«Edited by ANNE MASON Research Fellow, Centre for Health Economics University of York and ADRIAN TOWSE Director, Ofﬁce of Health Economics Radcliffe ...»
EQUITY AND FAIRNESS IN HEALTH AND HEALTH CARE 105II: Fairness in health care – which types of inequalities are acceptable?
According to Elster (1992) ‘the tasks of the major theories of justice can be stated as justifying deviations from equality’. What is a fair, acceptable – or just – type of inequality in health or health care then? This is not only an issue of an equity–efﬁciency trade-off, like a degree of inequality as expressed in social welfare function (SWF) weight. It also deals with the reasons behind an inequality.
Le Grand (1987) argues that some types of inequalities in healthcare use are not inequitable as they result from different choices or preferences.
Following Le Grand’s conception of equality of choice sets, those inequalities in health that emerge from an equal choice set (opportunities) are considered fair or acceptable.
In the textbook Strained Mercy, Evans (1984) holds that ‘“need” also carries signiﬁcant ethical overtones; its allegation asserts an obligation on others...
that someone ought to do something’. This leads me to question whether we feel the same obligation on others – completely independent of the cause of their needs? While we do have mercy, when and why is it strained?
To me, Le Grand’s arguments and those of many other authors who suggest that people should be held responsible for their own health-related behaviour – particularly so if well-informed – deal with fairness rather than equity per se. Are people equally entitled, and are all needs equally worthy of collective funding?
My message here is simply that I think that – when we think of it – a large proportion of the reasons people give for justifying deviations from equality in access to health care deal with issues concerning the causes of the need for health care (see Dolan and Olsen 2001; Olsen et al., 2003).
III: Equity in health – equity of which health stream?
As for Aki and Paul’s second core question – equity of what? – their emphasis throughout the paper is on health, and more precisely total health measured by expected lifetime QALYs (ELQ) as the distribuendum. This is associated with the ‘fair innings’ argument: reduce inequalities in ELQ by distributing QALY gains to those with shortest ELQ.
While I agree with Aki and Paul – and Alan – that ELQ is the most important stream of health in which to reduce inequalities, there are other potentially relevant streams. I know Aki and Paul agree, and they have even written a paper, as well as a reply to Nord (2006), on this (Dolan and Tsuchiya, 2005, 2006). Two alternative streams of health that might be ‘distribuendum
● prospective health: differences in individuals’ no-treatment health proﬁles are important for both ethical and equity reasons: if QALY gains are distributed to the most severely ill, inequalities in prospective health are reduced
106 THE IDEAS AND INFLUENCE OF ALAN WILLIAMShealth gains: an important equity issue deals with the degree to which ● there is diminishing social value of increasing QALYs gained (Olsen, 2000).
IV: Fairness in health – which types of inequalities in health are acceptable?
In the following I shall stick to ELQ as the most relevant stream in which to reduce inequalities in health. But which types of inequalities in ELQ do we ﬁnd acceptable, or fair – and for which ethical justiﬁcation?
Aki and Paul refer to Alan’s survey (Williams, 1988b) in which he questions whether or not the societal value of a unit of health should be regarded as equal across all types of patients. In other words, while equity involves that ‘a QALY is a QALY is a QALY’, what Alan opened up here was to ask under which circumstances people might think that ‘a QALY is not a QALY is not a QALY’. Interestingly, of the eight different ‘discriminators’ Alan used in this pilot survey, the one that most respondents would take into account was whether ‘people who have taken care of their own health should get preference over those who haven’t’. The other discriminators dealt with age, consequences on others (caring and ﬁnancial), and deprivation (social class). Although the setting of the questionnaire was NHS priorities, and thus fairness in the distribution of health care, it appears that the issue Alan sought to explore was which patient characteristics respondents would discriminate for or against, i.e. weighting of health gain units.
Again, I think ‘fairness in health’ also deals with the ethical justiﬁcations we give for accepting inequalities in health, i.e. under which circumstances we would think that ‘a QALY is not a QALY is not a QALY’.
