Memorandum submitted by the Leeds Institute
of Diagnostics and Therapeutics, University of Leeds (HO 15)
1. In addressing the questions posed in
relation to the Government's use of evidence in policy-making
regarding homeopathy; (1) What is the policy? (2) On
what evidence is the policy based? it is important to recognise
that clinical evidence is characterised by its incremental and
sometimes less than perfect nature. Many clinical decisions are
made on the balance of probabilities suggested by the evidence,
rather than clear, unequivocal evidence to support particular
decisions. Scientific evidence is supplemented by clinical experience
and knowledge when clinical decisions are made in practice. Evidence
used for policy-making is no different in that it too can only
represent the best information available at a particular time,
and equally must be balanced against other types of evidence,
including economic and ethical considerations.
1. The proposed "evidence check"
for homeopathy should therefore be undertaken within the context
of our wider understanding of the nature and value of evidence
in other clinical spheres.
2. In particular, it is important to have
a clear framework for assessing the existing evidence which takes
into account the gradations of certainty that are apparent in
any systematic review of the literature. One such framework would
be that provided by the British Medical Journal's "Best Health"
http://besthealth.bmj.com This Web-based project
aims to inform patients and practitioners of the extent and certainty
of existing scientific evidence for particular treatments and
conditions. The following is extracted from this website;
2.1. Best Health looks at medical
research that is published in journals all over the world. It
does this by using Clinical Evidence, a collection of the
best research evidence for doctors. Clinical Evidence gives
doctors and other health care workers a good, up-to-date summary
of what's known and what isn't about treating a wide range of
clinical conditions. It's published by the BMJ Publishing Group.
2.2. Clinical Evidence looks at
all the evidence and decides how well treatments work, whether
the research is good enough and how serious the side effects are.
Sometimes no one knows for certain whether a treatment works because
the research that's been done isn't good enough. Or it could be
that not enough research has been done.
2.3. Best Health adds to the Clinical
Evidence research. It makes the evidence easy to read. It
also enables patients to see the same research evidence that doctors
see. Clinical Evidence gives doctors and other health care
workers a good up-to-date summary of what's known and what isn't
about treating a wide range of clinical conditions. It's published
by the BMJ Publishing Group.
2.4. We follow a strict process to develop
each topic on Best Health. Here are the key steps:
2.4.1. Step 1: Selecting a topic
2.4.2. Best Health covers serious,
long-lasting illnesses that affect many people in the UK. It also
looks at more minor conditions that affect a lot of people, such
as coughs and colds. We are guided by national health statistics,
doctors and patient groups. The conditions we look at have been
included in Clinical Evidence.
2.4.3. Step 2: Asking the right questions
2.4.4. We cover the treatment options for
each condition and give background information to explain the
condition itself. Best Health works with the Clinical
Evidence team, an international team of doctors, and patient
groups to find out what matters most to doctors and patients.
They might ask questions such as: What does the research say about
exercise helping people with heart failure? What are the side
effects of treatments for childhood asthma?
2.4.5. Step 3: Finding the evidence
2.4.6. All our information is based on research
evidence and high-quality medical papers. Here is how we gather
2.4.7. Information about treatmentsThis
information in Best Health is based on Clinical Evidence.
To answer each question about a treatment, the Clinical Evidence
medical information specialists do a thorough search for studies
that measure how well treatments work. First the information specialists
look for the best types of studies (called systematic reviews)
and other good-quality studies called randomised controlled trials.
If there are none of these studies, the information specialists
look for other studies and say how much they can be relied on
and what problems there are with the research.
2.4.8. Once the research has been collected,
the information specialists weigh up the evidence and take out
the studies that aren't good enough. They do this using a method
developed by experts in how research is carried out.1 2 .This
thorough research helps us find out which treatments work best
for a condition, and also why certain treatments work. If you
would like to read more about how we search for and select studies,
see the Clinical Evidence website (http://clinicalevidence.bmj.com).
