Abstract
The provision of lifesaving treatment subject to severely limited resources can lead to serious moral dilemmas and legal challenges on many levels. The issue becomes particularly acute in a crisis such as the current Covid-19 pandemic. In this context, legal questions arise regarding withdrawal of treatment, particularly withdrawal of ventilation. Israeli law prohibits withdrawal of ventilation, a situation that affects the prioritization criteria for patients who can benefit from ventilation. This issue is discussed in the context of the Covid-19 pandemic where the legal prohibition may lead to a situation in which patients will not have access to ventilators. In addition, a theoretical framework is proposed that includes not only beneficence, nonmaleficence, and distributive justice but also a relational approach to autonomy and justice. We also propose potential solutions under the current legal framework. Based on the legal analysis, it is proposed that Israeli law be amended so that more patients will have access to ventilation during a pandemic. It is argued that clinicians cannot act beneficently to prolong life if they ignore patients’ fundamental rights or the distributional effects of ethical policies. Autonomy and justice and their accompanying liberal and relational perspectives should remain relevant even during a pandemic, a time when resources are scarce.
Keywords
Introduction
The provision of lifesaving treatment subject to a severe limitation of resources creates serious challenges on many levels. The issue becomes particularly acute in the context of a pandemic such as the Coronavirus. 1 High morbidity and mortality rates demand optimal availability of both medical equipment (ventilators) and trained intensive care unit (ICU) medical staff. During the outbreak of the Coronavirus pandemic, there were reports in some countries that the number of patients in need of ventilation far exceeded the number of available ventilators. 2
There was fear of such a scenario occurring in Israel. The first wave of the Coronavirus pandemic appeared in Israel in March 2020. At the end of that month, it reached a peak of 728 new confirmed cases per day. 3 By mid-May, the government managed to gain control over the spread. 4 However, starting from July 2020, Israel has been experiencing a second wave of the pandemic, in which between 1600 and 2000 new confirmed cases are reported every day. 5 The number of seriously ill and ventilated patients has also risen. As of the end of March 2020, the number of seriously ill patients was 98 and the number of ventilated patients was 64, while in early August 2020, the numbers had risen to 355 and 100, respectively. 6 Although the number of ventilated patients has not yet exceeded the number of available ventilators, 7 the experience in other countries indicates that such a scenario is not out of the question in Israel. In addition, the increase in the number of infected and quarantined medical staff may lead to a shortage of manpower in the health-care system. 8 In view of this state of affairs, both in Israel and other countries, the issue of allocating ventilators should not be ignored. 9
The potential shortage of ventilators and trained medical staff led Israel, as well as other countries, to issue ethical guidelines for the prioritization of access to ventilators. 10 Experts in philosophy, bioethics, law, and related disciplines have also contributed to the discourse on this issue. 11
Most of these guidelines embrace the bioethical principles of beneficence and distributional justice. In other words, they adhere to the utilitarian goal of saving as many years of life as possible. 12 While equal distribution of medical resources is often mentioned as a guiding principle, 13 the various versions of the guidelines have differed in how they evaluate the relevant considerations (such as personal medical history, cognitive state, short- and long-term prognosis, and age). 14 The leading bioethical principle of autonomy also receives some limited attention. 15
The extreme circumstances in which decisions need to be made subject to severely limited resources raise questions as to the influence of the law on provision of treatment. In particular, the issue of withdrawing lifesaving treatment arises when a decision must be made of whether to disconnect a patient from a ventilator when their condition deteriorates and the prospects of survival are low in favor of other patients with a better prognosis. This is a difficult moral dilemma facing physicians around the world during the Coronavirus pandemic. The issue may be less complicated when the law allows physicians to disconnect patients when ventilation becomes futile and merely prolongs life for a short period of time. However, if there is a legal prohibition against withdrawal of treatment, or if the patient’s consent is required, then the law constitutes a constraint for a physician wishing to save as many lives as possible.
In this article, we provide a critique of the legal prohibition against withdrawal of lifesaving treatment which highlights patients’ rights and distributive justice. Israeli law serves as the example for the analysis. It includes an inclusive and absolute statutory prohibition on disconnecting dying patients (who are defined as having 6 months or less to live) from ventilators, even when it is agreed that the treatment is futile. 16 The main argument presented in this article is that the withdrawal of ventilation is morally and legally justifiable when it accords either with the wishes of the patient (or those of the surrogate decision-maker) or when the wishes of the patient/family are unknown and ventilation is futile (i.e. the patient’s life will be prolonged for only a brief period).
The article proceeds as follows. The second and third sections set out the relevant legal and ethical positions in Israel. The fourth section presents the legal implications in the context of the Coronavirus pandemic and offers a theoretical framework for discussion which we believe should include not only beneficence and distributive justice but also a relational approach to autonomy and justice. In the fifth section, we provide potential solutions under the current legal configuration. In the sixth section, we discuss the conclusions that emerge from the experience with the Coronavirus pandemic.
The legal framework
In Israel, there are two Parliamentary Acts that deal with the provision of medical treatment. The first is the Patient Rights Act 1996 (hereafter: the 1996 Act) 17 and the second is the Dying Patient Act 2005 (hereafter: the 2005 Act). 18 The 1996 Act applies to patients with Covid-19 who are not dying at the time of diagnosis, while the 2005 Act applies to patients who are. The prohibition to disconnect patients from ventilators, which is the subject of this article, appears only in the 2005 Act. As will be discussed below, this situation is the source for several legal difficulties.
Provision of treatment
The 1996 Act determines a patient’s fundamental rights in the context of medical treatment. 19 It holds that any individual has the right to receive medical treatment, subject to the basket of health services provided by the public health-care system. 20 During normal times, this creates no difficulty. However, in the circumstances of a pandemic, public health-care systems may not be able to deliver the basket of services to all, according to their mandate. For example, while in normal times patients are connected to ventilators based on need, during a pandemic this may not be possible for all patients if there is a shortage of ventilators.
