Controversial Provision of Treatment: Conscientious Objection to Treatment Perceived as Futile

The most frequent reason for ICU clinicians judging treatment to be “inappropriate” is a belief that the patient is receiving excessive or unbeneficial medical treatment. ^{Footnote 17} In extreme cases this can lead to physicians and nurses declining to care for that patient. For example, in 2008, a dispute arose over the medical care of Samuel Golubchuk, an 84-year-old man with multiple organ failure and brain damage. ^{Footnote 18} Medical staff wished to discontinue treatment, whereas Mr Golubchuk’s family wished treatment to continue. It was reported that three physicians in the ICU refused to care for the patient because they regarded his treatment as unethical. ^{Footnote 19}

It is important to note that some authors have argued that refusal to provide treatment on grounds of futility is not a true CO (because it is based on professional norms/standards rather than personal values). ^{Footnote 20} However, the boundary between clinical disagreement and moral disagreement is often blurred. A determination that treatment is futile or unbeneficial can clearly be influenced by personal values, ^{Footnote 21} and, as is highlighted in this case, can lead to individuals declining to be involved in treatment. The ATS guideline includes moral objections to treatment (on the grounds of perceived futility) as a form of CO. ^{Footnote 22}

Controversial Nonprovision of Treatment: CO to Premature Limitation of Treatment/Palliative Care

The opposite ethical concern can arise when physicians or nurses have ethical objections to limitations of treatment. ^{Footnote 23} In a survey of 690 European intensive care professionals, approximately 43 percent of physicians and 53 percent of nurses indicated that they would go against a patient’s wish to refuse treatment if they felt that the patient would benefit from it. ^{Footnote 24} It is rare that such an objection would be absolute; physicians or nurses may accept limitation of treatment in some cases, but regard the outcome in a particular case as too uncertain (or too positive) to be comfortable forgoing support. It is of note that in the United States, 5 percent of physicians indicated an in-principle objection to withdrawal of life support. ^{Footnote 25} Objections of this nature sometimes impact on patients. In 2002, the United Kingdom courts heard the case of Miss B, a 45-year-old woman with long-standing spinal cord problems, who requested to be disconnected from the ventilator after she became quadriplegic. ^{Footnote 26} Intensive care physicians declined her request, although she had been assessed as competent by psychiatrists. They were prepared to institute a gradual process of reducing breathing support, and wished to give Miss B a further period of rehabilitation to see if she changed her mind. However, after more than a year of receiving treatment against her wishes, Miss B eventually sought (and obtained) court support to be allowed to die. ^{Footnote 27}

Clinicians in intensive care may or may not refer to their concerns using the language of “conscience” (they may refer to treatment options as “not clinically indicated” or “unethical”). Their concerns may be specific to particular cases, or they may be in-principle concerns about particular treatment options in intensive care. Regardless, the ethical question is the same: should they, or should they not object?

Autonomy

First, CNO respects patients’ rights to access treatment options that have been deemed legally permissible, and endorsed by members of the profession. This argument is based on respect for the patient autonomy, and patients’ freedom to make decisions about their own lives. Autonomy presents a strong justification for allowing patients to refuse treatment, even where the physician judges that the treatment would be in their best interests. This might ground a case for CNO in the context of limitation of treatment/palliative care. For example, two clinicians described their personal response to a case of ventilator disconnection in a paralyzed man, similar to the Miss B case. ^{Footnote 31} One of the physicians had strong pro-life beliefs, and clearly felt deeply conflicted about the decision to remove the ventilator. However, he was also reported to support individuals’ right to refuse treatment. ^{Footnote 32}

Autonomy might not provide as strong an argument in favor of providing treatment that the patient requests. Because autonomy is often construed in negative terms; that is, with regard to refusing treatments, patients are not usually thought to have a right to demand treatment. ^{Footnote 33} However, in situations in which other physicians would provide that same requested treatment, a physician’s refusal does appear to compromise the patient’s autonomy.

