NATIONAL INSTITUTE FOR CLINICAL EXCELLENCE
RESPONSE TO COMMENTS ON THE INSTITUTE'S GUIDANCE
MADE BY THE DRUG AND THERAPEUTICS BULLETIN THE BRITISH NATIONAL
FORMULARY AND OTHERS
1. The Institute rejects allegations that
its guidance is "flawed". The processes adopted by the
an independent, peer reviewed systematic
review of the totality of the available knowledge is commissioned
from one of the NHS R&D's Health Technology Assessment groups
and then quality-controlled by the National Centre for Health
consideration of written evidence
from relevant professional groups and patient/carer organisations;
consideration of evidence from patient
and clinical experts invited to the relevant meetings of the Appraisal
detailed consideration, on two separate
occasions, by its independent Appraisal Committee of all the evidence
submitted to the Institute; and
the opportunity for the consultees
to request an appeal before a panel whose members have not been
involved with the appraisal.
These arrangements are more public, thorough
and inclusive than any other form of scrutiny (nationally or internationally)
of which we are aware.
2. The Institute considers the evidence
not just from the perspective of doctors, or even a wider definition
of health professionals, but in a way that takes account of the
views of those with the condition and those who speak for them.
This means more than simply asking for their views on draft documents.
It involves actively seeking evidence from them and providing
them with the opportunity to talk directly to the Appraisal Committee,
as the evidence is being interpreted.
3. The Institute rejects the suggestion
that it has contradicted the Summary of Product Characteristics
for any pharmaceutical. Whilst at some date in the future it may
be appropriate to do so, this would be confined to established
technologies which have been in use for many years, and where
there is good evidence for their clinical and cost effectiveness.
There are no circumstances where the Institute would promote a
new, or recently licensed technology outside its authorised indications.
4. The Institute rejects criticisms that
its guidance should incorporate a complete account of the use
of a particular technology. For pharmaceuticals these are fully
described in the Summary of Product Characteristics which is subject
to the approval of either the Medicines Control Agency or the
European Medicines Evaluation Agency (depending on the grantee
of marketing authorisation). It should be noted that neither the
Drug and Therapeutics Bulletin, nor the British National Formulary,
provide the full prescribing details that are to be found in manufacturers'
Summaries of Product Characteristics.
5. The Institute provides two forms of guidance
for health professionalsappraisals of the use of individual
(or groups of related) health technologies, and clinical guidelines
on the management of particular conditions. Each represent important
forms of guidance, and clinicians welcome both. The guidance on
individual health technologies is based on clinical and cost effectiveness
and provides clinicians with an indication as to when they are
most appropriately deployed in the management of a particular
condition. Clinical guidelines, also based on clinical and cost
effectiveness, cover a wider area of clinical practice and are
much more lengthy in the time needed for their construction. Where
appraisals are undertaken, on a related area of clinical practice,
in advance of a guideline, the result of the appraisal is incorporated
by the guideline developers.
6. No appeal was made by any group representing
health care professionals against any of the appraisals criticised
by the Drugs and Therapeutics Bulletin.
In considering the criticisms of the Drugs and
Therapeutics Committee, it is worth noting some of the characteristics
of the Institute's approach which might lead to differences in
the way in which evidence is interpreted by the two organisations:
before the independent advisory Appraisal
Committee begins its deliberations, evidence is actively sought
from patients and their carers, health professionals and manufacturers
or sponsors, in the case of a clinical or diagnostic procedure;
the Appraisal Committee considers
evidence of both clinical and cost effectiveness and views from
the perspective of clinicians and patients;
patient/carers and health professionals
themselves nominate representatives to attend and inform
the Appraisal Committee and all consultees are provided with the
opportunity to comment, both on the evidence and on provisional
consultees have the right of appeal
against the final proposed guidance to the NHS before it is issued.
Limitations of the trials included
in the pooled analysis, making the subsequent interpretation "highly
Link between complications requiring
antibiotics and subsequent hospitalisation and death. Antibiotic
therapy is often used for minor infections. The observed reduction
of 6 per cent did not reach statistical significance.
