|
Pharmacoeconomics and Market Access in Europe:
Case Studies in Scotland and The Netherlands
Keith Tolley MSc, Director, Market Access Unit Mapi Values, Macclesfield, UK, and Maarten Postma, University of Groningen,
Groningen, The Netherlands
(This paper is based on the Workshop that the
authors organised during the last European
Conference of ISPOR in Florence (6-8 November
2005). Maarten Postma is a member of the Dutch
”Commissie Farmaceutische Hulp” (CFH), Keith
Tolley is an economic assessor for the New Drugs
Committee of the Scottish Medicines Consortium
(SMC). The CFH advises on the reimbursement of
new drugs, whilst the SMC issues guidance on
the use of new drugs and indications to the
National Health Service in Scotland. The views
expressed in this article are the responsibility of
the authors and not do necessarily represent the
opinions of the CFH or NDC/SMC.)
Introduction
Pharmacoeconomics has become established as
a valuable aid to drug reimbursement, pricing and
market access decisions for new drugs in many
European countries. However, to be useful for
such purposes, the methods used should be
robust and the results of pharmacoeconomic
evaluations timely. To aid robustness, guidelines
for the conduct of pharmacoeconomic studies
have been produced in numerous countries -
many of these guidelines have been summarised
and are available on the ISPOR website [1].
Timeliness for market access decisions means
that pharmacoeconomic evidence is required at
or around launch of a new product.
In this article we provide an overview of two
established European agencies that use pharmacoeconomics
as part of a system of rapid technology
appraisal to aid reimbursement and market
access decisions for new drugs at launch: the
Scottish Medicines Consortium (SMC) and the
Netherlands ”Commissie Farmaceutische Hulp”
(Committee for Pharmaceutical Help; CFH).
Scottish Medicines Consortium
The Scottish Medicines Consortium came into
being in 2002. It provides guidance and recommendations
to Health Boards in Scotland on the
use of all new pharmaceuticals, covering new
chemical entities and new indications for existing
licensed drugs (launched since 2002). The SMC
appraisal represents a two-stage process.
The first stage involves a company submission to
the New Drugs Committee (NDC) of the SMC of
the clinical and health economic data supporting
the new product/indication. This submission is
reviewed in detail by an NDC review team consisting
of a clinical assessor, an economic assessor
and a lead assessor (the latter is the only one
of the three who is a full member of the NDC).
Final draft guidance on the clinical and costeffectiveness
of the product, and a recommendation
on use (which can be one of accepted for
use, accepted for restricted use, or not accepted)
is drawn up by the NDC. The NDC stage involves
only consideration of the clinical and cost-effectiveness
evidence supporting the drug.
The second stage of the process involves consideration
of the draft guidance by the SMC
Committee including a response by the manufacturer
(who are sent the draft guidance after the NDC meeting), but also taking into account other
factors such as budget impact, level of unmet
need or innovativeness of the drug, precedence
regarding earlier decisions for drugs in the same
class/patient group, and patient advocacy submissions.
The final guidance is issued by the
SMC and posted on the SMC website about a
month later.
It is not the intention here to go into detail about
the process involved in an SMC submission - this
is adequately covered on the SMC website at:
www.scottishmedicines.org.uk. However, a
number of observations can be made about the
NDC and SMC process and the methods/
approaches adopted for pharmacoeconomic
analysis.
The SMC review at least provides
a test bed for the health
economic analyses, and the first
indication of the strength of the
economic case for the product.
Firstly, the whole process takes about 4 months
to complete from receipt of manufacturer submission
to final guidance being issued on the
SMC website (assuming no appeal). This rapidity
has the advantages of timeliness associated with
delivery of guidance to the health boards in
Scotland soon after the introduction of a new
drug, and allows manufacturers with positive recommendations
rapid access to the market (especially
as manufacturers can submit prior to
launch and plan to receive guidance straight after
launch). The downsides are the limitations of clinical
data at launch for pharmacoeconomic analysis
producing high uncertainty in the cost-effectiveness
results, a greater risk of false negatives
or false positives due to speed of appraisal, and
the possibility that cost-effectiveness of a product
can change over time i.e. generally improve, for
example due to dose reductions over time, or new
clinical/economic evidence. Although not currently
undertaken, this latter concern could be somewhat
addressed via the use of a formal review
period, whereby the original recommendation is
reviewed by the SMC say 2-3 years later if there
are reasons to do so (e.g. new evidence).
Secondly, whilst the process is rapid, it does not
mean it lacks robustness in the technical requirements
for the pharmacoeconomic submission.
The guidelines for the pharmacoeconomic submission
are available on the SMC website and are
based on (but do not completely follow) NICE
methodological guidance [2]. The preference is
for net cost per QALY data for the new product (although other forms of economic evaluation are accepted if appropriate),
with choice of comparator(s) representing the drugs most likely to be displaced
in Scotland. If, as typically occurs, there is a lack of head to head data
against such comparators this has to be overcome through the use of indirect
comparisons involving systematic review, meta analysis or other forms
of evidence synthesis. In addition, the time horizon has to be appropriately
specified and, if lifetime, adequately modelled from the shorter term data.
