Need for Health and Health Care


Any society must come to a decision concerning the allocation of health resources, of which access to medical services, or health care, is the clearest focal point. For many theorists and ordinary citizens health care services are a ‘special’ type of good that should not be distributed on the market-based principle of ability to pay. Rather, it is often said, health care should be distributed on the basis of ‘need’; if there was ever a place for Marx’s dictum ‘each according to their need’ it would seem that health care would be a good candidate. Bernard Williams (1973, p. 240), for example, suggested that ‘‘the proper ground of distribution of medical care is ill health: this is a necessary truth.’’ Of course not all societies have organized themselves entirely on this basis, but virtually all countries include a significant element of distribution of health care resources on the basis of some notion of need, whether as the main criterion for allocation, as in most European countries, or in services for the elderly, the poor, and the military personnel, as in the US. But if health care is to be distributed according to need it is necessary to explain what a need for health care means.

It would seem that, because the purpose of health care, broadly speaking, is to promote health, the need for health care must be derived from the need for health. Therefore, one ought to start with a prior question, what is the need for health?

Here it is argued that ‘distribution according to need’ names a general approach to health policy as opposed to distribution on the basis of ability to pay, rather than a specific principle of distribution. One reason for this claim is that all of the most prominent candidates for specifying a principle of distribution of health resources according to need face difficulties. Accordingly, a policy maker wishing to allocate resources according to health need will be compelled to balance a number of need-related considerations, among other relevant concerns, rather than follow a specific principle of distribution.

The Concept Of Health

To discuss different ways in which one can be said to need either health or health care, it is necessary, first, to clarify what is meant by health. But what health is continues to be highly contested. Nevertheless, without claiming to have resolved any of the difficult questions on which these debates center, it is possible to give a rough outline of a concept of health for the purposes of this discussion.

Consider, first, two well-known but rather different definitions of health. According to Christopher Boorse’s (1977) definition, health is the absence of disease. How disease, in turn, is defined is one of the most important parts of Boorse’s account of health and merits much more discussion than can be accommodated here, but suffice it to say that disease, in his view, is a deviation from the ‘normal’ functioning of certain parts and processes of the organism. Even disabilities and injuries would fall within the scope of this definition of disease, and as such it leaves a much narrower range left for health compared to how it is ordinarily understood.

In contrast to this definition, the World Health Organization (WHO) adopts a much wider definition of health, according to which health is ‘‘a state of complete physical, mental and social well being and not merely the absence of disease or infirmity’’ (1946).

Although it seems right that health should be closely related to well being, the WHO’s definition goes too far: in this view, health problematically appears to be indistinguishable from well being or happiness. Nevertheless, this definition draws our attention to a different aspect of health beyond Boorse’s definition in terms of the absence of disease: that of ‘positive’ health achievement. One can imagine other such positive health achievements, for example, athleticism or living a healthy lifestyle. Furthermore, certain aspects of health such as physique, physical strength, or endurance, can today be enhanced through various drugs and procedures – such enhancements could be said to constitute improvements to one’s health, regardless of their impact on the presence or absence of disease.

It cannot be resolved here which of these two, or indeed of the many other definitions that have been proposed, represents the most appropriate account of what health is. However, a disease model of health seems more appropriate in the context of a discussion of health need. Even if it is allowed that ‘merely’ being free of disease is not the ‘best’ level of health one can achieve, and that there are other health states that are superior, it is more difficult to see the ‘need’ for this latter form of health achievement. It might be proposed, then, that the need for health is best captured in terms of the need to be (reasonably) free of disease. For our purposes, the disease concept is narrowed down further, to encompass only such deviations from normal functioning that are harmful to the person.