2.4.9. Information about conditionsThe
information that we provide to explain medical conditions is based
on high-quality original medical papers and textbooks chosen by
our information specialists. On each page of the site, you will
find the details of the sources of information we have used.
2.4.10. Step 4: Making sense of the evidence
2.4.11. The research evidence for each treatment
is studied and summarised by a doctor who is an important expert
in a particular specialty. Each topic is then checked by at least
three more doctors. Then, a leading expert provides advice on
how doctors can use this research evidence. We ask people with
the condition to tell us what they think the important questions
are about their condition and treatments.
2.4.12. A team of experienced medical writers
makes sure this evidence can easily be understood by the general
public and writes the extra information that explains each condition.
2.4.13. Deciding which treatments workWe
group treatments into categories according to how good the evidence
is that they work. We use slightly different language to describe
the categories than you'll find in Clinical Evidence, but
the treatments are grouped in the same way. Here is an explanation
of what each category means:
|What it means
|Treatments that work
|There's clear evidence from randomised controlled trials that the treatment works. Also, the evidence shows that the chance of problems is small compared with the benefits.
|Treatments that are likely to work
|There is some evidence that the treatment works. But we can't be as certain that the treatment works as we can for those listed under "Treatments that work".
|Treatments that work, but whose harms may outweigh benefits
|There's some good evidence that the treatment works. But there's also good evidence that it can have serious side effects. Doctors and patients need to weigh up the benefits and risks according to what each person needs and wants.
|Treatments that need further study
|We don't know if the treatment is effective because there is either too little research to tell or the quality of the research is not good enough.
|Treatments that are unlikely to work
|There is evidence that the treatments probably don't work. But we can't be as certain that the treatments don't work as we can for the ones in the group "Treatments that are likely to be ineffective or harmful".
|Treatments that are likely to be ineffective or harmful
|Clear evidence shows the treatments don't work or will be harmful.
2.5. Step 5: Presenting the answers
All the information on Best Health is edited by a
team of editors and checked by our doctors. The information about
drugs has been reviewed by a team of qualified pharmacists working
in association with PharmacyHealthLink. PharmacyHealthLink is
a leading national charity that works to improve the health of
the public through the expertise of pharmacists and their staff.
2.6. Sources for the information on this page:
Sackett DL, et al. Clinical epidemiology: a basic
science for clinical medicine. Little, Brown and Co, Boston, USA;
Jadad A. Randomised controlled trials. In: Assessing the
quality of RCTs: why, what, how and by whom? London, UK; 1998.
3. Using this framework, the BMJ group have assessed
2,500 commonly used treatments and their summarised findings
are shown in the figure below.
4. Again, it seems appropriate that the committee's deliberations
regarding the evidence relating to homeopathy are conducted with
reference to the larger picture regarding the imperfect and emerging
evidence base informing policy for commonly provided treatments
within the NHS.
5. While evidence-based policy is a laudable goal, something
to be strived for, it can only happen in the prevailing climate
of imperfect and emerging knowledge. If, as seems likely, an unacceptable
gap is identified between the level of reported use of homeopathy
and the evidence available to help inform the public or the NHS
of its value, the most objective and ethical way forward would
be to support the generation of high quality research findings
to close this gap. The current structures of the NIHR are adequate
to facilitate this.
6. The above recommendations for further research echo
those made in relation to homeopathy in the GO-Science Review
of the Department of Health:
]. Flagship trials should be run in the most
promising areas, chosen on plausibility, and patient demand. [
The Health Technology Assessment Programme provided a framework
that should be as applicable to research on homeopathy as to any
GO-Science Review of the Department of Health, Annex 1 (2008).
Government Office for Science: Department for Innovation, Universities
and Skills; Paragraph 3.16.
7. DECLARATION OF
The author of this submission, Professor Katharine Thomas, is
an academic researcher at the University of Leeds; she is not
a homeopathic practitioner, and has no financial interest in the
provision of homeopathy.
Katharine Thomas BA (Hons) MA
Professor of Complementary and Alternative Medicine Research
Leeds Institute of Diagnostics and Therapeutics
University of Leeds