In this context, the 1996 Act does not address the issue of a physician’s discretion in the decision of whether to withdraw lifesaving or to withdraw life-prolonging treatment. 21 Thus, it does not provide any guidance in the case that the continuation of ventilation is viewed as futile by the physician and there are other patients in need of ventilation.
The other relevant piece of legislation is the 2005 Act which regulates end-of-life decisions. 22 It applies only to patients who meet the definition of a dying patient, namely, a patient who is suffering from a terminal illness and is expected to die within 6 months, irrespective of whether they receive treatment. 23
Prior to the 2005 Act, there were several attempts made to regulate end-of-life decision-making by means of designated legislation, but none of them was successful. In the absence of a specific piece of legislation, this domain was regulated by general legal rules and they still apply today. For example, hastening the death of a dying patient or assisting his/her death is prohibited in Israel according to the penal code. 24 Furthermore, a patient has—to a considerable degree—the right to refuse treatment. 25 Finally, the Basic Law: Human Dignity and Liberty 1992, which has gained constitutional status over the years, guarantees the right to life, human dignity, and autonomy. 26 Although these general legal rules are relevant to end-of-life decision-making, they do not provide specific guidelines to physicians in the case of end-of-life patients. 27
Consequently, the task of establishing specific rules was left to the courts, whose rulings made it clear that active euthanasia and physician-assisted suicide are prohibited. 28 However, the district courts upheld dying patients’ explicit requests not to receive artificial ventilation and to terminate it if it has already been started, provided the patient is in the final stage of a terminal illness. 29 The justification for these rulings was twofold: the patient’s right to autonomy and the perception of the withholding or withdrawal of life-prolonging treatment as passive euthanasia, namely, as an omission rather than an act. 30
Against this background, a national committee was appointed by the Israeli Ministry of Health in 2000 in order to propose legislation that would regulate the treatment of dying patients. Known as the Steinberg Committee, it included clergymen, various professionals (such as physicians, nurses, and social workers) and academics from various disciplines (law, philosophy, and the social sciences). 31 Despite the differing views of its members, the Committee’s aim was to reach a broad consensus on the main issues. 32 This was by and large achieved, though several members presented a minority opinion. The Committee’s final report was presented to the Minister of Health in 2002. A governmental bill was then drafted, 33 and in 2005, the Act was passed by the Israeli Parliament. 34
The drafters of the Act faced the difficult task of bridging between the two opposing approaches. 35 One supported a liberal position that emphasizes the dying patient’s right to make autonomous end-of-life decisions. The other was a religiously oriented approach, which emphasizes sanctity of life. 36 Interestingly, the views of the physicians, though cited in the Committee’s report, did not receive much attention in the Parliamentary discussion. In an effort to gain the approval of both sides, the drafters of the Act argued that it reflects a compromise between the values of the liberal tradition and those of Jewish law. 37
The 2005 Act states that it regulates the provision of treatment to dying patients based on the principles of sanctity of life, patient autonomy, and quality of life. 38 More specifically, the 2005 Act states that end-of-life decisions are to be based on the patient’s medical condition, his/her wishes, and his/her degree of suffering. 39 This is based on the assumption that people wish to continue living, unless it is proven otherwise beyond any reasonable doubt. 40 This assumption can be refuted in one of the ways prescribed by the Act, and primarily by means of the patient’s explicit statement to that effect, 41 or an advance directive that he signed before losing his/her mental capacity. 42
Once it is proven that the patient does not desire life-prolonging treatment, then that wish is to be respected. 43 This is one of the most important rules set in the 2005 Act. If the patient is legally competent, then reasonable efforts should be made to persuade him/her to accept treatments of the following types: oxygen, nutrition, fluids, routine treatments, and palliative care. 44 If the patient is legally incompetent, then the withholding of the aforementioned treatments is prohibited, notwithstanding his/her wishes. 45 Excluded from this rule are incompetent patients in the final stage (defined as having 2 weeks or less to live irrespective of medical treatment) 46 who are in considerable pain and wish not to be treated. 47
Despite the recognition of a dying patient’s right to refuse life-prolonging treatment, the 2005 Act makes clear that deliberate killing, assisted suicide, and any other action that intentionally and actively shortens a dying patient’s life are prohibited. 48 It specifically states that withdrawal of a continuous medical treatment that is likely to lead to death (such as disconnecting a dying patient from a ventilator) is prohibited, regardless of the patient’s wishes. 49 Nevertheless, if a continuous treatment is unintentionally or not unlawfully terminated (e.g. due to a power cut), it is legally permissible not to renew it, if this accords with the patient’s wishes. 50 The 2005 Act also permits the discontinuation of a cyclic life-sustaining treatment (such as hemodialysis) when that is the patient’s wish. 51 This rule also applies to treatments that by their nature are continuous but were planned in advance—using technological means—to be a cyclic treatment. 52 For example, installing a timer on a ventilator transforms a continuous treatment into a cyclic treatment that can be legally terminated.
In short, Israeli law prohibits the intentional withdrawal of ventilation from dying patients under any condition. In particular, the wish of the dying patient not to be ventilated, his/her mental and physical suffering, and the interests of third parties (including the interest of public health) do not play a role in this context.
