3 Health Technology Appraisals: cost
effectiveness and other issues |
Cost effectiveness and value-based
19. Technology appraisals are the function for
which NICE is best known. As NICE put it in its memorandum to
In this programme we assess the clinical and cost
effectiveness of (mainly new) health technologies. These are primarily
new pharmaceutical and biopharmaceutical products but have also
included procedures, devices and diagnostic agents. This programme
is intended to ensure that all NHS patients have equitable access
to the most clinically- and cost-effective treatments that are
available. In doing so we work openly and transparently with the
life sciences industry. We have, to date, published 264 technology
appraisals with 34 planned or already published in 2012-2013.
20. One of the issues that was extensively discussed
in written and oral evidence was the issue of cost effectiveness.
The Association of the British Pharmaceutical Industry (ABPI)
expressed concern that NICE gave too much weight to cost effectiveness
in examining new drug therapies:
The focus of NICE in HTA [health technology appraisal]
decision making is we believe disproportionately on cost effectiveness.
Although NICE considers a range of factors, the ultimate decision-making
metric remains as the Incremental Cost Effectiveness Ratio. Appraisal
Committee deliberations are therefore focused primarily on the
perspective of health economists rather than that of clinicians
and patients. A more balanced view of the various elements of
value is required and a full account should be given by Appraisal
Committees of what factors were taken into consideration and,
importantly, the impact of those factors on the final decision,
so that stakeholders properly understand how conclusions were
21. Professor Johnson of Cancer Research UK was
sympathetic to that analysis:
There is a very tight focus a lot of the time in
the way that NICE considers health technologies on the immediate
costs and benefits and not the rather broader consideration of
what effect it might have in other parts of the patient pathway.
A failure to take into account the broader value of particular
healthcare interventions does restrict their vision of what constitutes
"worthwhile" for the NHS.
22. Professor Smith, on behalf of the Nuffield
Trust, argued that
The principle of costeffectiveness is one that
has proved very hard to challenge. It arises from the desire to
spend a limited budget as effectively as possible. That gives
rise to the decision rule of costeffectiveness. The work
of NICE, which is internationally recognised, has embedded that
within our health system.
23. Laura Weir, of Patients Involved in NICE,
thought, however, that the issue of cost effectiveness had been
...it is a common misconception about NICE, probably
fuelled by the media, that it weights costeffectiveness
more heavily than clinical-effectiveness. The truth is it has
to see them both together; those calculations have to be done
together, and that is why we have the QALYthe quality adjusted
life year. It will weigh the cost of a medicine against the quality
and length of life that a treatment will give. Where medicines
only offer a marginal clinical benefit but they cost a lot, that
is when patient access schemes will often come to the fore and
where patient groups will lobby companies to reduce their price,
because, after all, if the price is lower, there is a higher chance
that patients will be able to get that medicine on the NHS.
24. Although the high reputation of NICE, both
in the UK and abroad, is largely based on respect for its work
on assessment of the cost effectiveness of new drugs, it is the
broader concern about the implications of this focus on health
economics which lies behind much of the recent rhetoric about
25. A move to value-based pricing was included
in the Coalition's Programme for Government:
We will reform NICE and move to a system of value-based
pricing, so that all patients can access the drugs and treatments
their doctors think they need.
In the Foreword to a consultation on moving to value-based
pricing issues in December 2010, the then Secretary of State wrote;
The current system of pricing medicines has tried
to achieve a balance between reasonable prices for the NHS and
a fair return for the industry to develop new medicines. However,
it does not promote innovation or access in the way this Government
is looking for. Also, too often, the NHS has been in the position
of either having to pay high prices that are not always justified
by the benefits of a new medicine, or having to restrict access.
We are determined to create a system that gives patients
access to the most effective medicines. There must be a much closer
link between the price the NHS pays and the value that a medicine
delivers. Pharmaceutical companies need a pricing system that
is more stable and transparent, and that gives clear signals about
priority areas, so that research efforts are directed to greatest
effect...This consultation document outlines a new system which
aims to recognise and reward innovation, in particular by encouraging
a focus towards genuine breakthrough drugs which address areas
of significant unmet need.
26. Chapter 4 of the consultation document set
out the current state of plans for value-based pricing:
The key principle of value-based pricing is to ensure
NHS funds are used to gain the greatest possible value for patients.
So the Government would set a range of thresholds or maximum prices
reflecting the different values that medicines offer.
The value of new products would be assessed and their
benefits compared with the benefits that could be gained if the
funds required were used to help patients elsewhere in the NHS.
