Part I - Well-Regulated Jurisdictions

1 Advance Directives in Israel

Miriam Ethel Bentwich

1.1 Introduction

This chapter explores advance directives in Israel, arguing that, paradoxically, the Israeli state’s formal legislative instruments to protect the autonomy of terminally ill individuals by empowering them to make decisions about the scope, nature and extent of medical treatment they wish to accept (or to reject) as they approach the end of life, and affirmed their right to die with dignity, have in fact led to the under-regulation of the use of advance directives. The chapter first discusses the legal and political-religious context that led to the enactment of legislation – the Dying Patient (DP) Act¹ – which constitutes the main legal framework for the use of advance directives in Israel.² The main attributes setting the legal framework in which advance directives are incorporated into this law are next explained, followed by a discussion of the state’s guidelines for the formulation of advance directives. Finally, the chapter demonstrates how the restrictive nature and the supposed strict regulation of Israeli advance directives have actually led to their under-regulation. Some reasons for this include the practical difficulties posed by the actual implementation of advance directives, the many “shades of grey” faced by healthcare providers when interpreting the law, and legal critiques of, and precedents questioning the DP Act.

1.2 Legal, Religious and Political Background Leading to Legislation

Prior to the introduction of the Patient’s Rights Law (1996),³ Israel had refrained from addressing the issue of end-of-life care, as evident by its ambivalence towards the introduction and enactment of specific legislation to address the matter. In fact, Israeli penal law perceives euthanasia – even for the sake of preventing suffering – to be an act of killing. Similarly, until the 2005 passing of the DP Act, under penal law, even passive euthanasia – namely, the act of refraining from carrying out actions designed to prolong the life of a DP – which was carried out without the explicit permission of the court constituted an act of manslaughter.

The draft of the Patient’s Rights legislation originally included a clause entitled “The Right to Die with Dignity”. That section (originally, section/clause 10 in the preliminary law proposal) reads as follows:

A terminal patient is entitled to [can be understood as ‘having a right to’] die in dignity and in accordance to their viewpoint and belief, and as much as possible, in the presence of a person that the patient requests. The physician and the medical institution will assist in fulfilling this patient’s right, and will refrain from [doing] anything that might infringe this right.⁴

However, this section was subsequently removed, supposedly due to opposition from some ultra-orthodox Jewish parties; notably, the final law that was passed did not include any specific reference to end-of-life care.⁵ Indeed, until the enactment of the DP Law – particularly with respect to the matter of prolonging the lives of terminally ill patients – since the mid-1980s and even more so in the 1990s, the main progress made to address this issue was achieved through the courts.⁶ Some pivotal cases, such as those of Tzadok, Binyamin, Tzadi, Gilad, Bivas, Ploni (“John Doe”), pushed towards greater acknowledgement of patients’ right to personal autonomy in the context of end-of-life care.⁷

The courts’ decisions regarding these cases should be understood against the backdrop of the “constitutional revolution”, a term often associated with Aharon Barak, the Deputy Chief Justice and later Chief Justice of the Israeli Supreme Court. Chief Justice Barak’s employment of this term suggests his preference for a court activism approach.⁸ This approach gained favour after 1992 when the “Basic Law: Human Dignity and Liberty” was enacted to express and confirm the judiciary’s commitment to upholding the state’s main human rights principles. Thus, the governing premise of Justice Barak and many other Supreme Court justices was that the enactment of this Basic Law had opened the way to the court for further interpretations and applications of this law, in specific realms beyond the original written letter of that law. Indeed, during the years 1992–5, various verdicts in the regional courts and in the Supreme Court had expressed a heightened activism approach, which laid stress on strengthening aspects of dignity and liberty in cases addressed by the courts. In fact, the “constitutional revolution” and its emphasis on the court activism approach were explicitly mentioned in the pivotal verdict of the Hamizrahi Bank case of November 1995, a few days after prime minister Yitzhak Rabin was assassinated. Rabin’s murder sent shockwaves across the world, for it held potentially major global political implications; its media reporting was so extensive that it overshadowed the court’s verdict on the Hamizrahi Bank case, which attracted scant media coverage at the time.

At any rate, the case of Miriam Tzadok v. Beit Hae’la in 1992 is considered fundamental to the end-of-life care issue for a number of reasons.⁹ It was the first case regarding end-of-life care that concerned an older person with dementia. The case revolved around the withholding of a feeding tube and raised questions about the authority of healthcare professionals to refrain from offering life-prolonging treatment. Some former cases had focused mainly on the issue of withdrawing ventilator support as death approached. In this case, however, and while still cognitively competent, the patient Ms Tzadok had left clearly written instructions (effectively, directives) requesting that medical staff should refrain from feeding and keeping her hydrated by artificial means, when her dementia had progressed to an advanced stage. The court ruled that although Ms Tzadok was no longer capable of expressing her wishes by the time her case reached the court, her directives – written while she was still cognisant – represented her true desires. It was therefore decided that Ms Tzadok’s desires be respected, and that the right to personal autonomy be recognised, in the interests of preventing her further suffering. In accepting Ms Tzadok’s written directives, the court not only upheld the principle of personal autonomy as a critical value, but by extension, confirmed the patients’ right to refuse life-prolonging interventions. Indeed, the court went even further in applying this value in the context of a non-terminal disease.

Other rulings similarly applied the value of personal autonomy when justifying the withdrawal of ventilator life support for patients with amyotrophic lateral sclerosis (ALS). A landmark case in this respect was the 1996 ruling on the request made by Itay Arad, an ALS patient. Arad had sought the court’s permission that he be allowed to be disconnected from a ventilator, once he reached the stage where he was unable to live without requiring the constant support and care of others.¹⁰ The court granted his request, thus upholding his autonomy and dignity, and two years later, his disease having progressed to the stage where he was dependent on medical support to maintain his life, Mr Arad was disconnected from the ventilator, as he had explicitly requested.

Another noteworthy case was that of Yael Shefer (1993);¹¹ the verdict in that case was interpreted as permitting the withdrawal of life support, and was subsequently cited in nine different cases, including that of Arad. Shefer’s case concerned a toddler with Tay-Sachs disease, who had, in fact, passed away before the court could publish its verdict. The court had been asked to decide on the issue of withholding various life-sustaining treatments for the toddler. The written deliberation by Justice Elon of the Supreme Court in this case became of importance because of the decision laid down by him. Judge Elon’s deliberation was based on both modern Israeli law and the Jewish religious (Halakhic) perspective, derived from biblical commandments, and subsequent Talmudic and rabbinic teachings.¹² Indeed, when referring to the possibility of withdrawing ventilators, Judge Elon offered an in principle interpretation of the Jewish religious law that supported this action be taken, if the patient was suffering. Of interest also is that Judge Elon’s written deliberation explicitly addressed both types of euthanasia – active and passive. The latter he referred to as legitimate in principle, including from the Jewish religious perspective. As we shall see in the following, Elon’s religious perspective in the Shefer case departs some distance from the religious viewpoint practically fostered in the letter of the law that was enacted more than a decade later.

