Intended for healthcare professionals

Education And Debate

The rationing debate: Rationing within the NHS should be explicit: The case for

BMJ 1997; 314 doi: (Published 12 April 1997) Cite this as: BMJ 1997;314:1114
  1. Len Doyal, professor of medical ethicsa
  1. a St Bartholomew's and The Royal London School of Medicine and Dentistry, Queen Mary and Westfield College, London E1 2AD


    Much recent discussion has revolved around whether the rationing of health care that is occurring within the NHS should be explicit or implicit.1 Many commentators argue in favour of implicit rationing, for a range of reasons. Opinion appears to be divided between those who claim that implicit rationing will (a) be inevitable since there are no clear criteria on which to base explicit rationing, (b) make patients and providers happier, (c) make the administrative and political processes of healthcare provision run more smoothly. I provide reasons for rejecting each of these contentions, arguing instead that explicit rationing is vital for the moral management of health care.

    The argument from confused criteria

    The creation of an internal market in the NHS appeared to place explicit rationing on the agenda of healthcare providers. Rationing had always occurred within the service but previously it had been camouflaged under clinical judgment. Now purchasers were to draw up plans showing how much was to be allocated to what type of care and why. Providers were to audit clinical staff to ensure that their work conformed to agreed criteria of effectiveness and all was to be open to public scrutiny.

    The expected transparency has not occurred. Health authorities have generally not come clean about their inability to meet demand and have awarded block contracts primarily on the basis of past expenditure, with shortfalls shared between existing clinical services. As a result, the realities of rationing within the NHS have remained where they always were–with clinicians making decisions on the basis of varied and conflicting criteria, often dressed in the guise of clinical necessity.2

    These developments have led to a weary resignation that any ambition to make rationing explicit within the NHS is hopelessly optimistic. It is argued that there are no clear rules according to which rationing should occur and a lack of political will to implement what criteria there are. For example, the health committee of the House of Commons proclaims: “There is no such thing as a correct set of priorities, or even a correct way of setting priorities.”3 Klein concurs: “Given the plurality of often conflicting values that can be brought to any discussion of priorities in health care, it is positively undesirable (as well as foolish) to search for some set of principles that will make our decisions for us.”4 So does David Hunter: “Rationing will always be a messy affair. We should not seek to deny the mess but accept it.”5

    Thus, since “ought implies can” and since explicit rationing seems practically impossible, we are said to be stuck with implicit rationing.

    The argument from the utility of ignorance

    The second argument against explicit healthcare rationing derives from health economics6 and emphasises the emotional consequences of explicit rationing. Explicitly to confront individuals with the fact that because of scarce resources they will not receive health care which they need will make them more unhappy than believing that there is no clinical option but to take what is offered. This distress will be compounded if they discover that other patients deemed more worthy of resources will receive treatments denied them.7 Two noted health economists have described the “deprivation utility” of being kept in ignorance in such circumstances.8

    This idea can be extended to healthcare rationers themselves. Telling patients they will not receive appropriate clinical care for economic reasons is stressful, more so than pretending that the treatment will be futile or just not mentioning it at all.9 As a result, it is again argued, on utilitarian grounds, that implicit rationing makes more sense than that which is explicit.

    The argument from bureaucratic and political effectiveness

    A third defence focuses on the bureaucratic and political difficulties that are said to accompany explicit rationing. Attempting to strike the right balance between competing claims for funding within health authorities is not easy, and the same argument holds for central government attempting to weigh up conflicting demands on the public purse. In such circumstances complaints by the public about the management of explicit rationing will certainly make life more difficult for those responsible for organising health care. Much better then to continue the myth that decisions about the allocation of such care are based on clinical criteria alone.

    Hunter, for example, has supported such mythology: “The public is more likely to accept rationing decisions made by doctors rather than managers and politicians.”5 Clinical discretion in rationing is essential given the diversity of individual cases. If the public becomes aware that more general value judgments–say about cost effectiveness, moral desert, and quality of life–are behind rationing decisions then such discretion may be undermined. Letting the cat out of the bag would then advantage articulate patients who will know how to play the now transparent system. “Lack of visibility,” Klein argues on the same note, “may be a necessary condition for the political paternalism required to overcome both consumer and producer lobbies.”10

    Similar arguments have been developed by Mechanic, who worries that explicit rationing might jeopardise the stability of the political process surrounding health care. He speaks of the “many disaffected people” created by the knowledge of why resource decisions are being made about them and others; whose responses would not be “conducive to stable social relations and a lower level of conflict; and who are likely to confront government and the political process with unrelenting agitation for budget increases.”11 Much better for people to believe–even if it is false–that rationing decisions are inevitable for purely clinical reasons. Then clinicians, managers, and politicians can get on with the job of making decisions in what they believe to be are the best interests of patients. Explicit rationing “will inevitably result in acrimony difficult to manage politically.” 11

    Clear criteria for explicit rationing do exist

    It is hard to believe that anyone really thinks that we should not at least try to understand the criteria which should be used in rationing decisions–to make them explicit in this sense and to compare them with criteria actually used. Refusing to make this attempt is tantamount to giving up the possibility of evaluating either the justice or the efficiency of the rationing process, of accepting that healthcare resources should be distributed in ways which might do as much harm as good.