This comparison is normally expressed with a cost-effectiveness
threshold. It requires the use of a "common currency"
for quantifying benefits in a consistent and comparable way across
the full range of health-related conditions.
One (but not the only) option would be to use Quality
Adjusted Life Years (QALYs), the 'currency' currently used by
NICE in its technology appraisals. A QALY is the amount of health
represented by a year of life at full health. It gives an idea
of how much extra length of life and the quality of life a person
might gain as a result of treatment - and health gains from different
treatments can be expressed in terms of the number of QALYs to
which they are equivalent. If we used the QALY in value-based
pricing, the threshold would be expressed as a cost per QALY gained.
In the existing system, a standard cost effectiveness
threshold is applied to all new products, with some flexibility
to take account of additional relevant factors, including societal
preferences. However, the mechanism for taking wider factors into
account is not completely transparent, and this may lead to perceptions
that important factors are not adequately reflected in the assessment
By contrast, under the new system of value-based
pricing, the Government would apply weightings to the benefits
provided by new medicines, which would imply a range of price
thresholds reflecting the maximum we are prepared to pay for medicines.
These thresholds or maximum prices would be explicitly adjusted
to reflect a broader range of relevant factors so they could be
used to calculate the full value of a new product.
The Government proposes that the price threshold
structure is determined as follows:
i. there would be a basic threshold, reflecting
the benefits displaced elsewhere in the NHS when funds are allocated
to new medicines;
ii. there would be higher thresholds for medicines
that tackle diseases where there is greater "burden of illness":
the more the medicine is focused on diseases with unmet need or
which are particularly severe, the higher the threshold;
iii. there would be higher thresholds for medicines
that can demonstrate greater therapeutic innovation and improvements
compared with other products;
iv. there would be higher thresholds for medicines
that can demonstrate wider societal benefits.
Designing the new system to be both stable and transparent
would allow companies to predict well in advance how prospective
products may fare, and to focus their research efforts on the
treatments that society values most. Companies would be informed
of these weightings - allowing them to orient their research and
development investments appropriately.
In introducing this system, the Government would
move away from a system of negotiations once every five years
under the PPRS [Pharmaceutical Price Regulation Scheme] to a more
stable framework. This greater predictability and transparency
will give companies greater certainty for making long term investment
27. It is notable that, despite the fact that
the Coalition Government has now been in office for over two and
half years there remains very little detail about how an alternative
"value-based" pricing system might work.
28. We asked Sir Michael Rawlins, the current
chair of NICE, how he thought value-based pricing would work.
He told us:
In the past we have looked at costeffectiveness
in relationship to the incremental costeffectiveness ratiothe
increased amount of money you get in terms of the increased numbers
of quality adjusted life years gained. We have had a threshold,
but we have always encouraged our advisory committees to be flexible
about the use of the threshold, and there may be circumstances
when they feel they ought to be giving a positive recommendation
even when things are above our threshold range. Endoflife
care is an example.
This has all been subjective and it has not been
objective. Maybe you could reasonably criticise us for not having
tried to create an objective approach rather than a subjective
approach, but I think part of valuebased pricing is to produce
some objectivity into it. One would weight the QALY. One might
weight double the value of the QALY for endoflife
treatments, for example. That is one component to it and that
is how I see valuebased pricing at that end.
The second component is what they call the economic
perspective. When we are looking at costs and benefits, we look
at it for the health service... There could be a case for going
wider. One could take into account the benefits to carers more
or time off work or unemployment. The perspective can be very
wide. I am agnostic about this whole perspective business, because
there are difficulties about how far you will want to go and when
you have a lot of unemployment what is the sense of compensating
that in a macro scale. So I am more neutral about the perspective.
29. In the current system, NICE makes its assessment
based on the manufacturer's suggested price. As it says in the
NICE does not have a role in relation to the pricing
of medicines, though the scrutiny of its appraisal process may
encourage drug companies to set prices which satisfy its cost-effectiveness
criteria. However, if NICE concludes that a drug may offer some
benefits but that these are not sufficient to justify the price
at which the drug is available, NICE's only option is to recommend
that the NHS restricts its use of that drug. There is no scope
for NICE in England, or its parallel bodies in the rest of the
UK, to enter into pricing negotiations or to recommend an NHS
30. Although this is the theoretical position,
the Committee understands that, in practice, discussions take
place between the sponsors of a new drug and NICE about the price
at which the drug would satisfy cost effectiveness criteria and
the Committee agrees with NICE that this flexibility is desirable.