The aforementioned courts’ rulings and deliberations that fostered the principle of personal autonomy in order to allow passive euthanasia to end suffering were mainly written from a secular perspective. Some representatives of the ultra-orthodox parties within the Israeli parliament regarded these rulings as a “step too far”; their understanding of end-of-life care was primarily grounded in values that stressed the sanctity of life, a philosophical and religious standpoint which emphasised a commitment to preserving and saving human life, regardless of the individual’s desires. This perspective offers, at the most, quite limited scope and space for recognising human suffering as a legitimate justification for overriding the commitment to the sanctity of life. Due to the complexity of the Halakhic law and its interpretations through varied rulings by rabbis with expertise in this law, it is well beyond the scope of the current chapter and its author’s expertise to delve into the specifics and nuances of these interpretations.¹³ However, there is no dispute that the Halakhic law – certainly from an ultra-orthodox religious perspective – is framed less in terms of regard for autonomy and individual rights, but rather, privileges the principle of the sanctity of life as a fundamental value that (in many of the cases) took precedence over other values. It is also not disputed that for the majority of Israeli Jewish citizens (and the 20 per cent of Arabs residing within Israel), the Halakhic law does not play an important role in their perceptions and lifestyle values.

Against this backdrop, in 2000, Rabbi Shlomo Benizri, the Health Minister of the ultra-orthodox Shas political party, convened a committee to establish guidelines for a new law to regulate end-of-life care. This regulation can be interpreted as a backlash against what was considered a “liberal-secular” stance, for its support of most court decisions that acknowledged the rights of individuals to refuse life-prolonging treatment, and for that right to be respected and lawful. Rather than appoint a secular bioethicist, physician, or legal expert, Benizri settled on Avraham Steinberg, a rabbi and physician, as committee chairperson. In a demonstration of his political power, Benizri used his position as Health Minister to appoint an individual with a self-declared ultra-orthodox rabbinical viewpoint as chairperson to a committee that set the legal framework of the DP Act. Steinberg’s committee held discussions for two years before submitting its main recommendations for the suggested law’s framework, eventually leading in 2005 to the enactment of the DP Act. Scholars such as Carmel Shalev have commented on the final version of the legislated DP Act, noting in particular that in substance, this law is conceptualised within a Halakhic framework rather than a secular-liberal perspective.¹⁴

1.3 DP Act: Main Attributes

Israel’s DP Act constitutes the main legal framework for advance directives, a legal instrument for a terminally ill patient to explicitly request the extent and nature of medical treatment and interventions at the end-of-life stage. Patients with a terminal illness are enabled to leave written instructions as to whether they would wish their life to be prolonged (or otherwise) at the point when their suffering becomes unbearable. However, before delving into the details of these directives under this law, it is important to highlight some main attributes and definitions that inform the framework in which advance directives are employed.

A good point to begin with is the definition of the law’s goal, as set out in section 1(a) which pertains to “the regulation of medical treatment given to the DP, while striking the right balance between the value of the sanctity of life and the value of a person’s autonomy and the importance of the quality of life”. Still within the “Goal” section of the law, section 1(b) states that it is “based on the values of Israel as a Jewish and democratic state, as well as on core values in morality, ethics and religion”. Thus, in two respects, the DP Act resonates quite well with its legal and political background, as previously described. First, the DP Act acknowledges the tension between the so-called liberal secular viewpoint, which emphasises individual autonomy and quality of life (i.e. where suffering is unbearable to the point that the patient cannot be said to enjoy any quality of life), against the more “religious” perspective which cherishes and privileges the principle of the sanctity of life. Second, the legislature explicitly states that religion (in particular, Judaism) constitutes an important foundation of the law, in line with the reason for which the Steinberg’s committee was appointed in the first place.

Following on from the aforementioned tension is the matter of the very definition of a DP according to this law, which distinguishes between two types of DP: a DP and a DP in the “final stage”. Section 8 of the law defines a DP as originating from “a patient suffering from an incurable medical problem with a life expectancy of no more than six months, even with the provision of medical treatment” [emphasis added]. In contrast, a DP in the “final stage” refers to “a patient with acute systems failure [of his/her body] who, even with the provision of medical treatment, has a life expectancy of no more than two weeks” [emphasis added]. Furthermore, the law distinguishes between a “competent” DP and an “incompetent” DP, the determination of which both fall under the jurisdiction of this law. Competent DPs possess adequate mental and cognitive capacity¹⁵ to autonomously reach an informed decision, as well as having the ability to communicate that decision to medical staff. By contrast, the incompetent DP is without the mental or cognitive capacity to reach an informed decision, or may be unable to communicate their decision to medical professionals. While both a DP and a DP in the “final stage” may choose not to prolong their life, the law is clear that accompanying treatment and nourishment can be withheld only from a DP who is in the “final stage” of life, as will be discussed later. Given that a DP in the “final stage” refers to a patient with acute systems failure of his/her body (also known as “acute multi-organ dysfunction”), for all intents and purposes “final-stage” DPs lack the capacity to express their “choice”, since they are usually sedated, a state which implies that they cannot, and therefore do not conform with the definition of “competency”, as defined by the DP Act (see note 15).

Sections 9–10 of the law require that a senior attending physician explain to the patient their terminal condition, and to ask the patient how they wish to proceed, should the physician decide that the patient’s condition conforms with the DP status as previously defined. In the event that the patient is already incompetent, whether as a DP or if they are DPs in the “final stage”, the physician will contact the patient’s family and explain the situation to them.

Israeli’s legal definition of a DP is quite limited, as it does not extend to the inclusion of patients with chronic illnesses yet without a clear prognosis, such as dementia or ALS. Such a narrow definition seems to be in conformity with more orthodox religious standpoints that regard the sanctity of human life to be a fundamental religious and philosophical principle – a key foundation on which the DP Act rests. The narrowness of this definition prompts the question of autonomy for patients who fall outside the jurisdiction of this law: what rights can they exercise in terms of their ability to fulfil their desire not to have their lives prolonged? This is especially pertinent in instances where patients lack the ability to express their choice (due to incapacitation arising from their disease) since their advance directives cannot be followed, as they do not fit the letter of the law as DPs.¹⁶

Yet another important attribute of this law concerns the forbidden and allowed actions that may or may not be taken in weighing the decision whether to prolong (or otherwise) the life of a person deemed to be either a DP or a DP in the “final stage”. For both categories of patients, section 19 of the law explicitly forbids euthanasia, which is understood as “an action aimed to kill or that its probable result will lead to death”. Section 20 also explicitly forbids physician-assisted suicide. In defining euthanasia as actions taken in the knowledge that the probable result will lead to death, Israeli law extends the scope of activities defined as euthanasia. For example, administering morphine in such dosages that will likely kill the patient might be interpreted as an act of euthanasia, even if the intention is to provide comfort and to ease the patient’s pain.

