Dichotomising Cost-Effectiveness undermines its ethical justification and potential contribution to health

Jack Dowie�LSHTM

Mette Kjer Kaltoft�University of Southern Denmark

IHEA Basel 2019�Panel on South-West Interventions

ISPOR Value & Outcomes Spotlight March/April 2015 Vol. 1, No. 2

And cost-effective procedures

are not necessarily effective

The most relevant ‘central’ threshold [in the NHS] is estimated to be £12,936 per QALY…�“ if anything, likely to be an overestimate”

Claxton Report

Background: Cost-effectiveness analysis involves the comparison of the incremental cost-effectiveness ratio of a new technology, �which is more costly than existing alternatives, �with the cost-effectiveness threshold.�

This indicates whether or not the health expected to be gained from its use exceeds the health expected to be lost elsewhere as other health-care activities are displaced. The threshold therefore represents the additional cost that has to be imposed on the system to forgo 1 quality-adjusted life-year (QALY) of health through displacement.

Claxton 2015, Abstract

The

NICE BIASES

The purpose of the NICE appraisal is to decide whether the new technology

‘works well’ (is clinically effective)

and

‘good value for money’ (is cost-effective).

​

At no stage of the scoping or appraisal process is an innovation that claims to be cost-effective and 'good value for money', but not 'clinically effective' in relation to the comparators, eligible for consideration

The full 4 quadrant CE Plane

​

with Claxton NE

quadrant rotated and placed in SW

quadrant

So

• All innovations below linear ICER are Cost-effective
• If threshold is infinite (ICER line Vertical), no SW innovations are cost-effective, only those in Eastern hemisphere
• If threshold is zero (ICER line horizontal) all SW innovations are cost-effective and none of those in Northern hemisphere

Terminology

• ‘Incrementally CE’ if in NE quadrant
• ‘Decrementally CE’ if in SW quadrant
• BUT both are CE
• May be helpful for operational reasons to characterise the differing origins of CE, but the two cannot be separated for policy purposes without abandoning the CE principle.
• Separating incremental and decremental cost-effectiveness is as meaningful as separating right-handed and left-handed ambidexterity

​

This argument is not an empirical one

• Not about the NICE threshold (whether 20 or 30k)
• Not about the NHS threshold (whether 12xxx or some other number), or
• Not about where ‘displacement’ occurs in order to fund a new, more expensive innovation

​

And it is not dependent on use of the QALY

The argument applies whatever the metric of effectiveness, but does assume there is one !

5 reasons advanced for not going SW � (beneath a linear ICER)

• Infringes rights
• Harms health (‘lossaversionitis’) (Gandjour)
• Can’t show will actually produce benefits (Sendi, Gafni, Birch)
• Kinked ICER (O’Brien) or MAD (Kent)

WTA ≠ WTP valid at societal level

• Prospect Theory (Tversky& Kahneman) and ‘Psychic numbing’ (Slovic) are legitimate at societal level

​

Infringes rights

• “SW interventions are simply wrong because they take away from them something people already have.”
• …. even if it were to be agreed that current recipients would not be forced to move on to the less effective treatment because it is now the cost-effective one, this argument lacks any justification when extended to those who acquire the same condition in the future.
• Having never enjoyed the effectiveness of the old treatment, they cannot have a right to it taken away from them.
• Those who become ill later cannot ethically be favoured, in relation to other sufferers, simply because they suffer from this disease or condition, rather than from some other one.

​

Harms health

• “SW interventions will produce ill health which will require treatment and impose extra costs”
• Gandjour argues that even prospect of SW interventions will create depression, anxiety etc.
• He may be right, but if ‘lossaversionitis’ is produced then treatment for it needs to be entered into the allocation system and the cost-effectiveness of interventions for it assessed.

​

Required proof can’t be provided

• “SW interventions should not occur unless it can be shown that there will be a net increase in health”
• ‘NICE cannot be expected to reflect what is likely to be marked variation between local commissioners and providers in how they react to an effective reduction in their budget as a result of positive guidance. Given NICE’s remit, it is the expected health effects (in terms of length and QoL) of the average displacement within the current NHS (given existing budgets, productivity and the quality of local decisions) that is relevant to the estimate of the threshold.’ p.8
• We see no justification for imposing higher specificity requirements regarding displacement on SW interventions than on NE ones.

​

Kinked ICER

• “Some SW interventions are acceptable, but only those under a (very) kinked ICER”

​

WTA ≠ WTP (loss aversion)

• “People require much greater compensation for a given loss than they are willing to pay for the same gain”
• Not in dispute at individual level
• Reproduced in studies aggregating individual preferences
• But why should we accept the societal inequity that will result by using this in public policy?

​

‘Psychic numbing’ is legitimate

Asymmetric value function of (descriptive)�prospect theory

SW innovations are everywhere!

• Just presented as SE ones
• Even by NICE itself…

Innovation – yes, please but only if dearer!

• NICE is charged with objectives other than maximising the increase in public health and among its other obligations is to support innovation.
• But this turns out to be biased support, in that no support can be provided for the development of technologies which are simply cost-effective.
• No innovation in the SW quadrant can meet the filter test of clinical effectiveness administered prior to the test of cost-effectiveness.
• So while NICE has a remit to support the adoption of innovative new technologies, in practice the support is confined to those that will cost more.

​

NICE is failing to uphold this critical responsibility to all NHS patients.

…it is the unidentified and unrepresented NHS patients who bear the true (health) opportunity costs. Although finding reasons to approve new drugs is undoubtedly politically expedient, this cannot be ethically literate, because the interests of NHS patients, whether they are identifiable or not, are just as real and equally deserving of the type of care and compassion that can be offered by a collectively funded health care system. It is to be hoped that NICE will begin to place the unidentified NHS patients who bear the real opportunity costs at the heart of its deliberative process… (p.6)

Claxton spot on … except for SW blind spot

Thanks for attending