If a day's reduction is not a compelling
reason for otherwise healthy adults, it is "questionable"
that it could be so regarded for those in the at risk population.
Zanamivir is licensed for the treatment of influenza
Type A & B in adults when influenza is circulating in the
community. The Institute has appraised it twice: in November 1999,
following a rapid review, and in October 2000 following a formal
In the first (rapid) appraisal, the trial data
revealed that there was a reduction in duration of the symptoms
of influenza in the pooled intention to treat population of about
a day. The rapid review committee was interested to understand
the effect of zanamivir on patients with pre-existing conditions
(such as cardiovascular, renal and respiratory diseases) and in
the elderly, where it is known that influenza can cause complications
and death. Although such patients were included in the studies
referred to above, they were not sufficiently numerous to enable
any conclusions to be reached. The committee was also concerned
to understand the likely impact of widespread prescribing of zanamivir
on the ability of general practitioners and primary care professionals
in general, to maintain a balanced range of services. There was
relatively little information available to enable this impact
to be accurately estimated.
The committee concluded that the benefit for
otherwise healthy adults of a day's reduction in the symptoms
of influenza, whilst important for some individuals, did not constitute
an efficient use of NHS resources. In particular, the committee
was concerned about the possibility that widespread prescribing
in the adult population during an epidemic would have a detrimental
impact on the ability of primary care to maintain a full service.
The committee called for further information on this and on the
effectiveness of zanamivir in the "at risk" groups.
This was communicated to the manufacturer and the Institute agreed
to review any further relevant data in 2000.
A second appraisal was undertaken in 2000, as
part of the Institute's formal appraisal programme, and as recommended
in the original (rapid) guidance. An assessment report was commissioned
from the West Midlands Development and Evaluation Service at the
University of Birmingham, through the NHS Research and Development
Programme. Following a review of this report, the Appraisal Committee
concluded that no new material evidence had emerged to suggest
a change to the existing guidance and proceeded to consult on
this basis. During the consultation period, the manufacturer,
Glaxo Wellcome, presented the results of a further clinical trial.
This was the first trial specifically recruiting an at risk population
to be reported. The West Midlands team were asked to produce a
supplementary assessment report taking account of this trial.
The West Midlands team, in their original assessment
report, used data from a meta-analysis produced by Glaxo-Wellcome
to report antibiotic usagethis showed a 6 per cent reduction
in antibiotic use for respiratory complications. The new trial
showed similar reductions (5 per cent) which did not reach statistical
significance. In their submission Glaxo Wellcome presented an
integrated analysis that gave an estimate of a 6 per cent reduction
in complications requiring antibiotic use in the at risk population
(95 per cent CI 0 per cent to 11 per cent). The West Midlands
team did a meta-analysis using virtually the same data set . Using
a fixed effects model this gave a similar estimate of a 6 per
cent (95 per cent CI 0 per cent to 11 per cent) reduction in complications
requiring antibiotics, which is on the borderline of statistical
significance. The West Midlands team concluded, however, that
it was in their view a true effect. The Appraisal Committee considered
that this "pooled analysis" was scientifically appropriate
and in accordance with the practice adopted both nationally and
internationally. It considered that reductions in complications
requiring antibiotics was a reasonable intermediate end point.
It therefore concurred with the conclusion of the assessment report
authors that by reducing complications requiring antibiotics,
it is probable that hospitalisation is also reduced and that deaths
are avoided. The Appraisal Committee therefore advised that zanamivir
should be offered to patients in the at risk population.
This new trial was not available to the regulatory
body at the time zanamivir was licensed and could not, therefore
have influenced the references to the effect of zanamivir in at
risk groups in the Summary of Product Characteristics. Because
the Institute is required to take account of all the best available
evidence, it was appropriate for it to use the new trial to inform
its revised conclusions. Nevertheless, the Institute's advice
is within the product's authorised indications.