Justice

Second, CNO addresses or reduces one potentially unjust feature of some medical decisionmaking: the apparent variability of physician decision making, and influence of personal values. I noted significant variation between physicians in intensive care in their end-of life decisions. This appears to affect provision of intensive care, as well as withdrawal of intensive care. Such variability can mean that whether or not patients are able to access admission to intensive care, or (conversely) the option of palliative care, depends on which physician happens to be on duty. ^{Footnote 34} However, the patient or the patient’s family will often be completely unaware that such variation exists, and may not share with the physician the values that have led to the decision. This variation appears arbitrary and unjust. If clinicians know that their peers would offer a particular treatment option, they should arguably also offer that treatment option, or at the very least make the patient/surrogate aware of the different views that exist. ^{Footnote 35}

Moral Uncertainty

Third, CNO is justified on the grounds of moral uncertainty. When we make a moral judgment that a particular course of action would be right (or wrong), we should take into account the possibility of error. ^{Footnote 36} Differing views in the cases described highlight that moral uncertainty exists. For individual clinicians, the fact that their assessments of the ethics differ both from society’s (as evidenced by this course being lawful) and from those of peers (who are prepared to provide a particular service) means that they should take seriously the possibility that their own view is mistaken. What they should do given this uncertainty is more difficult. There are different types and levels of moral uncertainty, as well as different theoretical approaches to dealing with such uncertainty. It is difficult to know how to evaluate the probability of error, or how much weight to give to different possible outcomes (assuming that a particular moral judgment is correct). However, plausibly in the case of genuine moral uncertainty, medical decisions should be guided by the values of the patient, not by the values of the provider. Clinicians should be prepared to set aside their own moral assessment and not object. ^{Footnote 37}

Incoherence

One potential concern is that CNO is incoherent or contradictory. I have argued that an individual clinician ought to take a course of action that they sincerely believe they ought not to take. A critic might ask whether it makes any sense for a clinician in the case example to believe simultaneously that they should and should not provide treatment to Mr. Golobchuk.

There are several potential responses to this concern. First, the nature of any true moral dilemma is that there is a conflict between moral requirements. The existence of contradictory norms is not surprising in these cases. Second, one way of characterizing the two requirements is that individuals have a pro tanto reason not to act in a certain way (for example, not to provide ostensibly futile treatment), whereas they have an all-things-considered reason to acquiesce and provide the requested treatment. Does this avoid incoherence? It may be that the individual clinician strongly believes that he or she has an all-things-considered reason not to withdraw treatment or to provide futile treatment. In that case, we could see the argument in favor of CNO as providing something that the professional should consider seriously prior to formally objecting. It provides some additional considerations that the clinician may not have factored in to an all-things-considered judgment.

Moral Distress

I noted at the start of this article that one consequence of the difficult decisions encountered in intensive care is the high rate of moral distress. It might be thought that CO provides a mechanism for alleviating moral distress (individual providers may choose not to participate in decisions that they would find distressing), whereas conversely, CNO would potentially increase distress. Should the medical profession therefore be more encouraging of objection?

Whether a particular policy or norm would lead to more or less distress among professionals is an empirical question. There are no existing data (to my knowledge) to assess whether promoting CNO would lead to more moral distress among staff. If so, there would then be an ethical question to answer: Whether this distress outweighed the reasons in favor of CNO. However, it is not necessarily the case that endorsing or promoting CNO would increase distress, and it is possible that it would reduce it. Moral distress is often said to reflect a sense of disequilibrium resulting from being unable to take a course of action that one recognizes as ethically appropriate. ^{Footnote 42} The arguments in favor of non-objection encourage professionals to be cognizant of their peers, of the variation in views that are held, and of the potential moral uncertainty at stake in such assessments. This would provide them with positive reasons to believe that they are doing the right thing in not objecting, and potentially help relieve their sense of discomfort.

Religious Objection

The examples I gave in this article were based on secular objections to provision or limitation of treatment in intensive care. Some may feel, however, that this focuses on easier cases of non-objection; religious cases of CO potentially pose more of a challenge. When a professional has an in-principle objection to a particular practice on the basis of core religious beliefs, it may be harder to imagine that that person will be swayed by arguments such as those outlined previously.

For example, the argument from autonomy will not necessarily convince a health professional who believes that the act in question is subject to an absolute religious prohibition. Likewise, the argument from moral uncertainty relies on giving some credence to the possibility that alternative moral assessments are possibly true. However, at least some religious adherents will believe strongly that their particular moral judgment is objectively morally correct, and that any alternative viewpoint is mistaken.