The DP Act not only forbids these extended forms of “euthanasia”, but also strongly mandates the conditions and manner in which “passive euthanasia” may take place. The legislature has placed significant limits on non-life-prolonging actions, even if they do not involve performing actions intended to bring about the death of the patient (e.g. lethal injection, orally administered poisonous pills). A major restriction, in this respect, is that in some medical contexts, the law does not permit certain actions of withdrawing or withholding treatment. Withholding and/or withdrawing treatment are the main actions in which “passive euthanasia” takes place, and the Israeli legislature has forbidden the withdrawal of a “continuous” treatment (e.g. ventilators). In fact, the distinction between “continuous” and “cyclical” treatments in the DP Act arises from the fact that withdrawal of continuous life-sustaining treatment is prohibited. The rationale here is that a cyclical treatment (e.g. dialysis, chemotherapy) can/could be not renewed; thus, stopping such treatment is not regarded as the withdrawal of an existing treatment, but rather as withholding the next or new cycle of treatment.¹⁷

That being said, the Israeli DP Act does allow for the withdrawal and withholding of nourishment (feeding and, in case of severe suffering, or harm hydration), palliative care, and secondary/accompanying treatment from incompetent DPs in their “final stage”. In an article published about the Halakhic law basis for the DP Act, Rabbi Steinberg outlined the justification for allowing such exceptions in cases where patients have reached the “final stage”.¹⁸ Rabbi Steinberg draws on key Halakhic rulings to argue that refraining from giving the aforementioned treatments is allowed, in situations where three or more systems in the human body have failed, and it is clear from a medical perspective that death is inevitable sooner than later. As it stands, the DP Act defines DP patients in the “final stage” as patients with “acute systems failure [of his/her body]”, and it is clear that this definition finds resonance with Halakhic stipulations that there must be failure of three or more systems in the human body. Rabbi Steinberg’s distinction between “final stage” DPs and DPs not in the “final stage” allows for the withdrawal of particular treatment in the DP Act, in accordance with pivotal Halakhic rulings.

Interestingly, the DP Act departs from general definitions of the age of consent in a manner distinct from the general legal system. The DP Act allows people from the age of seventeen to autonomously decide whether they wish to accept or reject life-prolonging interventions, as opposed to the general age of consent (and adulthood) in the Israeli law which is legally understood to be when the individual reaches eighteen years of age. This stipulation is relevant for both competent and incompetent DPs (see sections 5 and 14).¹⁹

Another interesting facet of the DP Act that has ramifications for the employment of advance directives is section 56 of the law that permits individual doctors to declare themselves conscientious objectors. This status permits doctors to withdraw from any involvement in the enactment of an advance directive (which will be discussed in further detail in the next section). Under these circumstances, the physician is obliged to refer the patient to another physician who is able to respect the patient’s choice.

Finally, the law stipulates that the default assumed decision concerning whether a DP wishes to accept or reject life-prolonging treatment is that they would wish to prolong their life. However, according to the law, a DP’s wish should be equally respected, as much as possible, regardless of whether they choose to prolong their life or not.

1.4 DP Act: Advance Directives

Advance directives are one mechanism for informing medical staff of the wishes of an incompetent DP regarding the manner in which they should be treated, within the legal framework set by the DP Act (sections 5, 14). The law also stipulates other alternative mechanisms for achieving this end, including assigned durable power of attorney by the patient (when still competent) to a close relative, an institutional or national committee defined as within the law. The latter is particularly relevant in cases where there are disputes, such as between parents of a minor or between two close family members in instances where the patient did not leave an advance directive. In the absence of these mechanisms, a final mechanism is a decision by a senior attending physician, based on a close relative’s explicit declaration of the patient’s expressed wish to prolong or end their life.²⁰ This latter option was intended as a last resort when all other options were either unavailable or inapplicable. However, as we shall see, this mechanism was actually applied more frequently than originally expected, at least according to the letter of the law.

Before delving into the manner in which advance directives are completed, it should be noted that, under sections 34(a) and 39(a) of the law, both advance directives and an assigned durable power of attorney by the patient to a close relative are limited for a period of five years, and must be renewed every five years by the patient in order to maintain validity. For certain patients (e.g. patients with dementia), the renewal of advance directives is practically impossible, once the patients’ cognitive and/or mental condition renders them unfit to make an informed decision. Advance directives completed five years previously cannot be renewed, and will become invalid once more than five years have passed since they were originally completed or renewed.

Advance directives are required to be written on a specific designated form which, once completed, must be signed by the patient and two witnesses, and then mailed to the MOH (sections 32, 44), where it is then placed in a national registry, which is designed to record all submitted advance directives in Israel. The form may be completed by a person defined as a DP (while still competent), but it could also be filled out by a healthy person before becoming a patient, let alone a DP (section 31). This form includes thirteen specific actions pertaining to not prolonging the life of an incompetent DP, and eighteen such particular actions for a DP in the “final stage” (of which the first thirteen actions are similar to those available for DP not in the “final stage”). In other words, the scope of legally recognised and applied advance directives is limited to these particular possible actions for not prolonging the DP’s life. As noted earlier, the default assumption of the law is that people wish to prolong their life as long as possible (although a DP’s decision in this regard should be equally respected whether they want to prolong their life or otherwise). Accordingly, the advance directives mechanism used under the DP Act is formulated in such a way that healthy people or patients (while still competent) must specify for each available action whether they wish to refrain from having that action taken.

Furthermore, the specified actions are quite particular. For example, the person filling out the advance directives form cannot simply state that they wish to refrain from having any form of resuscitation. There are five different actions pertaining to resuscitation, only one of which is defined as “performing resuscitation in any possible way”. Other specified actions of resuscitation include, for instance, providing resuscitative medicine, performing defibrillation, and inserting a tracheal tube. Other actions that do not pertain to resuscitation include the following:

1. 1. Being hooked to a ventilator

2. 2. Starting a dialysis treatment and determining whether to continue treatment if it is futile

3. 3. Performing large surgical operations (e.g. amputation of a leg or arm, removing an internal organ)

4. 4. Performing small surgical procedures (e.g. amputation of a finger)

5. 5. Performing radiations for a malignant disease

6. 6. Inducing chemotherapy for a malignant disease

7. 7. Administering antibiotics in a situation of severe sepsis that cannot be treated by usual antibiotic treatment

8. 8. Conducting diagnostic tests (e.g. blood, X-ray)

Treatments that individuals may choose to refrain from (or not) only in the case of becoming a DP in the “final stage” include the following:

1. 1. Regular (accompanying) treatment such as antibiotics or transfusion of blood and its products

2. 2. Treatment for a secondary disease, such as the provision of insulin for a diabetic patient

3. 3. Use of a feeding tube (nutrition)

4. 4. Pushing of fluids (hydration)

5. 5. Palliative care

Finally, it is interesting to note that, as part of the advance directives form, future incompetent DPs will have an opportunity to define the physical and/or cognitive conditions that for them would amount to “unbearable suffering”, and which would cause them to wish to refuse further life-prolonging treatment. In this section of the form, built-in options include conditions that are not necessarily terminal, and certainly do not comply with the legal definition of a DP, such as severe dementia or quadriplegia. Given that these particular options do not comply with the lawful definition of the DP, they cannot then legally be directly applied by the medical staff, merely on the basis of their stipulation on the form. Rather, these built-in options in the official advance directives forms are at worst, simply moot, or at best, might be used as a basis for appealing to the court; notably, the court, in turn, is obliged to act within the boundaries set by the DP Law. It is therefore unclear to what extent (if any) such personal definitions of what constitutes “unbearable suffering” may be applied in practice, given the narrow definition of a DP, as stipulated under the law.

