Monday, November 02, 2009

Are QALYs Discriminatory?

In 'QALYfing the value of life' (J. Med. Ethics, 1987), John Harris claims that it is unjust "discrimination" to allocate scarce medical resources to the patients that would benefit most from them (in terms of "Quality-Adjusted Life Years", or QALYs). Instead, he says, we should try to save (or, rather, postpone death for) as many people as we can, without regard for how much different individuals stand to gain from continued life. Since each life "counts for one", Harris argues, postponing death for two 90-year olds (by a month) is more important than postponing a teenager's death by scores of years.

There's something strange about insisting that each person's life has "equal moral value", without bothering to assess how much each person stands to gain from continued life. One does not show equal concern towards two people by being indifferent to whether one receives a papercut or the other is beheaded. We should, of course, be indifferent to which of two people receives an equal benefit (else we would be treating the favoured person's interests as more important than the other person's). But it is no kind of favouritism to prefer that a greater benefit be bestowed upon whoever is able to receive it. Egalitarians are simply confused to suggest otherwise.

Harris claims that it is "ageist" to allocate medical resources efficiently, since it means we would sooner give a dose of life-saving antibiotics to a 20 year old (who can expect to gain 60+ years) than to a 90-year old with the same disease (who only expects to gain a couple of months from being cured of this particular ailment). I do not think that this should be considered "ageist". Again, we are not treating the elderly as less worthy of receiving an equal benefit. Rather, we are saying that, given a choice between offering a slight benefit to one person (who may be elderly), or else offering a much greater benefit to another person (who may be younger), the latter option is obviously preferable, on perfectly impartial grounds.

(Similar remarks apply to Harris' objections to the 'quality-adjusted' component of QALYs as being "discriminatory" towards those with a lower quality of life.)

Harris' central confusion is revealed in the following claim (p.121):
If for example some people were given life-saving treatment in preference to others because they had a better quality of life than those others... this would amount to regarding such people as more valuable than others on that account.

This simply isn't true. Again, we do not ordinarily think that equal moral concern for persons entails being indifferent between a smaller benefit for one or a larger benefit for another. Rather, we may regard people with equal concern (i.e. treat them as being equally "valuable" in themselves) by treating their welfare interests as mattering equally. And treating their interests equally should lead us to prefer that treatment go to whoever would benefit most from it.

Simply put: to say that someone would gain more from continued life is not to say that the person is more valuable than another. It is just to say that what the person gains is more valuable to that person than the alternative outcome is to the other person.

P.S. In principle, we need to assess possible harms and benefits in context: it's not as though a healthy year of life has constant value to all persons at all times. As I argue in 'Gambling Life for Immortality', the next forty years of healthy life are worth much more to me than a subsequent forty (even if they're equally healthy, etc.), in light of what I want to achieve during my lifetime. So the real problem with QALYs is that they are insufficiently discriminating! But of course practical policies cannot be so perfectly fine-tuned, so QALYs may be the best practicable guide to efficient resource allocation. They're certainly a vast improvement over Harris' proposed alternative.


  1. Richard,

    I'll have to read Harris' article to know whether he makes this distinction. But we should distinguish discrimination from unjust discrimination.

    Using age as a criteria for determining which life to save (the 20 year old versus the 90 year old) is clearly discriminatory in the simple sense - it is using age to decide about treatment. But is it unjust discrimination (UD)?

    That depends on the conception of justice that you use.
    Here are two different ways of cashing out the injustice claim about QALYs (focussing on age for simplicity).
    1. Unjust discrimination involves the use of morally irrelevant criteria to make important decisions. For example deciding not to provide treatment on the basis of race or gender.
    Harris might be arguing that QALYs are morally irrelevant to treatment decisions therefore it this is UD.
    2. Unjust discrimination involves the use of membership of a protected minority group for decision-making even if that membership might be relevant to a particular decision. For example on this view it would be UD to deny employment to a female applicant to a position as a nightclub bouncer on the basis that they were female rather than male. (Assume that this is relevant here - I accept it might not be). Harris might be arguing that the elderly, or those with reduced current or future quality of life belong to such a protected minority.

