Are priorities arbitrary? Evidence from the European health basket

The public debate about priority setting in health care is usually carried out by nationally based commissions that produce reports emphasizing one or other abstract principle or developing empirically based normative guidelines (Ham, Reference Ham1997; Sabik and Lie, Reference Sabik and Lie2008). The European HealthBASKET programme offers a tool to identify these core values. In developing an inventory of health care services provided by public funds in Europe, this programme provides an empirical base for comparing different member states.

This comparative studyFootnote ² emphasises the similarities between states: most countries exclude plastic surgery, travel medicine and non-conventional treatments such as acupuncture from the list of state supplied health care services (Schreyogg et al., Reference Schreyogg, Stargardt, Velasco-Garrido and Busse2005).

Despite significant differences in the institutional structure and organisation of national health services, variations in the services provided appear to be fairly minor. This observation, however, is only valid if we consider a very broad conception of service provision which focuses on the so-called ‘depth’ of the health basket. If we focus instead on the ‘height’ of the basket – the proportion of the cost of goods and services charged to or reimbursed by the state or public institutions – the story is somewhat different. The depth criterion enumerates any good or service which is subject to some form of reimbursement and therefore not exclusively paid for by the patient. The height criterion takes into account the cost for the patient and the co-payment amount, introducing a monetary dimension. Thus, health baskets with the same depth can have very different heights.

Applying the height criterion, we can see that there are differences between European countries, though some groups such as children, adolescents, pregnant women and people in need of hospital treatment, benefit from high coverage everywhere. Some groups are prioritised over others through differential reimbursement or state financing for either all or some of the care that they receive. For example, France has, historically, accorded priority to the most seriously ill, the costs of whose care are entirely met. Since 2000, the poorest have also received total cover.Footnote ³ A recent proposal aims to abolish this level of cover for the seriously ill group, replacing it with a threshold of health spending defined in absolute terms or as a percentage of income.

Observation of health care baskets in Europe, in respect of both depth and height, shows that the same treatment can be financed very differently depending on the type of patient concerned. Health services all over Europe have created, developed or activated prioritisation policies designed to privilege certain patient groups by protecting them from excessive financial outlays. This form of prioritisation has intensified with the introduction of market reforms designed to reduce expenditure (Saltman, Reference Saltman2002; Wynand and Van de Ven, Reference Wynand and Van de Ven1996). Though health provision has always been targeted at certain groups, the erosion of solidarity has forced governments to justify current prioritisation policies and develop new ones.

The replacement of universal social protection policies with targeted ones has not led to a focus on one target group. Different groups such as the poor, the elderly or those with the most severe illnesses have been targeted in different circumstances. This variety in prioritisation mechanisms suggests that there are several different, yet potentially legitimate, conceptions of state health coverage at play in European health systems.

It is tempting to consider any explanation that relies on such historico-political factors to be arbitrary and thus devoid of any underlying economic logic or justification. On the one hand, economic enquiry has suggested that the medical profession plays a role in the creation of clinical needs (Cookson and Dolan, Reference Cookson and Dolan2005). Such clinical need is attributable to differences between doctors and patients in their knowledge of and attitudes towards medicine. According to economic theory, asymmetries of information can produce an artificial clinical demand which is not justified by the patient's health status (Evans, Reference Evans and Perlman1974; Leonard et al., Reference Leonard, Stordeur and Robertfroid2009). Technological innovation can also be exploited by the medical profession in order to promote profitable treatments that lack efficacy. On the other hand, medical innovations, especially in surgery, can provide new treatments which do become publicly funded (Korica and Molloy, Reference Korica and Molloy2010). One example is cataract surgery, now regularly offered to the elderly and which imposes a financial burden on the state. Technological innovation and its utilisation by the medical profession play a more important part in the increase in health expenditure than ageing itself and also contribute significantly to longevity and to quality of life (Dormont et al., Reference Dormont, Grignon and Huber2006). Notwithstanding the existence of corporatist interests or political lobbying, clinical need does not inevitably lead to overutilisation. The measures that contribute to the height of the health basket may appear arbitrary to some economists, but they are not in the eyes of the populations that support them.