Interestingly, in their SWF, Aki and Paul assign equal weight to the two groups ‘where the assumption is that neither party is responsible for the differences in their lifetime health’. They seem to imply that it is fair to assign unequal weights had one party been more responsible than the other.
Furthermore, based on some preliminary surveys, Aki and Paul ask if it could be that the fair innings argument will work across social classes, but not across gender. This corresponds with results I got in a Norwegian survey where only 5.8% would assign more weight to the health of the group with shortest life expectancy when this was men versus women, while 24% would do so when it was the lowest social class. In the case of smokers versus nonsmokers, among those who were prepared to discriminate, more of them opted for the group with highest life expectancy. Again, this highlights the importance of looking at the cause behind an inequality. As Aki and Paul also accept, it is not only the degree of inequality in ELQ that matters for the elicitation of subgroup weights in a SWF, but the characteristic of the subgroups.
EQUITY AND FAIRNESS IN HEALTH AND HEALTH CARE 107
POLICY IMPLICATIONSThere is sufﬁcient evidence to prove immense inequalities in health – across gender, social classes, regions, etc. In my view, the two most crucial questions
1 for which subgroup characteristics do we consider inequalities to be most unfair?
2 which policy sector should be held responsible for reducing the unfair inequalities?
As to the ﬁrst question, social class seems to be the answer. The ideology behind public health care was based on the view that inability to pay is an unacceptable reason for denying people access to health care.
I think the second question depends on whether the observed inequalities in health are caused by inequalities in access to health care. If the answer is yes, the responsibility clearly lies within the health sector, and the policy implication might be one of allocating relatively more healthcare resources to those groups with lowest ELQ, e.g. by expanding healthcare provision in socially deprived areas. If the answer is no, I think the health sector responsibility issue is not that obvious. If health inequalities are caused by systematic variations in the determinants of ill-health, one might argue that it is the relevant host sectors of these determinants (e.g. housing, work safety) that should be held responsible, rather than suggesting that it is the responsibility of the health sector to repair these inequalities.
Even if all current healthcare resources were allocated to the lower social classes, we might still not achieve complete equality in ELQ. However, if it turned out to be technically feasible to reduce inequalities in ELQ by a radical reallocation of NHS resources towards the lower social classes, such a health policy might still not be considered fair.
FURTHER RESEARCH PRIORITIESFirst, concepts: Alan and Richard Cookson in their Handbook of Health Economics chapter (Williams and Cookson, 2000) held that in economics, ‘“fairness’ is taken almost unthinkingly to mean reducing inequalities’ (italics in original). As emphasised above, to me fairness has more connotations about which sorts of – and levels of – inequalities most people would consider acceptable. I think there is a need for some clariﬁcations regarding what we mean by fairness in health and health care. Those who are prepared to explore this controversial issue of letting personal health responsibilities matter, might wish to consult Lake’s book Equality and Responsibility (2001), which attempts to bring together ideas on equal distributions of goods with ideas on what people are responsible for.
Aki and Paul set out three main research directions. The ﬁrst is more descriptive including ‘the determinants of the variation of ELQ’. I would say
108 THE IDEAS AND INFLUENCE OF ALAN WILLIAMS‘the variation in the determinants of ELQ’, and draw attention to a framework that I developed together with Paul (Dolan), Jeff Richardson and Paul Menzel (Olsen et al., 2003). The causes, or determinants, of (ill-) health are of three
1 genetics (the biological lottery) 2 environment (the social lottery) 3 health-related lifestyle (explained by social conditioning and preferences).
These determinants are then located on a continuum with different degrees of individual control, and hence responsibility, for own health. My point is that while these three sets of determinants are analytically the same for all of us, there are huge variations in how they hit us, and hence impact on our health; e.g. the higher social classes have been luckier in the social lottery (or chosen a healthier environment, if you so prefer). I think that more analytical descriptive research on variations in the determinants of ELQ is needed – a prime example of which is the book edited by Bob Evans (1994), simply and neatly called Why are Some People Healthy and Others not? The determinants of health of populations. So, yes, this research direction is important – particularly if policy focused and evidence based.