In a situation of medical emergency, the 2005 Act states that a physician can provide a lifesaving treatment to a dying patient without the patient’s consent if it is impossible to obtain consent due to the emergency circumstances. 53 However, if the physician knows that the wish of the dying patient—as defined by the 2005 Act—is that lifesaving treatment be withheld under these circumstances, the physician should refrain from providing this treatment to the patient. 54 A different rule applies to a dying patient in the final stage (i.e. a patient who has 2 weeks or less to live). In this case, the physician can withhold lifesaving treatments, unless she/he knows that the patient has already expressed a wish to receive treatment in this situation. 55
Finally, the 2005 Act determines that if a dying patient who wishes that his/her life be prolonged asks to receive exceptional treatment, even if the physician believes it has no justification, his/her wish should be respected. However, this rule does not apply to medical treatments that are not expected to prolong the patient’s life or might cause him/her or another person considerable harm. 56
The status of the patient’s family
Both the 1996 Act and the 2005 Act hold that the decision-making authority, including in the end-of-life stage, belongs solely to the patient if she/he is legally competent. Family members cannot make decisions on behalf of the patient and the physician is not obligated to consult with them. 57
As for patients who are not legally competent, the 2005 Act states that a family member will have the authority to make end-of-life decisions on behalf of the patient if the patient—while still competent—signed a power-of-attorney which designates that family member as a formal surrogate decision-maker. 58 However, if the patient did not nominate a relative to have power-of-attorney while competent (nor did he leave any advance directive), then the 2005 Act states that the authority to make end-of-life decisions on behalf of the patient is granted to the treating physician. In making such decisions, the physician must consult with a close relative of the patient, who is defined by the 2005 Act as: (1) someone with a familial ties or with an emotional attachment to the patient and who is dedicated to him/her; and (2) someone closely familiar with the patient based on a continuous relationship during the provision of treatment or prior to it. Thus, while according to the definition a close relative is usually a family member, this is not necessarily the case. 59
For patients who do not fall within the 2005 Act, the Legal Capacity and Guardianship Act 1962 holds that people at any stage of life can provide a relative with lasting power of attorney which gives them the authority to make health decisions on individual’s behalf in the case they lose their legal capacity in the future. 60 If they did not nominate a relative with lasting power of attorney before losing their mental capacity, then the courts are authorized to appoint a legal guardian for the patient. 61 If there is no legal guardian or a lasting power of attorney, then the physician will make the decision by default, particularly if there is insufficient time to ask the court to appoint a legal guardian. 62
Transparency
Both the 1996 Act and the 2005 Act impose a duty on physicians to provide detailed information to legally competent patients 63 or to the patient’s surrogate decision-maker if the patient is legally incompetent to make decisions. 64 Thus, physicians must obtain informed consent prior to treatment.
When the patient is competent, then providing information to family members is permitted if explicit or implied consent was given by the patient. This is stated in both the 1996 Act and the 2005 Act. 65 When the patient is incompetent, the Legal Capacity and Guardianship Act states that the patient’s surrogate decision-maker can be provided with any information that is relevant to the decisions he is authorized to make. 66 When the patient is dying and incompetent, the 2005 Act imposes a duty on physicians to inform the surrogate decision-maker (i.e. a close relative or his/her legal guardian) that they intend to act according to the advance directives left by the dying patient or the power of attorney she/he provided. 67 The 2005 Act also states that physicians must provide them with medical information about the patient and allow them to review his/her medical records. 68
The applicability of the law to withdrawal of ventilation
As stated above, the 2005 Act applies only to patients who meet its definition of a dying patient. Moreover, the prohibition on withdrawing ventilation is found only in the 2005 Act, but not in the 1996 Act. Therefore, the question arises as to whether the prohibition also applies to other patients (who are terminally ill yet whose life expectancy exceeds 6 months, such as amyotrophic lateral sclerosis (ALS) patients and patients who can be cured if given effective treatment).
Scholars are divided on this question. On the one hand, there are those who adopt a formalist or conservative approach, in which they argue that there is a lacuna in the law with regard to patients who do not fall within the definition of a dying patient according to the 2005 Act. In other words, they argue that there is no legal rule that applies to these patients and therefore it is reasonable to assume that the prohibition on withdrawing ventilation applies to these patients as well. 69 This approach is based on an a fortiori argument: if it is not permitted to disconnect a dying patient from a ventilator even when that is their wish, then clearly it is not permitted to disconnect a patient if they can potentially benefit from ventilation (and wish to remain connected).
On the other hand, there are those who adopt a liberal approach, in which they argue that the law prior to the 2005 Act remains in force with regard to patients who do not fall within the 2005 Act. There are a number of reasons that can be used to justify this view. First, the 2005 Act does not apply to non-dying patients. Section 1 of the 2005 Act states that its aim is to regulate the provision of treatment to dying patients, namely, patients whose life expectancy does not exceed 6 months. Arguably, this means that the 2005 Act does not apply to patients who are expected to live longer (whether or not they suffer from a terminal illness such as ALS). Second, if the 2005 Act’s sole aim is to regulate the provision of treatment to dying patients, then all the conditions set out in the Act are relevant to dying patients only and do not apply to non-dying patients. More specifically, the prohibition against withdrawal of ventilation set out in section 21 of the 2005 Act does not apply to terminally ill patients whose life expectancy exceeds 6 months and to those who do not suffer from a terminal illness. Third, if the legislature had intended to apply the prohibition against withdrawal of ventilation in the case of non-dying patients, it would have drafted section 21 accordingly or alternatively it would have amended the 1996 Act. Fourth, the 2005 Act addresses “life-prolonging” treatment rather than lifesaving treatment. The implication is that the underlying assumption of the 2005 Act is an end-of-life situation. Therefore, it can be argued that the 2005 Act does not address the provision of lifesaving treatment to a patient who can potentially be cured or is in need of ventilation in order to overcome a temporary respiratory situation.
Therefore, if we accept the liberal perspective, which suggests that the 2005 Act’s prohibition to withdraw treatment does not apply to non-dying patients, is there a law that does apply in this case? As described in the previous section, prior to the 2005 Act, Israeli district courts permitted physicians to withdraw ventilation, subject to the following conditions: the patient is suffering from a terminal illness; she/he is in the end-of-life period; she/he has expressed a wish that his/her life not be prolonged; and there is a medical consensus that treatment is futile and can only prolong life. 70,71 Accordingly, it can be argued that patients who do not meet the 2005 Act’s definition of a dying patient have the right to request the withdrawal of life-prolonging treatment. However, this does not provide guidance to clinicians who from a professional standpoint believe that the patient should be disconnected from the ventilator, even if the patient wishes to remain on it. For example, according to this legal position, an ALS patient with a life expectancy of over 6 months can ask the court to respect his/her wish to terminate the ventilation but on the other hand a physician who believes that ventilation should be terminated has no legal authority to do so if the patient wishes to remain on the ventilator.