It is difficult to provide an economic rationale for the criteria upon which prioritisation is built because they are unrelated to questions of efficiency. Targeted solidarity is not justified in instrumental terms: priorities are chosen and defended independently of their contribution to overall social welfare. Rather than contributing to welfare, priority-setting criteria are amongst the measures that help to define it. This may be seen from the fact that prioritisation policies attempt to distinguish essential from secondary concerns within the health system. They do not, therefore, assume the sort of indifference that would normally be associated with arbitrariness. Instead, prioritisation draws on implicit hierarchies of social and medical values that are rooted in the history of different national systems of social protection.

Priority setting in health policy draws on the moral aspects of solidarity that are endemic to European culture (Maarse and Paulus, Reference Maarse and Paulus2003). This social dimension would lead us to expect different health systems to share fundamental normative commitments, though it remains to be seen how these commitments inform concrete policy measures.

The justification of priorities and the normativity of health systems

The justification of prioritisation in coverage and access to health services thus frequently rests on implicit moral ideals embedded in an equally implicit conception of a common good for society and its members. The latter is susceptible to multiple potential definitions, but can be generally understood as that resource or set of resources to which all human beings must have access to live with dignity in their community.Footnote ⁴ By citing measures currently in use in Europe we aim to show that their priorities do reflect citizens’ sense of justice about health care; in these attempts to define priorities, health policy is translating moral ideals in general currency, into institutional mechanisms.

Another approach to charting such values draws on a descriptive typology of health care need. Hasman et al. (Reference Hasman, Hope and Osterdal2006) argue that three different principles, commonly used to underpin various institutional configurations of targeted solidarity, can be employed as criteria of three conceptions of health care need. These are: (1) the criterion of the poor initial state, premised among other things on the human desire to extend life, to have a life worth living, to end one's life well; these sorts of considerations justify allocating resources ‘regardless of life expectancy after treatment’. (2) The criterion of minimum normal functioning range, based on the argument that normal functioning is necessary for social participation, which justifies targeting the poorest for whom poor health can limit participation in a normal social and work life. (3) The criterion of significant gain, where ‘gain’ denotes measurability of improvement in health states. Here, need prioritisation can only be justified for needs meeting the criterion of demonstrable improvement according to some measure of significance.

This typology has limited analytical utility since it is largely descriptive of existing practices. It proposes three principles each with a different normative basis for the assessment of health care need. However, it does not interrogate or analyse the underlying structure of these complex social norms. This leaves unexplored the question of how the typology is to be grounded in the different moral conceptions of what is good for human beings on the one hand, and political concepts of rights, equality and justice on the other.

The key underlying issue for any needs-based typology is the interrelation of the biological and moral dimensions implicit in concepts such as ‘poor initial state’ and ‘normal functioning’. As stated, the first two principles do not show a basis for identifying these moral and biological dimensions. They are therefore principally of use in picking out working social conceptions of health care need in which the normative content is fixed by some inchoate mixture of moral and political notions. The third principle of significant gain is somewhat different. It brings health care needs under a cardinal scale through the biologically based notion of gain, whilst locating the moral dimension in the notion of significance. However, Hasman et al. define the latter as ‘capacity to benefit’, which is, again, a mere placeholder for the same inchoate mixture of moral and political notions. Nevertheless, despite its analytical limitations and its inability to provide a universal criterion for what is to count as a health care need, the typology does have empirical utility. The three principles, deployed in different ways, can be found at work in the measures that fix the height of the health basket in European countries.

Those in the poorest health (the first criterion) are prioritised to varying degrees in different European countries. The criterion of the poor initial state is, for instance, only implicitly adopted in the UK and Italy where accident and emergency units provide the method for bypassing hospital waiting lists. It is more explicit in France, where patients suffering from 30 listed conditions are free from paying the usual charges in the form of the ‘ticket moderateur’. In Germany and Belgium, long-term care is regulated in such a way as to guarantee a strict upper limit on patient spending. Holland, which uses a system of competitive insurance, excludes serious conditions such as chronic illness and disability from the insurance system and provides state coverage for them, Finally, Switzerland has put in place special measures to protect those affected by genetic conditions.