The Institute recommended the supply of zanamivir
through the use of patient group directions after discussions
with the Department of Health. In doing so, we were mindful of
the need to offer advice which would have the effect of minimising
the additional workload in primary care (but particularly on general
practitioners) whilst offering qualifying patients rapid access
to the medicine. Any medicine which might be the subject of a
patient group direction will have contra-indications. Prescribers
and those who supply medicines through patient group directions
will wish to be aware of the safety profile of any drug they offer
to patients. "Black triangle" medicines are not excluded
from patient group directions and clearly, had the Department
of Health considered that what we were suggesting was an inappropriate
use of patient group directions (because of concerns over antibiotic
resistance), they would not have agreed to it. As far as the safety
concerns which were noted, the Institute's position is that prescribers
should be aware of the contra-indications and warnings referred
to in the Summary of Product Characteristics, and any additional
safety information published by the manufacturer or the Committee
on Safety of Medicines.
NICE guidance recommends use "in
ways not recommended by the manufacturer". Specifically,
by suggesting that patients whose blood glucose remains high despite
an adequate trial of metformin plus sulphonylurea combination
should be offered either glitazone in combination with metformin
or sulphonylurea as an alternative to starting insulin.
There is "no evidence"
to support these unlicensed indications and therefore the guidance
The thiazolidinediones are clinically effective
as second line agents for glucose lowering in people with Type
2 diabetes. The review of studies showed this was also the case
for these agents in first-line therapy. However, the Institute's
guidance does not recommend the use of thiazolidinedione monotherapy,
either first or second-line because it is not a licensed indication.
The guidance recommends the use of combination
therapy when monotherapy alone does not achieve optimal glucose
control. The guidance also recommends that the first option for
combination therapy should be metformin and sulphonylurea, unless
there are contraindications to, or problems with their tolerability.
The thiazolidinediones are clinically effective for glucose lowering
as add-on agents in people with Type 2 diabetes. Currently there
is no evidence of improved clinical outcomes using combination
therapy with thiazolidinediones compared with sulphonylurea/metformin
combinations. However, evidence suggests that combination therapy
with thiazolidinedione and either metformin or sulphonylurea could
be offered if patients are unable to take metformin/sulphonylurea
combination or if blood glucose remains high, despite an adequate
trial of the latter. In this context, improved control of blood
sugar is considered both nationally and internationally to be
a reasonable intermediate outcome.
The decision whether to continue with oral therapy
or to commence insulin therapy when for example glycaemic control
is considered inadequate must be based on expert clinical judgment
coupled with individual patient choice. The Institute emphasizes
that its guidance was not discordant with the licensed indications.
3. RITALIN FOR
NICE guidance is "of limited
value" because it makes no attempt to advise on use of the
other licenseddexamfetamine. As a result, healthcare professionals
"have no way of knowing whether NICE intends that dexamfetamine
should be used as an alternative to methylphenidates reserved
only for certain (undefined) groups of children or not used at
There "is little difference
in effectiveness between dexamfetamine and the methylphenidates"
and the former is much cheaper than the latter.
The guidance "fails to answer
key questions or to provide comprehensive and complete advice
on whether, when and how drug therapy should be used in the NHS
to manage children with ADHD".
The Institute was not asked to undertake an
appraisal of dexamfetamine. The request that it received from
the Department of Health and the National Assembly for Wales was
clearly limited to the methylphenidates. The underlying reason
was the controversy, at the time, over the use of methylphenidate
in the management of Attention Deficit Hyperactivity Disorder
(ADHD). The fact that the Institute did not offer advice on the
use of dexamfetamine does not, in any way detracts from its value.
In addition, we received no suggestion from any of the clinical
experts or professional groups we consulted that, for the appraisal
to be of value it was necessary for it to include dexamfetamine.
The guidance concludes that "methylphenidate
is recommended for use as part of a comprehensive treatment programme
for children with a diagnosis of severe Attention Deficit/Hyperactivity
Disorder (ADHD)". The guidance goes on to describe the arrangements
which should be put in place to enable accurate diagnosis and
the health professionals which should be responsible both for
diagnosis and for initiating treatment:
Diagnosis should be based on a timely,
comprehensive assessment conducted by a child/adolescent psychiatrist
or a paediatrician with expertise in ADHD. It should also involve
children, parents and carers and the child's school, and take
into account cultural factors in the child's environment. Multidisciplinary
assessment, which may include educational or clinical psychologists
and social workers, is advisable for children who present with
indications of significant co morbidity.
