1. From here I shall take the “morally” as read. The assumption that distributions of healthcare resources can be justified is not uncontroversial. Right-libertarians deny that distributions in general can be justified independently of their genesis (Nozick, R. Anarchy, State, and Utopia. Oxford: Oxford University Press; 1974 Google Scholar). Others (utilitarians and luck egalitarians) deny that a distribution of healthcare can be justified independently of how it fits into a larger distributive pattern of some general distribuendum such as welfare ( Singer, P, McKie, J, Kuhse, H, Richardson, J. Double jeopardy and the use of QALYs in health care allocation. Journal of Medical Ethics 1995;21:144–50, at 149CrossRefGoogle ScholarPubMed), for the view that healthcare is a separate distributive sphere; see Walzer, M. Spheres of Justice. Oxford: Oxford University Press; 1982, 86–91 Google Scholar. I address the latter concern by stipulating throughout that there are no counterbalancing distributive inequalities in relation to non-healthcare-related goods.

2. By “older adults” I mean people aged 67+. That is stipulative. By “disabled people” I mean people with “a physical or mental impairment that substantially limits one or more major life activities” (http://www.ada.gov). Non-discrimination derives from a general non-discrimination condition. I discuss the disabled and older adults because, for good reason, the literature on healthcare rationing and discrimination focuses on these groups.

3. Harris J. It is not NICE to discriminate. Journal of Medical Ethics 2005;31:373–75, at 375. The reason I focus on ageist and ableist discrimination is that these are the two forms of discrimination that cost-effective healthcare rationing based on conventional cost-effectiveness measures is most likely to result in and result into significant higher degree than the other forms of discrimination. The reason I do not address racist, sexist, or religious discrimination in a healthcare setting is not that these forms of discrimination are somehow less morally problematic.

4. Ubel, P, DeKay, ML, Boran, J, Asch, DA. Cost-effectiveness analysis in a setting of budget constraints: Is it equitable? New England Journal of Medicine 1996;334:1174–77, at 1174CrossRefGoogle Scholar.

5. Bognar, G. Age weighting. Economics and Philosophy 2008;24:167–89CrossRefGoogle Scholar; Bognar G. Does cost effectiveness analysis unfairly discriminate against people with disabilities? Journal of Applied Philosophy 2010;27:394–408; Bognar, G. Impartiality and disability discrimination. Kennedy Institute of Ethics Journal 2011;21:1–23 CrossRefGoogle ScholarPubMed; Bognar, G, Hirose, I. The Ethics of Health Care Rationing. London: Routledge; 2014, 53–78 CrossRefGoogle Scholar; Gold MR, Stevenson D, Fryback DG. HALYS and QALYS and DALYS, oh my: Similarities and differences in summary measures of population health. Annual Review of Public Health 2002;23(1):115–34; Singer, P, McKie, J, Kuhse, H, Richardson, J. Double jeopardy and the use of QALYs in health care allocation. Journal of Medical Ethics 1995;21:144–50CrossRefGoogle ScholarPubMed. An extra life-year at full health has the value 1 QALY, while an extra life-year at a lower level of health has a value between 0 and 1, depending on the severity of the health condition (for criticism, see Schneider P. The QALY is ableist. Quality of Life Research, online first: https://link.springer.com/article/10.1007/s11136-021-03052-4 (last accessed 8 February 2022); Tännsjö, T. Setting Health-Care Priorities. Oxford: Oxford University Press; 2019, 143 CrossRefGoogle Scholar). In my view, there are conditions better than full health, for example, having an artificially enhanced health condition better than “natural” full health, and conditions such that it would be better not to exist, for example, being in constant and extreme pain with no prospect of improvement, and, accordingly, one should accept QALY value assignments to an extra life-year lower than 0 and higher than 1. While I discuss QALYs, my arguments apply mutatis mutandis to other measures of health-adjusted life-years such as disability-adjusted life years (DALYs). For overviews of measures of health-adjusted life-years, see Bognar, G, Hirose, I. The Ethics of Health Care Rationing. London: Routledge; 2014, 29–52 CrossRefGoogle Scholar; Gold MR, Stevenson D, Fryback DG. HALYS and QALYS and DALYS, oh my: Similarities and differences in summary measures of population health. Annual Review of Public Health 2002;23(1):115–34.