But it is also necessary to think about the question of time span. The definition of health given so far is silent on the question of how extended a period should be considered. It is clear that one may be afflicted by disease at any given time in one’s life. Moreover, the duration of any particular disease can vary greatly; some are short-lasted or can be cured, others are chronic. Clearly, the duration of any disease will matter greatly to how significant a departure from health one thinks that disease state represents. However, health is also a prerequisite for life itself – without health there is no life. So one could also conceive of health in terms of the time passed before the total loss of health, death, occurs.

Thus, time and duration seem of central relevance when health is discussed. One may therefore ask whether duration or longevity should be included in the concept of health. Is the duration of health – the length of life – a dimension of health? It would then follow that a shorter life would be a less healthy life. That would be true even if life had been lived ‘in full health,’ completely free of disease, in every moment up until the point of death. Alternatively, it could be said that length of life is simply health combined with duration. In that case, the length of life would not affect how healthy one would consider a particular life to be.

For our purposes, health is defined as including duration. That means that not only the absence of disease, but also a lifespan of a certain length, constitute the baseline of health achievement against which health need is measured.

The Logic Of Need

It is often argued that ‘need’ is a three-element relation: in case of human need x – a person – needs y – an object – in order to z – to achieve a purpose or goal. In this framework it is clear that one question of health need is what is needed to achieve health (health as z). Yet one can also ask what may be a logically prior question: What is health needed for (health as y)? This prior question will be considered first.

At least two central dimensions of one’s quality of life where one’s health will have a considerable impact can be identified. The first of these dimensions is well being. Disease is often accompanied or constituted by various forms of suffering such as pain, nausea, ‘feeling ill,’ or feelings of anxiety or depression, all of which have a very direct and to varying degrees negative impact on our immediate physical and mental well being. The second dimension is the ability to engage in ordinary human activities. Norman Daniels has discussed this in terms of the importance of ‘normal species functioning,’ a concept adapted from Boorse’s framework, for enjoying a normal opportunity range. The concept of normal species functioning is less clear than can be wished for, but at least a few relatively uncontroversial examples of normal functioning, come to mind such as having all major limbs intact, basic mobility, and being able to see and hear. These and other functionings will clearly be important for the pursuit of a wide variety of goals and projects. Many health conditions will be detrimental to or involve the loss of such functionings, and will hence negatively affect our opportunity range.

Nevertheless, the idea that health is needed for opportunity is not without difficulties. Consider, for example, the extent to which a condition such as paraplegia would affect one’s opportunity range. It has been pointed out that, a person living in a poor rural village with only dirt roads is likely to experience paraplegia as a much more disabling condition than a person living in a wealthy, urban environment with a well-developed infrastructure. In other words, the extent to which limited mobility or other functional impairments will restrict one’s opportunity range also depends on the nature and quality of one’s social and material environment, and not just on the health condition itself.

But even individuals living in the same environment may be affected very differently by the same health condition depending on their own particular circumstances, such as their resilience, ability to adapt, social support network, or their preferences . The level of health achievement that is needed in order to enjoy a reasonable range of opportunities will clearly vary across such individual circumstances as well as the social, cultural, and historical context.

Furthermore, longevity generally tends to be valued, and it is not uncommon to think that a certain length of life is a central aspect of a good life. It is not immediately clear exactly what it is about a shorter life that is unfortunate; after all, a premature death does not in itself, retrospectively as it were, alter the quality of life lived up to the point of the onset of death or the events that led to death (though that is not to deny that having advance knowledge of one’s own to be shortened lifespan is likely to affect one’s quality of life in various ways). But perhaps one could say that a shorter life is a life with less opportunity, both in terms of variety and the total ‘amount’ of opportunity. This diminished range of opportunities due to premature death would not affect the individual in the same way as diminished opportunities due to loss of functionings – perhaps it is not even quite correct to say that the diminished range of opportunities in the former case really ‘affects’ the individual’s lived life as such – but the loss of opportunity still represents an unfulfilled potential and, therefore, a shortfall.

Returning to the idea of need as a three-element relation (x needs y in order to z), it was noted that in addition to the question discussed in the previous section – what is health needed for? – one can ask ‘what is needed to achieve health?’ This is the central question of ‘health need,’ which will be considered next.