The second criterion, normal functioning range, is translated into policy in most European countries by measures intended to limit individual spending on health. However, above a certain threshold of health spending (defined in absolute terms or as a percentage of income), requirements for co-payment can lead to treatment being relinquished for financial reasons, which can in turn make it impossible to achieve the minimum level of health necessary for normal participation in social life (‘normal functioning’). The generalised limit on spending may be linked to measures aimed at the poorest members of society. Belgium has put in place a supplementary payment in addition to reimbursement for beneficiaries of so-called ‘intervention majorée’. This is distributed according to income, disability or dependence criteria. In Holland, the spending cap is fixed by the level of excess paid by insured patients, which is alleviated by fiscal compensation to correct for the poorest. France has not adopted spending caps, leaving patients at risk of financial ruin, though the state has instituted free universal cover for the poorest members of society since 2000. In the UK, the social criterion is less visible because it is directly integrated into the funding mechanism that links the department of health to the 151 National Health Service (NHS) primary care trusts. Their access to resources is determined by an index that includes information on age, sex, geographic location and a range of socio-economic indicators.

The final criterion, significant gain, is much more uniformly deployed in European health systems. Coverage of expenditure on medication that has been authorised for market is affected by its measured therapeutic value. The proportion of the cost borne by the state varies between drugs that are considered to be vital and which are provided for free, to other products that provide comfort without medical benefit and whose cost may be paid in full by the patient. Otherwise, patient reimbursement can be tied to the price of a generic substitute. The criterion of significant gain plays a crucial role in the lists of products issued by national agencies such as HAS in France or NICE in the UK. The lists delimiting state coverage are usually positive, but can be negative as in Germany and the UK. In the latter case, NHS doctors are prevented from prescribing drugs that are blacklisted and can only prescribe grey-listed drugs to certain patients. This criterion is discernible in the guidelines and rules of best practice associated with the rise of evidence-based medicine.Footnote ⁵ This conception privileges scientific measurement over clinical experience in the proof of therapeutic efficacy, and takes the biomedical paradigm and statistical analysis as the basis for priority setting. Despite developments in statistical techniques and attempts to objectify it, the criterion of significant gain remains anchored in the socio-political contexts in which it is applied. For example, empirical studies on treatments for obesity and tobacco addiction show that coverage varies significantly between countries: the UK offers full coverage, but France and Germany do not (Nguyen-Kim et al., Reference Nguyen-Kim, Or, Paris and Sermet2005).

Our analysis shows that there are no universally accepted criteria for establishing priorities, and that the different systems of health care provision retain their own identities. Nevertheless, the measures taken are not arbitrary since prioritising must have some objective and generally recognised basis. Priority-ordering must therefore meet two conditions: first, priorities must be community-based, and reflect a conception of the common good, and second there must be an objectivity guaranteeing that this conception is not localised to one community. We propose to bring together these two apparently conflicting requirements.

Societal values and the community

Empirical studies demonstrating that the depth of the health basket is relatively consistent across European countries tend to reinforce this mainstream analysis by confirming that there is widespread agreement on what constitute ‘essential’ services. However, our discussion indicates that different, perhaps incommensurable, principles determine the (equally important) height of the health basket. Though these principles cannot be simply derived from underlying moral and political notions, their diversity indicates that the traditional exclusive emphasis on economic efficiency is misplaced. The view that societal values are distortions that hamper the smooth functioning of health care markets does not do justice to the way that priority setting, as we have earlier noted, draws on a conception of the common good for a society, if only as the bald fact that economic calculation presupposes some such definition of the common good.Footnote ⁶

The typology of Hasman et al. (discussed in Part 2) offers a way to identify relevant societal values in the form of the three principles underpinning the criterion of the poor initial state, the criterion of minimum normal functioning range and the criterion of significant gain. Each principle gives primacy to a different aspect of human social and biological functioning, and each criterion finds application in some circumstances; but none dominates to the exclusion of the others. This diversity suggests that there are several ways that communities might choose to assess individual and group claims on the health system. However, in the absence of a meta-criterion to adjudicate between these different principles, the social sciences cannot provide guidance on which one to adopt. This means that choice of criterion inevitably defaults to a local decision. Indeed, some authors have suggested that there is a strong case for locating the ultimate source of prioritisation policies in the community. This idea is taken forward by institutionalists who advocate a move beyond the current dichotomy between welfarism and extra welfarism in health economics (Hodgson, Reference Hodgson2009; Mooney, Reference Mooney and Davis2001, Reference Mooney2009, Reference Mooney2010).