However, the arguments for CNO arguably still apply to religiously motivated concerns about provision or non-provision of treatment. The value of patient autonomy does not depend on the professional agreeing with the patient. In fact, the contrary is the case: Professionals’ support for patient autonomy only counts when they respect a patient’s choice despite regarding it as the wrong choice. In particular, patients have an absolute right to refuse treatment. It would be unreasonable to impose treatment on the patient that the patient does not want, on the basis of religious values that the patient does not share. Variation among clinicians in the options that they would offer to patients does not become justified when physicians have varying religious beliefs. On the contrary, diversity of religious beliefs ought to lead clinicians to be cautious about the way in which their personal values are influencing the professional judgments that they are making.

The question of moral uncertainty raises the issue of the epistemology of disagreement in the face of religious diversity. ^{Footnote 43} There are different views about how a religious adherent should respond to evidence of interfaith and intrafaith differences in normative judgments. Some have argued that humility in the face of religious diversity should lead to reduced confidence in one’s own belief and tolerance of the views of others. ^{Footnote 44} Others maintain that the lack of common ground for resolving disputes means that it is not irrational for individuals to remain committed to the truth of their religious beliefs to the exclusion of other beliefs. ^{Footnote 45} It is beyond the scope of this article to resolve those questions. However, separate from those epistemic questions is the issue of religious pluralism in democratic societies. Most democracies without state religions have explicit commitments to tolerance of diverse religious views. In societies such as the United States, Australia, and the United Kingdom, there is strong political and sometimes constitutional support for freedom of religion. Although such freedom may well provide a justification for permitting religious-based CO, it would also support the idea that those acting in public office, or in public service (such as healthcare) should not impose their personal religious views on those whom they are serving. It thus provides an additional argument in support of CNO.

Resource Limitations

One potential argument against CNO relates specifically to controversial requests for continued treatment, and is highlighted by the case of Samuel Golobchuk. In cases such as this, continued treatment may consume limited healthcare resources and may harm other patients (because they are unable to access treatment). ^{Footnote 46} If a clinician’s refusal is based on this sort of concern, should objection (rather than non-objection) be encouraged?

A full discussion of the role of resources in CO is beyond the scope of this article. Finite resources might provide a limit to medical decisions either in concordance with, or in opposition to, patients’ wishes. They may certainly provide a reason to decline a patient’s request for continued treatment.

However, resource-based treatment refusals ought to be based on a fair process that is transparent, consistent, based on relevant reasons, and, ideally, based on clearly articulated policy. ^{Footnote 47} They should not to be based on the conscience and values of the individual clinician – rather on the values and deliberation of society.

Unjust Norms

I have argued that a physician ought to provide a medical treatment option that is professionally and legally endorsed despite that physician’s personal moral beliefs that this would be wrong. However, there may be conceivable situations in which a physician has reason to question whether the relevant ethical and legal norms are themselves just. There are plenty of historical examples in which particular medical practices were generally endorsed, but which, in retrospect, were highly ethically problematic.

We could imagine, for example, a country that has strict religious or cultural norms against withdrawal of treatment, and against the wishes of a patient such as Miss B who strongly desires that treatment not be provided. Should the physician conscientiously object to the norm against treatment withdrawal? Alternatively, it may be that a particular health system has institutional policies that discriminate against patients with a particular illness or disability, and would deny such patients access to intensive care. In these cases, should a physician listen to his or her conscience and admit the patient to intensive care or decline admission?

It is important to note here that if withdrawal of treatment (in the first case) or admission to intensive care (in the second) is performed at the request, or for the benefit, of the patient, neither of these circumstances would fit with the definition of CNO that I proposed. The patient’s wishes are contrary to the norms of society. The arguments I outlined from autonomy, justice, and moral uncertainty would support objection rather than non-objection in such cases.

In other circumstances, in which a patient’s wishes are consonant with professional and legal norms that the physician believes are morally wrong or unjust, the ethically appropriate course is not to object to the treatment requested, but rather to campaign publicly for a change in those norms. ^{Footnote 48}