Health care appears to be the most obvious candidate for what is needed to achieve health. After all, health care is a means to improving health, and thus it would seem that a need for health is simultaneously a need for health care. However, not all health needs indicate a need for health care. Ordinarily, other basic needs such as food, water, sanitation, and shelter must be met as a minimal precondition for health; many health needs arise as the result of a failure to meet these other needs. In such cases, although health care might be necessary for short-term intervention, ensuring that these basic needs are met will clearly be more effective for overcoming population health needs in the longer term. Even in developed societies where basic needs are mostly catered for, it is argued that a level of health need arises as a result of poor quality housing, material insecurity, working conditions, and social exclusion. In many cases, targeting such ‘upstream’ causes of disease will be a better strategy for reducing health need overall.

A need for health, then, cannot be identified with a need for health care; only some health needs are at the same time health care needs. The concept of health care need will be considered once more in the last part of the article, but first the relationship between a shortfall in health and the need for health will be considered in more detail.

The Health Baseline

For the purposes of this discussion, health is conceived as the absence of harmful disease (understood very broadly). But it is also noted that what is to count as ‘harmful disease’ can vary culturally and individually. It is also suggested that longevity should be seen as a dimension of health. The notions of the absence of disease, and of living to a certain age, function not only as conceptions of health, but can also be conceived of as a particular baseline of health, against which shortfalls in health can be measured. Thus, premature death and the presence of disease both in different ways represent shortfalls in health achievement.

This baseline of health has a double function. On the one hand it provides an account of what it is to achieve health (as a means to a life of good experience and opportunity). On the other it provides a standard by which other things, such as health care, can be judged as meeting health need or not. The baseline of health, therefore, is central to the concept of need for health and health care. The question of what this baseline of health should be will be considered next.

On the most expansive conception of health need, the highest attainable health would be adopted as the baseline against which health need is measured. Consider the case of life expectancy. The life expectancy at birth in Japan, which is one of the highest in the world at nearly 84 years (CIA: The World Factbook, 2012), is usually used as the standard for the highest attainable life expectancy. Accordingly, if this life expectancy is adopted as the relevant baseline, any shorter life expectancy represents a health need. However, one might be skeptical of the idea that any shortfall from this very high standard of health is appropriately characterized as a health need. The UK, for example, has a slightly lower life expectancy at birth than Japan at approximately 80 years (CIA Factbook). But would one thereby say that the UK has a health need? This seems debatable.

One possible argument is that a shortfall in health is only a health need if it reflects a genuine possibility for health gain. But it is not clear that the highest known life expectancy attained by some is attainable by all. This will depend on what factors determine longevity and the extent to which these factors are within the scope of human control. Perhaps longevity is partly genetically determined. Other determinants, such as diet and lifestyle, are in principle within our control, but in practice it is hard to imagine a government imposing a particular diet on its citizens. One can see why one might think that only cases where there is a genuine possibility for improving health should be considered a health need: after all, to say that there exists a need seems to imply that something ought to be done. And to say that something ought to be done in turn seems to imply that something can be done – or so proclaims that familiar Kantian principle.

This issue can be set aside for now. Instead, consider a different reason to be skeptical that the UK has a health need in this case. One could argue that the highest attainable health is simply the wrong standard of health against which to compare our own health achievement for the purpose of identifying health need. Just as athleticism or other forms of positive health achievements go beyond what one would ordinarily say is needed, this ideal standard also seems to exceed what is required. Reserving the term ‘need’ for more substantial shortfalls in health seems more intuitive.