The idea that priority setting in health is a reflection of communal conceptions of the common good is central to Gavin Mooney's new ‘communitarian claims’ research programme. A central element of his critique of standard approaches is the insistence that the definition of ‘good’ health cannot be derived from either subjective utilities, or objective measures such as QALYs. According to Mooney, this is because, ‘Our abilities to manage to desire good health and in turn our ability to judge what constitutes good health are functions of community expectations with respect to both individual health and community health’. (Mooney, Reference Mooney2010: 137). Mooney builds on well-known critiques of rational choice theory such as Sen's (Reference Sen1991) analysis of adaptive preferences, which suggests that individuals may not have the resources to imagine or expect a ‘better’ health state as defined by some objective measure and consequently that subjective preferences cannot be seen as proxies for health care need. However, universal measures of individual and community health can also be problematic and undesirable. In order to support this claim Mooney draws on what he calls the ‘normative interpretation’ of communitarian political philosophy, which emphasises the non-instrumental nature of our commitments to the communities that we live in (Mooney, Reference Mooney2010: 126–127). According to Mooney, membership in a community justifies special obligations in the realm of health care on grounds of solidarity and because of the fundamental contribution that good health makes to human well-being. These obligations are non-negotiable and constitutive of what it means to be a citizen.

Here, Mooney faces the familiar objection made against other communitarian approaches that his term ‘community’ is underspecified: it could refer to all the citizens of a supranational organisation, a nation state, a province or members of a particular ethnic, cultural or socio-economic group. Mooney clearly does not regard this inherent vagueness in the notion of community membership to be problematic for his approach to health economics. His pragmatic response is that, in the context of health care policy at least, ‘community’ tends to be defined geographically since health care systems are usually administered at the national or provincial level (Mooney, Reference Mooney2010: 125).Footnote ⁷ Accordingly, the key to settling questions of overlapping membership of communities lies in identifying both the legal and informal conceptions of citizenship that individuals subscribe to; a member of a community, in Mooney's sense, is above all a citizen with specified rights, duties and expectations (Mooney, Reference Mooney2010: 126). This dual reliance on a geographical conception of community combined with an emphasis on citizenship provides Mooney with the tools for determining community membership on a case-by-case basis.

Current health policy is rule based and focussed on standards such as QALY league tables; it puts exclusive emphasis on health outcomes as measured against a predetermined maximand and ignores citizens’ sense of justice in health care. This perceived weakness leads Mooney to propose the direct involvement of the community in setting the objectives and methods of health care systems through a political agreement or constitution enshrining the rights, claims and priorities associated with health care provision (Mooney, Reference Mooney2010: 153–158). According to Mooney, such a constitutional mechanism challenges the dominant rule-based approach because, ‘. . . benefit maximisation or health maximisation says nothing about who gets the benefits or whose health is improved. . . The new paradigm involves the notion of a third party, in this case the community, determining what claims are and their relative strengths’ (Mooney, Reference Mooney2010: 159). Moreover, he stresses that democratic mechanisms will most probably not be used to resolve policy issues directly through public votes on specific questions. The intention, rather, is to seek public endorsement for the procedural bases upon which health care institutions act and according to which their action is subsequently justified (Mooney, Reference Mooney2010: 108–111). Such democratic endorsement of the basic normative foundations of health institutions can be linked to Mooney's notion of citizenship as described above.

It is germane to our argument that the recognition of the implicit normative content in health economics has not led Mooney down the path of searching for universal moral principles upon which to found health systems. There is understandable resistance to the idea that we should seek out universal rules to guide our thinking about health care. The emphasis, rather, is on what social actors deem to be legitimate within their communities and the task of the social scientist is descriptive rather than prescriptive.

At this stage, we might ask whether an appeal to community preferences as the basis of priority setting in health care can be sustained in the absence of some underlying moral commitments? Though Mooney stops short of explicitly enunciating universal principles, he does decisively reject full-blown relativism and makes implicit appeal to such commitments. In this rejection of relativism he refuses to endorse the view that any set of community preferences automatically embodies a legitimate conception of the common good. Indeed, he recognises that the adoption of a constitutional mechanism does not rule out the possibility that a whole community could be ‘wicked’. Mooney further admits that procedures cannot alleviate the problem of a wicked community. In the final analysis, a community's health care choices must be judged against a universal principle (or set of principles), whose content is left unspecified (Mooney, Reference Mooney2010: 137). A way to conceptualise these universal principles has been suggested by other institutionalists such as McMaster (Reference McMaster2007) and Hodgson (Reference Hodgson2008, Reference Hodgson2009), who have called for the integration of some non-utilitarian conception of need into health economics.

