Download NICE’s Selective Application of Differential Discounting: Ambiguous, Inconsistent and Unjustified

NICE’s Selective Application of Differential Discounting:
Ambiguous, Inconsistent and Unjustified
1,*
James F O’Mahony, MA & Mike Paulden, MSc, MA
2,*
* Both authors contributed equally to this work
1. Department of Health Policy and Management, Trinity College Dublin, Dublin, Ireland
2. Toronto Health Economics and Technology Assessment (THETA) Collaborative, University of Toronto, Toronto, Canada
[email protected], [email protected]
Abstract
2. Why is this a Problem?
The National Institute for Health and Care Excellence (NICE) recently recommended differential
NICE’s amendment presents a number of problems, the first of which is that the amendment
discounting of costs and health effects in the economic appraisal of healthcare interventions in
itself is ambiguously stated. However, the biggest problem with NICE’s guidelines relates to the
certain circumstances. The recommendation was published in an amendment to NICE’s guide
selective use of differential discounting only in certain cases. Other problems relate to NICE’s
to the methods of technology appraisal. The amendment states that differential discounting
choice of discount rates and their attempts to justify the amendment. These problems are
should be applied where “treatment effects are both substantial in restoring health and
described in more detail below.
sustained over a very long period (normally at least 30 years)”. Renewed interest in differential
discounting from NICE is welcome; however, the recommendation’s selective application of
2.1 Problems of Ambiguity
differential discounting raises a number of concerns. The stated criteria for applying
The language used in NICE’s amendment is ambiguous, making it unclear exactly which
differential discounting are ambiguous. The rationale for the selective application of
interventions qualify for differential discounting and whether differential discounting should be
differential discounting has not been articulated by NICE and is questionable. The selective
used in the base-case or merely in sensitivity analyses.
application of differential discounting leads to several inconsistencies, the most concerning of
which is the lower valuation of health gains for those with less than 30 years remaining life
2.2 Problems of Inconsistency
expectancy, which can be interpreted as age discrimination. Furthermore, the discount rates
Since differential discounting can yield more favourable cost-effectiveness ratios, but NICE’s
chosen by NICE do not appear to be informed by recent advances in the theoretical
amendment does not apply universally, this means the guidelines confer an advantage in
understanding of differential discounting. NICE’s apparent motivation for recommending
terms of cost-effectiveness that is enjoyed by only some interventions. Furthermore, as there
differential discounting was to ensure a favourable cost-effectiveness ratio for a paediatric
is no apparent reason to apply differential discounting selectively, there then seem to be a
oncology drug. While flexibility may be appropriate to allow some interventions that exceed
number of inconsistencies between eligible and ineligible interventions. Examples of apparent
conventional cost-effectiveness thresholds to be adopted, the selective adjustment of
inconsistencies are shown in Box B. Special consideration is given to the potential for age
appraisal methods is problematic and without justification.
discrimination in Box C.
1. What did NICE do?
2.3 Discount Rates not Justified by Recent Research or Public Consultation
NICE usually applies equal discounting of costs and health effects at an annual discount rate of
The discount rates adopted by NICE do not accord with the findings of important recent
3.5%. However, in July 2011 NICE amended their guidelines to recommend that differential
research on differential discounting. As such, they are without a sound methodological
discounting of 3.5% for costs and 1.5% for health effects be applied in special circumstances
justification. NICE did consult their Citizen’s Council regarding differential discounting, but it is
[1]. The guideline requires differential discounting be applied in cases in which “treatment
doubtful if this consultation offers any legitimacy to the amendment. Both these issues are
effects are both substantial in restoring health and sustained over a very long period (normally
considered in Box D.
at least 30 years)”. Importantly, this means not all interventions will be eligible for differential
discounting, only those matching the criteria.
Box B: Examples of Inconsistencies
Prevention vs Cure
Box A: What is Differential Discounting?
NICE’s amendment stipulates that differential discounting be applied to interventions for
“restoring health”, which presumably precludes preventative interventions. This leads to
Economic evaluations typically value future costs and health effects less relative to those
an apparent inconsistency whereby prevention is assessed on a less favourable basis
occurring in the present. This is called discounting and is applied in economic evaluations
than cure.
of healthcare interventions such as cost-effectiveness analysis (CEA). Discounting is
important, as cost-effectiveness ratios can be very sensitive to discounting
Cure vs Maintenance Therapy
The criterion in NICE’s amendment that health gains be “sustained” could be interpreted
Usually, discounting is applied equally to costs and effects. However, differential
as precluding interventions that required a maintained course of treatment rather than
discounting is the practice of applying relatively less discounting to health effects than to
offering a cure with a single course of treatment. If so, this would disadvantage
costs. Typically this results in more favourable cost-effectiveness ratios than when equal
maintained interventions such as anti-retroviral therapy relative to once-off cures.
discounting is applied, especially in cases in which costs are upfront but benefits occur
later, as in the case of prevention. This has implications for resource allocation as, in
The Many vs the Few
principle, interventions with more favourable cost-effectiveness ratios are more likely to
The amendment’s requirement that benefits be “substantial” could be interpreted as
be approved for use.
precluding those treatments that provide only a modest health gain per patient. This
again leads to apparently inconsistent treatment of interventions that have similar
Differential discounting is applied in very few countries worldwide. Typically those
aggregate health gains, depending on whether they are realised through modest gains
countries that apply differential discounting apply it universally to all interventions.
for many patients or large gains for a few patients.
