Has Oregon Found a Better Categorical Rule?

Although a right to assisted suicide for the terminally ill can be seen as a better categorical rule than permitting assisted Suicide for no one, there is a serious question as to whether this is actually the case. A number of commentators argue that it is unworkable to base a legal (or ethical) rule on the presence or absence of terminal illness, that the distinction between terminal and nonterminal illness does not give us a bright-line distinction in the way that the distinction between treatment withdrawal and assisted suicide does.

Some ot the arguments against a rule based on terminal illness lack sufficient merit. For example, it has been argued that there is no principled basis for permitting assisted suicide for terminal people with very serious, nonterminal illnesses. Indeed, if an important goal of a right to die is to permit people to avoid intolerable suffering, then people who have a longer time in which to live arguably have a stronger basis for having a right to die.

If our right-to-die rules were based strictly on case-by-case judgments, this objection would be very persuasive. However, as I have discussed, the distinction between treatment withdrawal and assisted suicide is dictated not simply by the usual theoretical arguments but also by the moral con-lcerns involved when the state makes life-and-death decisions based on its own assessment of people's quality of life. This moral concern takes us to categorical rules for right-to-die law. By their nature, categorical rules are impertect reflections of important moral principles. Accordingly, it is not surprising that we can find people in one category who we think should fall into another category. That is precisely the compromise that a society makes when it adopts categorical rules.

The more important concern about a line based on terminal illness is the point that there is some uncertainty about Iife expectancy predictions, even for seriously ill persons, Although we can precisely define terminal illness in terms of a life expectancy of six months or less, we cannot precisely determine when a person meets that definition. We have all probably known people who lived for many months after they were supposed to have died. Among patients certified for hospice coverage under Medicare as having a life expectancy of less than six months, 15 percent survived for more than six months. Accordingly, some patients may choose assisted suicide on the mistaken impression that they have only a few months to live. This is a serious concern, and it may prove to be a real obstacle to a widespread acceptance of assisted suicide, even if the right is limited to terminally ill persons.

On the other hand, there are good reasons to believe that it is not an insurmountable problem. First, concerns about uncertainty can be answered to a large extent by limiting assisted suicide to cases in which predictions of survival can be made with a high degree of certainty. For example, predictions about life expectancy are more reliable for patients with metastatic cancer than for patients with congestive heart failure, emphysema, or end-stage liver disease.Second, when making predictions about life expectancy in dying patients, physicians are much more likely to overestimate than to underestimate how much longer the patient will live. Third, our experience with treatment withdrawal suggests that uncertain prognoses are not a truly serious concern. Imprecise predictions of lite expectancy are also a concern for withdrawals of treatment. Patients may refuse life-sustaining treatment because they have mistakenly been told that they are terminally ill or that they will be dependent on artificial life support for the rest of their lives. For example, my example of a "permanent-ventilator" patient, Mr. Poe, might not really have been irreversibly dependent on a ventilator. Since there does not seem to be a real problem with patients ending their lives prematurely by treatment withdrawal because of mistaken prognoses, it is unlikely that there will be a real problem with patients ending their lives prematurely by assisted suicide.

Importantly, the point of having a categorical rule to distinguish between morally justified and morally unjustified deaths is not simply to have a rule that can be easily and precisely applied. Rather, the primary concern is to avoid having representatives of the state decide on a case-by-case basis whether someone's life is so miserable that it need not be prolonged. Even if it is difficult to determine whether someone is ternminally ill, the determination does not require case-by-case judgments about the value of someone's life. It only requires case-by-case judgments about the expected length of the patient's life.

In short, although there is some reason to question whether terminal illness provides a sufficiently bright line for purposes of a right to assisted suicide, there is very good reason to think that it does. On this question, Oregon's experience with a right to assisted suicide for the terminally will be instructive.