Deciding the Criteria Is Not Enough: Moral Issues to Consider for a Fair Allocation of Scarce ICU Resources

Abstract

During the first wave of the COVID-19 pandemic in Italy, practitioners had to make tragic decisions regarding the allocation of scarce resources in the ICU. The Italian debate has paid a lot of attention to identifying the specific regulatory criteria for the allocation of resources in the ICU; in this paper, however, we argue that deciding such criteria is not enough for the implementation of fair and transparent allocative decisions. In this respect, we discuss three ethical issues: (a) in the Italian context, the treating physician, rather than a separate committee, was generally the one responsible for the allocation decision; (b) although many allocative guidelines have supported moral equivalence between withholding and withdrawing treatments, some health professionals have continued to consider it a morally problematic aspect; and (c) the health workers who have had to make the aforementioned decisions or even only worked in ICU during the pandemic often experienced moral distress. We conclude by arguing that, even if these problems are not directly related to the above-mentioned issues of distributive justice, they can nevertheless directly affect the quality and ethics of the implementation of allocative criteria, regardless of those chosen.

1. Introduction

During the first wave of the COVID-19 pandemic in Italy, hospitals had to accommodate a large number of COVID-19 patients, who required highly specialized medical assistance: many patients affected by the SARS-CoV-2 virus needed ventilatory support, as they suffered from acute hypoxemia due to the occurrence of interstitial pneumonia. Although interstitial pneumonia is often potentially reversible, its acute course can last several days, and ventilatory support may be needed for weeks [1]. The care of patients with COVID-19 interstitial pneumonia in Italy required a large number of ventilators, specialized healthcare personnel, and intensive care beds. However, the availability of these resources was not always guaranteed, especially with regard to invasive mechanical ventilation. Because of these reasons, practitioners had to make tragic decisions regarding the allocation of scarce resources in the ICU. The scarcity of ICU resources was not the only scarcity faced during the pandemic [2]; it was the mainly debated topic since ICU treatments in this context were lifesaving. Many efforts were focused on identifying ethically justifiable criteria to answer the following question: provided that the 32nd article of the Italian Constitution states the same right to health for all Italian citizens1, who should be treated when not all can be treated at the same time? In the COVID-19 pandemic, such a question can be reformulated in James Childress’ words: “Who shall live when not all can live?” [4].

The two main documents drawn up during the pandemic addressing these issues were the “Clinical Ethics Recommendations for admission to intensive treatments and their suspension, in exceptional conditions of imbalance between needs and available resources drawn up by the Italian Society” of Anesthesia Analgesia Resuscitation and Intensive Care [5]2 and “COVID-19: clinical decision-making in conditions of Resource Shortage and the “pandemic emergency triage” criterion drawn up by the Italian National Committee for Bioethics [7]3. These documents led to a passionate debate, for example, on the role of age in allocation decisions [9,10,11] and the implementation of purely clinical or even extra-clinical criteria [12,13]. This debate was genuinely an ethical one: with this regard, in this Special Issue, Stefano Semplici provides an in-depth ethical reflection arguing that, in these situations, it is not science that can have the last word, but allocative decisions remain a matter of ethics and politics [14].

We agree that ethical debate and reflection are fundamental to identifying inspiring principles and operational criteria that strengthen the “ethical preparedness” of the healthcare system, that is, the set of all the actions aimed at guaranteeing the ethical justifiability of the measures implemented by the system to tackle the pandemic [8]; however, in this paper, we investigate three other moral issues emerging from both the Italian context and ICU clinical practice, which have been observed by one of the authors who worked as an ethics consultant in one of the provinces of northern Italy that was most affected by the pandemic4. We argue that, although these issues are not directly related to the topic of distributive justice, they directly affect the quality and ethics of the implementation of allocative criteria, regardless of those chosen.

In the next sections, we present and discuss the following topics: (a) in the Italian context, the treating physician, rather than a separate committee, was generally the one responsible for the allocation decision; (b) although many allocative guidelines have supported moral equivalence between withholding and withdrawing treatments, some health professionals have been considering it a morally problematic aspect; and (c) many health workers accountable for these decisions have been experiencing moral distress.