Against that background it is even less clear what substantive
change is implied by the concept of value-based pricing. Under
the proposed system, NICE will no longer make that recommendation.
It will still be asked for its expert advice, and it may run a
process more or less the same as the one it uses now. The recommendations,
and discussions about price with manufacturers, will be the job
of Government, that is (it appears) the Department of Health.
31. It is also clear that there will need to
be some extension or updating of the current PPRS in order to
provide for payment for those drugs already approved for use in
As it would not be feasible to carry out a value-based
pricing assessment for each individual branded medicine that is
already available, our intention is that branded medicines that
are on the market prior to 1 January 2014 would be covered by
new arrangements sitting alongside value-based pricing.
For branded medicines already covered by PPRS at
the end of 2013, a successor scheme to the PPRS will be required.
The details of this will be developed alongside value-based pricing.
Following the outcome of this consultation and in line with longstanding
practice in this area, the Department of Health will engage with
relevant representatives of the pharmaceutical industry on the
details of these arrangements.
32. There has been extensive
discussion of the principle of value-based pricing, but it remains
a source of concern to the Committee that so little progress has
been made on defining this nebulous concept. The practical implications
of the move to value-based pricing appear to be relatively modest:
with a limited number of health technology appraisals taking place
each year (around 30), the majority of drugs will for the foreseeable
future continue to be procured under a variant of the current
Pharmaceutical Price Regulation Scheme.
33. The consultation document
on value-based pricing was issued two years ago in December 2010,
and the response to the consultation was published in July 2011.
The Committee does not regard it as acceptable that the arrangements
for value-based pricing have still not been settled and that those
who will have to work with those arrangements are still unclear
about what value-based pricing will mean in practice. Industry
needs certainty about how it should bring its products to the
NHS, and patient groups and clinicians need to understand what
their role will be and how they can make their views heard. Given
the length of time since the consultation began, the apparently
modest implications of the proposed changes, and the fact that
the new regime is due to be effective from January 2014, we recommend
that the Department of Health should bring this uncertainty to
an end no later than the end of March 2013.
Cancer Drugs Fund
34. We discussed the operation of the Cancer
Drugs Fund during our inquiry. The scheme, introduced in 2011
and due to run till 2014, was designed to help provide access
to and funding for cancer drugs which have not been approved by
NICE for use in the NHS. Cancer Research UK told us:
The Cancer Drugs Fund has enabled around 21,000 patients
to access treatments they may not have otherwise had and has helped
to reduce the perception that cancer patients are not able to
access treatments. However, this is a finite amount of money for
a limited period. This means that it is especially important that
the new value based pricing system offers an authoritative solution
to funding clinically- and cost-effective drugs on the NHS. The
Cancer Drugs Fund is due to come to an end in 2014 and it is currently
unclear what this will mean for drugs that are paid for via this
route. It is essential that thought is given to how the removal
of the fund will impact upon NICE's work and reputation going
Professor Johnson of Cancer Research UK said that
"it has done a great deal of good in terms of patients' access
to treatment, but at the moment we don't have the data on what
the outcomes and the impact have been on quality and duration
35. Sir Andrew Dillon, Chief Executive of NICE,
told the Committee that:
The effect...of the Cancer Drugs Fund is that it
has expanded the amount of money that the NHS has available to
fund the exceptional treatment decisions that it makes all the
time, both in relation to NICE guidance and circumstances where
individual patients present and there is an argument put forward
by the attending physician for access to a treatment that is not,
on a straight reading of NICE guidance, indicated for that patient,
and in other circumstances where, as a result of the local decision
making, for some reason, particular treatments are not routinely
made available. The Cancer Drugs Fund has expanded that and more
patients are likely to have had access, through those exceptional
treatment decisions, to treatments that either NICE has not actually
made a recommendation on, either because we are still in the process
of doing it or because it is not something that we have looked
at anyway, or where we have made a recommendation but there is
a strong case, in the view of the patient and their attending
physician, to get access to the drug.
36. Laura Weir, of Patients Involved in NICE,
In my role as Chair of Patients Involved in NICE
I don't just represent the cancer charities but quite a number
of conditions. I think the Cancer Drugs Fund was sold to us as
something that would help speed up access to cancer medicines
and make sure that people were able to access the medicines they
needed. I would say that that is applicable across all condition
areas... I think there are some lessons there for valuebased
pricing because, whatever our definition of "value"
is, it must fairly and consistently apply across all conditions.