This issue of whether withdrawal of ventilation applies to all patients in Israel has yet to be discussed by the Israeli Supreme Court. However, it can be assumed that the first (more conservative) approach would be adopted, namely, the courts would not permit patients to be disconnected from ventilators whether or not they are near the end of life. This assumption is based on four factors: first, in a relatively recent decision made by an Israeli district court, the Attorney General (AG) adopted this approach and the court accepted it 72 ; second, the Halachic prohibition on shortening a patient’s life by the withdrawal of a continuous life-prolonging treatment, as described above, has substantial influence on the courts’ response to legal questions of this type 73 ; third, the recent position paper issued by the National Council for Bioethics addresses the provision of treatment during the Coronavirus pandemic and implicitly adopted this approach 74 ; and fourth, the medical staff would find it morally and emotionally difficult to withdraw a dying patient from a ventilator. 75
The ethical framework
In April 2020, a month after the outbreak of the Covid-19 pandemic in Israel, a position paper on the prioritization of access to lifesaving treatments was published by a Joint Committee of the National Council on Bioethics, the Israel Medical Association’s Chambers of Ethics, and the Ministry of Health. 76 Its purpose was to establish ethical guidelines that address legal, bioethical, and religious aspects of treating patients with Covid-19, as well as to provide guidance to the attending physicians. The position paper was intended to guide decision-making in the situation of a significant shortage of resources and skilled manpower. 77 Notably, although the position paper declares that its aim is to establish legal guidelines, it is not an enforceable legal document. As such, it cannot contradict existing legislation and can only be used as a source for interpreting existing legal rules when that is necessary.
The position paper’s principles and guidelines can be summarized as follows: The sanctity of life takes precedence over any other value and it is to be assumed that an individual wishes to continue living, unless it is proven otherwise beyond any reasonable doubt.
78
In the prioritization of medical treatments, all individuals are to be considered equal. Therefore, discriminating between patients based on religion, race, gender, nationality, country of origin, sexual orientation, socioeconomic status, social status, family status, civil status, occupation, age, and so on is prohibited.
79
This is essentially the adoption of distributional justice. The main guiding principle in prioritizing patients in need of scarce resources (such as ventilators) is to save as many as lives as possible, according to the principle of maximum benefit to the greatest number of individuals. Therefore, it is permissible and justifiable to deviate from the principle of equality only when the treatment’s chances of success and the patients’ prospects of survival differ across patients.
80
Efforts should made to ascertain the patient’s wishes regarding invasive treatments (such as ventilation). Moreover, a competent patient should be informed of the chances of success of invasive treatments and the possibility of alternative treatments, especially with regard to palliative care. Nevertheless, in the case of an extreme shortage of resources (such as ventilators), the treating physicians can decide not to connect a patient to a ventilator regardless of his/her wishes. In other words, a decision not to provide ventilation to patients can be made without their consent in this situation.
81
Thus, while the guidelines consider the principle of patient autonomy to be relevant, ultimately it expresses a preference for the utilitarian principle, which calls for the saving of as many lives as possible. Family members have no special status in the decision-making process, unless they have been given power of attorney or are the patient’s legal guardian. Moreover, although entitled to be informed of a prioritizing decision that withholds treatment from the patient they wish him/her to receive, family members cannot object to such a decision. However, if the patient did not complete an advance directive nor did she/he grant power of attorney and she/he is incompetent to make decisions, then family members can be consulted regarding the patient’s wishes.
82
Even in times of emergency and a shortage of medical resources, as a result of which lifesaving or life-prolonging treatment is withheld from a patient whatever his/her wishes are, palliative care must be provided to the patient according to the accepted standards in the profession.
83
The position paper’s reference to the issue of withdrawal of ventilation is particularly relevant to the discussion. While the committee discussed this issue at length, it did not present an explicit position. Thus, the paper states that: “This is a very sensitive and controversial issue in Israel and society has not managed to reach a consensus in this matter.” When the position paper was published, there was no shortage of medical resources in Israel, and apparently the Joint Committee did not feel it necessary to resolve this dilemma. It also stated that if such a situation does emerge, the guidelines for prioritization that appear in the position paper will avert the need to make a decision on this issue. It also stated that, if necessary, the Joint Committee would convene in order to decide on the matter.
Finally, and most importantly, the position paper stated that it would appear that disconnecting a patient from a ventilator is illegal under section 21 of the Dying Patient Act. 84 This implies that the Joint Committee holds the view that section 21 of the 2005 Act applies to patients with Covid-19 whether or not they fit the definition of a dying patient. Moreover, it suggests that the Joint Committee would not allow disconnecting patients from ventilators in the case of a shortage of resources. However, it also refrained from declaring that it is unambiguously prohibited according to Israeli law.
At this point, it is important to reiterate that the 2005 Act explicitly prohibits the withdrawal of ventilation. It is also worth mentioning that the Joint Committee’s position paper cannot overrule a prohibition determined by law (assuming, of course, that section 21 applies to all patients with Covid-19, whether or not they are dying). The priority of the 2005 Act over the view expressed in the position paper is also relevant in the case of other issues discussed in the position paper (e.g. informed consent). 85 Therefore, any change in the current legal situation regarding the prohibition against withdrawal of ventilation or any other condition specified in the 1996 Act or the 2005 Act requires an amendment approved by the Knesset. As will be argued below, an initiative to amend the legislation, and in particular the prohibition against withdrawal of ventilation, is unlikely to succeed.
To conclude, the Israeli position paper, which was drafted by, among others, senior medical lawyers, adopted the utilitarian mechanism of saving lives at the expense of respecting patient autonomy. It also refrains from expressing opposition to the legal prohibition against withdrawal of life-prolonging treatment. We thus argue that even under the extreme circumstances of severely limited resources, Israeli law is highly relevant in shaping policy. In the next section, this point will be demonstrated by means of the law’s impact on how patients are treated during the Coronavirus pandemic.
Legal and ethical implications for the treatment of patients with Covid-19
The legal and ethical position described above has several important implications for the treatment of patients with Covid-19 which are unique to Israeli law.
If it is assumed that the prohibition against withdrawal of ventilation applies to all patients with Covid-19 whether or not they are legally defined as a dying patient, then a physician will not be able to disconnect a patient with Covid-19 from a ventilator, even if it is clear that his/her life expectancy does not exceed a few days and even if a waiting patient with Covid-19 with better recovery odds would definitely benefit from ventilation.
