The rationing debate: Maximising the health of the whole community. The case against: what the principal objective of the NHS should really beBMJ 1997; 314 doi: https://doi.org/10.1136/bmj.314.7081.669 (Published 01 March 1997) Cite this as: BMJ 1997;314:669
- John Harris, professor of bioethics and applied philosophya
- a Institute of Medicine, Law, and Bioethics, Universities of Manchester and Liverpool, Centre for Social Ethics and Policy, Manchester M13 9PL
Patients rationally want three things from health care. They want the treatment that will give them maximum life expectancy coupled with the best quality of that life, and above all they want the best possible opportunity of getting the combination of quantity and quality of life available to them given their personal health status. I believe that each citizen has an equal claim on the protection of the community as expressed by its public healthcare system, and this means that each is entitled to an equal chance of having his or her, necessarily individual, health needs respected by any publicly funded healthcare system.
Means and ends
It is common ground I suppose that we have to think about the ethics both of means and of ends. Even if it were to be accepted that the healthcare system ought principally to aim at maximising aggregate health gain, it does not follow that the most effective ways of achieving this are legitimate. If all seriously ill people were to be allowed to die this might dramatically improve the aggregate health of the community at large. I hope such a policy would not seem ethically defensible. Yet this is precisely what measures which use quality adjusted life years, or similar mechanisms, do: they systematically accord preference to those who have better health prospects, and, by selecting against those with worse prospects, tend to improve the aggregate health status of the whole community at the expense of the life chances of those with poorer prognosis.
We should notice that to make aggregate improvements a principal objective, even if not the only objective, is to imply the subordination of the health needs of individuals to something very abstract, and in some circumstances something very trivial indeed–namely, the improved health status of the whole community. For this could imply sacrificing the life of one person who was very ill and expensive to treat, if doing so would make even a tiny improvement to the aggregate health status, an improvement which no individual would even notice
Distributive justice must be built into any articulation of principal objectives for the NHS, but it cannot be enough to define the relevant principle of distributive justice in terms of a more equal distribution of health across populations, because such an objective could be achieved as much by levelling down as by levelling up. One method of allocating a scarce resource which apparently satisfies the requirements of justice is, of course, not to allocate that resource to anyone. All are then treated equally.
The fallacy of such a supposition is easily illustrated. The principles of justice, and indeed the principles of equality, are moral principles, principles that are designed to be more than impartial, that are designed among other things to respect and to do justice to people. In some sense this must involve some benevolent attitude to people which is often abbreviated as “respect for persons.” Such an attitude to others is as different as it is possible to be to that of simply showing an equality of lack of respect or an equal indifference to their fate.
So, neither the failure to allocate resources that would save lives or protect individuals nor the simple attempt to move towards a more equal distribution of health could be part of a claim to satisfy the requirements of equality or justice conceived of as moral principles (and how else are we to think of them?). This is because equality or distributive justice has at its heart the claim that people's lives and fundamental interests are of value, that they matter. Anyone who denied resources which would protect life and other fundamental interests is not valuing the lives of those to whom she denies these protections. Although she might be treating people equally in the sense of treating them all the same, she is not treating them as equals, as people who matter and hence matter equally.
Now this brings us close to the positive part of my account, because I believe it to be an integral part of any principle of distributive justice that people's moral claims to resources are not diminished by who they are; how old they are; how rich or poor, powerful or weak, they are; or by the quality of their lives. A principle of justice worth its salt covers young and old, healthy and sick, weak and strong, regardless of race, creed, colour, sex, quality of life, and life expectancy. Before further articulating the basis of this principle and what it means for the objectives of the NHS we must take a brief look at the concept of efficiency.
Efficiency in the delivery of health care is often defined in terms of maximising beneficial health care or of maximising health outcomes. These styles of definition of efficiency simply beg the question at issue. This question is: what is the good to be delivered by health care? They beg the question because they imply that the greater the health gain per treatment the greater the efficiency of that treatment. This implication is true in one context or application but false in another and it is the conflation of applications, either negligently or deliberately, which gives such plausibility as it has to the proposition that the NHS ought principally to maximise aggregate improvements in health status.
It is true that in order sensibly to maximise health outcomes you need an acceptable measure of success or failure. However, prioritising those outcomes you can best measure and calling it “maximisation of health outcomes” is letting the tail wag the dog. Any measure of what health care tries to maximise which counts life years after treatment faces a problem. The problem turns on the difference between selecting between different treatments for the same patient and selecting between different patients for the same treatment.
This distinction is of the first importance. If you are choosing between rival therapies for the same condition you would be wise to choose the therapy which maximises health outcomes. However, it is a fallacy to suppose that the measure of what is the best or most efficient treatment for a particular patient or condition can also be the measure of the most efficient or best way of distributing resources for care among patients when this amounts to prioritising patients for treatment rather than treatments for patients. The question of which is the most efficient treatment for this patient or condition is not the same as the question: which patients or groups of patients is it efficient or beneficial to treat? This is because there is an equivocation over the meaning of “beneficial” in the two contexts and a problem about incompatible ways of quantifying the size of benefit.
If the millionaire and the pauper both lose all they have in the stock market crash, on one way of thinking about the loss, each has suffered the same degree of loss, each has lost everything. On another, each has suffered a different quantity of loss measured by the total sum lost. There is no straightforward way of reconciling these different approaches. If we are searching for an equitable approach to loss it is not obvious that we should devote resources allocated to loss minimisation to ensuring that the millionaire is protected rather than the pauper. The same is true of health gain. Even if it is agreed that resources devoted to health care are resources devoted to minimising the loss of health or maximising the health gain, it could not be demonstrated that the person who stands to lose more life years if they die prematurely stands to suffer a greater loss than the person who has less life expectancy. Nor can it be shown that the measure of health gain must equate to the number of life years, quality adjusted or not, which flow from treatment.