These issues have the potential to undermine the proper application of allocative criteria in emergencies. For this reason, we conclude that we need to deepen the reflection regarding these issues not only in the academic debate but also within clinical practice; the scope is not just to avoid negative consequences impacting the well-being of the patient and the healthcare team but to also allow for the fair and transparent implementation of the allocative criteria.

2. Who Makes the Decision and What Are the Consequences of Distributive Justice?

In the Italian debate, little room has been given to the question of who should practically decide which patient should be given priority access to the ICU [16]. There seems to be a rather general and implicit consensus that—as in everyday medical practice—the therapeutic decision to admit a COVID-19 patient to the ICU requires a medical indication formulated according to clinical guidelines and good clinical practice; consequently, the physician should remain responsible for the assessment of COVID-19 patients and for the subsequent decision on prioritization. This is what actually happened in Italy, where there was no legally binding implementation of allocative criteria, and, moreover, the treating physician was the one who had to make the allocative decision in practice.

To be fair, it should be noted that both SIAARTI and INCB stated that the responsibility for making the choice must take place with a view to sharing responsibilities. On the one hand, according to INCB, “[i]t is important that the therapeutic decision concerning the different patients to be treated, according to the severity of their pathology, is as far as possible the result of consultation between several doctors […], to allow the sharing of the responsibility and burden of a decision” [7] (p. 8). On the other hand, the SIAARTI recommendations present “relieving clinicians of certain responsibilities in their choices” as one of the general purposes of the document [5] (p. 3). Moreover, SIAARTI made more explicit criteria than INCB for allocating decisions [8], and some understood them as a way to overcome the problem of who has to make allocative decisions [17].

However, both documents rest on the idea that physicians should play a central role in the allocative decision. As a matter of fact, in both the SIAARTI and INCB documents, there is no mention of the possibility of relieving clinicians of the burden of such a choice by delegating it to an external triage committee, as many documents in other countries do5. In the Italian academic and public debate, this has remained an unexamined question6. The external committee or the triage officer should not provide direct patient care and should only focus on the rationing decision based on the neutral application of the objective criteria suggested by guidelines [19]. For instance, according to the N.Y. State Task Force on Life and the Law “[t]o ensure that patients receive the best care possible in a pandemic, a patient’s attending physician does not determine whether his/her patient receives (or continues) with ventilator therapy; instead a triage officer or triage committee makes the decision” [20] (p. 5). Also in this line, during the COVID-19 pandemic, the California Department of Public Health states that “a group of triage officers should be appointed” [21]. Here, we argue that, regardless of the criteria suggested or required by the documents or guidelines, the choice of who should make allocative decisions—and, consequently, the mandatory or non-mandatory nature of the guidelines—is not morally neutral and can have consequences on the fairness of the resource allocation system, theoretically ensured by the allocative criteria within the guidelines. From this perspective, two documents can propose the same allocation criteria, but, depending on who makes the choice and on the binding nature of the guidelines, they can have different implementations.

Delegating allocative choices to an external triage committee may have several advantages. For instance, implementing an external triage may relieve the clinicians of the burden of the decision, having possible positive consequences on the psychological status of doctors and, therefore, increasing or, at least, not undermining the quality of healthcare during a pandemic. However, more importantly, from the point of view that we discuss in this paper, such a strategy may ensure greater homogeneity and consistency of allocating decisions within the healthcare system and, therefore, better control over the decided allocative criteria. In fact, it would be easier to respect the generally accepted formal equality principle that Aristotle articulated referring to Plato, namely, “treat like cases as like” [22]. Delegating the allocative decision to the physician’s conscience—especially when, in an emergency context, extra-clinical criteria may be necessary—could lead to the arbitrariness of the single physician precisely in the identification and adoption of extra-clinical criteria, inaugurating scenarios in which similar cases are treated in a dissimilar way or in which the criteria used are not explained or adequately justified.