The reintroduction of need

In Hodgson's work the ‘need for health’ is grounded in notions of physical, psychological and socially induced suffering, in a way that our desires or preferences are not (Hodgson, Reference Hodgson2008: 241). Health needs have two salient characteristics: they tend to be non-voluntary and unevenly distributed. Many illnesses, injuries or accidents that create health needs are unpredictable and independent of the choices or responsibilities of those who suffer them. In addition, needs are highly varied and idiosyncratic; the incidence of health problems does not affect everyone equally and remedies may vary (Hodgson, Reference Hodgson2008 243–245). The heterogeneity and uncertainty surrounding the framing of health needs more generally is replicated in the case of the need for health care, which is the more pressing concern for health economists, according to Hodgson. He recognises both the dynamics of changing health care needs due to demographic, technological and environmental factors and the complexity and diversity of the institutional frameworks that could potentially be adopted to cope with them.Footnote ⁸

However, the two parts of Hodgson's argument – the provisional institutionalist analysis and the advocacy of the concept of need – are independent of each other. He does not connect the recognition of the objective basis of health care needs to a specific normative basis for institutional design. In other words, Hodgson's recognition of health care needs does not lead him to propose a hierarchy of essential services, nor a mechanism for determining what such a hierarchy might look like; the distinction between need and preference, resting principally on moral intuitions, remains schematic.

Our discussion has already noted the complex framework of unarticulated moral and political concepts underpinning existing typologies of need such as that of Hasman et al. In what follows, we propose to clarify this picture by drawing on the philosophy of need. We propose a new direction for the institutionalist programme in which need for health care can provide the moral foundations for the provision of health care in general, and prioritisation in particular.

One might already think that any claim on the health services is legitimated by the needs of the claimant (relative to the needs of others) and therefore that need forms the moral basis of all such claims. This is not a new idea in health economics; it is implicit in much of the work of extra welfarists and premised on the observation that need and desire do not always coincide (Brouwer et al., Reference Brouwer, Culyer, van Exel and Rutten2008). However, the tendency to focus on epidemiologically defined health outcomes has tended to reduce the concept of need employed in health economics to what Mooney calls ‘simple need’. Simple need is measured solely in medical or biological terms by assessing the extent of the health problem or the capacity to benefit from treatment, and assumes that health gains should be weighted equally, irrespective of who is benefiting from them. This definition of need makes no place for the social components of health and is consequently (in our opinion correctly) rejected by institutionalists.

One obvious alternative would be to adopt a social definition of needs and see them as entirely constructed towards socially defined goals, with no more than a contingent relation to simple need. This view would conceptualise health needs in the same way as needs for transport or leisure. The latter position is superficially consistent with Mooney's approach in that it would allow community preferences to determine need entirely. But this view of needs, even if intelligible for use in policies elsewhere in the social spectrum, is open to serious objections when applied in the domain of health care. Determining who to treat and how to treat them on the basis of such socially constructed needs detaches decision-making from ethics by yoking it to social contingency. Moreover, the constructivist position equates health needs with aggregated subjective assessments of utility and therefore runs the risk of committing the welfarist error at the community level. This is because nothing ensures that such collective assessments of health status are any more valid from a medical standpoint than individual ones. Furthermore, if one were fully to adopt the communitarian perspective, one would be unable to make comparative assessments of different health systems. As we have already seen in Part 3, this is clearly not the intention behind current institutionalist approaches.

In summary, institutionalists argue that the prioritisation of resources in health care should respond to objective, medical-biological needs, not to the subjective utilities of individuals. At the same time the normativity of need statements in the domain of health care must reflect societal values; this carries the risk of re-introducing utilities as aggregated individual preferences. We encounter here the instability inherent in discussions of human need. On the one hand, a concept of universal, medical-biological need will fail to take account of individuals’ values and life projects. On the other hand, a wholly socially constructed concept of need, one constructed out of the contingent preferences of individuals, has lost any claim to universality.

‘Thick’ health care need and vital needs

We have seen that the recognition of health needs presents a promising source of a non-arbitrary moral justification for the delivery of health care in a situation where resources are limited. Yet notions of need that are reduced to the medical-biological determinants of health and emptied of their social content cannot hope to resolve the problems associated with priority setting. Similarly, in constructivist notions that define need in terms of the preferences of the majority, the threat of relativism arises. In order to base priority setting on firmer foundations a more sophisticated ‘thick’ conception of need is required, one that incorporates both universal and socially contextualised elements. Health care needs must be structured Janus-faced: looking towards their communitarian context and at the same time connecting this to general human concerns. The task of this section is to show how health care needs, while based on a central core of objective truth about human beings as biological and social creatures, are at the same time ethically thick in having different value expressions according to social context. For this we turn to David Wiggins's concept of vital need (Wiggins, Reference Wiggins and Wiggins1987).