Treatment with methylphenidate should
only be initiated by child and adolescent psychiatrists or paediatricians
with expertise in ADHD, but continued prescribing and monitoring
may be performed by general practitioners, under shared care arrangements
Careful titration is required to
determine the optimal dose level and timing for methylphenidate.
The drug should be discontinued if improvement of symptoms is
not observed after appropriate dose adjustment.
A comprehensive treatment programme
should involve advice and support to parents and teachers, and
could, but does not need to, include specific psychological treatment
(such as behavioural therapy). While this wider service is desirable,
any shortfall in its provision should not be used as a reason
for delaying the appropriate use of medication.
Children on methylphenidate therapy
should receive regular monitoring. When improvement has occurred
and the child's condition is stable, treatment can be discontinued
at intervals, under careful specialist supervision, in order to
assess both the child's progress and the need for continuation
The guidance does not suggest that methylphenidate
is more effective than dexamfetamine. Nor does it, explicitly
or implicitly, recommend that it should be used in preference
to dexamfetamine. It offers clear advice that methylphenidate
is both clinically and cost effective when used as part of a properly
organised treatment programme. It is a matter for parents, children
(where appropriate) and prescribers to decide which is preferred,
in individual cases.
Trials assessed the effects of these
drugs on patients' cognitive functions and "global measures".
There is "considerable doubt" about the relevance of
these indicators for patients.
The guidance offers "no (other)
convincing rationale for recommending the use of (these three
drugs) in the NHS".
The Institute recognised the difficulties associated
with diagnosing Alzheimer's Disease. Our guidance notes that there
is no measure of quality of life for use in patients with cognitive
impairment that is universally satisfactory. However, the RCT
evidence reviewed by the Appraisal Committee showed that all three
drugs produce significant improvement in cognitive function.
Patient groups encouraged the Institute in the
view that any intervention for Alzheimer's disease needs to be
evaluated in terms of the significance which patients and carers
place on its benefits. In its evidence to the Institute, the Alzheimer's
Society submitted the results of a survey which considered the
views of almost 1000 people with Alzheimer's who had been prescribed
donepezil. From this survey, the Society concluded that ".
. . it is the changes in behaviour and mood that appear to be
most frequently reported by people affected and their carers."
This survey result was supported by the oral evidence given by
those who spoke for patients during the appraisal and by the clinical
experts consulted by the Appraisal Committee. This encouraged
the Appraisal Committee to take the view that there is a relationship
between the measures of disease progression used in the clinical
trials and the global effects reported by patients.
In addition, the guidance carefully contextualises
the use of these medicines, by describing the arrangements which
should be in place for assessing patients, prior to commencement
of therapy. It advises that only those patients with mild to moderate
disease, with an MMSE score of 12 or above, should be considered
for therapy with these medicines.
It concludes by recommending that patients who
continue on the drug should be reviewed by MMSE score and global,
functional and behavioural assessment every six months. The drug
should normally only be continued while their MMSE score remains
above 12 points, and their global, functional and behavioural
condition remains at a level where the drug is considered to be
having a worthwhile effect. When the MMSE score falls below 12
points, patients should not normally be prescribed any of these
three drugs. Any review involving MMSE assessment should be undertaken
by an appropriate specialist team, unless there are locally-agreed
protocols for shared care.
The guidance is therefore both appropriately
cautious and adequately targeted to provide sufficient advice
to clinicians, as they exercise appropriate judgement in individual
5. RILUZOLE FOR
A suggestion that the Institute failed
to take due note of the assessment report's observations that
"There is limited evidence of a modest benefit in tracheostomy-free
survival for patients taking riluzole." and "Even under
the most optimistic assumptions, riluzole at best only postpones
death for a few months".
Guidance wrongly quotes the Assessment
Report's calculation of the relative reduction in hazard ratio
(17 per cent at 18 months as opposed to 12 per cent at 18 months).