This leads us to point out another aspect: implementing an external triage may also be useful to protect the “publicity requirement”, namely, one of the “formal constraints of the concept of right” proposed by John Rawls [23] (p. 115), which exists in every moral theory that addresses justice [24]. According to Rawls, the publicity requirement, a condition implicit in Kant’s categorical imperative, requires that a principle of justice needs to be publicly recognizable as one of the fundamental rules of society. Such a constraint arises spontaneously from a contractarian conception of society: the parties, called to decide the fundamental terms of their association in an initial position of equality, assume that everyone knows about the principles of justice and that they are the result of a public agreement. In order for society to impose “just” rules on individuals, such rules should be at least public, transparent, and known. Applying this consideration to the allocative issues, it seems implausible to argue that the parties choose to submit to a non-transparent allocative process, such as that deriving from leaving the allocative decision to the doctor. Moreover, the transparency and publicity of the allocative process may have desirable effects and support the stability of social cooperation or, at least, avoid an excessive distrust of science and the healthcare system during emergencies. In this respect, quite curiously, promoting transparency is one of the purposes of both the SIAARTI and INCB documents. However, as we tried to show, it may fail to be respected if we want to let the physician make allocative choices.

A final and related possible advantage of the external triage committee strategy could be that it may contribute to involving other helpful experts in the decisions due to the fact that physicians are not generally trained for making such tragic choices. Delegating these decisions to the doctor could lead to scenarios in which the allocative criteria are not adequately justified or that the choices are then delegated to more experienced physicians who, by virtue of their power relationship and experience, may influence such decisions without a title to do so. It is clear that, if the external committee was composed only of doctors not prepared to face such dilemmas, the same criticism could also emerge for the strategy of the external committee. However, it can be argued that an external committee might be better equipped to address the aforementioned problems.

It is clear that the external committee strategy has significant implications, as well as problems that need to be addressed. For example, one of the issues concerns whether the committee refers only to a hospital, whether it refers to several structures, or whether it refers to a more or less vast territorial area. Moreover, some may argue that the implementation of external committees may undermine the patient–physician relationship. From this perspective, the patient and their family would not feel taken care of by the doctor who abdicates the responsibility of care [25]. To be carefully considered, this argument should be supported by qualitative and quantitative evidence: clearly, the burden of proof is up to the proponents. An argument against this hypothesis may be that the doctor–patient relationship is traditionally a dual one that often leaves no room for other considerations from outside. The fact that it is the doctor who makes the decision considering not only the patient’s interest but also the interest of the healthcare system itself could, in turn, be detrimental to the relationship.

This is not the best place to answer “all things considered” to the question of who should make the allocative decisions. From this perspective, further empirical research would be useful to compare healthcare systems in order to understand how allocative issues were addressed where the external committee was and where they were not. However, in light of the arguments proposed, we limit ourselves to observing that the choice of such a considerable system affects the quality of the implementation of the allocative policy.

3. The Distinction between Withdrawing (WD) and Withholding (WH) Medical Treatments and Its Implications for a Fair Allocative Distribution

The next ethical issue comes directly from clinical practice. We observed that, during the COVID-19 pandemic, several practitioners considered withdrawing (WD) and withholding (WH) treatments for morally different reasons, with a higher moral weight attributed to the former. Our anecdotical evidence is in line with a well-known tendency in clinical practice, which does not only occur during an emergency: in a US study, it was found that US physicians find withdrawing life-sustaining treatment more psychologically difficult than withholding it (61%) and that withdrawing life-sustaining treatment is typically more ethically problematic (59%) [26]. The debate about equivalence and nonequivalence matters because there is a disconnect between ethical theory and the views of health professionals: while many doctors and nurses observe an ethical difference between WH and WD, most professional guidelines endorse the equivalence thesis [27], namely, the thesis according to which there is no prima facie moral difference between WH and WD [28]. This perspective has received considerable support from philosophers and bioethicists, and it can be defended from different perspectives. As a consequence, some argue that physicians who do not accept the equivalence thesis are, to some extent, psychologically biased [29].