As Batifoulier et al. (Reference Batifoulier, Latsis and Merchiers2009) note, Wiggins is unsatisfied with the reduction of need to preference because of two characteristics of needs: their objective dimension, and their necessity based on the claims of an individual to avoid suffering. In what follows we argue that health care needs are to be seen as a subgroup of vital needs.

Although much has been written about the concept of need and about its relation to health care (Culyer, Reference Culyer1995, Hurley, Reference Hurley, Culyer and Newhouse2000), we draw here on Wiggins's concept of vital need for several reasons. First, it is analytically rigorous. As already noted, he carefully distinguishes need from want or preference. He also distinguishes absolute need as the ethically relevant form under discussion, from needs that are instrumental in realising contingent human goals and projects (Wiggins, Reference Wiggins and Wiggins1987: 6–10). Secondly, he links vital need (through harm) to suffering, which extends the scope of the account to (what philosophers call) individual moral psychology (see e.g. Wiggins, Reference Wiggins and Wiggins1987: 11, 14). Thirdly, his conception of need is non-Hegelian and non-Marxian (Wiggins, Reference Wiggins and Wiggins1987: 2–3). An important consequence of this is that the way that needs are given contextual shape by a community is not conceived of as dialectical.Footnote ⁹ Instead, Wiggins sees needs (and the rights that they engender) as being integral to the implicit moral consensus or ‘system of social morality’ that all societies must possess. They are discovered a posteriori when individuals question the circumstances under which the non-fulfilment of a need would constitute grounds for reconsidering their adherence to the society in question.Footnote ¹⁰ As a proposed mechanism of need-construction this is more realistic than an abstract dialectical model, and consequently, as we note in our conclusion, more accessible to empirical investigation. It also provides compatibility between the psychology of individual subjects and their social being. Fourthly, and last, through its definitional criterion of suffering harm Wiggins's account draws together the three levels of human experience as classically defended by the moral philosopher Bernard Williams (Reference Williams and Williams1995: 185–186). Williams has argued that judgements in terms of thick ethical concepts draw their evaluative significance from the social level of shared practice, the psychological level of individual attitudes and feelings and the objective level of facts such as the human capacity for suffering.Footnote ¹¹

Wiggins defines a vital need as one which must be met if the individual is to avoid suffering harm.Footnote ¹² The notion of harm is directly normative, being an impairment to human functioning and well-being.Footnote ¹³ Suffering, in addition to the descriptive sense of ‘undergoing’ something, is also normative. Wiggins links it directly with wretchedness (Wiggins, Reference Wiggins and Wiggins1987: 11), which brings out the psychological dimension of undergoing harm and its consequences. Vital need is thus seen to be a thick ethical concept incorporating social and psychological values. It also incorporates norms of human functioning. Vital needs have as their basic core what is demanded by invariant facts about human nature, the ‘laws of nature and invariable environmental facts, or facts about human constitution’ (Wiggins, Reference Wiggins and Wiggins1987: 16). But needs are also basic when they arise from what is invariant about humans as subjects. Need claims have both social and psychological normative force, as we have seen, but this normativity has itself an objective basis in the human condition; harm to human beings is harm to a species of creatures with minds and bodies, capable of psychological and physical suffering. These facts ground the normative principle that harm to and the suffering of members of the species should be limited: it registers something which is ‘non-negotiable in the various ideals about human harm and flourishing that conditions our sense of the socially (humanly) possible’ (Wiggins, Reference Wiggins and Wiggins1987: 16).

Vital need and ‘common humanity’

The reference point for the construction of vital needs remains the facts, biological, psychological and social of our common humanity as described by Bernard Williams (Reference Williams and Williams1973: 232): ‘those who belong to the species homo sapiens . . . [all] have the capacity to feel pain, both from physical causes and from situations represented in perception and in thought; and the capacity to feel affection for others’. This conception of common humanity connects need with moral psychology. It grounds a strong claim on the part of theorists such as Wiggins and Williams that reference to common humanity involves more than a methodological principle; it introduces a conception of what is ‘non-negotiable’. The normativity of vital need takes account of both the individual subjective meaning of harm and suffering, and its intersubjective meaning. It thus respects the connection between the significance of suffering for an individual and the morality of their society, made explicit by its societal values.

