«Edited by ANNE MASON Research Fellow, Centre for Health Economics University of York and ADRIAN TOWSE Director, Ofﬁce of Health Economics Radcliffe ...»
The fourth edition of The Economics of Health and Health Care by Sherman Folland, Allen Goodman and Miron Stano was published in 2004. There is one chapter out of 24 on ‘Equity, efﬁciency, and need’. It devotes little space to the debate on ‘what is equity’, and the term ‘equity’ does not appear in the glossary. The main part of the chapter assumes the answer to the question of ‘equity of what’ is utility, as opposed to health or health care, but later parts explore alternative theories including equal opportunities and maximin.
The last textbook reviewed here is the second edition of Economics of Health Care Financing: the visible hand by Cam Donaldson and Karen Gerard, which was published in 2005. There are two chapters addressing equity and fairness.
The ﬁrst of these, on economic objectives of health care, reinterprets the ongoing debate on ‘what is equity’ as one concerning horizontal equity, and point out that there is another, vertical version of equity, namely the equity of ﬁnancial contribution to health care: in essence, the rich should contribute more towards the ﬁnance of health care than the poor.† The authors state their own view ‘that equity objectives should be focused on health care [such as equity regarding access, utilisation, and/or ﬁnance] rather than health’ (Donaldson and Gerard, 2005, p. 87; square brackets added). There is also a brief mention of procedural justice. The second chapter on equity reviews evidence on the distribution of contribution to healthcare ﬁnance and of opportunities to utilise health care.
From this brief review ‡ we will conclude that there are three core questions in equity and fairness in health and health care (also see Wagstaff and van Doorslaer, 2000).
First core question: what is equity of health care?
Equity in health care can be deﬁned either as the equality of expenditure, * However, Dolan and Olsen (2002) continue to discuss trade-offs between efﬁciency and equity in the context of health-related social welfare functions. The implication is that the deﬁnition of efﬁciency in use is not Paretian but utilitarian.
† This formulation is problematic. For example, ceteris paribus, horizontal equity requires that those with equal income make equal contributions to the ﬁnance of health care, and vertical equity requires that those with unequal need receive unequal health care.
‡ There is also a forthcoming textbook by Stephen Morris, Nancy Devlin and David Parkin. Judging by the list of contents on a promotion ﬂyer by Wiley, there will be one chapter out of eleven devoted to ’equity in health care’.
88 THE IDEAS AND INFLUENCE OF ALAN WILLIAMSaccess, utilisation, or health beneﬁt, controlling for need;* or, alternatively, equity in health care can be equality of health. This debate goes back to Mooney (1983), and was revisited by Culyer and Wagstaff (1993). The health economists’ community seems to be split roughly into those who argue for equity as equality of access (who may use utilisation as a proxy for access) and those who argue for equity as equality of health. The latter include those who appreciate equal access as an immediate objective as long as it contributes towards achieving equal health. Each school is related to a large empirical literature overlapping with public health, reporting the evidence on the extent to which equity is or is not achieved.
Second core question: equity of what?
Those who support the idea of health as the distribuendum face a twopronged attack. Besides the debate above against those who believe in the equity of healthcare services as opposed to the equity of health, they also face those who assume that the distribuendum is utility, not health. This question as to what to equalise across the relevant population relates to ideology. The welfarist tradition has held that the distribuendum addressed in economics is individual utility. Non-welfarism in health economics began by putting forward health as the distribuendum, but this may be because health was counted as one of the basic capabilities that people should have opportunities to achieve (Culyer, 1989; also see Cookson, 2005). On the other hand, challenging consequentialist foundations of standard welfare economics has led to questioning distributive justice as the only criterion for equity, and to arguing for procedural justice in health care (Tsuchiya et al., 2005).
Third core question: what is ﬁnancial equity?
Financial equity is deﬁned as unequal contribution to healthcare ﬁnance, in proportion to income. It seems to be a much less controversial topic,† and a large part of the literature concerns methods on how to operationalise this concept, and evidence on the extent to which different countries and healthcare systems achieve ﬁnancial equity.
Overview of Alan Williams’ contribution to the literature Let us look at Alan’s views on the three core questions identiﬁed above. Alan’s position was that what matters is the equity of health, or equality in lifetime experiences of health across different population groups. Equity in terms of health care (access or ﬁnance) is an instrumental means that contribute * And of course, there is the debate on what is need. The two rival deﬁnitions are: the capacity to beneﬁt, and ill-health per se.
† It also seems possible to deﬁne ﬁnancial equity as contribution to healthcare ﬁnance in proportion to the volume of heathcare services consumed, or in proportion to the size of health beneﬁt achieved, but neither seem to be debated. This is probably because the former is in effect the free market, i.e. the default model in microeconomics, and the latter may be interesting but in reality totally impractical.
BEING REASONABLE ABOUT EQUITY AND FAIRNESS 89towards achieving the more fundamental equity objective, equality of life time health. After all, the efﬁciency objective of any healthcare system is to maximise health, not to maximise healthcare delivery per se. Health care is of value to the extent that it improves health. If so, why should the equity objective be about health care as opposed to health? A good example is the Handbook of Health Economics chapter with Richard Cookson on ‘Equity in health’ (Williams and Cookson, 2000). All the ﬁgures representing different approaches to equity are about health between two parties, because health is the distribuendum.
Alan’s work on equity in health can be divided roughly into three groups.