Arguing in favor of the moral distinction may be relevant in cases where patients decide to refuse already-implemented lifesaving treatments and the physician does not intend to withdraw it or when a patient has treatment unjustly withheld because doctors know (or think) that, once they have started treatment, it will be very difficult to stop. Physicians who support the moral distinction often argue that withdrawing treatment is equivalent to causing death and, therefore, killing the patient—which, in several ethical traditions, is almost always considered morally wrong—whereas withholding the same treatment is more like letting the patient die, which may, in some circumstances, be permissible. Here, we do not provide arguments against this view or in favor of the equivalence thesis, which has been extensively debated in the aforementioned literature. We limit ourselves to arguing that, in this context, physicians supporting the moral difference between WH and WD may unduly frustrate the autonomy of the patients who want to refuse medical treatments after an informed and rational reflection of their condition or even give rise to medical futility and aggressive treatment cases, which may be against the principle of non-maleficence.

Nevertheless, distinguishing WH and WD from a moral point of view may also pose some problems with the fair allocation of scarce ICU resources. Consider the following example: an elderly patient arrives in the ICU with little chance of surviving but still a chance sufficient to consider that treatment proportionate to her; at that time, a bed in the ICU is still available. ICU doctors agree that it would seem problematic to argue that the elderly should not be treated. After a few hours, a younger patient who needs ICU treatment is taken to the hospital: she is more likely to survive and, therefore, to benefit from the treatment. However, the doctors truly believe that the ICU treatment, which is still proportionate for the old patient, should not be withdrawn; they feel that this is more ethically preferable than withholding treatment from the younger patient. The doctors’ beliefs reported in this example are in line with a survey conducted in 2006, where only one-fifth of surveyed Italian intensivists were willing to withdraw treatment from a patient with a lower probability of survival for other patients needing admission [30].

Generally speaking, resource allocation considerations appear to play a much smaller role in decisions about treatment withdrawal in intensive care [27]. However, embracing the moral distinction between WD and WH in emergency times may have negative consequences on the fairness of our allocative system. It forces us to accept the “first come first served” criterion, which is problematic from an allocative point of view during crises [31]. Such a criterion seems to be in contrast with the “medical need” criterion, according to which those with higher hopes of benefiting from the treatment should be treated. In in times of crises, medical criterion may also be translated into saving as many lives as possible. This criterion is accepted as a priority by many guidelines and documents, including the Italian ones already mentioned, and it can be defended from both a utilitarian and a contractarian perspective [8].

Furthermore, by accepting the first come first served criterion, we would unduly prioritize people who arrived at the hospital earlier because they were more advantaged, e.g., they live closer to the hospital or they are more informed about the course of the disease, regardless of other relevant considerations, such as the odds of survival [32]. Of course, a minority of scholars consider such a distinction ethically defensible because it respects the intuitions of health professionals and the understanding of the nuance and context of medical decision making [33]. However, allowing physicians to follow their intuition may be extremely costly in terms of fairness for the healthcare system.

To sum up, to promote a fair distribution of scarce resources, we should also consider problems regarding practitioners’ perceptions and attitudes toward withholding and withdrawing treatments in the ICU during crises. Therefore, it is necessary to promote ethical reflection among doctors and nurses to help them manage their own moral intuitions in clinical practice; enhancing ethical reflection among practitioners is one of the core purposes of the Ethics Consultation Service (ECS), which is, an advisory service that is designed to assist patients, families, and all healthcare professionals in identifying, analyzing, and resolving ethical dilemmas. ECS is particularly widespread in the US: in 2018, ECSs were found in 76% of all general hospitals and in 96% of hospitals with >100 beds [34]. It is not the task of this paper to discuss the ECS, but, here, we only claim that promoting ECSs in the Italian context may also be positive for guaranteeing fair allocation in pandemic contexts [25]. This recommendation is even more evident and urgent when we consider the broader issue of moral distress.