1.5 Discussion: Over-regulation Leading to Under-regulation in Practice

A key feature emerging from the description of the Israeli DP Act is its extensive regulation in general, and specifically with respect to advance directives. Such extensive regulation also expresses the quite narrow boundaries of practically allowed passive euthanasia in Israel and their corresponding advance directives. This narrowness can be attributed to both the definition of the DP and the limited actions allowed to be taken in order to comply with the patients’ stated desire that their life should not be prolonged.

In fact, this extensive regulation – with its emphasis on the limited scope of allowed actions that will not prolong a DP’s life, including in the advance directives form – is in stark contrast to the manner with which end-of-life care was handled by the courts before the passing of the DP Act. As previously noted, in the absence of a general official regulation prior to the enactment of the DP Act, the courts simply ruled on a case-by-case basis. No less importantly, these rulings generally covered a wider diversity of patients, including those who did not suffer from terminal illnesses (e.g. dementia, as in the case of Miriam Tzadok), or alternatively, did suffer from a terminal disease but did not have the precise “date of expiry”, that is, that the patient was expected to pass away within six months of receiving the prognosis, as required by the DP Act.

Similarly, these courts’ verdicts allowed a wider scope of actions for not prolonging the life of patients than the DP Act permits. For instance, this law explicitly forbids any individual from disconnecting a DP from a ventilator, or the withholding of tubal feeding or liquid pushing, except for DPs languishing in their “final stage”. In contrast, as already discussed, prior to the DP Act, the courts allowed the disconnection of dying persons from ventilators upon the patient’s request (including any written instructions in advance), as well as confirming requests for artificially administered nourishment and liquids to be withheld from patients without a clear life expectancy of two weeks, at most.

Moreover, a gap also exists between the law’s presumption to represent a so-called Jewish religious perspective, and the particular interpretation of this perspective by Justice Elon, prior to the enactment of the law. As it may be recalled, Justice Elon’s interpretation – also based on Halakhic law and rulings, as was the DP Act – nevertheless did not reach the same conclusion, that the Halakhic-inspired definition of a DP should be contingent upon a specific time frame (i.e. the up-to six months’ life expectancy prognosis). In addition, and again in stark contrast to the DP Act’s underlying interpretation of the Halakhic law, Justice Elon acknowledged the possibility (in principle) that in the case of severe suffering, DPs could be disconnected from ventilators. This gap is an indication of the complexity and multifaceted nature of the Halakhic law, and undermines its somewhat monolithic representation, when it was driven and advocated by the political power of the ultra-orthodox political parties.

Against the backdrop of this observed contrast between the DP Act and many former courts’ rulings, some legal experts have argued that precisely because of the law’s limited inclusion of patients that it covers, it is entirely possible for different individual interpretations of the law. Instead of the assumption that patients who are not covered by the law are not allowed to request that their lives should not be prolonged (including through advance directives), the suggested interpretation seems to run in the opposite direction. According to some legal experts, this suggested interpretation could allow the courts to revert to the case-by-case rulings that were the norm prior to the passing of the DP Act, since this law (presumably) does not apply to patients who are not covered under its jurisdiction. Such an interpretation also emphasises the right to autonomy, which had been upheld in the Patient’s Rights Law, as well as the general spirit that celebrated patients’ autonomy (and which characterised most of the courts’ rulings), before the DP Act was enacted.²¹

Fundamentally, this interpretation serves to illuminate the unbalanced manner in which the law addresses the tensions between a liberal-secular perspective and a more religious standpoint in favour of the latter, as already noted in this section, and at the end of the “Legal, Religious and Political Background Leading to Legislation” section.²² Admittedly, in practice, it appears that this interpretation has not secured the support of the courts.²³ Hence, the law’s framework regarding the definition of a DP, as well as the allowed and forbidden actions in order to refrain from prolonging life was upheld by the courts, regardless of whether or not the patient was a DP, as stipulated by the DP Law.

Nevertheless, this interpretation reveals the paradoxical possibility that the extensive regulation employed by the law could effectively lead to limiting its application (i.e. under-regulation). On the one hand, there is extremely restrictive regulation regarding the definition of a DP, and the limited permissible actions that may be taken to prolong their life (as demonstrated in previous sections), all of which originate from and are influenced by a particular and strict religious standpoint. On the other hand, there is the more liberal and secular legal perspective, which undermines the restrictive nature of the DP Act, as evident in its suggestion that precisely because of the narrow definition of DPs, the law holds little relevance for many patients. These legal experts suggest therefore that for patients who do not meet the narrow definition of DP, decisions regarding end-of-life matters should be delegated back to the courts, while not binding them to the letter of the DP Act.

In fact, although this particular legal interpretation remains underutilised by the courts, other aspects concerning the application of the DP Act already reveal its under-regulation in practice. While these aspects are supposedly varied, I argue that – at least to a certain extent – they echo the same tensions between a law that is substantively grounded in religious philosophy, and in its praxis, which is influenced by medical professionals and more secular perspectives of patients and their families. At the very least, the latter side of the tension leads to the under-utilisation of different aspects in the restrictive regulation embedded within the DP Act, and could ultimately lead to the very undermining of this regulation. In other words, such under-utilisations of the DP Act can be interpreted as an under-regulation of it in practice.

Specifically, the discussion here briefly addresses three such aspects: (1) practical difficulties in implementing advance directives; (2) “shades of grey” in healthcare providers’ interpretation of the law; and (3) other legal decisions and interpretations undermining the regulation set by the law.

As evidenced by the conclusions of two studies conducted approximately ten years after the enactment of the DP Act, advance directives have rarely been submitted by the potentially relevant population. One study, based on official records of the Ministry of Health, found that as few as 3 per cent of older people had completed advance directives forms.²⁴ Another study, based on a random sample of 720 people over fifty years of age, found that only 4 per cent of the respondents indicated they had written advance directives, and only 5 per cent had appointed a durable power of attorney.²⁵ Recall if you will that completion of the advance directives form and the appointment of a durable power of attorney were supposed to be the main mechanisms through which future incompetent DPs could ensure that their wishes would be respected within the boundaries of the law. Yet, if the low proportion of research respondents who employed these mechanisms in the two aforementioned studies reflect the overall reality, then the main regulative vehicles employed by the law for achieving this end appear to be considerably under-utilised. Furthermore, it would seem that the “last resort”, that is, the most under-regulated mechanism, might be the most favoured, possibly against the spirit of the highly regulated law. In the absence of any written documentation, the option for a senior attending physician to act on the best considered decision – on the basis of a declaration made by a close relative that the patient did not want to receive life-prolonging treatment – becomes the most potent mechanism.

In seeking to explain the under-utilisation of advance directives, the studies concluded that low public awareness about DP directives and cultural-religious values represented possible factors underlying their reasoning. Both studies emphasised the need to increase the general public’s awareness of the purposes and benefits of advance directives. In fact, the second study (with a random sample of 720 people from the general public), found that only seven respondents (1 per cent) had been contacted by a healthcare provider to discuss their preferences for end-of-life care, and almost half of the respondents (43 per cent) thought that the mass media did not offer sufficient information regarding end-of-life care choices.²⁶

In terms of the influence of cultural-religious values, the majority of respondents (73 per cent) in the second study offered the view that from a Jewish-religious perspective (not necessarily reflecting their own viewpoint and certainly not based on expertise in Halakhic law) every effort should be made to preserve a life. A small minority of respondents who identified as religious or ultra-orthodox were more likely to deny the need for advance directives,²⁷ thereby echoing yet again the unresolved tension leading to the under-regulation of the DP Act: in this case, a stricter religious proclivity that is embedded in the law, but in actuality, a viewpoint not shared by the majority of Israelis. Furthermore, it has been suggested that at the very least, the lengthy advance directives form contains too much legalese and medical terminology, and that the complexity of the form makes it difficult for laypersons to comprehend.²⁸ This factor may help to explain its low utilisation.

