When health-care systems become strained in the coming weeks, there won’t be enough ventilators to go around. The Office for Civil Rights in the Department of Health and Human Services warns that doctors cannot use disability-based distinctions to decide who gets access to one: “Persons with disabilities should not be denied medical care on the basis of stereotypes, assessments of quality of life, or judgments about a person’s relative ‘worth’ based on the presence or absence of disabilities.”

The problem is that many states have adopted treatment protocols that appear to contradict that declaration. Washington state’s policy provides that doctors should consider a patient’s “baseline functional status” — explicitly including “loss of reserves in energy, physical ability, cognition and general health” — before deciding whether to provide lifesaving treatment. Alabama is even more explicit, directing hospitals to “not offer mechanical ventilator support for patients” with such conditions as “severe or profound mental retardation,” “moderate to severe dementia” and “severe traumatic brain injury.”

To many in the public, rules like these make intuitive sense. When sufficient medical services aren’t available to every wounded soldier on a battlefield, doctors engage in triage. Those with the most severe wounds may be left to die — not because the doctors value their lives less, but because the doctors can save more people by concentrating on soldiers with less severe, but still life-threatening, wounds.

Let’s be clear: Virtually nobody — including us — believes that doctors must provide scarce lifesaving treatments to patients who truly cannot benefit because they will die in the immediate term no matter what. To that limited extent, the battlefield triage context may offer an analogy to the rationing decisions that hospitals may soon be forced to make in responding to the coronavirus. Medical systems need not withhold lifesaving treatment from those who can actually use it for the quixotic purpose of giving it to people who cannot — even if those who cannot benefit sought treatment first.

But the rules in Washington and Alabama, as well as Tennessee and Kansas — two other states that have drawn recent complaints — are not limited to that narrow context. And there are many crucial differences between the battlefield triage situation and the ones covered by those policies. For one thing, health systems are unlikely to be presented with a choice of saving one disabled patient or saving many more nondisabled ones. At the moment two patients seek a single ventilator, we can save only one of them. Unlike the battlefield, the choice here is not between saving more lives and saving fewer. The choice is which life we should save — and whether a preexisting disability should be the basis for that choice.

Still, shouldn’t we prioritize those who will live longer lives, or lives of higher quality, after they receive lifesaving treatment? That’s the approach taken in many states’ policies, including Alabama and Washington.

Let’s examine “quality of life” first. Denying people with preexisting disabilities lifesaving treatments for distinct conditions, based on the concern that the disability limits the quality of their lives, raises deep and serious ethical questions. May a doctor deny such a treatment to a patient who mopes his way through life, on the theory that the happy-go-lucky patient in the next room will benefit from it more? Or to make the point even sharper, what if poor people or members of stigmatized minority groups experience less happiness in their lives than those who are better off? Would it be acceptable to prefer the wealthy for lifesaving treatments because they will obtain a greater benefit from it?

These problems are compounded by the vexatious nature of any effort to measure the quality of life — and by the well-documented differences in the views of disabled and nondisabled people on the quality of life with a disability. A striking fact is true of both physical and mental disabilities: People without those disabilities tend to believe that the conditions inherently limit happiness and the quality of life. But people with those disabilities can live full lives and tend to report life satisfaction that is close to, if not the same as, that of nondisabled people. People with disabilities are substantially underrepresented in medical professions. To allow (largely nondisabled) health professionals’ views of the quality of life with a disability to dictate whether disabled people receive ventilators is to invite bias into life-or-death decisions.

Bias does not just affect judgments about the quality of life with a disability. It also renders suspect judgments about disabled persons’ expected length of life. Studies have repeatedly demonstrated that physicians maintain (conscious or unconscious) bias about disability, as detailed in the National Council on Disability’s latest bioethics and disability report series to policymakers. Such bias can lead them to underestimate disabled patients’ life expectancy. Even if, in the abstract, doctors could ethically rely on expected length of life — and, again, would we permit them to deny lifesaving treatments to members of racial minority groups if the actuarial tables demonstrated that members of those groups live shorter lives? — their concrete judgments about the effect of disability on life expectancy are likely to be unreliable.

Denying ventilators to coronavirus patients with preexisting disabilities is thus not at all the same as making a battlefield triage decision to do the greatest good for the greatest number. There are a variety of ways a health system might, without any discrimination, allocate scarce lifesaving care during our current crisis: It might choose to prioritize treating health-care workers and others who can help treat or slow the spread of the virus. As the HHS explained during the George H.W. Bush administration, a system may consider “the cost of medical procedures, the length of hospital stays, prevention of death, and prevention of contagious diseases,” or, indeed, “any content neutral factor that does not take disability into account or that does not have a particular exclusionary effect on people’s with disabilities.” But health systems should not deny needed treatments simply because a patient has a preexisting disability.