6. Tsuchiya A. QALYs and ageism: Philosophical theories and age weighting. Health Economics 2000;9(1):57–68.

7. Harris, J. QALYfying the value of life. Journal of Medical Ethics 1987;13:117–23CrossRefGoogle ScholarPubMed; Harris, J. It is not NICE to discriminate. Journal of Medical Ethics 2005;31:373–75CrossRefGoogle ScholarPubMed.

8. Bognar, G. Does cost effectiveness analysis unfairly discriminate against people with disabilities? Journal of Applied Philosophy 2010;27:394–408 CrossRefGoogle Scholar; Brock DW. Ethical issues in the use of cost-effectiveness analysis for the prioritization of health care issues. In: Anand, S, Peter, F, Sen, A, eds. Public Health, Ethics, and Equity. Oxford: Oxford University Press; 2004:201–23Google Scholar; Brock, DW. Cost-effectiveness and disability discrimination. Economics and Philosophy 2009;25:27–47 CrossRefGoogle Scholar; John, TM, Millum, J, Wasserman, D. How to allocate scarce health care resources without discriminating against people with disabilities. Economics and Philosophy 2017;33(2):161–86CrossRefGoogle Scholar. It is sometimes argued that CER is discriminatory because QALY values of different states reflect the fact that as a result of discriminatory bias people overestimate the badness of disability and advanced age (https://ncd.gov/sites/default/files/NCD_Quality_Adjusted_Life_Report_508.pdf,) p. 40; Brock DW. Ethical issues in the use of cost-effectiveness analysis for the prioritization of health care issues. In: Anand S, Peter F, Sen A, eds. Public Health, Ethics, and Equity. Oxford: Oxford University Press; 2004:201–23; Persad, G, Wertheimer, A, Emanuel, EJ. Principles for allocation of scarce medical interventions. The Lancet 2009;373(9661):423–31, at 427CrossRefGoogle ScholarPubMed; Sinclair S. How to avoid unfair discrimination against disabled patients in healthcare resource allocation. Journal of Medical Ethics 2012;38:158–62; for a reply to Sinclair, see Whitehurst, DGT, Engel, L. Disability discrimination and misdirected criticism of the quality-adjusted life year framework. Journal of Medical Ethics 2018;44:793–95.CrossRefGoogle ScholarPubMed I sidestep this concern. It cannot show that CER based on non-biased assignments of QALY values is discriminatory (Bognar G. Cost-effectiveness analysis and disability discrimination. In: Cureton, A, Wasserman, D, eds. The Oxford Handbook of Philosophy and Disability. Oxford: Oxford University Press; 2020)CrossRefGoogle Scholar, and where certain disabilities (chronic depression, and very old age) are concerned, it becomes less relevant.

9. Incompatibility is consistent with treatments being perfectly equally effective from a medical point of view across old and young patients and across disabled and non-disabled persons, for example, a patient receiving a bypass operation will gain the same number of extra life years whether the patient is blind or not. (However, the QALY value of these additional years will be lower for the blind patient than for the non-blind person, ceteris paribus. Also, some disabilities, e.g., depression, come with co-morbidities, e.g., an increased risk of suicide) Incompatibility is concerned with cost-effectiveness and not medical effectiveness even though, of course, the latter affects the former. One might think that healthcare resources should be distributed based on medical effectiveness. However, that is not an objection to incompatibility. Incompatibility is also consistent with its being the case that many medical treatments, for example, very expensive cancer treatments, are cost-ineffective independently patients’ disability status or age. Finally, incompatibility is consistent with there being medical treatments such that they are more cost-effective if given to older or disabled patients than if given to young or non-disabled patients (see note 25). I thank an anonymous reviewer for pointing out the need to clarify these points.