If section 21 of the 2005 Act applies to all patients, then withdrawal of ventilation is prohibited even in the following situations: (1) the patient made clear in advance that when death becomes inevitable and imminent she/he wishes to be disconnected from the ventilator; (2) the patient’s relatives ask for—or consent to—the withdrawal of ventilation; (3) the patient’s wishes not to receive life-prolonging treatment at the end of life are unknown; or (4) ventilation is futile and is expected to cause the patient more harm than good in the short or long term. 86
The impact of the law is made clearer by the fact that a physician’s decision to ventilate a patient is not based solely on medical considerations. Our empirical research indicates that a physician’s decision to ventilate a dying patient is often the result of familial pressure to prolong the patient’s life. 87 Furthermore, in circumstances of medical emergency (such as when the patient arrives at the hospital with acute respiratory difficulties), and in the event that the patient’s wishes are unknown, physicians will in many cases choose to ventilate the patient. 88 This suggests that physicians prefer to prolong the patient’s life even when his/her prognosis is grim and she/he does not request it. Physicians find it emotionally difficult to simply let the patient die. 89
It follows that the legal prohibition against withdrawal of ventilation coupled with physicians’ reluctance to let patients die leads to the reality in which there are many patients in Israel who are ventilated for non-curative reasons and patients who continue to be ventilated notwithstanding the medical futility of the treatment and regardless of their wishes.
Therefore, in the context of treating patients with Covid-19, the legal prohibition against withdrawal of ventilation may prevent the saving of as many lives as possible. It does not allow for the cessation of futile ventilation in order to free up ventilator for patients with Covid-19 with good recovery prospects, even when ventilators are in short supply. 90 As such, the law does not accord with the ethical principle of beneficence (and nonmaleficence) and the utilitarian aim of saving as many lives as possible. 91 This claim requires elucidation.
The principle of beneficence requires physicians to act in the best interests of their patients. It is closely connected to the principle of nonmaleficence, which requires physicians not to harm patients, whether intentionally or as a result of negligence. 92 During a pandemic, these principles apply in the broader context of public health. They require that treatment decisions be guided not only by the resulting benefit to an individual patient but also by the benefit to society as a whole. 93 In the context of a shortage of ventilators during a pandemic, they justify the withdrawal of treatment if it is futile and its provision to other patients with good recovery prospects.
The same conclusion is reached based on a utilitarian perspective, 94 which in our context means saving the greatest possible number of lives. 95 This is a widely accepted principle during a public health crisis such as a pandemic. 96 Relying on a utilitarian approach when there is a shortage of ventilators means that when ventilation is no longer beneficial to the patient it is morally justified to disconnect him/her so that patients with a better prospect of recovery can be connected. 97
However, this raises the bioethical issue of acts and omissions and more specifically the issue of killing, which is considered an act, and letting die, which is considered an omission. 98 The question then becomes whether withdrawal of life-prolonging treatment, such as ventilation, should be viewed as an omission, that is, letting the patient die, which is morally permissible (because it does not violate the physician’s duty not to cause harm), or whether it is an act that kills the patient and hence should be viewed as morally—and legally—prohibited (because it violates the duty not to cause harm).
There are various approaches to the acts/omissions debate. There are those who argue that defining withdrawal of life-prolonging treatment as an omission is a moral fiction. They argue that one must admit that physicians are killing their patients when they switch off the ventilator and that this constitutes a violation of their duty not to cause harm to the patient. 99
Other scholars focus on causation in distinguishing between act and omission. Andrew McGee argues that unlike acts of euthanasia, such as injecting a lethal drug into a patient, withdrawing artificial ventilation is morally permissible because the main factor in the patient’s death is her underlying disease. 100 According to this view, by withdrawing treatment, the physician does not cause a death or hasten its occurrence but merely lifts a barrier that postpones the inevitable. Thus, the physician, by connecting the patient to a ventilator, prevents nature from taking its course. Ultimately, disconnecting the patient from a ventilator does not cause death; rather it simply changes its timing. 101
Other scholars reject this analysis and argue that when the patient depends on a ventilator to continue living, then switching it off causes the patient’s death. They argue that what explains the death of a patient who depends on artificial ventilation is not the terminal illness but the act of turning off the ventilator, which is the proximate cause of death. They add that medical technology has a double effect: It enables physicians to prolong life but also to cause death. 102 Others reply that technology provides the means to “hold death at bay,” a view that should lead to the conclusion that physicians do not violate their duty not to cause harm when they withdraw ventilation. 103
The discussion thus far has shifted the bioethical debate from causation to the physician’s moral responsibility. It demonstrates that the line between the withholding of treatment which is permitted and the act of withdrawing treatment is a fuzzy one. McGee, for example, argues that if the patient is ventilated manually rather than mechanically, then the physician’s decision not to reinflate the oxygen bag is defined—strictly speaking—as the withholding of treatment. Thus, the fact that the patient is connected to a machine that performs the ventilation does not essentially change the moral calculus. 104
The discussion of the physician’s moral responsibility is related to his/her moral duty to treat the patient and not to cause him/her harm. Arguably, a physician is morally responsible for the patient’s death (and thus causes the ultimate harm) only if she has an obligation to provide treatment or continue its provision and fails to do so. If the physician has no duty to provide effective medical care, then there is no causal link between her conduct and the death of her patient. In this context, the physician would not be morally responsible for the patient’s death regardless of whether she withholds or withdraws life-prolonging treatment, and she would not be violating her duty not to cause harm. Thus, the question is whether the physician has a moral duty to provide artificial ventilation when a patient is dying. The answer depends on how one views the principles of autonomy, sanctity of life, and quality of life. For Beauchamp and Childress, physicians in principle have no moral duty to provide futile treatment, that is, treatment that does not improve the patient’s health. 105 According to this approach, one might also accept that physicians have no duty to initiate ventilation (withholding of treatment) or continue its provision (withdrawal of treatment) if the treatment is futile, that is, when it will neither cure the illness nor prolong her life substantially.