Another practical difficulty appears to stem from the absence of discussion and communication about end-of-life preferences. Some studies carried out among staff of nursing homes in Israel found that staff members were generally unaware of residents’ preferences for their end-of-life care.²⁹ These studies also revealed an overall lack of communication and support resources to facilitate discussions around such care, and which would include conversations with patients about their wishes for end-of-life care. The conclusions of another more recent study, focused on hospital staff, echoes these findings, certainly insofar as lack of sufficient communication with patients regarding their end-of-life care. Therefore, these studies highlight additional barriers to the implementation of advance directives, while stressing the under-regulation of the law – particularly its advance directives mechanism – in practice.³⁰

With respect to the shades of grey in healthcare providers’ interpretation of the law, at least some elements of the DP Act that are supposedly strictly regulated are not necessarily so, despite being very thoroughly detailed. For instance, the very definition of a DP includes the requirement that the patient has a prognosis of six or fewer months to live. But this raises the question of what happens if – according to the medical literature – the patient receives a prognosis of six- to eight-months life expectancy; how should the physician interpret this range when deciding whether this patient should be deemed a DP? Similarly, let us take for instance, an individual with early-stage dementia, who, while still competent, and wishing to avoid the poor quality of life that is projected with the progression of this disease, completed an advance directive requesting that physicians refrain from all possible actions denoted on the form, including being administered antibiotics when in the “final stage”. If this person is hospitalised due to acute pneumonia, where it is uncertain whether the pneumonia can be successfully treated by currently available antibiotics, but without which the pneumonia will possibly cause their death within seven to ten days, how should the medical staff interpret their condition? Should they view the patient’s condition as a DP in the “final stage”, based on the possibility of pneumonia being acutely terminal, or just as a person with dementia, who is not considered as a DP, let alone a DP in the “final stage”? In the former scenario, the law allows physicians to withhold different medical treatments, including antibiotics. Such a case is hardly clear cut as high dosages of antibiotics may resolve the acute pneumonia, thereby returning the patient to their previous state, as persons with dementia who are not considered to be DPs. The physician is placed in the difficult position whereby s/he is required to make a clear-cut decision on whether the pneumonia is terminal, despite the inherent uncertainty as to whether it is treatable. Thus, these two examples again illustrate how the very specificity of a law that is mainly grounded within a particular and quite restrictive religious cosmology contradicts the uncertainty of some diseases, and might actually lead to problems in interpretation. That the law allows room for clinical discretion in these matters potentially undermines the regulative framework set by the law.

Although the Israeli MOH has issued guidelines for Implementing the Dying Patient Law in 2008, which summarise key provisions of the law and provide instructions to healthcare practitioners on the implementation of advance directives, a study conducted in 2014 found that many physicians remained unaware of the specifics of the law.³¹ The majority of physicians surveyed could not provide the correct definition of a DP or the age of competency under the law. This suggests that the complexity and specificity of the law may cause difficulties in its implementation; in other words, stricter regulation (and possibly over-regulation) is, paradoxically, leading to under-regulation.

Finally, a third aspect of this paradox is from a judicial perspective: namely that other legal decisions and interpretations may undermine the regulations set out in the law. In 2014, the court addressed the matter of the legality of a “workaround” suggested by the attorney general to circumvent the restrictive scope of the law. The case concerned a patient with ALS who requested that they be allowed to be disconnected from a ventilator.³² The employed workaround was that, instead of physically disconnecting the patient from the ventilator, the percentage of oxygen in the ventilator was gradually decreased until it reached the normal rate in the air (21 per cent), which would not sustain the life of the DP in this case. Here, the court simply confirmed the attorney general’s suggested workaround to be lawful. Still, in granting the patient’s request, the ruling is essentially at odds with the DP Act in at least two respects.³³ First, this ruling effectively (even if not literally) recognised an ALS patient as being a DP, even though the patient clearly did not have a prognosis of a life expectancy of less than six months. Second, the verdict essentially undermined the legally established principle that non-cyclical treatments should not be stopped or withdrawn. To be sure, this court’s ruling is not considered an official legal precedent, as it was issued by a lower court and, therefore, has no binding force for future court verdicts. However, this case cogently demonstrates how the strict regulation embedded within the law has led even the Attorney General of Israel to support the undermining or under-regulating of the law in practice, even if this pertained only to a single case. I would also suggest that this case echoes the same tension between a law that is grounded in strict religious faith, and its actual praxis, that was influenced by the less religious perspectives of patients and their families. In this case, there was an attempt to resolve this tension, but had the tension not existed, the attorney general would not then have needed to resort to an action that undermined the letter of the law.

1.6 Conclusion

Israel represents a particularly interesting case study with respect to the legal framework and its implementation insofar as advance directives are concerned. The Israeli case highlights how the interplay of politics and religion may be highly influential in the formulation of the legal framework for advance directives. Such formulation is manifested in relatively restrictive and highly regulated advance directives, certainly when compared to the courts’ rulings prior to the enactment of the DP Act. As this chapter has demonstrated, in practice, this restrictive and highly regulated form of advance directives paradoxically leads to an under-regulation of those very same directives. Unfortunately, as long as Israeli politics are plagued by a disproportionate power for religious political parties (as demonstrated in the case of Steinberg’s committee appointment and its influence on the phrasing of the DP Act), it would seem that the core reason underlying and leading to this paradox is largely unresolvable.

2 Advance Medical Directives in Singapore: A Faltering Policy for End-of-Life Care?

Tracey Evans Chan

2.1 Introduction

Advance directives rose to prominence in Singapore sometime in the mid-1990s when the National Medical Ethics Committee (NMEC) issued a report recommending enabling legislation that would recognise and give effect to individual patient decisions, made in advance of terminal illness, refusing certain forms of life sustaining treatment while the individual in question still possessed the capacity to do so. After an extensive consultative and legislative process, including deliberations by a Select Committee of Parliament, the Advance Medical Directive Act 1996 was passed and came into force on 1 July 1997. Advance directives considered in this volume encompass statements in which a competent person makes an advance decision in respect of healthcare, social welfare and other personal matters, which are to be implemented in the event that that person loses decision-making capacity at some later date. Pursuant to this definition, this chapter will consider the advance medical directive (AMD), which is a legislative instrument, and the common law advance decision. It then analyses reasons for the limited impact these have had on advance medical decision-making in Singapore, and observes the consequential shift to less formal and more consultative advance care planning (ACP) processes promoted by the Agency for Integrated Care in Singapore. Notwithstanding this development, it argues that there remains a legitimate role for an expanded scope of advance directives to offer a better range of options for end-of-life care to meet the needs of an evolving and diverse populace in Singapore.