The ﬁrst group of work is theoretical, which overlaps with political and moral philosophy. Besides his Handbook of Health Economics chapter this group includes a paper subtitled ‘A guide though the ideological jungle’, where libertarian and egalitarian views are contrasted, and discussed in the context of public and private funding of health care (Williams, 1988a). Then, there is an interesting exchange with the utilitarian philosopher John Harris who criticised ‘QALYfying’ peoples’ life, and the use of simple aggregation rules across different patients in healthcare resource allocation (Harris, 1987). Alan’s reply was succinct and summarised the differences between Harris’s position and his own (Williams, 1987). In a paper presented to the Royal Institute of Philosophy Conference on ‘Philosophy and Medical Welfare’ in 1987, and published as a monograph in the following year, Alan questions whether or not the societal value of a unit of health should be regarded as equal across all types of patient, and proposes several possibilities to be explored in survey work. This list includes discrimination (or differential weighting) by age, by family responsibilities, by social worth (or talent), by individual choice and behaviour, by deprivation, and (just to see whether people support it) by willingness and ability to pay (Williams, 1988). Leading on from this, the most important piece of theoretical work by Alan in this area is probably the one on the fair innings argument (Williams, 1997); this will be discussed in more detail below.
The second group of work overlaps with public health, and is about ﬁnding out the facts (the extent and determinants) of existing inequalities in lifetime health across different population groups within different countries.
A large part of this enterprise was carried out in collaboration with members of the EuroQol Group. The MVH (Measurement and Valuation of Health) study reported population norms for the UK using the EQ-5D instrument (Kind et al., 1998). More recently, Szende and Williams (2004) looked at the differences in self-reported health using the EQ-5D instrument across 15 different countries; they disaggregated this by age, gender and educational attainment; and reported inequalities in health at the individual level and disaggregated by the dimensions of EQ-5D.
The third group of work involves eliciting values: i.e. ﬁnding out what people think about the existing inequality in lifetime health across population groups, and to what extent people will trade off efﬁciency in order to improve equity.
90 THE IDEAS AND INFLUENCE OF ALAN WILLIAMSThis belongs to an area of work that some call ‘empirical ethics’. Alan often used participants of public health or health policy conferences he attended, and audiences of health economics lectures he gave, as guinea pigs (see for example the relevant sections in Williams, 1988b; or Williams et al., 2005). In a research area where large-scale surveys of representative samples of the general public are often seen as the only sound way ahead, Alan was impatient because imperfect data now has its uses and perfect data may never materialise. In a more formal project funded by the Economic and Social Research Council (ESRC) on ‘Measuring preferences regarding equity and variations in health’, Alan (and both of the authors of this paper) contributed towards quantifying the trade-off that members of the public will make between improving overall average health and reducing inequalities between different population groups (Shaw et al., 2001; Dolan et al., 2002; Dolan and Tsuchiya, 2003; Williams et al., 2005). The ﬁndings from this project will be revisited below.
THE FAIR INNINGS ARGUMENT IN HEALTH ECONOMICSHistory of the concept and the Health Economics 1997 paper In 1987, Alan had a debate in the Journal of Medical Ethics with John Harris. It therefore came as a surprise to some of us who were aware of this bit of history that at the ﬁrst International Health Economics Association conference at Vancouver in 1996 Alan gave a plenary talk, which was based on John Harris’ idea in his 1985 book The Value of Life. This was the fair innings argument, and the talk was subsequently published in Health Economics (Williams, 1997). The quote from TheValue of Life which Alan used summarises the essence of the fair innings argument: ‘while it is always a misfortune to die when one wants to go on living, it is not a tragedy to die in old age; but it is on the other hand both a tragedy and a misfortune to be cut off prematurely’ (Harris, 1985, p. 93; cited in Williams, 1997; italics in original). However, Alan has been thinking about discriminating against elderly people in a transparent and systematic manner at least as far back as 1987 (Williams, 1988b).
Similar ethical views have been discussed in the bioethics literature by Jonathan Glover (1977), Norman Daniels (1988), Michael Lockwood (1988), and Daniel Callahan (1990; see Tsuchiya, 2000a, for a review). The innovation by Alan introduced a clear trade-off with efﬁciency, so that the intensity of this argument can be expressed in terms of the magnitude of overall health gain that can be sacriﬁced for it, which was in stark contrast to the approaches taken by the philosophers.
In the 1997 paper in Health Economics, Alan argues that everybody should be entitled to some common target quantity of lifetime health, ideally measured in terms of QALYs. If some die without having achieved this fair innings, then they are in some sense ‘cheated’. Then, he introduces an inequality-averse health-related social welfare function, deriving a rate of substitution between a marginal health improvement to one group (whose lifetime health is expected
BEING REASONABLE ABOUT EQUITY AND FAIRNESS 91to exceed the fair innings) and a marginal health improvement to another group (whose lifetime health is expected to fall short of the fair innings); this value can then be used as the basis of an equity weight in healthcare priority setting.
The paper used the example of UK men from different socio-economic classes. The ﬁrst stage was static, and based on quality-adjusted life expectancy at birth (QALE(0); this is based on the usual life table adjusted for the level of health at different ages): men from social classes I and II have a QALE(0) of 66 years, whereas men from social classes IV and V have a QALE(0) of 57 years. Therefore, other things being the same, a marginal health gain to the former will be valued less by society than a marginal health gain to the latter.
So at this point, the fair innings argument is being used to generate social class weights.
The next stage was to make this dynamic. At any point in time, within any given population subgroup, expected age of death is an increasing function of age due to the survivor effect. Thus, if 20 year olds and 60 year olds from the same socio-economic background are compared, the 60 year olds will have a higher expected age of death (and therefore expected lifetime QALYs, or ELQ). Applying the fair innings argument to this situation will lead to the conclusion that, other things being the same, the social value of a marginal health improvement to 20 year olds is larger than the same to a 60 year old. This then is the fair innings age weighting (within a given socio-economic class).