This point can be accommodated if a more modest level of health is adopted as the relevant baseline, for example, a level of health that it is reasonable or realistic to expect to attain. Alan Williams has expressed a related view with respect to length of life, arguing that ‘we are each entitled to a certain level of achievement in the game of life,’ and that anyone exceeding this level, which he refers to as a ‘fair innings,’ ‘has no reason to complain when their time runs out’ (Williams, 1998, p. 319). It is possible to extend and apply this concept of a ‘fair innings’ to the standard of health; the idea is that because it is clearly both possible and desirable to improve health beyond this level, a person or a population that has reached this standard of health has attained a fair or sufficient level of health, and therefore does not have a need for health.

Although it remains true that there is a sense that someone who has lived beyond the age of the ‘fair innings’ could understandably still claim to have a need for health, just as a wealthy person could claim a need for more money, there is a sense, in both cases, in which one could say that their needs have been met, and what they claim to need is a form of luxury or excess. On this account, need is assimilated to something like basic need. It is true that one can have further needs even when basic needs have been met, but for political purposes it could be that only basic needs call for action.

It may be that this notion of a fair innings of health does not lend itself equally well to all dimensions of health or all levels of analysis; or perhaps one must approach the notion of sufficiency differently with regard to such different dimensions of health rather than speak of sufficiency of health overall. For example, perhaps only moderate levels of pain will be accepted as ‘within’ our standard for being ‘sufficiently’ pain free, whereas our standard for a sufficient length of life could be significantly lower than the known human potential; and having achieved sufficiency in one dimension of health may not imply sufficiency in a different dimension. Clearly, the notion of a fair innings of health requires more work. Nevertheless, one can make sense of the idea that a shortfall from or failure to achieve the highest attainable level of health does not have to indicate a health need.

If this idea of a fair innings of health is accepted, how should one go about determining what level of health it would be reasonable to expect to attain? Health outcomes are partly determined by one’s social, material, and economic environment. The quality and nature of this environment in turn depend on a society’s level of affluence and on how its resources are distributed. The question of what level of health it is reasonable to expect to attain can only be answered with References: to these further substantive issues, and as such is hardly normatively neutral.

On the global level, there are enormous inequalities in material standards of living. Hundreds of millions of people live in extreme poverty lacking adequate nutrition, clean drinking water, sanitation, and access to basic health care. Whereas these levels of extreme poverty are avoidable, it is perhaps less clear what level of material living conditions would be generally attainable if global resources were distributed more fairly. The standard of living is not the only important determinant of health, and health achievement is unlikely to improve exponentially with improvements in the material standard of living; nevertheless, the realistically attainable standard of living is likely to impose some constraints on the level of health one can reasonably expect to attain. For example, it seems dubious that the exceptionally high standard of living found in Monaco, where citizens are generally extremely wealthy, is attainable for all. Life expectancy at birth here is the highest in the world at nearly 90 years (CIA: The World Factbook, 2012) – but insofar as this health achievement is a result of their wealth and high standard of living, it is not realistically attainable for the world’s population as a whole.

The question of what standard of living will be generally attainable aside, within any society there will be other important decisions to be made about how much priority should be given to the promotion of health over other things that are valued. Such questions of priority are likely to arise in many different contexts, but one can illustrate the point by considering the case of reducing or eliminating health risks. How much effort should be expended on this task? Some interventions furthering this objective could have prohibitive costs in other areas of life. For example, road accidents, being one of the top 10 causes of death worldwide (WHO (World Health Organization), 2008), constitute a severe health risk. However, even if banning the use of cars altogether were to improve our health overall, there are obvious reasons why it is neither desirable nor practicable to go through with such a proposal.

The answer to the question of what level of health it is reasonable to expect to attain will depend on other normative judgments, such as ‘what is a fair distribution of resources?’ and ‘how important is health compared to other dimensions of quality of life?’ Depending on what answers are given to these and related questions, one will have different ideas about the appropriate baseline of health achievement against which shortfalls in health should be measured, and therefore, about what counts as a health need.