1
4. Why does this Matter?
Box C: Age Discrimination
By apparently attempting to avoid a political controversy over an expensive drug, NICE
Possibly the most important example of inconstancy stemming from NICE’s amendment
adjusted their appraisal methods in an arbitrary way, which creates the potential for
relates to age discrimination. The restriction to interventions with effects lasting 30 years
unjustified differences in how different interventions are appraised. These differences
or more precludes interventions for those with a remaining life expectancy less than 30
potentially lead to unjustified differences in cost-effectiveness and, by extension, unjustified
years from the benefits of differential discounting. Since those with shorter remaining life
differences in coverage decisions. In doing so, NICE has compromised their objectivity and
expectancy are often, although not always, older people, the amendment leaves NICE
undermined their scientific credibility.
open to criticism of age discrimination.
The charge of age discrimination is one that NICE has faced before regarding the use of
quality-adjusted life-years (QALYs) as a metric of health gain [2]. These accusations have
Box E: Stop Press! April 2013 Change to Discounting Guidance by NICE
previously been soundly refuted, as NICE’s use of the QALY values health gain equally in
Since the acceptance of this poster NICE have made a further amendment to their
individuals of any age [3]. However, this defence does not apply to NICE’s selective use of
discounting guidance. In April 2013 NICE published a complete revision to their methods
differential discounting.
of technology appraisal [7]. The recommendation for differential discounting in special
circumstances has been removed, which is welcome. However, in its place, the new
guidance now states that equal discounting at 1.5% for both costs and health effects
should be applied in special circumstances very similar to those in the 2011 amendment,
3. So why did NICE do this?
whereas the standard reference case discount rate of 3.5% remains for other
interventions.
NICE apparently adopted selective differential discounting to achieve a more favourable costeffectiveness ratio for mifamurtide, an expensive drug used to treat bone cancer in children
and young adults. The drug’s high price meant it had a poor cost-effectiveness ratio and so
While the new guidance avoids some specific problems related to differential
should have been rejected by NICE. However, such a rejection would clearly be politically
discounting, the selective use of lower discount rates remains a problem, as the
unpopular. It appears that NICE selectively changed their methods guidance to achieve a more
inconsistencies between otherwise similar interventions remain. We urge NICE to
favourable ratio so the drug could be approved and thus avoid controversy.
reconsider the use of selective discounting.
References
Box D: Ignoring Evidence and Manufacturing Legitimacy
1. National Institute for Health and Clinical Excellence (NICE). Discounting of Health Benefits in
Special Circumstances; Available from: http://www.nice.org.uk/media/955/4F/
Clarification_to_section_5.6_of_the_Guide_to_Methods_of_Technology_Appraisals.pdf.
[Accessed June 19th, 2012].
The discount rates chosen by NICE do not reflect recent research on what the
appropriate differential between discount rates might be [4,5]. This research has shown
that given the decision rules employed by NICE, any discounting differential should be
2. Harris J. QALYfying the value of life. J Med Ethics 1987;13(3):117-23.
linked to the rate of growth in the cost-effectiveness threshold. However, NICE ignored
3. Paulden M, Culyer AJ. Does cost-effectiveness analysis discriminate against patients with
short life expectancy? Matters of logic and matters of context. Theta Collaborative Working
Paper Series. Toronto: University of Toronto; 2010.
this methodological literature and recommended apparently arbitrary rates.
4. Claxton K, Paulden M, Gravelle H, Brouwer W, Culyer AJ. Discounting and decision making in
the economic evaluation of health care technologies. Health Econ 2011;20(1):2-15.
NICE referred the issue of discounting to its Citizen’s Council of lay people [6]. They asked
the council to consider the issue of differential discounting in the context of a
5. Paulden M, Claxton K. Budget allocation and the revealed social rate of time preference for
health. Health Econ 2011;21(5):612-8.
hypothetical childhood cancer drug; an example that shared a marked similarity to
mifamurtide. While the Citizen’s Council endorsed the application of differential
6. National Institute for Health and Clinical Excellence (NICE). NICE Citizens Council meeting:
How should NICE assess future costs and health benefits; Available from: http://
www.nice.org.uk/media/06B/B8/Citizens_Council_report_on_Discounting.pdf. [Accessed
October 17th, 2012].
discounting, the consultation arguably unhelpfully conflated issues of greater willingness
to pay for interventions for children and for severe diseases with questions of
discounting. As such, it is not clear if the consultation adds any legitimacy to NICE’s
7. National Institute for Health and Care Excellence (NICE). Guide to the methods of technology appraisal 2013; Available from: http://publications.nice.org.uk/pmg9 [Accessed May 13th
2013].
amendment.
2