37. Professor Haslam, the incoming Chair of NICE,
expressed similar views:
...the Cancer Drugs Fund has led to more patients
receiving more drugs. We don't know what the outcomes have been
and we don't know what the impact of that has been on other sections
in the health service...If [the Fund] is going to be reviewed,
I would want to look at it across the board...there are other
conditions that are as serious as cancer and we should not discriminate
against those because they do not have as frightening a name.
38. The Cancer Drugs Fund was
established to help provide cancer treatments which would not
otherwise be available in the period up to January 2014, when
it was considered that the introduction of the new value-based
pricing system, with its perceived greater flexibility than the
current NICE approach, would mean that it would no longer be required.
From the evidence of our inquiry, the Committee considers that
three things need to be done before the Fund ceases to operate:
- There needs
to be an assessment of the outcomes for those patients whose treatment
has been paid for by the Fund, to see what impact it has had;
- If there is clear evidence of
beneficial outcomes, then that evidence needs to be built on in
constructing the new value-based pricing scheme, and applied to
treatments for conditions other than cancer;
- A defined funding mechanism
needs to be developed which will allow drugs which have been paid
for by the Fund to continue to be available to individual patients.
Information about clinical drug
39. Another issue that we discussed in the context
of health technology appraisals was the availability to NICE of
all the relevant information about a drug. It has recently been
argued that some pharmaceutical companies keep information about
drugs trials out of the public domain and thereby undermine the
quality of information available to regulators, clinicians and
patients about the efficacy and safety of drugs.
40. When we raised this with Sir Michael Rawlins,
he acknowledged that pharmaceutical companies are not legally
obliged to publish all the available data about drugs, and that
"More importantly, they are not under a legal obligation
to publish negative results".
He also acknowledged that NICE was not necessarily in a position
to examine all the data about a new drug down to the level of
individual patient data:
I am not sure, to be honest, whether we would ever
have the resources to do individual patient metaanalyses
on every single thing we looked at. It is a big task to do that.
What we do want to know is the summary statistics, the summary
findings, the details of how the study was done and so on and
so forth. For our purposes, looking particularly at the new products,
necessarily always requiring individual patient data might be
over the top.
41. Sir Andrew Dillon, Chief Executive of NICE,
told us that realistically there were few levers that NICE could
use to ensure that it had all the information about drug trials
that a pharmaceutical company held in its records:
We don't know what we don't know. We get the medical
directors of pharmaceutical companies to sign a statement that
indicates, to their knowledge, that they have supplied all the
data that is relevant to the appraisal that we are undertaking.
Once we have that, it is then very difficult for us to be able
to make a judgment that that may not be the case. Clearly, if
there is some indication from elsewhere that there is data that
the company is not making available, then we already pursue the
companies. We pursue it anyway. That is something that we would
do as part of the normal course of the appraisal. In the end,
if there was something that was absolutely fundamental to a robust
and credible set of recommendations being formulated, based on
data that was not being released that we had reasonable confidence
existed, then we could halt the process until that data was made
available and we could make the reason for halting the appraisal
42. The issue of transparency of research evidence
about drugs goes much wider than just the effects on the NICE
appraisal process. Much of what we discussed would be a matter
for the MHRA as regulator, as Stephen Whitehead of the ABPI said,
and the legal framework is set by European law. In July 2012 the
European Commission produced a proposal for a Regulation of the
European Parliament and of the Council on Clinical trials on medicinal
products for human use. The Commission envisages that this will
repeal the 2004 Clinical Trials Directive, which set the framework
within which present UK regulations on clinical trials are cast.
We are pleased to note that our colleagues on the Science and
Technology Committee have announced an inquiry which will examine
these wider issues.
43. Professor Haslam, in his pre-appointment
hearing, summed up the situation succinctly:
This whole complex issue of unavailable data...is
an extraordinarily important area. I find it impossible to come
up with a good argument as to why all data should not be released.
But at the same time I very much recognise that the pharmaceutical
industry has done a huge amount of good. You only have to look
at a condition like AIDS to see the remarkable changes there have
been for HIVpositive patients in the last few years. Therefore,
we mustn't demonise the pharmaceutical industry. But, following
this particular discussion about open data, they are in a difficult
position, where they have to rebuild trust with professionals
and the public. One of the ways of doing that would be to be much
more open with their data.