2.2 The Legal Framework: Types of Advance Directives in Singapore

2.2.1 The Advance Medical Directive

The first advance directive introduced into Singapore law is the AMD, so named to make clear that it pertained only to medical decisions.¹ The impetus for enabling legislation was the need to address difficult ethical and legal issues related to the use and withdrawal of new medical technology that eventually proved futile in preventing death. Affording a patient the ability to indicate her treatment preferences in advance would better guide her physicians in making difficult medical decisions, especially in the face of differences of opinion amongst family members. It was also recognised then that various other jurisdictions facing the very same issues had introduced and legalised advance directives to allow patients the means to be spared futile medical treatment that only prolonged the dying process, and avoid placing such decisional burdens on their families.²

At that time, the Singapore courts had not yet addressed the legality and operation of advance directives as a matter of common law. Waiting for a sufficient body of precedents to build up to provide sufficient legal guidance was unpredictable and would take too long. Furthermore, the reception of common law precedents from other jurisdictions may not adequately take into consideration the principles and beliefs of the local community. The NMEC therefore recommended that legislation be introduced to remedy this deficit in the legal framework pertaining to advance medical decisions.³

2.2.1.1 The AMDA Framework: Making an AMD

Section 3 of the Advance Medical Directive Act (AMDA) provides that anyone who is not “mentally disordered”⁴ and at least 21 years of age may make an AMD in the form prescribed. In the schedule to AMDA Regulations, the prescribed form stipulates that the maker of a directive should not be subjected to “extraordinary life-sustaining treatment” if (a) the person suffers from a “terminal illness” and (b) is unconscious or incapable of exercising rational judgement. Two adult witnesses are required for a valid AMD, one of whom must be a medical practitioner (section 3(2)), and there are various disqualifications for witnesses on the bases of some interest in the estate of the person after her death, or conscientious objection to acting on an AMD (section 3(3)). It is the responsibility of the medical witness to take reasonable steps to assess the person executing an AMD and ensure that she is (a) not mentally disordered, (b) is at least 21 years of age, (c) doing so voluntarily, without undue inducement or compulsion and (d) has been informed of the nature and consequences of making an AMD.⁵ Finally, the duly executed AMD must be registered with the Registrar of AMDs appointed under the Act, failing which the directive may not be acted upon (section 5).

2.2.1.2 Capacity to Make an AMD

There is some uncertainty about the capacity threshold required in order to execute an AMD. Section 3 refers to a person who is not mentally disordered. This phrase is not defined in the AMDA, but the amending legislation that introduced it, the Mental Health (Care and Treatment) Act 2008, defines mental disorder as “any mental illness or any other disorder or disability of the mind”.⁶ Likewise, the required medical practitioner witness is expected under section 4(a) and (d) to take reasonable steps to ensure that the patient “is not mentally disordered” and “has been informed of the nature and consequences of making the directive” respectively. The threshold of mental disorder looks like a status-based conception of capacity, for not all persons suffering from a mental disorder or disability may lack the functional capacity to make an AMD.

Section 5(1) of the Mental Capacity Act 2008 (MCA 2008) adopts a functional conception of capacity, where an incapacitated person must be unable to (a) understand the information relevant to a decision, (b) retain that information, (c) use or weigh it as part of the process of making a decision, or (d) communicate his decision. Confusingly, this latter concept of capacity appears elsewhere in the AMDA, albeit in terms reminiscent of English common law’s description of functional capacity:⁷ under section 10 of the Act, the medical practitioner who contemplates acting on an AMD must not have reasonable grounds to believe that “the person was not, at the time of making the directive, capable of understanding the nature and consequences of the directive”. Likewise, section 19(1)(c) protects a medical practitioner from liability for a decision made in good faith and without negligence as to whether the patient was “capable of understanding the nature and consequences of the directive”. There is no mention in this section of a decision as to whether a patient was mentally disordered at the same point in time, although section 19(1)(d) refers to a decision by a medical practitioner as to whether a directive was valid, which may incorporate the absence of mental disorder prescribed by section 3 of the Act.

This apparently inconsistent conception of capacity within the AMDA could be resolved in two ways: The first is that mental disorder operates as a threshold floor to making an AMD; all mentally disordered persons are precluded from making an AMD, even if they possess sufficient functional capacity to do so. The requirement for functional capacity implied by section 19(1)(c) is then an additional requirement for persons who are not mentally disordered. This interpretation is problematic. It questionably excludes persons by reason of mental disorder, whether permanent or temporary, who are nevertheless functionally capable of making an AMD, and therefore devalues their autonomy. It is submitted that mental disorder should be purposively read, instead, as a mental disorder that prevents a person from being capable of understanding the nature and consequences of making an AMD (i.e. a functional capacity requirement, by reason of a mental disorder). The NMEC in its report recommending enabling legislation did not envisage such nuanced distinctions between status based and functional capacity, but simply required that persons be legally competent to execute an AMD.⁸ This interpretation would also render the capacity or competence requirement under the AMDA consistent with that which prevails more generally under the MCA 2008.⁹

2.2.1.3 Safeguards in Making an AMD

There are several provisions that protect the integrity and voluntariness of executing an AMD, which is based on the principle of respect for patient autonomy.¹⁰ Section 14(1)(a) makes it an offence to procure or obtain the execution by another person of an AMD, by means of “any deception, fraud, mis-statement, unconscionable conduct or undue influence”. It is also an offence to falsify or forge someone else’s AMD, or wilfully conceal or withhold knowledge of the revocation of an AMD.¹¹ Curiously, there is no proscription of deception or undue influence in a patient’s decision not to execute an AMD. This may have been an oversight, or perhaps reflects an implicit deference to the usual physician-led decision-making together with the patient’s family.¹² A conviction under section 14 will deem the AMD revoked and of no effect, while in the interim, no AMD may be acted upon unless it is determined that the AMD was validly and voluntarily made in accordance with the Act.

Section 15 makes it an offence for anyone likely to have medical care of any patient from asking if that patient has made or intends to make an AMD. The underlying rationale was to address a concern that patients who signed AMDs would be neglected by doctors in their medical care. The NMEC was of the view that the medical management of a patient should not be influenced by the existence of an AMD, and so a patient should not be forced to disclose to an admitting hospital or attending physician that she has made one.¹³ This seems to be an overreach as patients may not be aware of the need for and facility of an AMD. Thus, the duty or right of a medical practitioner to discuss or explore the concept of directives and the objectives and provisions of the AMDA are preserved in section 15(2) provided they are consistent with good medical practice, held in the context of a physician–patient relationship and in furtherance of the purposes of public education. Further professional ethical guidance in this respect, under the wider rubric of ACP (which may involve the execution of an AMD), was issued by the NMEC belatedly in 2010.¹⁴

Section 15(3) also requires that all information pertaining to a patient’s making of an AMD, or communications relating to an intention to make one, must be kept confidential by the medical practitioner or other medical worker having care of the patient. This might appear to contradict the recommendation of the NMEC¹⁵ and the Select Committee that the patient be encouraged to involve family members when making an AMD, and physician counselling and advice should extend to these family members.¹⁶ However, section 15(3) can be read harmoniously with this expectation if the patient consents to the inclusion of family members in discussions concerning the execution of an AMD.¹⁷ Finally, section 16 prohibits any person from making an AMD a condition for insurance or receiving medical or healthcare services.