Resource utilisation and utility data should be robustly collected and sufficiently
relevant to the Scotland/UK context, and a full and transparent sensitivity
analysis should be conducted. Adherence to good decision analytic/
modelling practice (e.g. in the use of Markov models) is also very important.
There should be a clear link between the clinical evidence submitted and the
economic evidence. Currently, NDC requirements are (arguably) not as stringent
as NICE, for example, probabilistic sensitivity analysis is not mandatory
and there is not a set reference case that needs to be followed. However,
unlike NICE, if no economic case is submitted the product will get an automatic
negative recommendation.
Thirdly, the onus is placed on the submitting manufacturer to be convincing
and transparent in the pharmacoeconomic case presented. Rejection of the
economic case is generally on one of two grounds:
a) The submission is transparent and pharmacoeconomic submission of
good quality, but the incremental cost per QALY is considered too high (i.e.
typically above £25-35,000 per QALY, although no formal threshold applies).
Not surprisingly, this reason for rejection is relatively uncommon, and
b) There are flaws or a lack of transparency in one or more key aspects of
the pharmacoeconomic submission, such that the assessor cannot determine
the actual cost-effectiveness of the product. Most rejections are on this
ground (i.e. the economic case has not been demonstrated).
Hence, it is important for the manufacturer to make sure they are using sufficiently
robust pharmacoeconomic methods (but not over-complex) and following
SMC guidelines. There is likely to be sympathy to limitations in data
faced by manufacturers, but this should not preclude efforts to produce a
sufficiently robust pharmacoeconomic model for the submission. Planning
early can help overcome some of these problems.
 Figure 1 presents the recommendations issued by the SMC over the period
Jan-October 2004. The impact of the two stage process can be clearly seen
with just less than 70% of drugs reviewed over this period receiving a positive
recommendation, but only 42% being accepted on the clinical and economic
evidence presented at the NDC stage. As mentioned other factors
come into play at SMC stage, but also some of the changed decisions may have come about as a result of improved clarity in the pharmacoeconomic
case on the part of the manufacturer.
It is not currently clear what impact SMC guidance is having on clinical practice
even in Scotland (although this is being investigated in a research project
due to report in 2007), let alone outside of Scotland. However, anecdotally
it appears to be having an impact. As the UK is often the first to launch
market, the SMC is likely to be the first HTA type body a manufacturer of a
new drug will have to submit to. Hence, the SMC review at least provides a
test bed for the health economic analyses, and the first indication of the
strength of the economic case for the product. This means it is important for
the manufacturer to be prepared with a robust economic model, especially
with the advent of the NICE single technology appraisal programme likely to
raise the stakes in the UK and beyond.
Commissie Farmaceutische Hulp (CFH),
The Netherlands
From January 1st 2005 in the Netherlands, a favourable cost-effectiveness
result has become a further criterion for reimbursement, after quality (good
laboratory practices), safety, and effectiveness have been established. In
particular, manufacturers of new drugs who claim added “therapeutic value”
for their product must, in addition to a therapeutic file, submit an economic
file to the Council for Health Insurance (”College voor Zorgverzekeringen”;
CvZ). These drugs aim for a listing on section 1B of the Dutch reference pricing
system (”Geneesmiddelen VergoedingsSyteem; GVS). If newly registered
drugs are deemed comparable to existing drugs, they are clustered in section
1A and a fixed maximum price applies. An economic submission is not
required for such drugs, which are typically ”me-too's” and generics.
However, to attain 1B listing an economic submission is required - which
should consist of a cost-effectiveness or cost-utility analysis with preferred
outcomes in net costs per life-year gained or QALY gained, but not a costminimisation
analysis. The ”Commissie Farmaceutische Hulp” (CFH), advises
CvZ on reimbursement issues and pharmacoeconomists have participated
on the CFH for several years now.
The “cost-effectiveness” 4th hurdle that was erected in The Netherlands in
2005 was not a sudden event. For some years previously an arrangement
had been in place whereby manufacturers were free to add an economic file
to their reimbursement claim. However, this was not mandatory.
Nonetheless, in many cases manufacturers took the opportunity to submit a pharmacoeconomic case. For example, economic files were submitted prior
to 2005 to underpin reimbursement claims for clopidogrel (Plavix®) and
pimecrolimus (Elidel®). These voluntary submissions enabled all parties
(the manufacturers, CvZ and CFH) to gather experience on how to handle the
pharmacoeconomic aspect of reimbursement submissions.
In dealing with the economic files, CFH and CvZ have made use of 19 guidelines
for pharmacoeconomic research, derived from the Institute for Medical
Technology Assessment in Rotterdam that were developed in 1999 [3].
