Is NICE ageist?

In the UK, new health technologies are assessed by the National Institute for Clinical Excellence (NICE). NICE determines the cost incurred for each additional quality-adjusted life-year (QALY) that the new technology provides over and above the currently standard treatment. Though there is considerable flexibility in the process, technologies which offer a cost-per-QALY of £20 000-£30 000 or less would normally be recommended for use. The thought is that, given a fixed total health budget, use of technologies with a higher cost-per-QALY will generally decrease aggregate health by displacing more cost-effective interventions.

One criticism levelled at NICE maintains that its methodology is ageist. Since younger people typically have a longer life expectancy than older people, a life-saving treatment will tend to produce more QALYs in a younger person. So too will a quality-of-life-improving intervention, since it will improve quality of life over a longer period. The NICE approach might be said to systematically favour younger people.

In this issue, Stevens and collaborators (see page 258) respond to this charge. They concede that the cost-per-QALY approach could disfavour the elderly, but argue that it will do so only in rare cases—cases that have never occurred. These would most likely be cases of extremely expensive interventions that cure imminently fatal conditions and restore normal life-expectancy. Moreover, even if such a case did occur, NICE might nevertheless recommend the intervention for use. Stevens and collaborators note that NICE's expert advisory committees have considerable leeway to consider factors besides cost-per-QALY. They also point to various other features of the NICE process that tend to protect against ageist decisions.

In a commentary (see page 263), John Harris and Sadie Regmi respond to this defence of NICE by arguing that the NICE approach is ageist in theory even if not in practice. They claim that it expresses the view that old people ‘are not worth the expenditure of resources’ and uses ‘arbitrary’ considerations, such as one's baseline life expectancy and quality of life, to inform decisions.

Harris and Regmi are surely right to note that a resource allocation process could be ageist ‘in theory’. But the points made by Stevens and collaborators might yet have significance, for ageism in practice may matter too. Harris and Regmi draw an analogy between ageism and racism, and suggest that NICE is rather like a racist person who, despite having racist beliefs, never acts in a racist way. But surely this racist is a less bad sort of racist than one who is thoroughly racist both in thought and action. Even if NICE's methodology is ageist, the fact that this ageism rarely if ever finds its way into NICE decisions may mitigate the problem.

Moreover, as Harris and Regmi acknowledge, one might dispute whether NICE's methodology really is ageist even in theory. Baseline quality of life and life expectancy are arguably relevant to determining the amount of benefit that an individual will derive from a treatment, and it could certainly be questioned whether the amount of benefit produced by an intervention is an ‘arbitrary’ consideration.

Distribution of vaccines in a pandemic

Concerns about the allocation of scarce healthcare resources are also discussed elsewhere in this issue. Hugh McLachlan (see page 317) considers how scarce influenza vaccines should be distributed during a pandemic. The UK's current pandemic plan distinguishes between seven different priority groups, with healthcare staff receiving the highest priority, followed by providers of essential services, followed by those at high medical risk, and then all elderly persons. The plan explicitly aims to minimise the health impact of a pandemic.

Both this aim and the priorities it yields are mistaken according to McLachlan. He favours a model in which the state distributes vaccines according to non-outcome-based duties such as the duty to treat all impartially and the state's duty of care towards those it employs to do dangerous things. He argues that, following this approach, vaccines would be distributed first to ‘those who are at risk of catching the pandemic flu in the line of their duties of public employment’. Any remaining vaccine would then be allocated by means of a lottery in which all others are given equal chances of receiving a vaccine.

McLachlan is right to question the assumption that minimising the health impact of a pandemic should be the only consideration in distributing scarce vaccines. However, I wonder if he has gone too far in the opposite direction. For surely the UK government's concern to minimise health impact remains one important goal of pandemic policy. McLachlan acknowledges this. But he appears to assume that this concern does not amount to a duty, or at least, not a duty as powerful as the other duties he discusses. For example, he claims that, leaving aside public employees exposed to danger, influenza vaccines should be distributed according to an equal-chances lottery. This appears to assume, perhaps unjustifiably, that the state's duty to treat all persons equally outweighs any imperative to minimise negative health impact.