|Home | About | Journals | Submit | Contact Us | Français|
At the 5th International Conference on Priorities in Health Care in Wellington, New Zealand, 2004, one resonating theme was that for priority setting to be effective, it has to include clinicians in both decision making and the enforcement of those decisions. There was, however, a disturbing undertone to this theme, namely that doctors, in particular, were unjustifiably thwarting good systems of prioritising scarce healthcare resources. This undertone seems unfair precisely because doctors may, and in some cases do, feel obligated by their professional ethics to remain uninvolved either in deciding priorities and in some cases in enforcing them. I will argue that the professional role of a doctor ought not be considered inconsistent with the role of a priority setter or enforcer, as long as one crucial element is in place, a rationally coherent and broadly justifiable regime for prioritising healthcare. Given this I conclude both that prioritisation and doctoring are not incompatible under certain conditions, and that the education of healthcare professionals ought to include material on distributive justice in healthcare.
Prioritisation decisions are quite rightly made at several different levels: the macro governmental level; the meso level—embodied, for example, by the district health boards in New Zealand or health trusts in the UK—and, finally, the micro level of individual healthcare professionals.
Each of these levels can interfere and obstruct each other in different ways, and any one of them can prevent prioritisation schemes from succeeding. Ideally, what ought to happen is that the government gives funding and broad directions on prioritisation, the meso level focuses more clearly on what will and won't be funded within these constraints and, finally, doctors and other healthcare professionals carry out these priorities, as regards both what and who get treated. Though each level of decision making is important in terms of prioritisation, doctors have a special role; for the most part, they are in the hard position of interacting with those who are disadvantaged by prioritisation, and they have some capacity to thwart prioritisation through such behaviour as misreporting, or treating despite the regulations.
One of the papers given at the 2004 International Conference on Priorities in Health Care was by Daniel Wikler, who argued that the case of priority setting in Oregon,1 usually held up as a triumph of a rational open approach to priority setting, failed to achieve its objectives as regards cost containment.2 Wikler argued that this occurred for the most part because of non‐compliance by the doctors; if the doctors felt that patients were unfairly missing out, they reallocated resources to treat these patients.
A similar, possibly apocryphal, story circulated at the conference about an event in New Zealand. There, one of the explicit methods of healthcare prioritisation used by district health boards is to set limits on the number of certain sorts of operation, such as heart bypasses, so that doctors can then allocate the operations according to need or benefit. In other words, given 100 operations a year, the doctors then choose according to various criteria the 100 patients who deserve the treatment most. There is, of course, an exception—emergency patients who need a heart bypass get them without taking away from the 100 heart bypasses available. When this was first put in place in one district health board they apparently used up their allocation of non‐emergency heart bypasses by October. Although technically the non‐emergency heart bypasses ceased for the remainder of the year the number of emergency heart bypasses was said to have tripled, with the result that the district health board did more heart bypasses in total than ever before. While I couldn't verify this story, it is nonetheless not outside the bounds of possibility and serves as a pertinent reminder of what could happen if healthcare professionals do not support a system of prioritisation.
What this example indicates is that the direct involvement and engagement of doctors in priority setting is unavoidable. Without their understanding and consent, prioritisation simply won't work. There are two powerful reasons for this. First, if they don't support the system, they are uniquely placed to game the system, thwarting cost containment and prioritisation measures.3 Gaming the system in this context involves intentionally misdiagnosing or misrepresenting the severity of a patient's condition, either to obtain treatment for them that they would not otherwise have received or to bump them up a queue.4 Second, it is unlikely that any system of prioritisation will function well without the individual knowledge and expertise of individual clinicians and their abilities at assessing what is needed in each case.
While some defend gaming, and in certain circumstances it does seem laudable—for example, if someone was being denied treatment because of their skin colour in a racist prioritisation scheme—in general, it seems likely to have negative consequences. Gaming, given the background of a reasonably fair prioritisation scheme, undermines fairness and equity. Furthermore, gaming puts patient safety at risk, unless the entire care team is in on the game; gaming involves deceiving the healthcare funders and management, and also other healthcare professionals and potentially even the patients themselves. This could lead to the provision of inappropriate treatments by other healthcare professionals and even dangerous nocebo effects for the patient. Significant inefficiencies are likely to occur as a result of gaming, since providing the right kind of healthcare technology relies on accurately predicting what healthcare will be needed—difficult to do when you are being misled by one of the key sources of information. Finally, it may well prove to be self‐defeating, since other physicians with different values might also engage in gaming and, worse still, since each clinician has contact with only a small percentage of the overall number of patients, gaming might well lead to patients gaining priority over or taking resources from patients that even the gaming physician would agree ought to receive priority.