Important guidelines include the use of an appropriate perspective, the
choice of appropriate comparator, the discount rate to be used, and the need
for an adequate time horizon. CFH/CvZ determine an overall score based on
adherence to these guidelines, which contributes to the final assessment of
the quality of the pharmacoeconomic claim. For example, the company submission
for the anti-platelet drug clopidogrel adhered to more than 60% of
the guidelines. Ultimately, it is the CvZ who advise the Ministry of Health -on
the basis of the CFH-advice - on the quality of the pharmacoeconomic file
submitted. In principle, this could mean an advice to the Minister to intepret
the economic file as adequate (for example, adhering to all important guideline
items), even if the evidence for the drug is that it is not sufficiently cost-effective. For the latter consideration, there are only informal thresholds used
in the Netherlands, often around c =20,000 per life-year or QALY gained.
The pharmacoeconomic guidance of the CFH/CvZ can be found on
www.cvz.nl, but is only available in the Dutch language.
The main similarity between the CFH model and the NDC/SMC approach is
in the timing of the assessment, with the pharmacoeconomic evaluation
being based on the clinical data available at or soon after registration. Given
this timing, in both countries only Phase III clinical studies are usually available
to judge the clinical and cost-effectiveness of new drugs, which is
opposite to the situation for full NICE appraisals which may be conducted
several years post registration when observational data on drug utilization
and outcomes is more likely to be available.
In addition, some of the key differences and similarities of the systems
applied by CvZ/CFH- and NDC/SMC are:
-
Both operate with a pharmacoeconomic checklist and use explicit “state of
the art” guidelines for pharmacoeconomic research, although with some
differences in content;
-
For both the time from starting the assessment to actual advice being
issued is 3-4 months;
-
SMC-guidance is intended as a recommendation to local authorities in
Scotland on the use (and indirectly the funding) of a new drug, whereas CvZ/CFH advice is specifically for determination of the reimbursement status
of new drugs;
-
SMC evaluates all new drugs and indications (whether hospital or
non-hospital), whereas CvZ/CFH is primarily directed at the non-hospital
market, and within this only those drugs that claim added value compared
to the existing drugs for a specific indication;
-
For CvZ/CFH a user-friendly electronic version of the pharmacoeconomic
model is required, whereas NDC/SMC does not typically request the manufacturers
electronic model;
-
CvZ/CFH only advises on the quality of the economic file, whereas NDC
attempts an assessment of whether the drug can be considered costeffective
or not;
-
SMC may be overruled later by guidance from a full NICE appraisal, so in
a sense it may be perceived as a preliminary recommendation; CvZ/CFH
provide a final advice unless the manufacturer re-appeals; and
-
SMC allows input from experts and patient advocacy groups, whilst this
only occurs in exceptional circumstances for CvZ/CFH.
Of course, both the CvZ/CFH and the SMC can issue negative recommendations.
This may not be due to the pharmacoeconomic submission alone. The
clinical evidence may be limited to show therapeutic benefit, often due to
weak study design, too small numbers or absence of evidence on “hard endpoints”,
i.e. mortality and clinically relevant morbidity. Furthermore, the evidence
in the clinical studies is not against the comparator drug or technology
used in actual practice, leading to problems reconciling the clinical and
pharmacoeconomic sections. Also, all assumptions (e.g. for resource use)
in the economic model should be based on robust, ideally published data
sources, which is not always the case.
Table 1 shows the outcomes of the evaluations for 3 selected drugs with
economic files for CvZ/CFH, SMC and NICE. This illustrates how, due to a
range of factors (both evidence based and other factors), outcomes can vary
even across decision making bodies that have a similarly high technical standard
for pharmacoeconomic submissions.
The status of the CvZ/CFH-advice is changing as a new health insurance system
is being adopted in the Netherlands from January 1st 2006, with more
independence in decision making being given to individual health insurance
companies. Indeed, a number of individual insurance companies have already “overruled” CvZ/CFH advice and reimbursed methylphenidate OROS.
Additionally, an extension of the reimbursement system to require pharmacoeconomic
files for in-patient drugs to show added value is expected in The
Netherlands.
Conclusions
Recent developments at NICE is a new programme of rapid single technology
appraisals of new drugs/indications at launch [4]. This has started in
2006, initially for selected anti-cancer drugs. It is therefore interesting, and
also to provide learnings for NICE, to assess experiences at SMC and CFH
regarding the use of pharmacoeconomics as part of a process of rapid technology
appraisal. Also, one of our aims in the original ISPOR workshop was
to assess similarities and differences between the SMC and CFH, and so we
have attempted to cover this objective in this article.
References
1. www.ispor.org/peguidelines/index.asp. Last accessed March 27, 2006.
2. NICE: Guide to the methods of technology appraisal, NICE, April 2004.
3. CvZ: Guidelines for Pharmacoeconomic Research. Amstelveen: CvZ, 1999.
4. NICE: NICE to issue faster guidance to the NHS, NICE Press Release, Nov 3rd, 2005.
|