Three Concepts Of Health Care Need

Next, consider the question of need for health care. It was established that health care is not always needed to achieve health. But it is necessary to look at the relationship between need for health and need for health care in more detail. Here, three concepts of health care need which each limits the concept of need in different ways are considered: presence of disease, capacity to benefit, and cost-effectiveness of treatment.

The idea that the presence of disease equals a need for health care is very straightforward: if a person is sick or injured, it seems natural to say that he or she is in need of health care. However, not all diseases can be treated or cured. Although one could still consider such cases to be health needs, it is perhaps less clear whether one can say that there is a need for health care in these cases. Arguably, it seems strange to say that there is a need for health care if no health care exists, or if health care provision is at such a primitive or underdeveloped level that it would be harmful rather than beneficial. Many medical practices common in the past are now known to be either inefficient or in fact harmful, such as lobotomy or bloodletting; it cannot be said that there was ever a genuine need for such services.

However, it seems more appropriate to say that such cases represent a need for health care in general, even if there is no specific treatment available at a given time that would be of benefit. Furthermore, one can point to examples where it might be said that effective health care ‘ought’ to have been available. For example, not much effort has been spent on developing modern effective treatments for a group of debilitating diseases often referred to as ‘neglected tropical diseases.’ This group of diseases primarily affects poor populations in the developing world, and has typically received little attention from the pharmaceutical industry; there is reason to believe that more funding and research could lead to significant improvement in treatment options. In cases such as these, it also seems right to say that there is a need for health care, even if currently no specific treatment exists.

In other cases, treatment is available, but for different reasons a particular individual may be unable to benefit from the treatment. For example, a treatment may be contraindicated for patients outside a particular age bracket, patients with other, preexisting health conditions, and so on. These patients would not benefit from the treatment in question. It therefore seems somewhat counterintuitive to say that these patients ‘need’ this particular treatment.

For reasons such as these, some would reject the proposal that the presence of disease itself is sufficient for there to be a need for health care. That brings us to our second proposed definition of need for health care, as ‘capacity to benefit (from treatment).’ This definition is often favored by health economists. According to this view, a patient is only in need of a given treatment if the patient can benefit from that treatment. Thus, on this view the patients in the examples above could not be considered to be in need of that particular health care treatment.

In many ways this definition of health care need is intuitive. At the same time, narrowing down the concept of health care need in this way does not seem to take anything away from our sense that something ought to be done. As has already been suggested, it seems important to distinguish between the need for a particular treatment or intervention (or the lack thereof), and a more general need for health care. Furthermore, the reasons why a given treatment will not be effective also seem to matter to our judgment. In some cases, for example, the treatment being effective is contingent on the patient complying with certain behavioral requirements, for example, quitting smoking or losing weight. In this case, it seems somewhat more intuitive to say that the patient needs the treatment, even if he or she is failing to comply with the requirements in question. Alternatively, imagine that the effectiveness of a treatment was contingent on the patient being well nourished before the start of the treatment. In cases where lack of resources meant patients were inadequately nourished, it also seems incorrect to say that the patient had no need of the health care treatment in question.

The ability to offer decent health care may also be limited by resource shortage and competing needs. Many countries limit the availability of health care in accordance with the cost-effectiveness of the various treatments or interventions. Sometimes certain treatments will not be offered, even if they can improve a patient’s health, because the cost is considered too high relative to the health benefits it would yield. Our third proposed definition of health care need incorporates considerations of cost-effectiveness, such that a patient is considered to be in need of a given treatment only if that patient will benefit from that treatment, and that treatment is considered to be cost-effective. Thus, a patient does not need a given treatment if that treatment is too expensive or yielding too little health benefit to be cost-effective, even if the patient could benefit from the treatment.