However, there is a major objection to doctors being involved either in priority setting or in enforcing prioritisation decisions. This is that their professional ethics, the role‐based obligations imposed on doctors, may seem to conflict with making prioritisation decisions where this would disadvantage one of their patients. Healthcare professionals are typically expected to do their utmost for each of their patients, but priority setting seems to precisely neglect this, as the very notion of priority setting involves choosing to do the utmost for some patients at the expense of others. Thus it might be argued that healthcare professionals are obligated to resist prioritisation as contrary to their professional ethics.
In general, doctors are seen to have strong duties to aid people without discrimination. What precisely the professional obligations of doctors are to a particular patient, however, are less clear. While many people make reference to the Hippocratic oath in this regard, it is unclear even which version they are referring to. It is almost certainly not the original Greek Hippocratic oath, since this forbids, among other things, doctors providing surgery.5 The modern Hippocratic oath is not entirely clear on the point of what the obligations of a doctor are to individual patients. However, The modern version of the Hippocratic Oath does say:
I will apply, for the benefit of the sick, all measures [that] are required, avoiding those twin traps of overtreatment and therapeutic nihilism.6
This part of the Oath does give us a prima facie justification for supporting the claim that doctor's professional ethics require that they reject prioritisation. However, not all doctors take the Hippocratic oath, and there are several variations of it, so it is unclear that simply appealing to the Hippocratic oath resolves this question.7
We could, alternatively, look at the ethical codes of various medical associations. For example, the British General Medical Council lists the following duties:
This list, in particular the first duty, implies that doctors ought to focus their care on giving the best possible treatment to the present patient—that we should view doctors as obligated to do the very best possible for each specific patient. There is an intuitive plausibility to this, and it aligns with many common accounts of professional obligations more generally. For example, if we look at lawyers, they are typically considered to have obligations to their clients such that if representing a prospective client would constitute a conflict of interest, they ought to turn down the potential new client even at great financial cost to themselves.9 This is a common characteristic of a fiduciary relationship, which is typically seen as the type of relationship that exists between professionals and their clients.
However, for medical professionals this sort of relationship seems untenable over time: providing the best possible care for one patient inevitably means providing worse care for the next patient, since you will have spent a lot of time and resources on treating the first patient. Instead, it seems that part of being a healthcare professional is balancing the needs and interests of your various clients. This is especially the case when their needs come into conflict. This further guidance from the General Medical Council indicates this:
Conflicts may arise when doctors are called upon to make decisions about the use of resources and about patients care, and the needs of an individual patient and the needs of a population of patients cannot both be fully met. Dilemmas of this kind have no simple solution. When taking such decisions, a doctor should take into account the priorities set by government and the NHS and/or their employing or funding body. But they must also be clear about their own role. As clinicians, doctors must make the care of their patients their first concern, bearing in mind the effects of their decisions on the resources and choices available for other patients. As managers, doctors must allocate resources in the way that best serves the interests of a community or population of patients. In both roles doctors should use evidence from research and audit to make the optimum use of the resources available.10
While this offers one solution—namely, that as clincians doctors ought to give their patients utmost priority, while as managers they ought to serve the community interests—this sharp division of professional roles and obligations raises difficult questions. In particular, is it possible to play two professional roles with conflicting obligations? In any case, there seems no reason per se to prioritise the role of a manager over the role of a doctor, and so no reason to not claim that the doctor ought to give their utmost priority to their individual patients even if this contradicts their role as a manager.
Ideally, it would be better to establish whether doctors' professional ethics are ever compatible with prioritisation than to claim that their professional ethics are trumped by those of another role that they also play.
The most classic case of prioritisation in healthcare is the example of triage. This term is used to describe typical behaviour in situations, classically in wartime, where, because of a surfeit of patients and a deficit of doctors, the treatment of patients is prioritised into three categories, those who are treatable but who will probably survive without treatment, those who are untreatable and those who urgently need treatment.11 The wounded who need urgent treatment and are likely to gain from it are then treated first, followed by those who will probably survive without treatment; and finally those who are unlikely to be saved are treated. Of course, within these categories, prioritisation occurs as well, depending on the urgency and severity of the treatment required.
Triage occasionally still occurs in emergency situations. While obviously regrettable, it doesn't seem inconsistent with the professional ethics of healthcare professionals. (It is “regrettable” in the sense that it would be better, of course, if there were more doctors available, not regrettable in a moral sense.)
It might be asked why triage is not inconsistent with a doctor's professional ethics. I think it is the clear necessity of the situation confronting the doctor, who is clearly in a situation where not all can be helped, and in such a situation prioritising to try to maximise the number helped is not inappropriate. In part, this is an appeal to the Kantian moral principle that “ought” implies “can”. In fact, in situations of scarcity, prioritisation is not just appropriate, it is unavoidable; choosing to focus on only the patient in front of you regardless of the severity of their condition or the urgency of other patients' needs is itself indulging in prioritising.