10. Lippert-Rasmussen, K. Born Free and Equal? Oxford: Oxford University Press; 2013, 13–78 CrossRefGoogle Scholar.

11. “A socially salient group” is a group such that perceived membership of it makes a significant difference to social interactions across a wide range of different social contexts ( Lippert-Rasmussen, K. Born Free and Equal? Oxford: Oxford University Press; 2013, 30–6CrossRefGoogle Scholar). As defined here, the concept of direct discrimination is silent on whether instances of direct discrimination are always, often or only occasionally unjust. Sometimes “discrimination” is used in this purely descriptive sense. Thus, the 2005 NICE guidelines say that “where age is an indicator of benefit or risk, age discrimination is appropriate” (quoted by Harris, J. It is not NICE to discriminate. Journal of Medical Ethics 2005;31:373–75, 374CrossRefGoogle ScholarPubMed). However, on other occasions the term is used in a moralised sense so that anything discriminatory is ipso facto unjust or otherwise morally objectionable ( Lippert-Rasmussen, K. Born Free and Equal? Oxford: Oxford University Press; 2013, 13–53 CrossRefGoogle Scholar).

12. “Proportionate” can be cashed out in purely descriptive or (as is more often the case) non-descriptive terms. An example of the latter: disadvantages are proportionate if, and only if, they are of such a size relative to the benefits deriving from the relevant action that the action is not unjust.

13. Bognar G, Hirose I. The Ethics of Health Care Rationing. London: Routledge; 2014, 12. For an illustration of this way of challenging CER, see the HCFA’s objection to Oregon Health Services Commission’s rationing scheme ( Brock, DW. Cost-effectiveness and disability discrimination. Economics and Philosophy 2009;25:27–47, at 29CrossRefGoogle Scholar).

14. Bognar G, Hirose I. The Ethics of Health Care Rationing. London: Routledge; 2014, 12, 82; see also Bognar G. Cost-effectiveness analysis and disability discrimination. In: Cureton, A, Wasserman, D, eds. The Oxford Handbook of Philosophy and Disability. Oxford: Oxford University Press; 2020, 8CrossRefGoogle Scholar; Bognar, G. Impartiality and disability discrimination. Kennedy Institute of Ethics Journal 2011;21:1–23, at 8CrossRefGoogle ScholarPubMed.

15. Harris, J. It is not NICE to discriminate. Journal of Medical Ethics 2005;31:373–75, at 373 CrossRefGoogle ScholarPubMed.

16. Menzel P et al. Toward a broader view of values in cost-effectiveness analysis of health. Hastings Center Report 1999;29(3):7–15, at 10. Greg Bognar (Bognar G. Cost-effectiveness analysis and disability discrimination. In: Cureton, A, Wasserman, D, eds. The Oxford Handbook of Philosophy and Disability. Oxford: Oxford University Press; 2020 CrossRefGoogle Scholar) stresses that he uses the treatments-not-people response as part of a defense of cost-effectiveness analysis “as it is used in practice.”

17. Greg Bognar and Iwao Hirose hold: CER is not defeated by the fact that (arguably at least) justice is concerned with the distributive profile of health benefits and yet, as it is typically conducted, CER is distribution insensitive, because it need not be distribution insensitive (Bognar and Hirose 2014, 67; see section “Incompatibility: The Weighted QALYs Challenge”). It is not clear how they can take this position consistently with defending it against the ageism/disability discrimination objection based on how, in fact, it is conducted.

18. Bognar G. Cost-effectiveness analysis and disability discrimination. In: Cureton, A, Wasserman, D, eds. The Oxford Handbook of Philosophy and Disability. Oxford: Oxford University Press; 2020 CrossRefGoogle Scholar; Bognar, G. Fair innings. Bioethics 2015;29:251–61, at 259–61CrossRefGoogle ScholarPubMed; Johri, M, Norheim, OF. Can cost-effectiveness analysis integrate concerns for equity? International Journal of Technology Assessment in Health Care 2012;28:125–32CrossRefGoogle ScholarPubMed; Nord, E, Pinto, JL, Richardson, J, Menzel, P, Ubel, P. Incorporation of societal concerns for fairness in numerical valuations of health programs. Health Economics 1999;8:25–39 3.0.CO;2-H>CrossRefGoogle Scholar.

19. This raises the challenging question of what weight an extra QALY has when it accrues to a person who is both old and disabled.

20. Suppose pandemic involves only 1 person and that the effect of the vaccine lasts for only 40 years. We can give the vaccine either at birth or at the age of 40, but not both. Surely, it is not a better health outcome for that person to have the vaccine at the age of 40 rather than at birth as the weighted QALY view suggests.