2.2.1.4 Implementing an AMD

Terminal illness, together with incapacity,¹⁸ is the trigger for the eventual implementation of an AMD under section 9 read with section 10. It is defined in section 2 of the AMDA as:

an incurable condition caused by injury or disease from which there is no reasonable prospect of a temporary or permanent recovery where –

1. (a) death would, within reasonable medical judgment, be imminent regardless of the application of extraordinary life-sustaining treatment and

2. (b) the application of extraordinary life-sustaining treatment would only serve to postpone the moment of death of the patient.

Extraordinary life-sustaining treatment is defined in complementary fashion as treatment that would only postpone imminent death, but excludes palliative care – the provision of reasonable medical procedures to relieve pain, suffering or discomfort, and food and water.¹⁹ One immediate implication of the use of “terminal illness” as a trigger for implementing an AMD is the exclusion of patients in a persistent vegetative state and other prolonged disorders of consciousness.²⁰

There were suggestions before the Select Committee for the AMDA to define “imminent” in chronological terms, for example, within 6 to 12 months, but the committee preferred to leave this to medical judgement.²¹ Instead, the Act prescribes a rigorous process to certify terminal illness. The medical practitioner responsible for the treatment of a patient (‘attending physician’) must first certify in the prescribed form such terminal illness in order to search the AMD register to determine if the patient has made an AMD (recall, it is unlawful under the AMDA to directly ask a patient whether he has or intends to make an AMD). The attending physician is under a duty to do so where she “has reason to believe that the person is (a) suffering from a terminal illness; (b) requires extra-ordinary life sustaining treatment; and (c) is unconscious or incapable of exercising rational judgment”.²² If the search reveals a valid AMD made by the patient that remains in force, further certification of terminal illness is required from two other medical practitioners. Out of the three certifying physicians, two of them must be specialists recognised by the Director of Medical Services (DMS) for the purposes of the Act.²³ Where the two additional certifying physicians are not in agreement, the matter has to be further referred to a separate panel of three specialists, also appointed by the DMS under section 8 of the Act. Only a unanimous opinion of this separate panel will determine the patient terminally ill under the AMDA.²⁴ Otherwise, the AMD will not be acted upon.

Once a person who has executed a valid AMD has been certified as terminally ill in accordance with the procedure under the Act and is not pregnant with a viable foetus,²⁵ it is the duty of the attending physician to act in accordance with the terms of the AMD unless there is reason to believe that (1) the patient has revoked his AMD in any recognised manner or (2) was incapable of understanding the nature and consequences of the AMD at the time of its execution.²⁶ Recognising that persons do change their minds about preferences for end-of-life care, revoking an AMD is much easier than making one.²⁷ Section 7 allows revocation in writing, orally or by any other means of communication so long as there is at least one witness of this. The patient and the witness must notify the Registrar of AMDs as far as practicable of the fact of revocation in writing.

2.2.1.5 Consequential Provisions

In addition to imposing a duty on medical practitioners to implement an AMD once the statutory triggers have been satisfied, the Act makes several important consequential provisions. The physician who implements the AMD in good faith and without negligence shall not be subject to any civil or criminal liability, or professional disciplinary proceedings. A person acting under the instructions of such a medical practitioner is similarly protected in giving effect to an AMD unless he has knowledge of the revocation or intended revocation of the AMD in question.²⁸ Secondly, the implementation of a patient’s AMD does not constitute a cause of that patient’s death unless there was negligence in the certification of terminal illness.²⁹

2.2.2 The Common Law Advance Decision

Although Singapore has adopted the model of the United Kingdom’s Mental Capacity Act 2005 as its framework for handling the care and treatment of mentally incapacitated adults, it chose to omit sections 24 to 26 of that Act which deal generally with advance decisions to refuse specified treatment, which may relate to all medical treatment including the provision of artificial nutrition and hydration.³⁰ The drafter of the MCA 2008 chose to omit these sections presumably because the prior AMDA already deals with advance decisions in medical situations. However, because of the restrictive nature of the scope and operation of the statutory AMD, it is often asked if other advance decisions that fall outside the AMDA are legally recognised and enforceable at common law.

Although the provisions of the AMDA do not address this issue explicitly, it seems reasonably clear that the Act is not meant to preclude advance decisions under the common law. Section 13(1) states that that Act shall not affect the right of any person to refuse medical or surgical treatment, without attempting to restrict it to contemporaneous decisions. This standing common law right to make advance decisions concerning medical treatment is well recognised.³¹ In the local case of Re LP, the High Court recognised that “[g]enerally, a person who is sufficiently matured is entitled to give or withhold consent to any medical treatment and the doctors are entitled, if not obliged, to respect that person’s decision”.³² Although the court left open the question of whether living wills or American-type advance directives distinct from an AMD were applicable here,³³ its reasoning in authorising the amputation of the patient’s gangrenous legs held that the patient’s earlier statements to her doctors did not amount to a clear, express refusal of treatment presently proposed.³⁴ This appears to be an implicit endorsement of the right of a patient to refuse treatment in advance at common law. Furthermore, the NMEC’s proposals to legislate on AMDs made clear that the intention in doing so was to provide a clear and certain legislative route, while allowing for the continued development of ethical principles and common law in relation to advance directives and related issues.³⁵

Therefore, patients dissatisfied with the limited ambit of the statutory instrument could have recourse to a common law advance directive.³⁶ There are no prescribed formalities for this, although the common law places the burden on anyone seeking to rely on the terms of an advance decision to satisfactorily prove its existence and validity. However, where life is at stake, “clear and convincing” proof is required of its existence and continuing validity.³⁷ The ethical guidance from the Singapore Medical Council on end-of-life care stipulates that a patient’s wishes not to receive specific treatments must be respected,³⁸ while its Handbook on Medical Ethics states that, apart from an AMD, “any advance statement made by patients on care or treatment that they might accept or refuse in particular circumstances” is helpful in determining a patient’s values, concerns, wishes, and ultimately what is in their best interests.³⁹ Although the SMC’s Handbook on Medical Ethics pronouncements are more likely intended to assist in the determination of a patient’s overall best interests under the MCA,⁴⁰ the NMEC envisaged that discussions on ACP could lead to the drafting of an advance directive.⁴¹ In particular, the committee described advance directives in broader terms than an AMD, encompassing “oral and/or written instructions that convey treatment preferences in the event of a loss of decision-making capacity”.⁴²

However, this flexibility of form gives rise to potential evidentiary and interpretational uncertainties concerning the implementation of a common law advance decision. For example, there is no independent witness requirement at common law, and there is no equivalent statutory protection against civil or criminal liability or professional disciplinary proceedings in determining the validity of a common law directive. This may impede the respect for and implementation of such advance decisions. Unlike the AMD, there is also no statutory registration mechanism that might provide assurance that the terms of a written common law directive will be brought to the attention of attending physicians in the appropriate circumstances – but access to the AMD registry is an unwieldy process to begin with.⁴³

The Specific Licensing Terms and Conditions on Medical Records for Healthcare Institutions instead requires that a patient’s medical record include documents pertaining to her advance care plan if prepared.⁴⁴ As ACP may result in the documentation of an advance decision to accept or refuse medical treatment, which could arguably amount to a valid common law directive,⁴⁵ this must therefore be captured in the medical record as well. The National ACP Steering Committee and Agency for Integrated Care have also rolled out an ACP information technology system that captures key decisions on care options, and catalogues conversation transcripts and other supporting documents into a single record. Since April 2017, this system has been integrated with the National Electronic Health Record (NEHR), thus enhancing accessibility and availability across different institutional care settings.⁴⁶ It therefore becomes more likely that any common law advance directives made by a patient in the process of ACP will be brought to the attention of attending physicians throughout the public healthcare system with access to the NEHR. Nevertheless, under the Living Matters ACP initiative, the emphasis has been to deemphasise the completion of legal directives and promote iterative conversations and regular opportunities to review care plans to adapt to changing health and care circumstances.⁴⁷ There nonetheless seems to be less concern that any documentation on end-of-life planning will adversely influence the clinical care of the patient. In any case, the proscription on enquiring about the existence of an AMD does not apply to such instruments.