    Against the background of the explicit moral foundations of the NHS, such pessimism is curious. For in general terms, nothing could be clearer than the ethical principles at the heart of the health service. The most well known and important of these principles is that there should be equal access to health care within the NHS based on equal need.12 The first group of critics might be claiming that there is something inherently confused about the equal need-equal access equation. Conversely, they could be claiming that even if the formulation is clear in principle, in practice it is so bereft of organisational and procedural content that it is of little use to those who must work in the real world of managerial, economic, and moral expediency. Both of these claims are false.

    As regards the first claim, suppose that we define the need for health care as the requirement for specific clinical intervention in order to avoid or to minimise sustained and serious disability.13 What would it mean to suggest that we have no clear understanding of what this means in practice? It is what occurs in the delivery of health care at its best on a daily basis throughout the world. To be sure, there are disagreements about the appropriateness or efficacy of some interventions, diagnoses, and prognoses.2 But this does not detract from the clarity of what we do know or the success of the service that is often delivered. Those who like to emphasise the uncertainty of medicine will no doubt change their tune when they contract serious and treatable illnesses.

    There is similar clarity associated with moral arguments for providing access to appropriate health care on the basis of need. Our potential to flourish as individuals in whatever cultural environment depends on our ability either to participate within or to struggle against it. We require the help of others if we are to discover what we are capable of doing and becoming. Sustained and serious disability inhibits our capacity to interact with and learn from others and is thus in the interests of everyone to avoid if possible.13

    But this is just another way of underlining how vital it is for those so disabled that appropriate health care be distributed on the basis of need and on no other individual attribute.14 On a macro level this means that healthcare resources should be allocated to local populations on the basis of the most accurate needs assessments of which we are capable. This means that generally speaking, resources should be divided proportionally between the different types of disabling and treatable illnesses represented within such populations. Specific types of illness should not be discriminated against on the grounds of popularity or estimations of social worth. Rationing should take place within rather than between different areas of healthcare need.15

    On the one hand, disabling disease may strike any of us without warning and if health care is distributed on any other basis, most of us cannot know for sure whether or not we may individually qualify for it. On the other hand, it is in our interests that those known and unknown persons on whom we socially depend for our potential to flourish will also be kept as healthy as possible. Rationally, therefore, we should want for others what we desire for ourselves.15

    The concepts of equality of need and of access to health care based on it are also reasonably straightforward. Once it is accepted that the focus of any definition of healthcare need should be associated disability and that the macro allocation of resources should take place accordingly, the issue of equality on the level of micro allocation partly reduces to what levels of disability can coherently be deemed to carry with them the same moral entitlement to health care. It also partly concerns how we can ensure that those who are believed to be in such equal need can be assured an equal chance of benefiting from whatever clinical resources are available for its satisfaction.

    Triage is the procedural embodiment of the belief that some levels of disability caused by illness are morally similar enough to warrant the same priority of access.16 When triage is linked to a system of waiting which ideally gives each person within each category of urgency an equal chance of treatment–one based on a first come first served basis–then equal access to available resources will be seen in principle to be provided on the basis of what is accepted to be equal need.15 A trip to any well run accident and emergency department will provide ample practical illustration.

    Of course, people may accept such principles in theory yet argue that in practice they become so muddled as to reduce to confusion. Such arguments confuse substantive and procedural moral issues.17 Moral principles must be interpreted to apply them to specific problems, and unless there are procedures to optimise the rationality of such interpretation, confusion and injustice can indeed follow. For example, the assessment of healthcare need is often based on questionable methods, including ad hoc extrapolations from prior levels of clinical demand.18 Further, the traditional organisation of surgical waiting lists tells us more about the clinical preferences of surgeons and strategies for queue jumping than the just distribution of treatment to patients.19

    Thus confused organisational practices do not necessarily entail confusion within the moral principles which are supposed to inform them. They can also reveal the inability or unwillingness of rationers to take clear principles seriously, or to recognise the rights of patients in whose interests they are supposed to be acting. We should direct our energies to correcting this problem rather than wringing our hands about the inevitability of methodological and administrative chaos.

    Of course, further effort is required to show how theory and practice can be better integrated, and the clarity of theory will benefit as a result. This is the aim of current attempts to create uniform guidelines for clinical diagnosis and treatment, and similar research should be undertaken on various aspects of rationing–for example, triage and fair waiting patterns for different conditions, the non-provision of life saving treatment, the determination of clinical futility.20 Those who support implicit rationing rightly argue that it will be a difficult task.2 However, theoretical clarification and consistent practice will continue to elude us unless decision makers are encouraged to make explicit and publicly defend the criteria for rationing which they do use.