However, while triage does still occur, it is typically thought to occur only some of the time in healthcare provision. So even if doctors' prioritising in the situation of triage is acceptable, is it acceptable in general practice?
Although doctors won't always have more patients in front in them than they can treat at any one time, I think healthcare provision in reality is more like a triage situation than it might appear from the perspective of an individual doctor.
It is hoped that this section of the paper will in part be preaching to the already converted. Many people are well aware of our inability to provide adequate healthcare funding for all medical conditions. However, some people have argued that explicit prioritisation of medical resources is not needed.12
The most common argument offered against the need for healthcare prioritisation is that really all we need to do is to create a more efficient healthcare system, to remove bureaucratic inefficiencies, streamline staffing and cut the high wages of most medical professionals. This amounts to claiming that a present need to prioritise healthcare is not inescapable but instead is artificial, the product of a political and social system focused on growth and making money. There is clearly some truth to this claim. If we changed our current social and economic system, different levels of resources would be available for healthcare. Likewise, if we removed the constrictions medical associations usually put on the amount of new healthcare professionals who are registered each year, we would decrease the high cost of doctors and so decrease healthcare costs. Furthermore, if we rearranged our spending priorities, we might be able to both free up further money for healthcare and cut down on healthcare needs—for example, by avoiding going to war unless absolutely necessary, we could cut back on military spending and avoid the need to spend healthcare resources on new veterans. Another possibility would be to do more to reduce socioeconomic inequalities, since there is strong evidence that this directly affects health.13 Finally, health rather than healthcare ought to be seriously considered, preventive measures and simple lifestyle changes would do much to reduce future healthcare needs.
In many cases, these claims seems justifiable, and we ought to change these social practices to make more money available for healthcare funding. It might be thought that until these changes are made, healthcare professionals ought to attempt to thwart the system as a form of civil disobedience. Although sometimes this might be justified, as a general rule it ought to be avoided. Like most forms of civil disobedience, it should be reserved as a last possible action, once all other avenues have been exhausted. As I will explain later, the sort of healthcare system that will be justifiable will rely heavily on input from doctors and, importantly, will welcome the re‐examination of prioritisation decisions through official channels.
My claim is, however, that regardless of which of these measure we undertake, our action will nonetheless not provide for enough medical resources to provide every healthcare need anyone has, without at least the sacrifice of other great goods, and probably will not produce enough resources at all. For the reasons I am about to outline, cost‐cutting and efficiency gains will never be enough to prevent the need for some priority setting.
A basic fact of economics is the simple truth that humans have, for all practical purposes, unlimited wants and only limited resources to satisfy these wants. This requires us to prioritise what we want and to spend money on multiple competing interests. So, for example, when our society decides how it will spend its taxes, it has to consider healthcare and also roads, education, law enforcement, fire services, environmental protection, welfare, pensions and aid to Third World nations. We cannot simply work out how much we need to spend to provide optimum healthcare in isolation from these other demands.
Potential healthcare spending is close to limitless. This is because of the constant research into new medicine and procedures. At present, there is a vast amount of research into many areas of healthcare. This is increasing our ability to cure and diagnose illnesses and also our ability to spend money on healthcare. Several new areas of research, such as genetic therapy and transplant and xenotransplant operations, have the capacity to dramatically increase healthcare costs. To give just one example, it is estimated that the costs of pig xenografting will be US$10 billion/year by 2010 and that by then this market will have only barely begun to reduce the pool of dialysis patients.14 So in effect the level of healthcare we could provide our population is changing all the time, and the amount that we could spend is also increasing.
In other words, we could in general spend more money and aid more people; likewise, given any particular fixed budget, some people will probably miss out on treatment. In this situation of scarcity, any choice to use healthcare resources to aid someone limits someone else's access to healthcare. The implication of this actual scarcity in healthcare is that although the need for prioritisation may not be apparent to an individual doctor, in reality doctors are in a version of a triage scenario, with more patients than they can treat to the fullest level over time.