21. A similar challenge can be directed at disability-weighted QALYs. Suppose we have a choice between giving a greater benefit to a person at a time in her life when she is not disabled or providing a smaller benefit to that person at a stage in her life when she is disabled. If we use disability-weighted QALYs, cost-effectiveness analysis may recommend the latter even if it is worse for the person in question.

22. Harris, J. QALYfying the value of life. Journal of Medical Ethics 1987;13:117–23CrossRefGoogle ScholarPubMed; Harris, J. The age-indifference principle and equality. Cambridge Quarterly of Healthcare Ethics 2005;14:93–9, at 96CrossRefGoogle ScholarPubMed.

23. Brock, DW. Cost-effectiveness and disability discrimination. Economics and Philosophy 2009;25:27–47, at 34CrossRefGoogle Scholar.

24. John et al.’s description of the first leg of what they call the QALY trap involves a similar appeal: “If we want to value interventions that raise people’s quality of life, we are forced to give less value to saving the lives of those with lower quality of life” ( John, TM, Millum, J, Wasserman, D. How to allocate scarce health care resources without discriminating against people with disabilities. Economics and Philosophy 2017;33(2):161–86, at 164)CrossRefGoogle Scholar.

25. Bad non-disabled converters are not far-fetched. In their response to Harris, Rawlins and Dillon give the example of secondary treatment for osteoporosis. This treatment has a significantly higher incremental cost-effectiveness for patients aged 70 than for patients aged 50, because the former are at higher risk of complications of osteoporosis ( Rawlins, M, Dillon, A. NICE discrimination. Journal of Medical Ethics 2005;31:683–84, at 683 CrossRefGoogle ScholarPubMed).

26. Bognar, G. Impartiality and disability discrimination. Kennedy Institute of Ethics Journal 2011;21:1–23, at 6CrossRefGoogle ScholarPubMed. Admittedly, some might respond that this is an oversight on part of friends of the moral equals argument and that CER treats women as if they have a higher moral status. In my view, this response is quite implausible—especially when coupled with the view that this putative fact renders CER unjustified.

27. Some argue that developing a disability is not a misfortune, but an injustice, the reason being that it is because of ableism that disabled people enjoy fewer QALYs than non-disabled people do. CER using QALYs—even when they are assigned in a non-biased way—compounds injustice in that ableist injustice is, so to speak, built into the QALY values assigned to different states (e.g., Wasserman, D, Asch, A, Bickenbach, J. Mending, not ending. In: Clements, L, Read, J, eds. Disabled People and the Right to Life. London: Routledge; 2008:37–50 Google Scholar; John, TM, Millum, J, Wasserman, D. How to allocate scarce health care resources without discriminating against people with disabilities. Economics and Philosophy 2017;33(2):161–86, at 169, note 15CrossRefGoogle Scholar). In my view, there is something in this view, but the issues are complex. My arguments in this section apply mutatis mutandis to this way of compounding injustice as well.

28. Brock, DW. Cost-effectiveness and disability discrimination. Economics and Philosophy 2009;25:27–47, at 35CrossRefGoogle Scholar. For related but slightly different versions of the compounding injustice claim, see Kamm F. Deciding whom to help, help-adjusted life years, and disabilities. In: Anand, P, Peter, F, Sen, A, eds. Health and Equity. Oxford: Oxford University Press; 2004:225–42, at 240 Google Scholar and Harris, J. QALYfying the value of life. Journal of Medical Ethics 1987;13:117–23, at 119–20CrossRefGoogle ScholarPubMed.

29. Hellman, D. When Is Discrimination Wrong? Cambridge: Harvard University Press; 2008 Google Scholar.

30. It might be replied that being disabled and being a bad converter of healthcare resources are too tightly connected to allow compounders of injustice to concede that the latter is a ground for imposing a disadvantage on people whereas the former is not. I am not sure this reply is sustainable. In any case, it requires a principled account of the conditions under which connections are “tight”—one that avoids implying that being a man is too tightly connected to being a bad converter of healthcare resources.