    I suspect that Harris is making the first claim, and my guess (from some of his other writing) is that it is based upon a particular notion of what having an interest in something entails, and how we determine the strength of that interest.
    On one version of preference utilitarianism, the 90 year old and the 20 year old have an interest in life-saving treatment because they have (or would have if they were conscious and reflecting about it) a preference that their life continue.
    The strength of their interest depends on the strength of their actual or hypothetical preference for continued life. Let us assume that the 90 year and the 20 year old both want to live very much - on this account they have an equally strong interest in life saving treatment.
    It would then be UD to decide between them on the basis of age.

    I don't hold this view, because I think (as you appear to) that the strength of an interest is dependent both on the strength of a preference as well as the nature of the benefit that accrues from the preference. So on my account age would be morally relevant.
    But there is a separate question about whether there is benefit in a general rule that prohibits discrimination on the basis of age (which in many instances is not morally relevant). A simple rule that prohibits age discrimination might have better overall consequences than a complex rule that allows exceptions in certain cases etc.


  2. One issue that may push people back in the direction of Harris concerns disability. While you are right that we should not be indifferent to papercuts and beheadings, some believe we should be indifferent--in the sense of not using it as a basis for choosing one course over the other--to whether the life-years gained contribute to more or fewer qalys given the presence of certain disabilities. If we can produce 50 qalys by saving the life of a non-disabled person but only 25 by saving the life of a paraplegic, perhaps we should be indifferent to the magnitidue of the benefit here.

    The other issue that comes up in cases like the ones you present but which involve multiple people, are those where we can eiterh save 20 80 year olds for one more year each of perfect health, or else save one 20 year old for 19 more years of perfect health. Cost-effectiveness analysis using qalys says save the 20 65 year olds, but some may think it's wrong to do this.

    None of this is new, but it does shake things up in ways that might make some things Harris says more plausible. Although I agree with you that claims like the one from Harris you quote are, strictly speaking, wrong.


  3. obviously i changed the twenty 65 year olds to twenty 80 year olds to add drama. but i failed to conceal the legerdemain completely.... :)

  4. Dom, you write:

    But there is a separate question about whether there is benefit in a general rule that prohibits discrimination on the basis of age (which in many instances is not morally relevant). A simple rule that prohibits age discrimination might have better overall consequences than a complex rule that allows exceptions in certain cases etc.

    It is, I believe, incorrect to characterize the QALY approach as discriminating, justly or unjustly, on the basis of age. Using QALYs to allocate scarce resources just happens, as a matter of contingent fact, to favor the young over the old. Under different circumstances, deciding on the basis of QALYs might have the opposite implication. This would be so, for example, if life expectancy for people of a given age was more than one year greater than life expectancy for people one year younger, provided that health remained constant over time. Indeed, this is actually the case (I believe) for babies and young infants, so in practice allocating resources on the basis of QALYs would favour older people, provided they are young enough.

    More generally, in deciding whether a policy is discriminatory, it is insufficient to look at the distribution across the relevant variable (sex, age, race): it is also necessary to look at the process that generated that distribution. If more women than men are, say, admitted to writing school, this may simply be the result of gender differences in ability or personality; if the differences were reversed, the same process would result in more men than women being admitted. (Of course, in practice many policies and decisions do discriminate on the basis of sex, and often unjustly so. But establishing this requires more work than merely observing a correlation.)

  5. Hi Paul, my postscript notes one reason for thinking that 19 years for a young person might actually be a greater benefit (since they could use that time to complete more tasks of central life importance, like raising a family) than 20 times 1 year for an elderly person who hasn't any outstanding life goals that they could complete in this time.

    More often, people seem to use such examples to suggest that there are reasons of "fairness" to offer even a lesser benefit to the younger person who, by virtue of their early death, has a worse life overall. Perhaps this is what you had in mind. I'm skeptical that there are any such reasons of "fairness". If saving the twenty really would be better overall -- if that's what we'd choose from behind a veil of ignorance, or if we were to imagine living each person's life in sequence, or some other impartial process that weights each person's interests equally -- then I can't see the motivation for thinking it's somehow "wrong" to do this.

    Back to the main point, you write: "some believe we should be indifferent--in the sense of not using it as a basis for choosing one course over the other--to whether the life-years gained contribute to more or fewer qalys given the presence of certain disabilities."

    Some people believe all sorts of ridiculous things. The question is, what possible reason could one have to support such a claim? One might question whether the people implementing the QALY policy have accurately gauged the quality of life possessed by the disabled. But supposing that the next fifty years of life really are only half as good for the paraplegic as for the able-bodied person, what possible reason could there be for treating these options equally? For that is to treat the paraplegic's interests as twice as important as the able-bodied person's interests. How is this relevantly different from the papercut/beheading case?