In the context of a pandemic, the act–omission debate would imply that it is morally justifiable to withdraw ventilation from patients with Covid-19 so long as there is a medical consensus that it is futile. The arguments that support this conclusion in general (that the disease is the cause of death, that a decision not to manually ventilate the patient is permitted, etc.) are even more relevant in a pandemic.
Furthermore, in the context of a pandemic, a utilitarian approach may have the upper hand in the act–omission debate. If the aim is to avoid a situation in which the hospitals’ ICUs are overwhelmed by patients with Covid-19, then prohibiting the withdrawal of ventilation (whether or not it is perceived as an act of killing) cannot be morally justified.
Moreover, the prohibition against withdrawal of ventilation is also open to criticism from the perspective of justice. First, such a prohibition has unjust distributional effects on some segments of society, and in particular, the elderly and individuals with preexisting medical conditions. There are clear indications that these individuals face a significantly higher risk of developing a severe version of the COVID-19 illness. 106 Therefore, the probability that they will need ventilation if infected is presumably far greater than for other patients. Accordingly, a prohibition on withdrawal of ventilation is expected to disproportionately affect this population in the case of a shortage of ventilators. 107
The unjust impact of this rule is exacerbated based on the evidence of a positive relationship between poor health status and low socioeconomic status. Preexisting medical conditions that increase the risk for patients with Covid-19, such as obesity or smoking, are more prevalent among people from disadvantaged communities. 108 These are also individuals whose risk of infection is a priori higher, due to more crowded living conditions, a lack of information, and cultural norms, including religious practices and traditions. 109
Note that while a law which allows withdrawal of ventilation under certain circumstances is expected to primarily affect individuals who are at high risk of developing serious illness, this should not be considered as unjust from a distributional perspective. As discussed above, it is morally justifiable to disconnect a patient from a ventilator if the treatment is futile, notwithstanding his/her wish or that of his/her family. When this is the case withdrawal of ventilation is medically justified. Therefore, it should not be considered as having an unjust distributional effect. On the other hand, a law prohibiting the withdrawal of ventilation may deprive individuals who would benefit significantly from ventilation. Since individuals who are at high risk of developing a serious disease are also most likely to need ventilation and since these individuals often share the same characteristics (old age and preexisting medical conditions), such a law might be perceived as having unjust distributional effects.
A Rawlsian approach to justice can provide further grounds for criticism. Behind Rawls’ “veil of ignorance,” where no one knows his/her position in society, most individuals would decide to prohibit decisions that benefit one group at the expense of another, since they might end up being in the latter group. 110 In the context of life-prolonging treatment, it can be assumed that behind the veil of ignorance people would not desire a law that absolutely prohibits withdrawal of ventilation in a pandemic. By not knowing one’s personal risk of developing a severe disease that would require ventilation, individuals will prefer a law that permits withdrawal of ventilation when chances of recovery are slim and there is a shortage of ventilators.
The principle of autonomy also argues against the prohibition of withdrawal of ventilation since patient’s wishes are being ignored. As already mentioned, according to Israeli law, once a patient is ventilated, withdrawal from ventilation is prohibited, even if that is the patient’s wish. In these cases, the continuation of ventilation not only contradicts the principle of beneficence, it also compromises the patient’s right to autonomy, a fundamental principle in Western bioethics. 111 Moreover, the prohibition against withdrawal of ventilation indirectly compromises the right to autonomy of patients who would benefit from ventilation and wish to be ventilated. In the context of the Coronavirus pandemic, a patient with good chances to overcome his/her illness who wishes to receive treatment may not be able to if there is a shortage of ventilators.
Furthermore, another source of criticism emerges from a relational approach to patient autonomy. 112 Unlike the liberal-individualistic approach to autonomy, which emphasizes that individuals are self-directed human beings, relational autonomy stresses that an individual has intimate relationships with others 113 and that his/her decisions are shaped by those relationships and are their outcome. 114 According to this approach, and when the patient’s wishes are unknown, the clinician should consult with his/her relatives and should respect their decision since family members are bound together in a complex web of love and obligations, life projects, and tradition. 115
The relational approach to autonomy reveals the shortcomings of a rule that prohibits the withdrawal of ventilation. As noted above, this prohibition as set out in the Act applies regardless of whether the family has consented to or requested the termination of treatment. Therefore, a physician cannot disconnect a patient from a ventilator even if his/her family declares that the continuation of ventilation does not reflect the patient’s wishes, values, or interests. 116 For example, family members are likely to be well acquainted with the patient and may contend that had the patient been capable of making a decision and expressing his/her wishes, she/he would have asked that ventilation be withdrawn. Thus, although the family is best-positioned to know the patient’s preferences, values, and wishes, the clinician cannot heed them by withdrawing ventilation from the patient.
It follows that the law fails to acknowledge the close relationships most individuals have with their family. These relationships, normally characterized by mutual care, love, and commitment, are important building blocks of one’s personhood. A sweeping prohibition against withdrawing ventilation, irrespective of the family’s view, disregards the importance of familial relationships and denies the significant place they have in the patient’s identity and life experience. As such, it conceptualizes the patient as separate from his/her family, rather than as a social creature. 117 From a theoretical perspective, the prohibition fails to acknowledge the relational aspects of the right to autonomy. 118 This is also relevant when the family asks the physicians to connect a patient with Covid-19 to a ventilator knowing that this is his/her wish, but the physicians cannot do so due to a shortage of ventilators and the prohibition against withdrawal of ventilation.