Some very expensive and cost-ineffective cancer drugs for advanced stage disease are sometimes excluded on the grounds that they are not cost-effective. In cases where the cancer cannot be cured, treatment may nevertheless give the patient a few more months of life. In the UK there have been cases where these drugs were not offered through the National Health Service because they were deemed of too limited benefit to justify their very high cost. How many patients can avail of a given treatment can affect the price and hence the cost-effectiveness of that treatment. The so-called orphan drugs is a relevant example here. Orphan drugs are drugs for very rare conditions. If a condition is rare, market demand for the drug will be expected to be low, and it will be difficult for a pharmaceutical company to sell enough drugs to cover the expenditure involved in the research and development of the drug. Therefore, the price of such drugs is often very high, and they will rarely be cost-effective.

In this definition of health care need, the extent of need in a population will be relative to the society’s level of affluence. That leads to the interesting implication that as a society becomes wealthier, and can afford to relax the cost-effectiveness constraints, all else being equal, the total need for health care would in fact increase. Although it may seem a surprising result that the need for health care increases in accordance with a society’s increased wealth, this view also captures something of importance. For example, in a wealthy society, crooked teeth could be considered a need for dental services. But in a very poor society, the correction of crooked teeth would rather be considered a luxury than an actual need. Something seems right about this judgment. It is possible that what should be considered a need could be somewhat relative. Our sentiments will vary to some extent depending on what it is perceived as ‘reasonable’ to expect to achieve in a given context with the given level of resources. This echoes the arguments put forward earlier in the discussion about what level of health would constitute an appropriate baseline for measuring health need. Although it is relevant to know what the highest attainable standard of health is, one also ought to consider what kinds of conditions – including the level of provision of health care – will be necessary in order to reach this level of health, and the costs of bringing about such conditions.

Health Care Rationing And The Ranking Of Health Care Needs

There is something to be said for each of the proposed definitions of health care need that have been considered so far. But going back to the initial observation that the concept of need is often perceived as the most appropriate guiding principle for the distribution of health (care) resources, one may ask, what would a principle of distributing health care according to need look like on each of these three concepts of health care need?

For the purposes of this discussion, it will be assumed that not all health care needs can be met. How are needs ranked, according to each of the definitions of need? As will become clear each candidate will have different implications for which needs are the greater needs. Assuming that greater needs should be given priority over lesser needs, each definition will imply different strategies for rationing health care resources. Although it is not possible to go into detail for each of the concepts here, a few examples will be pointed out that demonstrate that distribution of resources on the basis of any of these concepts of need on its own will have distributive consequences that are unsatisfactory.

The first definition of health care need that was identified was health care need as the presence of disease. How would needs be ranked on this definition? It is useful to distinguish between a severe and an urgent health state, where severity reflects how poor a health state is, and urgency reflects the imminence of death. For simplicity, the questions of urgency will be put aside here. If one focuses merely on the severity of a health state, then the greater the health need (i.e., the worse the health state), the greater the need for health care.

Although it seems intuitive that those with the greatest health needs should also have the greatest needs for health care, it is unreasonable to give absolute priority to those with the worst health. The need for treatment can in principle be infinite; one can imagine cases where a health condition is very severe, and incurable, but where medical treatment can nevertheless be of (ever so slight) benefit. In such cases, there is potentially no limit to the amount of health care resources that could be spent in order to improve health, but without fully satisfying and hence eliminating the need. Therefore, a need for health care would remain, no matter how much health care is provided. This is the well-known problem of the bottomless pit. And the bottomless pit problem aside, some increments in health – for example, going from near-complete immobility except being able to wiggle one toe, to being able to wiggle two toes – may simply be too small to be a worthwhile expenditure. But ranking needs for health care entirely on the basis of the severity of the health state cannot accommodate such judgments.

Our second proposed definition of health care need, as capacity to benefit from treatment, avoids this problem. According to this view, need is synonymous with potential for gain; thus, the greater the potential for gain, the greater the need. Naturally, health states that are close to full health do not represent great potential for gain, and thus patients who are not very sick will not be considered to have a great need for health care. Here, the second definition is in agreement with the first definition. But patients who are very sick will only be considered to have a great need for care if effective treatment that can significantly improve the patient’s health is available. Considering the example above, it is clear why this definition is so appealing: if there is not much that health care can do, the need for health care is deemed minimal.