  6. Good stuff, Richard.

    I wonder, would you still want to say that two extra years for a 20 year old is a "greater benefit" than 20 more for an 80 year old? I guess I can imagine a reasonable person making this claim, but I'm not sure it's right. Nevertheless, I think there might be a case for giving the 20 year old two more years, even if it's not as large a benefit as 20 more for the 80 year old. (Two years is probably not enough time to accomplish major life goals, other than the goal of enjoying the time one has while one has it. But then an 80 year old can experience quite a bit of enjoyment in 20 years, no?)

    If we should save the 20 year old despite this involving the allocation of a smaller benefit to him than we could confer on another, would I want to say that this a judgment predicated on the notion of fairness? I'm not sure. I have to be motivated by some distributive concern not reducible to pure aggregation, but I'm not sure what I should label that concern in this case.

    With regard to your last question: imagine two people, identical in all respects except one has a disease that will kill him in one year, unless he gets the one pill you have to distribute, and the other will die from his disease in two years, unless he gets your pill. It seems plausible to say that the benefit you can give the one is half as good as the benefit you can give the other. Who should get your pill? Seems to me that flipping a coin is the right thing to do. But you seem to think that this would treat the interests of the gainer-of-one-year as twice as important as the interests of the gainer-of-two-years. Surely such an example shows that treating cases similarly even when vastly differential benefits are at issue *is not* akin to treating one party's interests as twice as important. Or no?

  7. Hi Paul, flipping a coin there actually does seem wrong to me. (As evidence, again consider what the interested parties would prefer from behind a veil of ignorance, or if they were to live out both lives sequentially, etc.) There's no justification for risking the less good outcome, that I can see.

    Mind you, I agree that someone who flips the coin needn't be internally motivated by the thought that Mr. One-year's interests are more important (or to be weighted more heavily) than those of Mr. Two-year. More likely, the person just has some confused (and overreaching) ideas about fairness, which leads them to neglect comparative evaluation of the degree to which each patient's interests are affected by their choice. But I do think that the result of this confusion is that the coin-tosser ends up treating the lesser benefit to Mr. One as though it were the moral equal of a factually greater benefit to Mr. Two -- even if he doesn't conceive of what he's doing in this way.

    As a methodological point: it may be difficult to make further progress by appealing merely to intuitions about particular cases, since our intuitions differ. Instead, can you identify any more general principle that might support and make sense of the claim that coin-tossing is justified in such cases?

  8. As a methodological point: it may be difficult to make further progress by appealing merely to intuitions about particular cases, since our intuitions differ. Instead, can you identify any more general principle that might support and make sense of the claim that coin-tossing is justified in such cases?

    Nope, no can do. I don't have any more general principles. I would, however, wish to resist the characterization of my stances as confused or overreaching about fairness. But if successful resistance requires my providing general moral principles to fend off your charge, I guess I'm outta luck.

    I'm even more out of luck if I'm not allowed to appeal to my intuitions in questioning the cogency of the appeals to a veil of ignorance, say. Yes, such a device usefully models some relevant considerations. But it may leave others out, especially if it leads to moral rationales privileging ex ante maximization of individual expected utility, for instance. Here I am swayed by Kamm's cases where it seems wrong to adopt a policy of depriving one of a chance to survive so that we can save another while also curing a third person's sore throat. But this again is an appeal to intuitions. Is it illicit? Is it illicit because I can't point to some other methodological device to vindicate it?

    I realize these are big methodological questions worth pursuing. But I don't have any answers beyond what strikes me as the right answer after reflecting on cases like the ones we've been discussing.

  9. Cool, an opportunity to expand on the infinitely valued life thing....

    Harris' conclusions are equivalent to those gained by assuming each moment of life has infinite value.

    If each moment is worth an unbounded amount, the only way to really improve total welfare is to give this infinite gift to more individuals, regardless of the individual in question.

    I like concrete examples, so I'm going to compare giving one hour to each the 20 year old and the 80 year old.

    Old people sleep a lot. The hour coincides with the octogenarian's afternoon nap. The 20 year old loses their virginity.

    While moving the nap certainly changes the example, stretching out the hour over time or over individuals, the choice is roughly going to be of the same kind - young people go out and do things. Old people often don't have the energy to do so.