4. Moral Distress and Fair Resources Allocation

During the COVID-19 emergency, several healthcare professionals, particularly those involved in the ICU, experienced moral distress [35]7. Although this notion is mainly used in reference to nursing ethics [37], here, we refer more generally to healthcare professionals. According to the most famous account provided by Andrew Jameton, moral distress is the emotional state that arises “when one knows the right thing to do, but […] constraints make it nearly impossible to pursue the right course of action” [38]. The symptoms of moral distress are anxiety, depression, peritraumatic dissociation, and burnout, which were experienced by ICU healthcare workers during the pandemic [35]. With regard to “constrains”, Elizabeth Eppstein and Sarah Delgrado argue that they can be internal or external [39]. When we discuss internal constraints, we refer to the fear of losing one’s job, self-doubt, anxiety about creating conflict, or a lack of confidence [40]. External constraints, instead, include power imbalances between members of the healthcare team, poor communication between team members, pressure to reduce costs, a fear of legal action, a lack of administrative support, and hospital policies that conflict with patient care needs [38]. The lack of resources clearly falls within the external constraints category. Pneumologist Annachiara Facchini, bringing her experience as a frontline doctor during the COVID-19 pandemic, certifies moral distress with the following claim: “we knew very well what it was right to do [namely, curing everyone], but we couldn’t do it” [41].

However, such a definition of moral distress is not enough to grasp the complexity of this phenomenon in allocative decisions. In fact, we can observe a further level of moral distress, which leads us to broaden the initial definition provided by Jameton. Granted that, in the emergency context, healthcare workers want to treat everyone, in light of the impossibility to do so, a decision about who should be treated is still possible. The possibility of such a decision gives rise to a new form of moral distress. Healthcare workers may, in fact, have moral doubts about such decisions, disagree with each other, or disagree with suggested or imposed guidelines. Based on this, we can introduce a second and broader definition of moral distress, according to which moral dilemmas are also encompassed, and it is “negative stress symptoms that occur due to situations that involve an ethical dimension and where the healthcare provider feels she/he is not able to preserve all interests at stake” [42]. In this way, when we discuss moral distress, not only are we able to refer to the impossibility of making what healthcare workers consider to be the right choice (first-level moral distress), but we are also able to refer to the impossibility of understanding which is the best possible choice (second-level moral distress).

What are the consequences of experiencing moral distress during a pandemic? Individuals are guided by their moral values to make decisions in morally challenging situations; moral distress seems to affect the moral compass and the identity of individuals, which might lead to a decreased capacity to follow and act upon personally held moral values. Such conditions arise in allocative scenarios, particularly in contexts where the general policies outlined—if any—are difficult to apply in practice, providing limited guidance for specific situations in choosing the “best solution” for whom to treat and whom to leave [43]. Therefore, moral distress during a pandemic can compromise not only the ability of healthcare workers to cure and take care due to the symptoms mentioned above but also their capacity to comply with the allocative criteria and guidelines. Again, this could lead to treating similar cases differently, compromising the aforementioned formal equality principle. Therefore, although moral distress is mainly addressed through the lens of the well-being of healthcare personnel, here, we should recognize that, in pandemic emergencies, there are also further reasons based on the fair allocation of ICU resources to tackle it. To avoid the negative consequences of moral distress with respect to allocative decisions, it is therefore important to face the latter through psychological and ethical support, which should be part of the ethical preparedness of a healthcare system. Preventing moral distress may help reduce the likelihood of misapplications of the proposed allocative criteria. Moreover, in this case, these arguments do not depend on the choice of the criteria per se, but on their implementation in complex situations such as emergencies8.

5. Conclusions

In this paper, we firstly underlined that the Italian debate on the rationing of ICU resources has focused on individuating ethically justifiable criteria for making allocative decisions. Nonetheless, we argued that a fair allocation decision is not only guaranteed by the identification of criteria, but it also depends on several other factors that should be carefully considered. With this regard, we discussed three ethical issues arising from clinical practice, namely, (a) who should decide which patient to bring to ICU, (b) the moral difference between WH and WD, and (c) moral suffering during the pandemic. We argued that these issues, if not properly addressed, can undermine the quality of a fair allocation decision regardless of the allocative criteria decided. In light of this, we claim that there is a need to deepen the reflection regarding the allocation issues not only from a policy and theoretical point of view but also from a clinical practice point of view. In this way, we may be able to avoid negative consequences impacting the well-being of the patient and the healthcare team, and allow for the fair and transparent implementation of the allocative criteria.