However, ranking needs on the basis of who can benefit the most can also be problematic. Consider the following example: Imagine two patients who both need a kidney transplant, but only one kidney is available. Patient A is 30 years old and expected to live for another 40 years after the transplant, whereas patient B is 40 years old and expected to live another 20 years. In this case, allocating the kidney to patient A will yield the greatest health benefit, and therefore patient A is considered to have the greater need. But at least some would object to distinguishing between and ranking the needs of these two patients in this manner; after all, patient B also stands to gain significantly from the kidney transplant. Furthermore, consider a different example: as before, one must decide which of two patients should be allocated a kidney transplant. But in this case, patient C will attain full health after the transplant, whereas patient D will only attain a lower level of health, because this patient also has a permanent disability (which is unrelated to the kidney disease). Say that, on a scale from 0 to 1, where 0 is being dead, and 1 is full health, the kidney disease is rated at 0.3. Without the treatment, both patients have 0.3 in health; although patient D also has a disability, in this case the disability does not ‘add’ to the severity of the overall health state (this would be true if, e.g., the kidney disease causes you to be constantly hooked up to a dialysis machine, in which case a disability like paraplegia would not add further disadvantage to the overall health state). If paraplegia is rated at 0.7, then patient D would only gain 0.4 (i.e., an increase in health from 0.3 to 0.7) in health as a result of the kidney transplant, whereas patient C would gain 0.7 (i.e., an increase from 0.3 to 1.0). The most effective use of the health resources in this example according to a health maximizing principle would be to allocate the kidney to patient C. This implication is a particularly controversial outcome of ranking needs on the basis of maximizing health benefits.

Finally, these two cases aside, this principle of ranking needs cannot help us distinguish between different health states that have equal potential for health benefit. That is, a patient whose health can be improved from 0.2 to 0.6 will be considered as equally needy as the patient whose health can be improved from 0.5 to 0.9. But here, the severity of the health state would seem to be a relevant consideration for determining which patient has the greatest need for health care; it does not seem right to rank these two patients as having an equally great need for health care, even if their prospective health gain is of the same magnitude.

The last of the proposed definitions of health care need defined need as cost-effectiveness of treatment. That means that the ranking of a need will depend on how much a patient’s health will benefit from treatment relative to the cost of that treatment. Even small health gains can be cost-effective, as long as the cost of the intervention is very low. One reason for ranking needs in this manner would be to get as much health as possible with our scarce resources; the money that can be saved by choosing more cost-effective treatments can in turn be used to pay for further treatments. As such there is an overlap with the previous definition of health care needs, which ranking of needs also pushes us to maximize health outcomes.

However, this approach can lead to many small and relatively trivial health gains being ranked as of higher priority than much more substantial health gains. This is exactly what happened in Oregon with the introduction of the Oregon Health Plan in 1990. The prioritization of health care services offered through the Medicaid health plan came about as a result of seeking to extending care to a greater number of people. But in order to achieve this, the system had to be made more cost-effective. An expert group, the Health Services Commission, compiled a list of prioritized health services, on the basis of, amongst other things, the relative cost-effectiveness of different services. The first version of the list ranked tooth-capping as of higher priority than life-saving appendectomy – a controversial result which has been the subject of much commentary and discussion since. Although cost-effectiveness is an important consideration, it seems that focusing solely on this aspect would miss other important considerations.

The discussion thus far has covered the ranking of health care needs in accordance with three different principles: severity of health state, capacity to benefit, and cost-effectiveness of treatment. The discussion has shown that a ranking of needs on the basis of any one of these considerations on its own is unsatisfactory. There is merit to all three of the considerations, and all of them ought to be taken into account in deliberations on how to allocate our health care resources. Indeed, in practice, most countries will give all three kinds of considerations weight when allocating health care resources. Sometimes those with the worst health will be prioritized, even if a treatment only provides a minor health benefit. Other times it seems important to provide a treatment even if it is not cost-effective to do so. For example, governments sometimes do provide orphan drugs, even if they are not cost-effective.