Another aspect of criticism is rooted in a relational understanding of public health ethics, according to which individuals should be treated as interdependent beings. 119 The prohibition against withdrawal of ventilation, regardless of whether there is any benefit to the ventilated patient, and its denial of potential benefit to other patients with Covid-19, also fails to acknowledge that individuals, as well as their interests, are interdependent. More specifically, it ignores the fact that keeping a terminally ill or dying patient alive when there is a shortage of ventilators might result in the death of a different patient with a good chance of recovery. From a relational perspective, adhering to the prohibition against withdrawing ventilation during the COVID-19 pandemic fails to consider the impact on other members of the community. 120
This critique is particularly relevant in the context of Israeli society. A primary motive for the adoption of a prohibition on withdrawal of ventilation is the desire to satisfy the demands of conservative religious-based parties in the Israeli Parliament (the Knesset), who often hold the balance of power. 121 According to the Jewish Halacha (Jewish law), actively and deliberately shortening a patient’s life is prohibited. 122 Life has intrinsic value that is not measurable. Thus, the value of 70 years of life is equal to the value of one second, and the value of the life of a sick individual is equal to that of a healthy one. 123
It is therefore reasonable to assume that the withdrawal of ventilation from any patient would not be acceptable to most religious communities, even if the patient is in his final stage of life and the goal is saving the life of another individual. While the religious beliefs of individuals should be respected, adopting a religiously driven prohibition against withdrawing ventilation might affect the welfare of patients with Covid-19 in need of ventilation, regardless of their religious affiliation. Moreover, it is reasonable to assume that the aim of saving as many years of life as possible is shared by many in Israeli society, including members of the conservative religious-based camp. It follows that preferring the interests and values of one community over another in Israeli society is not aligned with the relational approach.
Potential solutions
Given that a patient cannot be disconnected once ventilated, the issue that Israeli society faces is how to resolve a situation in which demand for ventilators exceeds supply. In practical terms, the issue is how to fulfil the aim of saving as many lives as possible when there is a shortage of ventilators and the law bars physicians from withdrawal of ventilation. From a bioethical and social perspective, the question becomes how to balance beneficence, autonomy, and justice within the current legal framework.
There are a number of options. One is to amend the law so that physicians can withdraw ventilation from patients whose condition deteriorates to a point where it is no longer of any benefit. This can be limited to circumstances in which resources are severely limited, as in the case of the Coronavirus pandemic. Either the Knesset can amend the 2005 Act to this effect or alternatively, the government could issue special time-limited regulations to permit withdrawal of ventilation only during a pandemic when demand for ventilators far exceeds supply. The proposed amendments are desirable since their primary purpose is to increase the number of available ventilators and therefore save additional lives, thus embodying the principle of sanctity of life, which is, as discussed above, fundamental to the Israeli legal system. 124 More specifically, these amendments can be adopted based on a liberal-individualistic approach, which accords with the Basic Act: Human Dignity and Liberty that guarantees an individual’s right to life and bodily integrity and is perceived as being part of Israel’s unwritten constitution. 125
However, the chances of amending the current legislation appear to be slim. It would be blocked by objections from the religious and conservative communities, and furthermore, politicians would prefer to avoid doing so, knowing that it is a divisive issue in Israeli society. The reason for such opposition is that changing the law may lead to physicians withdrawing ventilation and thus indirectly shortening the life of a patient, a situation that religious and conservative groups do not agree with. In addition, it is unlikely that amending the law would be supported by the medical profession. As our previous study indicated, physicians often find it difficult to switch off ventilators, even if they are legally permitted to do so. 126 Therefore, they have no interest in changing the law either. Moreover, some clinicians may explicitly object to a change in the law since they consider the withdrawal of ventilation to be an act of killing and therefore unethical. 127 Others might wish to avoid the emotional burden attached to the withdrawal of treatment, knowing that it will lead to the death of the patient. They might also argue that imposing the burden on them is unfair.
If in practice the law cannot be changed, are there any other options to potentially increase the number of available ventilators? One would be to adopt rigorous exclusion criteria for patients who need ventilation when resources become severely limited. 128 Taking the Coronavirus pandemic as an example, if the law prohibits withdrawal of ventilation, then physicians would have only one decision to make: whether or not to connect a patient with Covid-19 to a ventilator. At this point, a physician would have to decide, based on the exclusion criteria, whether a particular patient with Covid-19 should be excluded from the list of patients eligible for ventilation. These criteria, which would be based on medical considerations and professional assessment, would exclude individuals from the list of patients eligible for ventilation who if not for the pandemic would receive ventilation (such as patients suffering from advanced dementia or advanced-stage cancer). These exclusion criteria would have to be rigorously followed and enforced, in order to deflect attempts by the patient and/or family to circumvent the restrictions. 129
This approach, as explained in the second section, finds support in the 2005 Act, which states that a physician is not obligated to provide medical treatment to a dying patient, if the treatment is not expected to extend the patient’s life or might cause him or another individual considerable harm. 130
However, relying on strict exclusion criteria raises several issues. First, a question arises as to whether all ventilators belong to a single pool that is available to all patients (those with Covid-19 and those suffering from other medical conditions). If they do, then using strict exclusion criteria may require prioritizing ventilation among all those that need it, whether they suffer from Covid-19 or some other condition. If it turns out that patients with Covid-19 are allocated a certain number of ventilators from the total, then using strict criteria may lead to a disproportionate impact on disadvantaged groups (the elderly, those with disabilities/co-morbidities, etc.).
Furthermore, establishing exclusion criteria would probably not be sufficient to fulfil the utilitarian goal of saving a maximal number of lives since although the shortfall would likely be reduced it might not be eliminated. Thus, additional options need to be considered in order to further reduce the shortfall by means of circumventing the prohibition against withdrawing ventilation, even when the patient is already ventilated.
One of the options is to reclassify ventilation as a cyclic treatment whose renewal is subject to a considered decision rather than a continuous treatment that cannot be terminated, a strategy that is permitted by the 2005 Act. In practical terms, the Act permits the installation of timers on ventilators. 131 This technology would give physicians the option to decide—before the ventilator stops—whether to continue the ventilation (because the patient’s condition has improved) or to allow it to stop (if the patient’s condition deteriorates to the point where ventilation of no benefit). Notably, Article 13 of the 2005 Act makes it clear that a clinician can decide against the provision of ventilation, regardless of the patient’s wishes or those of his/her family. Therefore, Article 13 can be implemented by means of timers that periodically stop and restart the ventilation, thus giving clinicians the option of not restarting ventilation. 132 In addition, timers also facilitate the withdrawal from ventilation if that is the patient’s wish or the wish of the surrogate decision-maker/legal guardian.