If you and I are competitors for treatment and I will have a better health outcome from treatment than you, but both of us will make a health gain that is significant and important to us, automatically preferring to satisfy my needs rather than yours seems unfair. Why should my life be judged more worth saving because I am more healthy rather than more intelligent, say, or more useful? Arguments can (and have) been made on both sides, but to define need, for example, in terms of capacity to benefit and then argue that the greater the number of life years deliverable by health care, the greater the need for treatment (or the greater the patient's interest in receiving treatment) is just to beg the crucial question of how to characterise need or benefit.
Equally, to define efficiency in terms of “the maximisation of health outcomes” and then argue that efficiency demands that the NHS aims at maximising aggregate health gain across the whole community is just to beg the question as to how we should think of the gain or benefit to be delivered by the NHS. Efficiency is like motherhood and apple pie; no one can admit to being against it. Arguably health outcomes are maximised and a healthcare system operates efficiently when more people who can derive significant benefit from it are given their chance of access to health care.
I suggested at the start that patients want the treatment that will give them maximum life expectancy coupled with the best quality of that life and the best possible opportunity of getting the combination of quantity and quality of life available to them. Maximising aggregate improvements in health status of the whole community will not necessarily be a rational strategy for achieving these three objectives. Whether it is or not will depend on one's existing or probable health status. This in turn will depend on many things, including one's genetic constitution. If one principal aim of the NHS ought to be to give the people it serves what they want for themselves then this is unlikely to be the maximisation of aggregate improvements in health status. People tend to want the best for themselves and those they care most about, and a policy aimed at maximising aggregate improvements in health status will tend to favour those with the best prospects of large improvements, those with a “healthy” genome for example. People would only be likely to choose such a policy if they could be sure that they themselves would likely benefit.
NHS is there to protect life and liberty
Imagine an industrialised state that has big conurbations where millions of citizens are concentrated, many smaller towns, and thousands of tiny villages. It has vast sparsely populated tracts of agricultural land and vaster mountainous areas and wilderness where few people live. How should it distribute its access to health care? Probably it will place the major hospitals and medical schools in the centres of population, but smaller hospitals and medical centres will serve the smaller towns and isolated villages. For the remotest areas there will probably be an air rescue service or even a flying doctor or flying hospital service.
For geographical reasons if for no other, those in the most remote regions will be generally more expensive to treat. To fly the remote farmer and backwoodsman to the major centres of excellence for specialised treatment will be naturally more costly and hence less cost effective than to bus suburban commuters downtown. We will assume, what is probably true, that the funds devoted to servicing the health needs of citizens who are geographically remote from major centres would have treated more people had they been allocated to urban populations. Why do societies divert resources available for health care away from the more numerous city dwellers in a way which must adversely affect their ability to maximise aggregate improvements in health status or indeed to maximise numbers treated?
I believe the ends subserved by public healthcare systems are broadly the same as those which justify the high priority given to national defence. All governments and would be governments boast the strongest commitment to national defence. The question that is seldom asked is what is national defence for, what justifies its prominent place in national priorities? The simplistic answer is, of course, that without national defence there might be no nation and hence no national priorities. But pressed further it is reasonable to ask for the underlying values and interests it subserves.
Arguably protecting citizens against threats to their lives, liberties, and fundamental interests is the first priority for any state. When in 1651 Thomas Hobbes wrote “The obligation of subjects to the sovereign, is understood to last as long, and no longer, than the power lasteth, by which he is able to protect them he was providing an answer to this question. On this view, any citizen's obligation to the state and to obey its laws is conditional on the state for its part protecting that citizen against threats to her life and liberty. If we reflect on what citizens today want and need in the way of protection I believe we will find that in most contemporary societies the most important threats to life and liberty come not in the form of soldiers with snow on their boots but from illness, accident, and poverty. This is why it is arguable that the obligation to provide health care, and in particular life saving health care, to each and every citizen, regardless of its effect on the aggregate health status of the community, takes precedence over the obligation to provide defence forces against external (and often mythical) enemies.
There is a good principle which states that real and present dangers should be met before future and speculative ones. If this is right the healthcare system should have first claim on the national defence budget. I should make clear that no part of my argument assumes a given budget for health care; rather I argue that the budget could and should be larger, that the health budget has first call on the defence budget, but that whatever the budget is, there are ways of distributing the budget which are to be avoided because they are unjust.
Another feature of the state's obligation to defend its citizens which is often overlooked is its egalitarian nature. Just as each citizen owes his or her obligation to obey the law regardless of such features as race, religion, sex or age, quality of life, or prognosis, so the state must discharge its obligation of protection with the same impartiality. If we expect people to obey the law even though their life expectancy is short and the quality of their life poor, we must not deny them the equal protection that is an essential part of the social contract. I have suggested that the protection of the healthcare system is one of the principal elements of the state's side of this contract and that discrimination against those with poor quality of life or shorter life expectancy in the allocation of such resources is a betrayal, not only of those citizens, but of the social contract. Where all cannot be treated and priorities must be set the basis of prioritisation should not be the effect on the aggregate health of the whole community, for this will tend to discriminate against those arguably most in need of health care.
The principal objective of the NHS should be to protect the life and health of each citizen impartially and to offer beneficial health care on the basis of individual need, so that each has an equal chance of flourishing to the extent that their personal health status permits.