If my argument is correct so far, then we have established that prioritisation of patients is necessary and that doctors are a necessary part of this. We have also established that at least one form of prioritisation, namely, that carried out in triage, is not traditionally seen as contrary to doctors' professional ethics. However, it might still be argued that in general a fiduciary relationship is incompatible with prioritisation between clients, in this case patients. But is prioritising patients in a public healthcare system genuinely incompatible with a fiduciary relationship? The reason that fiduciary relationships are considered appropriate between professionals and their clients, as opposed to the normal market relationship, is because of the imbalance of power, knowledge and expertise between the professional and the client. This imbalance would make an arms‐length market relationship inappropriate, because the professional would be able to take advantage of the clients, and the clients, knowing that, would not be inclined to hire a professional in that area. In other words, the fiduciary relationship protects the client, but it also enables the existence of the professional's role. In many cases, allowing a professional such as a lawyer to prioritise between their clients would clearly undermine a fiduciary relationship: lawyers might simply swap to the highest paying customer in a dispute, for example. But the case of healthcare professionals seems different, at least when they are employed by a public health system. It is not clear that prioritising one patient's interest over another is a breach of this fiduciary relationship, given that both patients know that this prioritisation is going to occur and know the grounds for the prioritisation (be it need, greatest benefit, equality or some other factor). Indeed in so far as a prioritisation system for a public healthcare system is democratically influenced and supported, it would seem that healthcare professionals are obligated to participate in it.
I have argued that, really, doctors are currently and constantly in a triage situation, but that this is obscured by not having all the patients in need right in front of them right now. If this is correct, then, given that triage is not ethically unacceptable, a doctor's professional ethics is not inconsistent with priority setting and in fact might go hand in hand with direct engagement with prioritisation.
For prioritisation to be consistent with the professional ethics of medical professionals, however, needs to be supported in two significant fashions. First, doctors need to be convinced that the system of prioritisation, while not necessarily ideal, is at least ethically sound and justifiable. Their professional ethics will not allow them to support an arbitrary system of prioritisation. Constructing and adopting a reasonably acceptable prioritisation scheme will probably require some hard work, both by politicians and by theorists.15
While it is beyond the scope of this article to full describe an ethically sound and justifiable system of prioritisation, I would take the Daniels/Sabin accountability for reasonableness (A4R) framework as a fair stab at a broad criterion of acceptability, though of course substantive work still needs to be done. This framework is a proceedural account of fairness in the decision making process of prioritisation. It consists of four constraints on decision making:
While there are criticisms of the A4R framework as a comprehensive account of prioritisation of healthcare resources, as simply a set of minimal proceedural criteria for a broadly acceptable system, it seems robust.17
If these conditions are fulfilled, behaving as a conscientious objector by thwarting the agreed‐upon prioritisation scheme seems inappropriate, a matter of bad faith, since their input has already been recognised and there are available channels to argue for the reform of the present practices.
Adopting this framework will require that several politically difficult challenges be overcome, namely, that a general background public acceptability of explicit priority setting needs to be established both by public information campaigns and by ministers introducing and using the language of prioritisation in justifying hard medical decisions. Nonetheless, the alternative, though politically easier, is worse: an unjustifiably arbitrary healthcare system which, when considered reasonably, seems perverse in its decisions.
Second, healthcare professionals, in particular, need to be educated, both in the process of prioritisation and, more importantly perhaps, regarding the ethical arguments around its justifiability. If healthcare professionals believe it is ethically unacceptable to ever prioritise between patients, any prioritisation scheme is doomed to fail, regardless of whether it is actually consistent with their professional ethics.
I would like to thank Søren Holm for helpful comments on the development of this paper and Rosalind Hursthouse for comments on an earlier draft of the paper. Finally, thanks to Bob Brecher for his review and insightful criticisms.
A4R - accountability for reasonableness
iOf course, this is only one way to conceptualise the relationship between the various levels involved in the provision of healthcare, and alternative models have been proposed—for example, attempting to remove government from having a steering role. Nonetheless, in the UK and other countries with public healthcare systems, this model seems to accurately reflect how they operate.
iiA nocebo effect is the opposite of a placebo effect, that is, a negative response to an inactive substance or procedure; this could come about because the patient believes their condition is worse than it actually is, as the result of a clinician's gaming.
iiiA strong duty is one which may be seen as binding in a way that trumps or over‐rules other concerns. It can be contrasted with a weak duty, which is merely a prima facie obligation to do something, easily over‐ruled by other concerns.
ivHere I am treating the patient as simply anyone who has a medical problem of some sort which requires intervention or assistance. It should be noted, though, that one means of prioritisation that is used is to define people who are in genuine medical need as not being patients at all, through the use of screening criteria; that, in effect, stops them at the door. This has the politically useful effect of keeping patient numbers down—for example, on waiting lists for particular operations. Similar definitional tricks are used by governments to lower unemployment figures, for example. This hidden sort of prioritisation practice ought to be resisted. It is better to know the actual size and scope of particular healthcare problems, so that we can best deal with them, than to have artificially lower figures.
vA fiduciary relationship is one based on trust; it can be contrasted with the typical, arm's‐length marketplace relationship, which is based instead on mutual advantage.
viWhile this is a utilitarian priority, I do not intend to suggest that this is the only appropriate prioritisation scheme in the situation described, but just that priorisation is not inappropriate.
Competing interests: None.