Time to register randomised trials
BMJ 1999; 319 doi: https://doi.org/10.1136/bmj.319.7214.865 (Published 02 October 1999) Cite this as: BMJ 1999;319:865
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Dear Sir,
Horton and Smith make a cogent case for requiring all
clinical trials to be registered (BMJ, 2 October 199, p865-6). However,
this is not a complete solution to the underlying problem to which they
allude: the "chaotic" process by which randomised controlled trials come
into
being. It may prevent unnecessary duplication (as distinct from desirable
replication) but it will not address the real problem of health service
R&D: that is, its seeming irrelevance to those who manage resources
within the NHS (clinicians, health professionals who manage e.g. clinical
directors, and professional managers).
Within the UK the health service R&D endeavour has many features
(including the Cochrane Collaboration) in place which should promote
worthwhile work which impacts upon resource-use decisions nationally and
locally. However, this does not occur as well as it might because the mind
-set of those who control and do R&D is that of the scientist
rather than that of the resource manager. In essence, the scientists
largely define the issues to be explored and how the findings are to be
presented. Practitioners and other resource managers are expected to
absorb and act upon the 'wisdom' handed down to them via systematic
reviews and other means of dissemination. The model employed seems to
be that of knowledge changing attitude and attitude, in turn, changing
behaviour in a manner the scientists deem appropriate. Unfortunately, the
'knowledge, attitude and behaviour' model has been discredited in health
promotion circles where it has long been realised that changing
people's behaviour is a far more complicated process.
Horton and Smith place the systematic collection of data from
randomised controlled trials at the "intellectual centre" of NHS R&D
and laud the Cochrane Collaboration for its role in this. Undoubtedly,
trial data bases, systematic review and dissemination are important but
they are not the intellectual core of R&D; they are merely supporting
technologies. Moreover, to the onlooker involved in
resource management those engaged in these technologies might seem
obsessed with the finer points of their trade (e.g. tracking down elusive
work and the methodology of meta-analysis) to the point that it is all a
bit precious. Furthermore, much of the wisdom dispensed has little bearing
on the questions resource managers want answered because few reported
trials encompass the Peckham extended trial
protocol (1) and the wider humanistic issues (often expressed
qualitatively rather than quantitatively) necessary for a balanced
judgement.
These problems arise because those engaged in R&D wrongly think that
they should be doing 'science' when in fact the epistemological status of
health service R&D is different though no less rigorous or
intellectually challenging. In our book my collaborator, Dr Jo Walsworth-
Bell, and I show what R&D should really be (2). We develop a 'new
paradigm' for R&D and show how this naturally engages resource
managers who are the consumers of R&D and thus those who should direct
its agenda. Naturally, something as naive as the 'knowledge, attitude,
behaviour' model gets short shrift because the motivators for change are
deeply embedded in the 'new paradigm' which, as we show, has been
prototyped in the North West of England.
1. Peckham, M. Research and development for the National Health
Service. The Lancet 1991; 338: 367-71
2. St Leger, AS. and Walsworth-Bell, JP. Change Promoting Research
for Health Services. Buckingham: Open University Press, 1999.
Dr A. Selwyn St Leger
Consultant/Senior Lecturer in Public Health Medicine
University of Manchester,
Biostatistics Group,
School of Epidemiology and Health Sciences,
Stopford Building,
Oxford Road,
Manchester M13 9PT
Competing interests: No competing interests
As somebody working very closely with parents whose children were
involved in clinical trials at a hospital in the North of England, I have
to say that the question of being able to obtain copies of any protocol
what-so-ever in relation to those trials, has been a horrendous ordeal for
many of the parents concerned. After months of requesting the protocols
for two studies, they have yet to materialise. One has to ask the question
why if everything was above board?
It has transpired over the last few months that not only were some
parents unaware of the fact that their children were being entered into
any research project, they were also asked to subject their children to a
specific form of treatment that they were told was a kinder and gentler
way of treating neonates, whilst in a state of extreme distress as a
result of having had a premature child. The ethics surrounding this type
of coercing without parents having any real understanding of what was
happening and also being very vulnerable and open to persuasion as a
result of the fear of losing their child, are unacceptable and repugnant.
I feel that any research being conducted at any medical establishment
should be public knowledge. All hospitals should have available in their
foyers, details of any ongoing research, details of what the projects
entail, copies of the protocols for those projects and contact numbers of
independent advisers should anyone wish to know more or even volunteer to
be entered in to any research.
Sadly it is the continuous secrecy surrounding human research that
gives rise to suspicion within the members of the public. The phrase
"human research" itself conjures up pictures of atrocoties and lies
uncomfortably with most humans as very little is ever explained and they
are left with visions of puppies with masks on with cigarette smoke being
pumped into them.
Whilst we all appreciate that for medical science to move forward it
is imperative that research is undertaken, however unpallatable that may
be to certain people, it is also of equal importance that there is total
honesty and frank discussion between all parties involved, with an
independent advisor always at hand to allay any concerns and fully explain
all matters relating to a project. I also feel that it must be an absolute
that the medical profession to have always obtained full and informed
consent from the prospective participants or their guardians.
One further observation, it is patronising of the medical
establishment that we are deemed unqualified to "understand the medical
terminology etc..." and are therefore furbished with the bare minimum in
laymans terms. This in itself breeds an air of mistrust between the public
and doctors and also is wide open to abuse and misunderstandings.
Penny Mellor
Competing interests: No competing interests
Fine words, but doctors hold a veto over the disclosure of protocols
for the research they have proposed.
When researching alleged unethical research I have been refused
copies of protocols which patients have wanted to see and have also failed
to obtain, because: 'they are the intellectual property of the
researcher'.
What is the justification of the medical profession for refusing to
allow protocols of research projects approved by
research ethics committees to be seen by the subjects recruited to those
projects, or their parents or legal guardians, either at the time of the
recruitment or subsequently?
Why do chairs of ethics committees allow their submitting researchers
to exercise a veto over disclosure?
What steps can research subjects actually take to obtain copies of
such research projects short of instigating the kind of costly legal
action so much condemned lately by the NHS?
Is there not a case for automatic publication of every original
research protocol with the annual report of the research ethics committee
that approved the research so that patients in the locality can see
exactly what was proposed and compare it with what was executed?
Why the secrecy?
Brian Morgan
Conflict of interest: currently researching examples of such
refusals.
Competing interests: No competing interests
In response to the call for registration of randomised trials, the
administration for such a scheme is already in place.
All research conducted on human subjects within the NHS has to be
approved by either a Multi-Centre or Local Research Ethics Committee.
The administrators of these committees could easily send a copy of all
approved research protocols to a central database.
Very little extra cost - problem solved !
Competing interests: No competing interests
A good idea that doesn''t go far enough
The problem of publication bias, when negative results are swept
under the carpet, is undoubtedly serious. These proposals are to be
heartily welcomed as an attempt to address that problem. However, they may
fall far short of solving it. If a pharmaceutical company undertakes to
publish a trial whatever its results, what is to stop them publishing in
the BMJ or the Lancet when the results are good, and publishing in some
tin-pot little foreign language journal with a circulation of 200 and that
doesn't make it onto Medline or Embase when the results are bad? In the
latter case, the results may have been published, but will still be widely
ignored.
How about this for a bold idea: when the protocol is registered, a
specific journal should undertake to publish the results, and, more to the
point, the investigators and sponsor should undertake to publish in that
journal, whatever the results.
Competing interests: No competing interests