APPENDIX 4
Memorandum by Sir Iain Chalmers, Director,
UK Cochrane Centre (NC 22)
SUMMARY
There is a need for greater transparency in
the data available to the National Institute for Clinical Excellence.
All the data that have informed NICE's judgements should be made
public.
INTRODUCTION
1. I am the Director of the UK Cochrane
Centre, which forms part of the infrastructure supporting the
Cochrane Collaboration—an international organisation, registered
as a charity in England, which prepares, maintains and makes accessible
systematic reviews of research evidence on the effects of health
care interventions. The NHS Research and Development Programme
and the Chief Scientist Office in Scotland have provided very
important support for this Centre and other elements of the Cochrane
Collaboration. The output of the Collaboration's work—the
reviews of evidence published electronically in The Cochrane Library—have
been used extensively by the National Institute for Clinical Excellence,
and by other national and international organisations responsible
for developing guidelines and policies.
2. Within this work, one of my specific
interests has been the extent to which reporting biases have distorted
publicly accessible evidence about the effects of health care
interventions. Studies with "positive" results are more
likely than others to be selected for presentation at scientific
meetings; more likely to be reported in print; more likely to
be published promptly; more likely to be published as full reports;
more likely to be published in journals that are widely read;
more likely to be published in English; more likely to be published
in more than one report; and more likely to be cited in reports
of subsequent, related studies. Under-reporting research is scientific
and ethical misconduct. At best, it results in resources being
wasted on interventions that have no important beneficial effects;
at worst, it can result in avoidable suffering and death of people
using health services.
RELEVANCE TO
NICE'S WORK
3. It is possible to understand the demand
for commercial confidence of information about the effects of
drugs that have not yet been licensed by the Medicines Control
Agency or the European Medicines Evaluation Agency. In contrast,
withholding information about the effects of drugs (and other
health care interventions) that have been licensed is indefensible.
Patients, prescribers and policy makers should have access to
all the information about the effects of the drugs and other health
care interventions they may wish to consider. (Roberts I, Li Wan
Po A, Chalmers I. Intellectual property, drug licensing, freedom
of information, and public health. Lancet 1998; 352:726-729).
4. NICE is currently expected to serve the
interests of patients, the public more generally and the NHS (i)
without knowing the extent to which it is being provided with
all relevant information (and not a biased subset of it); and
(ii) without being able to make public all of the information
that has informed its judgements. While this state of affairs
persists, NICE cannot be expected to achieve the credibility that
is essential for it to earn the trust of the public.
5. Speaking at NICE's annual conference
at the end of last year, Lord Hunt stated that NICE's work is
transparent. Although the Institute's work is commendably transparent
in some respects, as indicated above, it is anything but transparent
in some crucially important respects. Even though it might be
difficult to encapsulate in publicly available documents the series
of judgements leading to the appraisal decisions made by NICE,
making public the data informing these decisions would help to
promote greater trust in the process.
6. Under current arrangements, NICE cannot
be certain that the information that it receives in confidence
from industry is all of the relevant information it should take
into account, nor can some of the information it does receive
be made publicly available to those who wish to understand the
background to NICE's appraisals and guidelines. Some of the people
working on analyses relevant to NICE have told me that they are
aware of important data having been withheld from NICE by companies
whose interests would not be served by submitting it.
RECOMMENDATIONS
7. The most appropriate response to these
causes for concern would be for NICE to insist that it receives
all of the information relevant to a particular judgement, and
that this information would be made public once the appraisal
or guideline was released. I gather that Ministers rejected this
at an early stage in the life of the Institute, but the Health
Committee could recommend that this should be required.
8. Short of this desirable level of transparency,
other steps can be taken to improve the current situation. First,
those companies that declare that they have submitted all relevant
information to NICE should be openly credited for having done
so. I gather from staff at GlaxoSmithKline, for example, that
it is the company's standard practice to submit all the data of
which they are aware.
9. Second, even if the results of controlled
trials are not made public, NICE should require companies to make
basic details of the relevant trials publicly available through
registration and the metaRegister of Controlled Trials, and for
an international Standard Randomised Controlled Trial Number to
be assigned (see www.controlled-trials.com). Not only does the
NHS Research Governance Framework now require registration of
all controlled trials, but also, over a year ago, the Association
of the British Pharmaceutical Industry recommended that all of
its member companies should follow Glaxo Wellcome's lead in registering
their trials.
PERSONAL STATEMENT
Although I am confident that these views are
shared by many others, this memorandum has been submitted on an
individual basis. If the Committee wishes me to give oral evidence,
I would be willing and able to do this after 17 February.
January 2002
| 