The question of how such different and often conflicting considerations should be weighed against each other is a task for another day; here, it should simply be noted that such issues cannot be resolved by a stipulative definition of the concept of need. Defining need for health care in terms of one of these considerations does not thereby undermine the force of any of the other considerations.

The question of need aside, there are several other considerations too that are relevant to the distribution of scarce health care resources. Health equity – which is itself an ideal that could be interpreted in many different ways – is one such consideration. For reasons of equity, it might be decided to disregard capacity to benefit or the cost-effectiveness of treatment, and treat like health states alike, regardless of, for example, age or preexisting disabilities. Health care resources can also be rationed with the use of waiting lists or lotteries, without giving particular groups or individual patients greater or lower priority as such. Alternatively, one could interpret health equity as requiring us to reduce inequalities in health outcomes; that could be a reason to prioritize treatment for a patient with worse health, even if the treatment is expensive and of only modest benefit. Desert could be another relevant consideration – it might be decided to give priority in health care to groups that have taken significant risks for the sake of the country, such as firefighters and military personnel.


The notions of need for health and need for health care are clearly important at what one might think of as high-level strategy for resource allocation. If the government announces that it will distribute access to health services on the basis of need, it is clear that it has rejected market-based pricing for services, and will allocate its services according to something like the burden of illness or disease. Yet there is a limit to how much can be done with the concept of need alone. It is not plausible that a health service should allocate services purely on the basis of need. More importantly, however, as the discussion has shown, the concept of need is neither self-evidently clear nor normatively neutral. Defining the concept of need already requires us to take a stand on complex moral questions; one cannot cut through these difficult issues simply by referring to need. ‘Distribution on the basis of need’ is the name of a social program rather than a principle of distribution, and many different detailed principles of allocation are broadly consistent with a needs-based approach.


  1. Boorse, C. (1977). Health as a theoretical concept. Philosophy of Science 44, 542–573.
  2. CIA: The World Factbook (2012). Available at:
  3. WHO (World Health Organization) (2008). The top ten causes of death, fact sheet 310. Geneva: WHO.
  4. Williams, A. (1998). If we are going to get a fair innings, someone will need to keep the score!. In Barer, M. L., Getzen, T. E. and Stoddart, G. L. (eds.) Health, health care and health economics, pp. 319–330. New York: Wiley.
  5. Williams, B. (1973). The idea of equality. Problems of the self. Cambridge: Cambridge University Press.
  6. Anand, S., Peter, F. and Sen, A. (eds) Public health, ethics, and equity. UK: Oxford University Press.
  7. Daniels, N. (2008). What is the special moral importance of health? Just health: Meeting health needs fairly, ch. 2, pp 29–78. Cambridge: Cambridge University Press.
  8. Dworkin, R. (2000). Justice and the high cost of health. Sovereign virtue: The theory and practice of equality, ch. 8, pp 307–319. USA: Harvard University Press.
  9. Kingma, E. (2007). What is it to be healthy? Analysis 67(294), 128–133.
  10. Marmot, M. (2006). Health in an unequal world: Social circumstances, biology and disease. Clinical Medicine 6, 559–572.
  11. Nordenfelt, L. (2007). The concepts of health and illness revisited. Medicine, Health Care and Philosophy 10(1), 5–10.
  12. WHO (World Health Organization) (1946). Preamble to the constitution of the World Health Organization as adopted by the international health conference, New York, 19–22 June, 1946; signed on 22 July 1946 by the representatives of 61 States and entered into force on 7 April 1948, Official Records of the World Health Organization 2.
Evaluating Efficiency of a Health Care System
Impact of Income Inequality on Health