    The disutility of ignorance: microrationing should be explicit

    The second argument made against explicit rationing embraces the utilitarianism of traditional health economics. Thus it suffers from the same blindness to issues of equity as other attempts to reduce rationing decisions to the aggregate calculation of preference–for example, QALYS (quality adjusted life years).15 The key argument is the same: explicit micro rationing will ultimately create more unhappiness–less utility–on the part of both patients and doctors than implicit rationing.

    There is nothing new about the idea that because patients may find certain types of information distressing, they should not be told it. Yet any benefit derived from deception will be sustained only while patients are kept in ignorance. If they discover that they have been deceived, their sense of betrayal will probably far outweigh any distress from being told the truth.21 Therefore, it can just as convincingly be argued that utilitarian clinicians should pretend that they take seriously the right of patients to be told the truth, even if in reality they do not. Indeed, evidence suggests that this is precisely what patients wish, including those who are terminally ill.22

    Similar arguments apply to the suggestion that patients will be less distressed if they are not told about the real reasons why they are denied treatments. Such a discovery could again lead to considerable unhappiness when the deceit is discovered; we cannot calculate the utilitarian outcome of deceit with any certainty. Of course, if we take seriously the right of patients to protest against rationing decisions then such deception will be unjustifiable in any case.

    That clinicians will be happier if they keep patients in the dark about the realities of rationing is just as questionable. This argument works only if it is assumed that the deception will always be successful. Yet, because we cannot be sure of the outcome, sustaining deception over time can itself be distressing, especially if patients begin to ask more direct questions about why they are not receiving care which they have heard is available to others. Also, because their professional guidelines so consistently emphasise the duty to respect the autonomy of patients, good clinicians are increasingly taught to feel uneasy about any form of deception not invited by patients in advance. In any case, to base a decision on the well being of the clinician rather than the best interests of the patient would be unacceptable.

    Macro rationing should be explicit

    Within a democracy an informed public can undoubtedly give administrators and politicians a hard time. Yet citizens should have explicit information about any policies which can dramatically affect their lives.

    Firstly, as JS Mill saw so clearly, unless citizens are given at least the potential for such influence, their own moral development will be damaged: they will not have the same personal stake in either learning about or conforming to the rules of their culture. More specifically, their moral commitment to democracy itself will be undermined. If we accept that democratic participation in public and political life is a good worth pursuing then it follows that the citizenry should be educated about the matters on which their participation is sought.23

    Secondly, informed democratic feedback can improve the effectiveness of public policies through allowing policy makers more accurately to assess the results of their labour. It also helps to make them more reflective, knowing that they may be held to account by those whose interests they are supposed to serve. Such accountability is particularly important in the light of the tendency for vested interests to dominate the formation and implementation of policy.13

    More informed public understanding and participation should aid rather than impede the efficiency, accuracy, and equity of healthcare rationing through enabling more accurate needs assessments, more effective audit, and more representative research.24 This will help to ensure that macro policy aims are being achieved and that the moral boundaries of acceptable rationing are not being exceeded in the name of expediency. Reasonable levels of understanding and participation will also help to minimise distress in the face of non-treatment. This is because the degree of scarcity and the reasons for it will be explicit, along with the knowledge of how and when resources are being distributed between different areas of clinical demand.

    The fact that the public has been oblivious to healthcare rationing in the past may well explain some traditional allegiance to the NHS. Such ignorance undoubtedly made the work of health care much easier than it would otherwise have been. Yet it also has led to injustices–for example, ageism and arbitrariness in the construction and management of waiting lists.19 25 That cat is now out of the bag, and the media will see that it is not put back. The argument for implicit macro rationing on the grounds of bureaucratic and political stability is just unrealistic.

    The same argument is also paternalistic, illegitimately conflating a professed concern with the public welfare with bureaucracies' love of secrecy. The key premise is that the public will not be able to understand and therefore not be able to accept the degree of indeterminacy and inaccuracy which necessarily accompanies decision making about healthcare rationing.

    As regards health care, there seems little convincing evidence that this is so. When anger and frustration do occur, it is usually in the face of the harm caused by what is perceived to be a mistake falling outside the boundaries of what is regarded as acceptable error. Citizens in the United Kingdom have traditionally drawn such boundaries generously. Mistakes and inaccuracy in themselves have usually been tolerated provided that they are publicly acknowledged and that serious attempts are made to detect why the problems arose and how they will be avoided in future.26


    I have argued that none of the arguments against explicitness in healthcare rationing are convincing. Attempts to clarify the moral principles on which rationing should be based are not doomed to failure. We already know what these principles are: we must now have the moral courage to develop them further to ensure that they form the explicit basis for rationing decisions at both micro and macro levels. There is too much secrecy in British public life already. It should be reduced rather than sustained within the NHS.


    I thank Lesley Doyal.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    14. 14.
    15. 15.
    16. 16.
    17. 17.
    18. 18.
    19. 19.
    20. 20.
    21. 21.
    22. 22.
    23. 23.
    24. 24.
    25. 25.
    26. 26.