Claims that there is "considerable statistical uncertainty"
about this effect.
The guidance should therefore have
been "clearer" in explaining that the evidence about
the effect of the drug extending tracheostomy free life is "equivocal".
The guidance did indeed contain the misprint
referred to above. This has been acknowledged by the Institute.
The DTB criticism refers to the apparent differences
between the Institute's guidance and the assessment report. Assessment
reports are commissioned by the National Coordinating Centre for
Health Technology Assessment, on the Institute's behalf. Their
purpose is to assemble the available evidence of clinical and
cost effectiveness. Their authors are not asked to interpret this
evidence, although they do, importantly, comment on its appropriateness
and its robustness. It is for the Appraisal Committee alone to
draw conclusions from the evidence and formulate the guidance.
The Appraisal Committee took careful account
of the assessment report. It considered its findings both with
the report's authors and with the patients groups and clinical
experts who attended its meetings. The Committee considered the
statistical analysis which had been undertaken on the clinical
trial data in the assessment report. The Appraisal Committee took
the view that, in the light of comments from the Motor Neurone
Disease Association, a three month extension to life without a
tracheostomy could not be considered as "modest". Having
considered the economic analysis, the Appraisal Committee concluded
that those patients should have the opportunity to receive this
6. ORLISTAT AND
NICE has issued guidance on two drugs
for obesity, but had given "no advice on how the two drugs
should be used in relation to each other. It is unclear as to
whether NICE regards them as interchangeable or whether they might
be suitable for particular groups of patients.
The guidance fails to offer "clear,
integrated advice on the use of these two treatments in the overall
management of patients who are obese".
"As a result, it is possible
that individual healthcare professionals and service providers
will have to formulate their own policies on selecting between
and using the two treatments: this could lead to marked variation
in practice throughout the NHS (which would be) counter to NICE's
remit to end confusion by providing a single national focus".
The guidance fails to mention "some
of the practical limitations" on the use of sibutramine which
"limit its value". The DTB noted that it had recommended
against the use of the drug for a variety of reasons, including
"the stringent requirements for monitoring during treatment".
These medicines were originally part of the
same appraisal. They had to be considered separately because sibutramine
received its licence later than anticipated and after orlistat.
The Institute will only issue guidance on pharmaceuticals once
the licensed indications have been approved. The aim of the appraisal
was to assess the role of both medicines in the management of
obesity. The Institute was not asked to develop a clinical guideline
on the management of this condition. Clinical guidelines are separately
requested and commissioned by the Institute though a different
While the guidance was not expected to be in
a clinical guideline format, in both guidance documents there
is an emphasis on considering these drugs in the context of an
overall approach to the management of obesity. The first recommendation
in the Sibutramine guidance states "
sibutramine should be prescribed
only as part of an overall treatment plan".
The third recommendation states that "
when treatment with sibutramine is
offered, arrangements should be made for appropriate health professionals
to offer concomitant advice, support, and counselling on diet,
physical activity and behavioural strategies". It goes on
to state "that sibutramine should not be prescribed unless
adequate arrangements for monitoring both weight loss and adverse
effects can be made available".
It was not appropriate for the guidance to go
into the details of these plans or arrangements. Nevertheless,
it does state that " there is no evidence to support co-prescribing
with other pharmacotherapy aimed at weight reduction". In
the implementation section the guidance again emphasises that
"to facilitate the appropriate use of pharmacological management
of obesity, training and resourcing of a sufficient number of
primary care staff (practice nurse, nurse practitioners and health
visitors) supported by community dieticians will be necessary
so that they can carry out initial patient assessments and provide
continuing advice and support for patients before, during and
after treatment". In the related guidance section, the reader's
attention is drawn to the Institute's other guidance in this field
including the current appraisal on the role of surgery in the
treatment of obesity.
In addition, the guidance clearly refers to
the principal contra-indications of sibutramine and how practitioners
should monitor patients to assess these side effects to avoid
or minimise potential harm.
In its written submission to the Select Committee,
the BNF stated that:
On rare occasions, the Joint Formulary Committee
has agreed to omit mention of NICE guidelines because of concerns
about the quality of the advice. The BNF has identified a number
of difficulties with NICE appraisals."