Although it has not yet been applied in practice, 133 this solution has several advantages. First and foremost, it is explicitly permitted by the 2005 Act, which reflects a consensus achieved among both religious and secular legislators. Thus, applying it when there is a shortage of ventilators is not expected to raise objections from the conservative segments of society. Furthermore, this option does not involve an active withdrawal of ventilation but rather an omission to reconnect a patient, a situation that clinicians might find more ethically acceptable and less emotionally stressful. 134 Finally, this option embodies respect for the autonomy of the dying patient, if indeed it is aligned with his/her wishes.
A further option is to stop cyclic treatments that are administered to a patient once they are connected to a ventilator. Apart from the ventilation, patients also receive medication to maintain their blood pressure as well as other essential functions of the body. Stopping these treatments when it is medically agreed that the patient will die soon in any case will hasten the patient’s death. Since, according to the Act, these treatments are cyclic rather than continuous, it is legally permissible to discontinue them. In fact, this option is explicitly mentioned in the Israeli position paper and is considered to be legal permitted. 135 It has also been explicitly accepted by the members of the Halachic Subcommittee of the Joint Committee who wrote that part of the position paper dealing with the religious aspects of the issue. 136
Another option is to decrease the level of oxygen provided to the patient to the level in the atmosphere. In practice, this is equivalent to withdrawal of ventilation. This option was endorsed by the Israeli AG and by a district court several years ago in the abovementioned John Doe case. 137 In this case, a request to be disconnected from a ventilator was submitted by a mentally competent patient in an advanced stage of ALS. 138 The patient was fully aware of his condition, of the nature of his request, and of the expected consequences, that is, that he would die within a few minutes of being disconnected. 139 In light of his mental and physical suffering and the fear that he would soon lose the ability to communicate, he expressed the wish to be disconnected from the ventilator. 140 Based on the “spirit of the Act” and the special circumstances of the case, the AG ruled that while switching off a ventilator is legally prohibited, gradually reducing the ventilator’s oxygen level to that of the atmosphere is not. 141 The AG justified this solution as striking the appropriate balance between the prohibition on disconnecting a dying patient from a ventilator and the principle of patient autonomy. 142 The court fully adopted the solution proposed by the AG, thus approving the use of a medical procedure that essentially enables a physician to terminate life-sustaining treatment, despite the prohibition set out in the 2005 Act. 143
At first glance, this option may be of help to physicians when it is believed that ventilation is futile and should be withdrawn for the benefit of a patient with better chances of recovery. However, this option is conditional on the patient wishing to end his/her life. Arguably, this may not be the case for most of the patients with Covid-19. They will likely want to continue receiving ventilation and therefore, this option is less relevant in our context.
Finally, there is the option of transferring a patient from one respiratory device to another that is less effective. Although this essentially involves withdrawal of ventilation, it was considered legal by the members of the Legal Subcommittee of the Israeli Joint Committee that wrote the legal part of the position paper. According to this Subcommittee, when the chances of a ventilated patient to recover and survive are only slight and his/her medical condition does not improve despite the ventilation, she/he can be removed from the ventilator and transferred to a different device, which, while less effective, still provides respiratory assistance. 144
There are several relevant issues here. First, a decision by a physician to transfer a patient from a ventilator to a different device may be legally permissible based on Article 13 of the 2005 Act. However, according to a strict interpretation of Article 21, it is prohibited to disconnect a patient from a ventilator. Second, the physician may face opposition to transferring a patient to a less effective respiratory aid either from the patient him/herself or from his/her surrogate decision-maker/legal guardian. In such a situation, it is questionable whether a physician would nevertheless transfer the patient. Empirical findings from our previous study reveal that family members often have substantial impact on a physician’s decisions, and therefore, this option may not be feasible in practice. 145
In sum, the aforementioned options, particularly those that fit into the existing legal framework, involve bioethical, social, and practical difficulties. Although they do not provide a comprehensive solution that would eliminate a shortage in ventilators during a pandemic, the exploitation of any or all of them may reduce such a shortage. As will be emphasized below, the main reason for such a shortage can be attributed to the obligation to respect a patient’s wishes.
Concluding remarks
Whatever solution is adopted, the decision should be subject, in our opinion, to two constraints. First, patients and families should be provided with palliative care in order to decrease their pain and suffering. 146 Second, physicians should maintain a direct channel of communication with the patient (if she/he is legally competent) and with his/her relatives (particularly if she/he is not) about available treatments and any change made in the patient’s treatment. 147 In short, decisions must be transparent for the patient and his/her relatives. They should be part of the decision-making process and their preferences should be obtained. Not only does this constitute appropriate moral conduct but, as described in the second section, it also fulfils professional/legal obligations. 148
Insisting that physicians should communicate any changes in the treatment provided to either the ventilated patient himself/herself or his/her relatives is important given the findings of our empirical research, namely, that physicians find it difficult to communicate bad news to patients and their families—they either try to avoid such conversations or use only the vaguest language. 149 There is no alternative to fully transparent conversations with patients and/or their relatives, whether in normal times or during a pandemic like the current one. When the condition of a patient deteriorates, physicians must ascertain his/her wishes. There will likely be patients and relatives who oppose any discontinuation of treatment even when it is clear that death is near. However, based on our research, there will be patients (whether during a pandemic or otherwise) and relatives who do not want to continue receiving ventilation if it merely prolongs life by a few days.
Overall, the basic message regarding any future crisis is clear: the withdrawal of ventilation should be legally permissible, provided it accords with the patient’s wishes or if the patient’s wishes are unknown and the treatment is futile. Moreover, society cannot adopt a utilitarian-based policy or act beneficently for the public by ignoring the legally recognized and secured fundamental rights of patients, particularly, the right to make informed decisions. In addition, the distributional effects of ethical policies should be taken into account in any discussion of this issue. Thus, as long as fundamental patient rights are in place, autonomy and justice from the liberal and relational perspectives should be given appropriate consideration even during a pandemic when the common goal is to save life in a situation of severely limited resources.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
