BJGP Back pages-Master[1]


738 British Journal of General Practice, October 2008

Book review
TRICK OR TREATMENT?
ALTERNATIVE MEDICINE
ON TRIAL
SIMON SINGH AND
EDZARD ERNST
Bantam Press, 2008
HB, 352 pages, £16.99, 9780593061299

The main conclusions of Ernst and Singh’s
very thorough and clever book are that:

• there is no evidence to justify any claim for
the specific effectiveness of any except a
very few specific complementary and
alternative medical (CAM) treatments in
specific conditions, nor of CAM treatments
overall;

• the perceived benefits are explained by
other dynamics of the therapeutic process,
particularly the placebo effect; and

• these benefits, even if substantial in terms
of the patient’s experience, should be
disallowed because they are unethical and
dishonest; they involve lying to patients
about the nature of the treatment.

These conclusions are supported by clear
discussion of the nature and virtues of
evidence-based medicine and the scientific
method that underpins it, and an overview of
research evidence in CAM that demonstrates
the absence of effectiveness. Many with
research experience in this field will argue
that this overview is partial and by no means
definitive, but my interest in the book is more
general.
The authors set great store by the pursuit

of truth and knowledge, and the role of
science in that enterprise. The book is
explicitly dedicated to the pursuit of truth,
and claims a balanced presentation of the
facts in pursuit of the truth of the matter, in
this case the validity of CAM as a
contribution to health care. I applaud that
aim. Truthfulness is an essential attribute of a
therapeutic relationship and an essential
attribute of the scholarship that informs
clinical practice. But it is in this regard that I
am critical of ‘Trick or Treatment? Alternative
Medicine on Trial’. It is full of data about
CAM, but it does little to advance our
knowledge of medicine, particularly our

knowledge of its role in human healing. That
is disappointing. And although I don’t doubt
the authors’ desire for truth, I do doubt their
hold on truth. They display great certainty in
their pronouncements, but there is a
difference between truth and certainty. Both
are precarious, but those engaged in the
pursuit of truth know it. Those possessed of
certainty often don’t. It takes wisdom and
discernment to tell the difference, and on
those counts I find this book lacking.
The book provides excellent counsel about

the shortcomings of CAM (and there are
many, if you take the whole nebulous field into
consideration), and its susceptibility to
popular and commercial exploitation. But a
recurring lack of truthfulness is the lack of the
perspective that would have been provided
by relating these to comparable problems in
conventional medicine. For example, risk is
an ever-present concern in all medicine, and
Ernst and Singh rightly highlight the risks of
CAM. Unfortunately they do not present
those risks in relation to the risks of
conventional medicine; nor discuss the rarity
of the CAM disasters they cite. Medical risk is
direct (harm caused by the intervention), or
indirect (harm resulting from neglect of some
more necessary or appropriate intervention).
In conventional medicine the predominant
risk is direct — adverse, for example, drug
reactions, anaesthetic or surgical errors. But
indirect risk also occurs. A simple example is
inappropriately prescribing an antibiotic for a
self-limiting viral infection rather than
explaining the condition and encouraging an
expectant and self-sufficient attitude;
compounded by the direct risk of an adverse
reaction and the further indirect risk of
developing antibiotic resistance. Iatrogenic
harm from direct risk in conventional medicine
is regrettably frequent, and can be fatal. In
CAM it is very rare. Indirect risk in CAM is a
more serious problem and comparable to
indirect risk in conventional practice.
This is one example of a lack of

truthfulness and balance in the book. If CAM
lacks compelling evidence of specific
efficacy, patients may legitimately prefer it to
a treatment of proven efficacy that carries a
greater degree of direct risk. More so if the
training of the practitioner guards against

indirect risk; increasingly the case in the best
established therapies.1

This use of the word ‘efficacy’ points up
another evasion. Throughout, the word
‘effectiveness’ is used in discussing the lack
of positive evidence from formal trials of
CAM. Formal trials usually test efficacy, the
ability of an intervention to do what is
intended in ideal test conditions.
Effectiveness is what happens in an actual
clinical context. ‘Efficacy’ tests narrowly
defined specific effects. Effectiveness is
more relevant to what happens in the real
world and really more interesting. Trick or
Treatment? Alternative Medicine on Trial is
generally challenging efficacy, not
effectiveness.
The book disappoints me in other ways

that are in a sense a retreat from truth, or
timidity in making an adventure in truth, a
truly scientific exploration of reality. Firstly, its
determined exposure of the negatives of
CAM neglects the positives. Surely we
cannot be so foolish as to assume that the
modern western medical scientific model
and paradigm are the whole truth, a
complete and sufficient answer to the ills of
mankind? In which case we need to enquire,
and the proper scientific attitude requires us
to enquire deeply into the phenomena that
CAM describes — insights into the complete
dynamics of illness and healing,
comprehensively documented in the best
CAM literature but wholly inadequately
researched, that lie outside the frame of
reference of the prevailing biomedical model,
and that could really enrich knowledge and
be a stepping-stone to truth.
Another disappointment is disregard of the

implications of what CAM reveals of the
power and importance of non-specific and
placebo effects; acknowledged on one hand
as a strength of the method, and on the other
dismissed as a form of trickery. Treatments
‘proved’ to be ‘nothing but placebos’ are
represented as a deceit practised on the
patient by dishonest or lazy doctors. (What
about the inappropriate prescribing of
conventional drugs?)
Fortunately this attitude is redressed by the

focus on the value of placebo in the edition of
the British Medical Journal that coincided with