    If I make the obvious extrapolation - I'd rather be the 19 year old than the 80 year old - multiplying quality of life factor by infinity puts the conclusion right back where Harris put it.

    Harris may not have believed that each moment has infinite value, but what he believe was logically equivalent, and thus equivalently absurd.

    However, there is a problem in that people have set points. Quadriplegics receiving the condition by accident don't significantly rate their post-accident life as worse. Age may actually be similarly irrelevant...or not.

  10. Hi, Richard. I'm very late to this discussion, but please consider this contribution.

    The premise of this discussion is that scarce health care resources must (or may) be allocated on a basis other than that of bidding in a market for such resources. In doing so, the discussion overlooks the ability of the recipient of such resources to contribute to their availability, and in turn overlooks the limitations of the validity of distributing goods and services constituting "health care resources" on a basis other than bidding in an open market. I'll give a real-life example to illustrate the problem here.

    My wife is 41 years old and almost completely blind. Her "QALY" might therefore be assessed at significantly less than 1.0 and one might argue that a 20-year-old, non-disabled person should receive life-saving medical treatment before her, all other things being equal.

    My wife holds a job at which she earns over $60,000 yearly, which is well in excess of the median average salary for an American adult. She is a net producer of medical resources because she pays for all her own medical expenses and also engages in voluntary and involuntary transfers of the wealth she creates (through her job) into medical resources to be used by others.

    The voluntary transfers are in the form of charitable donations to cancer research, because her father died of cancer. The involuntary transfers are in the form of taxation at the state and federal levels for use in Medicare, Medicaid, medical research grants and numerous other government-funded health care programs.

    Because my wife is over 40 years old and almost completely blind, one might argue that a 20-year-old, non-disabled person should receive life-saving medical treatment before her, all other things being equal. But what if that 20-year-old, over the course of his or her life, will not make a net contribution (i.e. individual contribution less individual consumption) of "scarce medical resources" as my wife would? The answer is that medical resources will grow more scarce. Thus, from a utilitarian or pragmatic view, the failure to account for the patient's net contribution to medical resources is a grave omission in any scheme for the allocation of such resources.

    Beyond pragmatism, one may question whether it is fair to use QALY (or any other formula that overlooks the patient's potential for net contribution to health care resources) for allocation decisions by any entity that is not funded completely by voluntary transfers of wealth. I'm not an ethicist, so I'll leave it to others to debate that point. If such a debate occurs, however, it should consider the idea that in an open market economy, no exchange takes place unless both/all parties believe they are better off making the exchange. A person accrues wealth in a market economy via one or more voluntary exchanges. Hence, the accrual of wealth is not per se illegitimate. It follows that the involuntary transfer of wealth from one person to another, whether it be for health care or any other expense, must somehow be morally justified. Over to you.

  11. John - right, this is a discussion about how to distribute whatever limited public funds are allocated for health-care. A radical libertarian might argue that there shouldn't be any such public spending in the first place. But few people accept such a view. And given that there is going to be some such public spending, we may consider how it should be spent. (Even a libertarian could say this: "I don't think there should be any public funds for this at all, but so long as there, at least make sure that the money is spent as efficiently as possible...")

    Regardless of how the public funds are allocated, people should (of course) be free to spend as much of their own money as they like on whatever medical treatments they like. So if your wife pays her own way, no problem. That's compatible with holding that public funds for healthcare should be distributed according to QALY cost-efficiency.

    (It's true that QALYs don't take into account every utilitarian consideration. But the thought is that they are the most practicable approximation available, and at least a step in the right direction compared to the sort of undisciplined allocation of public funds that is the most likely alternative.)

  12. I noticed from your discussion that your examples focus on age. My recent QALY work is looking at sleep. After accounting for death, quality of life is dependent on consciousness (missing a leg or eye rarely matter if asleep).

    Suppose you have two adults of the same age, one sleep eight hours and another who sleeps nine per night. The difference in consciousness between them is one hour per day or a half month per year. Over the years this difference adds up.

    Assuming that both adults have the same vitality (just different sleep need), society would favor the person who sleeps less, because most QALY measures prefer consciousness over unconsciousness. Therefore, might sleeplessness be beneficient?

    I don't want you to lose sleep over this, but the QALY concept at this time is still in development. Even its most vocal advocates are pragmatic skeptics. The more difficult question is how can we improve our policy decisions (using QALYs or other summary evidence).


Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)

Note: only a member of this blog may post a comment.