44. This issue also raises questions of professional
obligation for those researchers who are members of regulated
clinical professions. Sir Andrew Dillon told us that NICE requires
the medical directors of pharmaceutical companies to sign a statement
certifying that they have provided all information relevant to
a particular appraisal. Beyond that, the guidance from the GMC
for those doctors involved in drug trials is explicit:
You must report adverse findings as soon as possible
to the affected participants, to those responsible for their medical
care, to the research ethics committee, and to the research sponsor
or primary funder where
relevant. You must make sure that bodies responsible for protecting
the public, for example, the Medicines and Healthcare products
Regulatory Agency, are informed.
45. The Committee believes there
should be both a professional and legal obligation to ensure that
all regulators, including NICE, have access to all the available
research data about the efficacy and safety of pharmaceutical
products. All information arising from drug trials should be in
the public domain in an accessible and properly anonymised form,
including any negative information - as Stephen Whitehead of the
ABPI said, "negative trials often give you as much information
that is helpful as positive trials."
46. The Committee also recommends
that the pharmaceutical industry should introduce a new code of
practice covering research. This should include an obligation
to make public all data about drugs which are in current clinical
use once they have been through an appropriate peer review process.
These are measures that pharmaceutical companies can take now
without waiting for the new Clinical Trials Regulation to be approved.
47. The Committee also recommends
that the GMC reiterates its guidance on drug trials to its members,
and reminds them that failure to abide by these principles could
lead to fitness to practice proceedings being taken against them.
48. The Committee does not believe
it should be either legal or considered ethical to withhold research
data about pharmaceutical products. It is therefore concerned
that this simple principle is not universally applied in practice,
and also concerned by the implication of Sir Andrew Dillon's evidence
that NICE are making appraisals of drugs without having access
to all relevant data. The Committee welcomes the current review
of these issues by the House of Commons Science and Technology
Committee and recommends that Committee should examine the nature
of both the legal and ethical principles which should cover these
issues and how to make those principles enforceable in practice.
49. It is clearly very important that those who
have particular conditions should have their views represented
in any discussions within NICE on whether or not to approve drugs
designed improve that condition. Laura Weir from Patients Involved
in NICE told the Committee that:
...there is no clear role for patient groups. We
are invited to give oral evidence when NICE are approving a medicine
and we are also invited to give a written submission before then.
It is not clear whether NICE want us to literally comment on the
assumptions going into their model or if they want a case study.
Do they want to know how this drug has impacted on someone's life?
Do they want us to produce more qualitative or quantitative evidence?
It is very unclear where we fit in in the grand scheme of things.
50. Sir Andrew Dillon, Chief Executive of NICE,
disagreed with the suggestion that the role was not clear, but
accepted that the experience could be improved:
It is almost certainly the case that the experience
of engaging with NICE by individuals who might sit on our advisory
bodies, or by organisations that work with us across more than
one piece of work, is not necessarily always as good as they would
want. That is partly influenced sometimes by the nature of the
outcome, but it is also influenced by the challenge of individuals,
and indeed organisations sometimes with very substantial resources,
engaging with what is, I quite understand, in some circumstances
a complex and quite an intimidating process... We talk to the
[Patients Involved in NICE] group all the time about ways in which
we can improve their engagements, but we are very willing to do
more, if we can, within the resources that we have available to
change our processes where engagement is not clear or is not adequate
and to do our best to support individuals when they work with
51. NICE clearly does involve patients in its
work, but Sir Andrew acknowledged that there is room for improvement.
It is an area that needs to be kept under constant review. It
is important for the credibility of NICE and for the decisions
that it makes that the patient voice is effectively and openly
represented in all its work.
14 Ev 41 Back
Ev 49 Back
Q 11 Back
Q 68 Back
Q 36 Back
The Coalition: our programme for Government, Cabinet Office,
May 2010, page 25. Back
A new value-based approach to the pricing of branded medicines,
Department of Health,16 December 2010. Back
Ibid, chapter 4 Back
See, for example, Qq 20 and 21 Back
Q 125 Back
Consultation document, paragraph 2.8 Back
Ibid, paragraphs 4.2 and 4.3. Back
Ev 61 Back
Q 35 Back
Q 130 Back
Q 35 Back
Evidence taken before the Health Committee on 11 December 2012,
HC 831-i, Q 40. Back
See Ben Goldacre, Bad Pharma, Fourth Estate, September
Q 103 Back
Q 94 Back
Q 111 Back
Qq 43-44 Back
Evidence taken before the Health Committee on 11 December 2012,
HC 831-i, Q8. Back
Good practice in research, General Medical Council, April
2010, paragraph 16. Back
Q 41 Back
Q 13 Back
Q 123 Back