British Journal of General Practice, October 2008 739

Digest

the publication of Trick or Treatment?
Alternative Medicine on Trial.2 But here the
possibility of what might be learned from
CAM about our ability to stimulate self-
regulating and self-healing mechanisms
whose pervasive role in medicine Ernst and
Singh acknowledge, is tragically neglected.
Dismissing the results of the Bristol
Homeopathic Hospital clinical outcome study
on the grounds of explanations other than the
effect of homeopathic medicines (some of
which are tendentious and don’t reflect a
diligent study of the research paper), they
ignore the core fact that for whatever reason
many patients with a long and continuing
history of significant illness despite
conventional treatment did remarkably well.3

The question is not ‘do CAM treatments
show efficacy in formal trials’, but ‘do they
enhance patients’ whole wellbeing and
quality of life?’ And if so how? And by what
means can we make better use of whatever
is the answer to ‘how’? And what more can
we learn about health, illness, and the
human condition in the process? Surely this
should be the agenda of an academic
department of complementary medicine?
Surely its original remit to investigate the
scientific basis of CAM did not preclude
this? Surely it was not merely to investigate
whether CAM can be fitted into the same
box as conventional medicine; made to fit
the prevailing medical model?
To conclude, this book offers a salutary

critique of the problems that beset the
investigation and integration of CAM in
relation to conventional medicine, but by a
more genuinely truthful and truly scientific
examination of its subject it could have
enlarged our vision of health care.
Unfortunately it only serves to perpetuate a
kind of scientific tunnel vision.

Jeremy Swayne

REFERENCES
1. Healthwork UK. National Occupational Standards for

Homeopathy. Bristol: Healthwork UK, 2000.

2. Godlee F. Reclaiming the placebo effect. BMJ 2008;
336(3May): doi:10.1136/bmj.39567.551181.47
http://www.bmj.com/cgi/content/full/336/7651/0?rss=1
(accessed 5 Sep 2008).

3. Spence DS, Thompson EA, Barron SJ. Homeopathic
treatment for chronic disease: a 6-year, university-
hospital outpatient observational study. Journal of
Alternative and Complementary Medicine 2005; 11(5):
793–798.

DOI: 10.3399/bjgp08X342525

EDZARD ERNST — A RESPONSE

Trick or Treatment? Alternative Medicine on
Trial is not a book against alternative
medicine, it is a book in favour of good
evidence and single standards in health
care. Jeremy Swayne admits that
‘truthfulness is an essential attribute’ but
criticizes our book for lack of ‘wisdom and
discernment’, ‘lack of … balance’,
neglecting ‘the positives’ and disregarding
‘the power and importance of non-specific
and placebo effects’. He concludes by
accusing us of ‘scientific tunnel vision’.
These are strong words which require a
robust response.
Using our chapter on acupuncture as an

example, we can effortlessly demonstrate
that we bend over backwards to present
the current evidence as balanced and
truthfully as possible. We discuss the
history and complexities of clinical trials
and the best current scientific evidence in
much detail. The truth is that, according to
this evidence acupuncture might be
effective for some types of pain and
nausea. The truth is that even this evidence
is not fully convincing (NICE recently
recommended against using acupuncture
for knee osteoarthritis, the flag-ship
indication of acupuncturists!).1 The truth is
that it is not free of risks. The truth is that it
is expensive compared to many other
effective treatments.
The truth we present is based on the best

available evidence, which means that
sometimes we endorse therapies (for
example, several herbal treatments) and
sometimes we have to be very critical
about them (for example, magnet therapy).
But Swayne does not appreciate this type
of truth. Why? Presumably because, when
our truth is applied to his pet therapy,
homoeopathy, this form of treatment
appears like a sick joke.
In his endeavor to promote his type of

‘truth’, Swayne wants us to discuss the
failings and faults of conventional medicine.
We do not deny that these exist. Maybe,
one day, when we write a book about
conventional medicine, we will discuss
them in detail. But, writing about alternative
medicine, we think it is fair to mention them
only in passing.
And finally, Swayne points to the

importance of the placebo effect. But so do

we, even in much detail. The difference is
that we try to apply just one standard while
Swayne uses two. The placebo-effect is
clearly powerful, thankfully we both agree
on that. Swayne believes this justifies the
routine use of homeopathy and other
ineffective treatments. We point out that
you don’t need to administer a placebo to
generate a placebo response — effective
treatments do that too and they convey
specific effects as well. The logical
conclusion therefore is that, by using pure
placebos, we do our patients a grave
disservice.
So who is suffering from ‘tunnel vision’?

We suggest the readers of this debate go to
their library, take out our book Trick or
Treatment? Alternative Medicine on Trial
and make up their own mind.

Edzard Ernst

REFERENCE
1. National Institute for Health and Clinical Excellence.

Osteoarthritis NICE version — draft guidance for
consultation. NICE: London, 2007.
http://www.nice.org.uk/guidance/index.jsp?action=d
ownload&o=35844 (accessed 9 Sep 2008).

DOI: 10.3399/bjgp08X342534