31. Lippert-Rasmussen K. Is there a duty not to compound injustice? Law and Philosophy; forthcoming.

32. Obviously, these are not the only accounts to which one might have recourse. Ben Eidelson’s respect-based account of the wrongness of discrimination is worth mentioning ( Eidelson, B. Discrimination and Disrespect. Oxford: Oxford University Press; 2015, esp. 71–94 CrossRefGoogle Scholar). On his account, the use of CER might directly discriminate against disabled and older adult people if, say, healthcare planners would not have used CER had its implications for white, middle-aged men been like those it has for older adults and disabled people. However, this possibility does not point to CER being inherently discriminatory (as opposed to a particular use of it being discriminatory). In previous work ( Lippert-Rasmussen, K. Born Free and Equal? Oxford: Oxford University Press; 2013:193–216 CrossRefGoogle Scholar), I have defended a desert-based prioritarian account of the wrongness of discrimination. On this account, CER would probably not qualify as wrongful discrimination against older adults and disabled people per se. Indeed, on a lifetime prioritarian view benefits to old people count for even less than they do on the standard non-weighted QALY measures, so the desert-based prioritarian account, also, is not of interest here.

33. By this I mean her 2008 account of the wrongness of discrimination in general. In the previous section, I referred to her later, narrower account ( Hellman, D. When Is Discrimination Wrong? Cambridge: Harvard University Press; 2008 Google Scholar) of the wrongness of indirect discrimination.

34. Hellman, D. When Is Discrimination Wrong? Cambridge: Harvard University Press; 2008:8 and 48, 175n2Google Scholar.

35. Hellman, D. When Is Discrimination Wrong? Cambridge: Harvard University Press; 2008:8 and 48, 175n2, 29, 7Google Scholar.

36. Hellman, D. When Is Discrimination Wrong? Cambridge: Harvard University Press; 2008:8 and 48Google Scholar, 175n2, 35, 57.

37. Some readers might think that more evidence must be provided for the contention that the cultural meaning of CER is ageist or ableist to be plausible. Such skepticism, however, goes hand in hand with, rather than undermines, the overall argument that I make in this paragraph.

38. It is well documented that people generally prefer weightings that favor treating younger people. Presumably, this indicates that rationing against older adult people in the sort of case I have in mind here does not carry the cultural meaning of unequal moral status ( Bognar, G. Fair innings. Bioethics 2015;29:251–61, at 260CrossRefGoogle ScholarPubMed; Bognar G. Age weighting. Economics and Philosophy 2008;24:167–89, at 168; Williams, A. Intergenerational equity: An exploration of the ‘fair innings’ argument. Health Economics 1997;6(2):117–32, at 1283.0.CO;2-B>CrossRefGoogle ScholarPubMed).

39. Moreau, S. Faces of Inequality. Oxford: Oxford University Press; 2020:211 CrossRefGoogle Scholar.

40. CER of treatments does not affect deliberative freedom, since the treatments offered will be offered irrespective of traits such as whether one is disabled or an older adult. However, Moreau might dismiss this point on grounds like those to which I appealed in section “Incompatibility: The Treatments-Not-People Challenge.” Note also that the deliberative freedom Moreau is concerned with is not the freedom to engage in deliberation per se, but the freedom from having to engage in deliberation of a certain kind, i.e., one where one or more of one’s extraneous traits should be regarded as a cost.

41. There is also the question of whether CER that gives some slight priority to women over men because of the former’s greater life-expectancy would violate the deliberative freedom of men.

42. Bognar G. Cost-effectiveness analysis and disability discrimination. In: Cureton A, Wasserman D, eds. The Oxford Handbook of Philosophy and Disability. Oxford: Oxford University Press.

43. Perhaps not all healthcare goods are basic goods. If they are not, Moreau’s third way of not relating as equals does not speak to CER of these healthcare goods.

44. Mason, A. Levelling the Playing Field. Oxford: Oxford University Press; 2006:39–67 CrossRefGoogle Scholar.

45. A relevantly modified QALY here would be one where an extra life-year in a perfect job has the value 1 QALY, while an extra life-year in a less satisfying job has a value between 0 and 1 QALYs depending on how much less satisfying the job is. A less satisfying job will have a QALY value of 0.5 when one would be willing to trade 2 years in it for 1 year in the perfect job (everything else ignored).

46. Swift, A. How not to be a Hypocrite. London: Routledge; 2003 CrossRefGoogle ScholarPubMed.

47. Anderson, E. The Imperative of Integration. Princeton: Princeton University Press; 2013 Google Scholar.

48. Segall, S. Equality and Opportunity. New York: Oxford University Press; 2013:173–206 CrossRefGoogle Scholar.