
both treatment arms (53% and 72% for clinically meaningful
improvement of symptom severity with yoga and low FODMAP diet,
respectively at week 12) and in light of the above, it is rather likely
that these are completely accountable as placebo effect.
quence, this is a negative study (primary endpoint not reached) that
is falsely interpreted as positive (“patients may benefit ...”). Such
interpretative problems have been found with many other small
studies, not only in IBS.
Declaration of personal interests: P.E. received an unrestricted grant
from SymbioPharm, served as advisor for Allergan, Almirall, AstraZe-
neca, Boehringer, Biocodex, Ferring, GA, Heel, SymbioPharm, TEVA,
and UCB and as speaker for Almirall, Bayer/Steigerwald, Biogene,
Boehringer, Chiesi, Heel and Sanofi.
This article is linked to Schumann et al and Schumann and Cramer
papers. To view these articles visit https://doi.org/10.1111/apt.
14400 and https://doi.org/10.1111/apt.14686.
Department of Internal Medicine VI, University of Tubingen,
AP&T correspondence columns are restricted to letters discussing papers that have been published in the journal. A letter must have a maximum of 500 words,
may contain one table or ﬁgure, and should have no more than 10 references. It should be submitted electronically to the Editors via http://
Aliment Pharmacol Ther. 2018;48:105–110. wileyonlinelibrary.com/journal/apt © 2018 John Wiley & Sons Ltd