Vital need then provides the bridge to the thick concept of health care need. As a thick concept, health care needs incorporate values at all three levels: the objective, the psychological and the social.Footnote ¹⁴ By categorising health care needs as vital needs, it becomes plain that health care needs, while being basic as defined above, are not coextensive with needs for biological survival. Nor are they merely socially constructed. As a subgroup of vital needs to be met if harm and suffering are to be avoided, they are needs of human beings for their well-being as individuals in society. It is not our intention here to propose our own definition of well-being, or of harm, but to suggest that if a definition is to be anything other than ad hoc it will have to include two aspects we find in Wiggins's analysis. First, a clear statement of the ‘non-negotiable’ core conception of human well-being. This should refer to societal values, values embedded in our society, but it should also make explicit what is, and what is not, morally and psychologically envisageable for us under the richer conception of a common humanity sketched out here. And secondly, there should be a clear view of the ways in which the meanings of harm and suffering are understood by the different communities under consideration.

In Part 2 we referred to the typology of Hasman et al. (Reference Hasman, Hope and Osterdal2006), derived from a survey of existing practices. Its three principles or criteria each reflected a different normative emphasis in the assessment of health care need. However, it remained to identify and articulate the social norms underlying these principles. We argued that a needs-based typology must integrate the biological and moral dimensions underlying concepts such as ‘poor initial state’ and ‘normal functioning’, the bases of the first and second principles Hasman et al. identify. The third principle, of significant gain, brings together a biologically based notion of ‘gain’ and an evaluative notion of significance as ‘capacity to benefit’.

Armed with Wiggins's analysis of vital need, and proposing that health care needs should be placed in this category, we may now return to the Hasman typology to propose that part at least of its normative basis is supplied by the notion of harm. Wiggins's analysis of need ties directly to the avoidance of harm: accordingly the need for an ‘intervention’ (using Hasman et al.'s term) is a vital health care need if without it the individual will suffer harm.

Hasman et al.'s three principles all imply a reference to the notion of harm. The poor initial state and normal functioning criteria do so through the implicit claim that biological impairment is inevitably harmful to an individual's psychological and social well-being; the relation of need and harm here is in the reduction of present harm.Footnote ¹⁵ The principle of significant gain invokes an explicitly evaluative notion, the ‘capacity to benefit’, while however locating its evaluative dimension in the quantitative valuation of ‘health state forms’ using, again, biological and functional parameters. While such a quantitative, ‘objectifying’ approach might be thought to remove the need even for oblique references to human well-being, extending the analysis of need as the reduction of harm to this third principle not only preserves the richer interpretation of health care need, but also provides the basis on which benefit as the reduction of harm can be measured. Drawing on Wiggins's account of the dimensions of assessment of a vital need, the gravity of a need is the severity of harm or suffering arising at all three levels of evaluation (biological-functional, psychological and social) when the need is unmet.

As presented here, our theoretical proposal requires considerable further development to be empirically serviceable. First, however, we bring our analysis back to the work of Mooney. We have argued that health care need is a thick ethical concept; accordingly, the normative value that attaches to health care needs draws from all three levels at which we argue thick concepts derive their ethical import: the biological-functional, psychological and social. As noted earlier (Mooney, Reference Mooney2010: 137), Mooney explicitly links our desire for good health and our ability to judge good health to communal expectations. He is equally explicit in defending the view that good health makes a fundamental contribution to human well-being. Mooney's new paradigm of active community setting of health care priorities sees the legitimacy of community decisions as conferred by solidarity on matters of shared importance; it is in this area that concepts of need and, correlatively, harm and suffering, have their place. As a subspecies of vital need, health care needs inherit the non-negotiability of the avoidance of harm. At the same time, as a thick ethical concept, health care need rests on the three levels identified as sources of value (the functional-biological, psychological and social). Health care need thus has the flexibility to cope with the different socially conditioned conceptions of harm and suffering, and expectations of well-being of different communities. The impasse threatening Mooney's rejection on the one hand of universal moral principles, including a universal conception of the common good, and on the other of a full-blown relativism can therefore be avoided.Footnote ¹⁶