In giving oral evidence, Dinesh Mehta, Executive
Editor of the BNF is recorded as saying, in response to a question
about the quality of NICE guidance and the occasions on which
the BNF omits mention of it:
"That is very rare and so far the committee
has agreed on omitting just the one guidance on the management
of patients who have just had a heart attack".
The guidance referred to is the Institute's
clinical guideline entitled: Prophylaxis for patients who have
experienced a myocardial infarction.
The criticism made by the BNF in their evidence
to the Select Committee was:
initiation of drug therapy following
a myocardial infarction should take place prior to discharge.
The BNF took issue with the suggestion, in the NICE guidance,
that therapy could be initiated in primary care.
The Institute's guideline clearly states that
therapy with beta-blockers, antiplatelets drugs (aspirin) and
ACE inhibitors should be initiated whilst patients are in hospital.
The guidelines recommends that if this has not happened, then
primary care clinicians should initiate them as soon as possible
This is a common sense approach which reflects
the continuity that needs to exist between secondary and primary
care and; the continuing responsibility of general practitioners
for ensuring adequate care post discharge.
Criticisms of the Campaign for Effective and
Rational Treatment (CERT) and representatives of the pharmaceutical
industry regarding the NICE appraisal of trastuzumab (Herceptin).
women had died because NICE had delayed
issuing its guidance;
there were unexplained and inexplicable
delays in the Institute's appraisal of trastuzumab;
that the effectiveness of the drug
had been established unequivocally;
that the publication of guidance
on the other drug being appraised (vinorelbine) for breast cancer
was being delayed because of the changes to the timetable for
Trastuzumab is not a cure for cancer.
It is used to extend life for some women with breast cancer. It
was referred to NICE because there was uncertainty about its value
in clinical practice, which had resulted in postcode prescribing.
To suggest that women have died because NICE has extended its
timeline is both erroneous and offensive.
It has been suggested that the evidence
for the clinical benefit of trastuzumab is "incontrovertible".
The appraisal process will establish whether or not this is the
case. In addition, the process will establish the cost effectiveness
of the drug. Cost effectiveness is not established by simply taking
the annual value of the drug and multiplying by extension to life,
as was suggested by witnesses.
The Institute extended the timetable
for this appraisal following the responses it received from consultees
during the appraisal (see below). The reasons for doing so were
communicated to consultees on 13 August 2001 (see attachment)
and subsequently posted on the Institute's web site. NICE recognises
the importance of this appraisal to women with breast cancer and
their families and carers, and we regret the need to extend the
timetable. However, we believe strongly that we should take account
of all the available evidence and that we should allocate sufficient
time to do so. The consequences of a wrong decision would, obviously,
The appraisal of trastuzumab had
not reached the stage of an appeal, so it was wrong to suggest
that appeals were delaying publication. In any event, the opportunity
to appeal is an important part of the Institute's process.
During the consultation period, consultees
made it clear that they believed that there was evidence available
which had not been presented to the Appraisal Committee and which
should be taken into account. The Institute, on the advice of
the Appraisal Committee, therefore allowed an extension to the
appraisal in order to allow full consideration of this evidence.
As there was more of this additional
evidence for vinorelbine than for trastuzumab, the Institute separated
the timetables for the two appraisals to allow the Appraisal Committee
to consider the additional data as soon as it was available. Therefore
both appraisals are underway and there is no completed document
on vinorelbine ready for publication.
The Appraisal Committee met in October
2001 to consider the new evidence for trastuzumab. Following this
meeting a new provisional appraisal determination was produced
and circulated to consultees. Our process allows consultees four
weeks to consider the document. They received the document on
14 November and had until 14 December to submit comments.
The Appraisal Committee considered
the responses in January and produced a final appraisal determination
against which the consultees may appeal.
If no appeals are heard, the guidance
will be issued. The earliest date for the issue of guidance on
trastuzumab is in March.
Attachment: Letter to consultees from the apprasial
programme Director in the appraisals of trastuzumab and vinorelbine.
13 August 2001