Aki and Paul’s second direction involves economic theory and empirical work related to a health-related social welfare function. This is where I perceive them to intend to be ‘extending the Williams way’. Interestingly, within Alan and Richard Cookson’s handbook taxonomy of ‘Theories of equity in the distribution of health’, Aki and Paul’s type of SWF comes under the label ‘non-linear and smooth’. Like most people, including economists, I ﬁnd smooth curves much more appealing than linear or kinked ones, so I don’t blame them! However, while I ﬁnd iso-welfare curves nice in theory, I am pessimistic when it comes to measuring stable preferences in terms of people’s willingness to forgo overall health for more equal distribution. Aki and Paul acknowledge the complications and ask: ‘whether or not preferences elicited from the general public will be as well behaved as expected by theory?’ I believe such preferences are extremely sensitive to framing and context, including other streams of health such as severity levels and the absolute magnitudes of health gains. However, I share Aki and Paul’s interest in understanding the reasons that respondents give for their trade-offs, i.e. which types and levels of inequalities people consider to be fair.
The third research direction they outline is expanding into the economics of well-being. Well, it might well be a possible avenue for further work, particularly so if that is the interest of the researchers.
Then ﬁnally, I would like to thank Aki and Paul for a truly stimulating paper. I certainly think it has extended my way of thinking about equity and fairness in health and health care.
EQUITY AND FAIRNESS IN HEALTH AND HEALTH CARE 109REFERENCES Dolan P and Olsen JA (2001) Equity in health: the importance of different health streams. Journal of Health Economics 20: 823–34.
Dolan P and Tsuchiya A (2005). Health priorities and public preferences: the relative importance of past health experience and future health prospects. Journal of Health Economics 24: 703–14.
Dolan P and Tsuchiya A (2006) Severity of illness and priority setting: worrisome criticism of inconvenient ﬁnding? – A reply to Erik Nord. Journal of Health Economics 25: 173–4.
Elster J (1992) Local Justice: how institutions allocate scarce goods and necessary burdens.
New York: Russell Sage Foundation.
Evans RG (1984) Strained Mercy: the economics of the Canadian health care. Toronto:
Evans RG, Barer M and Marmor T (eds) (1994) Why are Some People Healthy and Others not? The determinants of health of populations. New York; A de Gruyter.
Kahneman D, Knetsch JL and Thaler R (1986) Fairness as a constraint on proﬁt seeking: entitlement in the market. American Economic Review 76: 728–41.
Lake C (2001) Equality and Responsibility. Oxford: Oxford University Press.
Le Grand J (1987) Equity, health and health care. Social Justice Research 1: 257–74.
Nord E (2006) Severity of illness and priority setting: worrisome lack of discussion of surprising ﬁnding. Journal of Health Economics 25: 170–2.
Olsen JA (2000) A note on eliciting distributive preferences for health. Journal of Health Economics 19: 541–50.
Olsen JA, Richardson J, Dolan P and Menzel P (2003) The moral relevance of personal characteristics in setting health care priorities. Social Science and Medicine 57: 1163–72.
Rawls JA (1958) Justice as fairness. Philosophical Review 67: 164–94.
Wagstaff A and van Doerslaer E (2000) Equity in health care ﬁnance and delivery.
In: Culyer AJ and Newhouse JP (eds) Handbook of Health Economics. Amsterdam:
Elsevier, pp. 1803–62.
Williams A (1988a) Priority setting in public and private health care: a guide through the ideological jungle. Journal of Health Economics 7: 173–83.
Williams A (1988b) Ethics and efﬁciency in the provision of health care. In: Bell
JM and Mendus S (eds) Philosophy and Medical Welfare. Cambridge; New York:
Cambridge University Press, pp. 111–26.
Williams A and Cookson R (2000) Equity in health. In: Culyer AJ and Newhouse JP (eds) Handbook of Health Economics. Amsterdam: Elsevier, pp. 1863–910.
Putting the ‘Q’ in QALYs... Paul Kind