By Jeremy Davis.
One of the most concerning aspects of our current crisis is the massive shortage of mechanical ventilators. Such scarcity has already led to rationing in some places; elsewhere, rationing will soon be unavoidable. This raises difficult ethical questions, perhaps the most central and urgent of which is: Which patients should we prioritize?
In their recent guidelines, The Italian Society of Anesthesia, Analgesia, Resuscitation, and Intensive Care (SIAARTI), holds that “we must aim at guaranteeing intensive treatments to patients with greater chances of therapeutic success,” which they say is “a matter of favoring the ‘greatest life expectancy’.” Similarly, the New York State Department of Health’s Task Force on Life and the Law holds that “an allocation protocol should utilize clinical factors only to evaluate a patient’s likelihood of survival and to determine the patient’s access to ventilator therapy.”
Thus far, the primary focus of the ethical commentary on access to ventilators has been their initial distribution—that is, when there are several patients who need one, who should get the next available ventilator? A distinct but related question concerns if and when withdrawing a ventilator is morally justified.
Withdrawing a ventilator is justified when further treatment is futile, or when the patient’s condition worsens considerably. In an interview with the New York Times, Dr. Matthew Wynia notes that, “if things are clearly getting worse, it’s really hard to justify a stance of once you’re on a vent, you own it, no matter how many people have to die in the meantime.” In such cases, there is no good served by continued treatment, so the rationale for withdrawal is clear, independent of the benefits that will follow from making the ventilator available to the next patient.
The more difficult question, however, is whether it is morally justified to withdraw a ventilator from a patient whose condition is stable—or, at least, neither futile nor worsening significantly—in favor of a patient, or perhaps several patients, more likely to survive. But is non-voluntary withdrawal of a ventilator from a patient who continues to benefit from it morally justified?
One could argue that a policy allowing doctors to withdraw patients from ventilators to reallocate them to other patients with better prospects might make it less likely that the greatest number of lives would be saved in the end. The New York State guidelines noted above claim that such a policy “could result in not providing any patient sufficient time to benefit from ventilator treatment if there is a constant influx of new patients who have a higher likelihood of survival.” Sheri Fink makes a similar point: “If physicians try to redirect resources—putting a patient on a ventilator for a few days, then giving it to someone else who appears to have better prospects—more people may die because few would get adequate treatment.”
This may turn out to be true, but it’s too hard to know with any certainty. It seems roughly as likely that this kind of policy would have precisely the reverse effect, and many more lives would be saved by continually reassessing the distribution of ventilators across all patients, whether currently on a ventilator or not.
Let us assume, however, that in at least some cases withdrawing ventilators would be more effective in saving lives. Does this make it morally justified?
Consider an example: A patient currently on one of the hospital’s few ventilators, whose condition isn’t worsening but is improving only very slowly, will require approximately four more weeks on a ventilator. But we have good reason to believe that another patient who is currently being treated in the hospital will soon require use of a ventilator, and will require it for much less time—say, two weeks. And his comparatively quick recovery will then free up that ventilator for a future patient who (we can fairly predict, based on the admission trends) will arrive in two weeks, and who will also require the ventilator for only two weeks. Should the ICU physician withdraw the ventilator from the first patient so that he can use it to save the other two?
If saving the most lives is all that matters, then the answer is clear: the physician should withdraw the ventilator. But this neglects the moral distinction between failing to aid—as when one is passed over for an available ventilator—and actively killing, which is how we ought to classify withdrawing a patient’s ventilator in cases like this.
In general, it is easier to justify a failure to aid than an active killing: while it may be permissible to let A die if we can thereby save B and C, it does not follow that we can kill A in order to save B and C. This principle applies elsewhere in medical ethics, and normative ethics more broadly. To borrow from a classic example, most would agree that it is justified to give five available organs to five patients in need, even if another needy patient would be left to die as a result. However, suppose we could kill one patient, harvest his organs, and distribute them to five needy patients. If all that mattered were saving the most lives, then we ought to do it. But surely harvesting one’s organs, even for the benefit of a greater number, is morally wrong.
The distinction between killing and letting die in these cases is not absolute. For example, if we could save 10 (or 20, or 50) lives by reallocating a ventilator—and in doing so, killing the patient currently assigned to it—this may be enough to justify it.
Recognizing the distinction between killing and letting die has several important practical implications. First, if it is harder to justify withdrawing a ventilator, then the issue of initial distribution is more morally significant than we might have previously thought. That is, the question of whom to assign to an available ventilator is more than just determining whom we ought to aid when patients outnumber resources. It also effectively removes that ventilator from the resource pool until either the patient recovers or further treatment becomes futile. Perhaps this serves as an additional reason to prioritize patients with better prospects over those for whom treatment is likely to be protracted and ineffective.
This distinction also plays a part in discussions about other novel policies, such as universal do-not-resuscitate orders for patients who code while on a ventilator. On one hand, such cases are failures to aid; on the other hand, that failure is intentional, so as to free up a scarce resource. Our ethical analysis of such policies must bear this distinction in mind.
Again, when continued treatment is futile and other needy patients are waiting desperately, withdrawal seems justified. We shouldn’t move heaven and earth for a patient with incredibly low prospects for survival when there is an ever-growing line of other patients who desperately need that same resource. But we must also recognize that once patients are reliant upon ventilators for their survival, our treatment of them is governed by different moral considerations. Any serious moral analysis of our current crisis must take this into account.
Author: Jeremy Davis
Affiliations: United States Military Academy
Competing interests: n/a