However, unless there is some standardisation of the language and intent behind such common law advance directives, uncertainties as to interpretation and circumstances of implementation will remain. In this respect, individual healthcare institutions are in the process of developing and implementing policies on how ACP documents are recorded and implemented.⁴⁸ ACP documents preferences and values on types of treatment depending on the stage of illness. Some of these preferences may be stated in directive terms, such as “DO NOT attempt CPR [cardiopulmonary resuscitation]”,⁴⁹ or indications that proxy decision-makers are to strictly follow the patient’s wishes.⁵⁰ On the other hand, the terms advance directive or decision are eschewed, and standard forms begin with the overarching statement that physicians “will always act in the patient’s best interests…”.⁵¹ This reference perhaps alludes to the best interests standard of proxy decision-making in the MCA 2008, where prior statements of wishes and feelings are to be considered but have no binding authority.⁵² This leaves some uncertainty about whether ACP documented preferences and wishes that are sufficiently precise and directive will be treated as having the binding effect of a common law directive. Nevertheless, the uncertainties in interpretation and confidence in implementation by healthcare professionals that originally concerned the NMEC in its 1995 report appear, to some extent, to have been addressed by the development of ACP policies, documents and protocols at the institutional level. Information technology has also afforded the means to overcome barriers to access for common law advance directives that are produced as a result of ACP. What is needed is clearer professional guidance on the recognition and implementation of common law advance directives, otherwise any documented preferences may simply be overridden by professional medical judgement of the patient’s best interests or family objection on similar grounds.

2.3 Assessing the Impact of Advance Directives in Singapore

If we look at absolute numbers, the take-up of AMDs in Singapore has thus far been very low. The Registry of AMDs does not publish statistics on AMD registrations and implementation, although the public is given periodic updates in the form of answers to questions filed in Parliament on AMDs. Between 1997 and 2008, a total of 10,100 AMDs were registered representing 0.4% of the resident population (three quarters of those between 2004–8), 19 were revoked and six were implemented.⁵³ By 2015, the last recorded disclosure on AMD statistics, 24,682 AMDs were registered and 10 AMDs were implemented, representing 0.7% of the resident population.⁵⁴ There are no apparent data on or research looking at common law advance directives, although in the parliamentary response in 2016 just mentioned, it was revealed that about 5,100 advance care plans had been completed between 2011 and 2015.

These numbers should not be surprising given the very restrictive ambit and formalities of the AMD. As mentioned previously, patients in a persistent vegetative state or other prolonged disorders of consciousness cannot depend on an AMD to refuse such treatment. Imminent death, as part of this definition, is also not defined in chronological terms but left to medical specialists to determine.⁵⁵ However, studies elsewhere and in Singapore reveal that it is very difficult to make a definitive prognosis of death even when it is very close.⁵⁶ Thus consensus between the required two or three medical practitioners certifying terminal illness will be less frequent. Extraordinary life-sustaining treatment hinges on terminal illness in order to be considered as merely prolonging the dying process. It further excludes palliative care, which includes the reasonable provision of food and water without specifying the means – commentators assume this exclusion will cover the reasonable provision of artificial nutrition and hydration.⁵⁷

The sum effect of these restrictions means that AMDs only cover situations of undeniable medical futility, for which the existence of an AMD adds little deliberative value to the decision to withdraw life-sustaining treatment.⁵⁸ Thus, local medical practitioners familiar with its terms have observed that there was no practical point in executing an AMD.⁵⁹ Part of the reason for this overly cautious legislative approach might have been the prevailing health policy zeitgeist at the time the idea of advance directives was first mooted by the NMEC in its position paper in August 1994.⁶⁰ This was not long after the Singapore Government had promulgated its White Paper on Affordable Health Care,⁶¹ which emphasised cost containment and sustainability in healthcare expenditure, along with individual responsibility for one’s health and to avoid over-reliance on state welfare or medical insurance. This juxtaposition, along with the common misunderstanding that AMDs were a form of passive euthanasia because they involve the withdrawal of treatment leading to death,⁶² may explain public concerns over neglect by medical practitioners once a patient had signed an AMD⁶³ or that the underlying motivation for recommending an AMD was healthcare cost containment.⁶⁴ These concerns led to stifling restrictions like making it a criminal offence for any doctor to ask if a patient had executed, or intended to execute, an AMD, thus making it that much harder for them to broach the issue as a legitimate part of end-of-life care. The NMEC has since clarified the ethics of communications in ACP, which may include the execution of an AMD or common law directive, thus providing a clearer path for physicians to raise these issues in clinical practice and, correspondingly, questioning the further need for such restrictions.⁶⁵

Even if the AMD were not so restrictive, it is questionable if patient responses would have been markedly different. Health decision-making practices in Singapore tend to adopt family-centric decision-making models that run counter to the individualistic, transactional model that the AMD is based on. Under the former model, many patients prefer to leave decision-making to trusted family members, and do not see it as their responsibility to make decisions, much less execute an advance directive.⁶⁶ An additional challenge is the not uncommon practice of collusion between families and physicians to withhold diagnosis and poor prognosis from a vulnerable patient, which also prevents active participation and independent decision-making.⁶⁷ Amidst such practices, the need and ability to exercise individual moral authority to take responsibility for one’s healthcare decisions do not arise at all. Others argue that in such a familial model of decision-making, advance directives are means of helping family members understand the patient’s voice in order to arrive at a family decision, which should not be dictated by the literal meaning of the advance directive.⁶⁸ Therefore it is not necessary in this thinking to institutionalise individualistic expression of prior wishes strictly by laws and regulation.

Finally, even if we were to put aside cultural values and practices, human psychology imposes various challenges in making good anticipatory decisions about future health scenarios with incomplete information. The literature raises doubts about people’s ability to accurately anticipate their reactions to serious illness, the stability of their life-sustaining treatment preferences even over short periods of up to two years (at least for a substantial minority) and their tendency to overestimate the stability of their attitudes and beliefs over time.⁶⁹ More recent psychological research on affective forecasting has led to better understanding of the systematic biases at play in making predictions about future preferences, apart from a lack of information about the future. One example is impact bias, which tends to predict greater intensity and longer lasting effects of future events than is actually the case. This is a result of the overweight of focus on the event itself, to the exclusion of the contribution of other factors to the predicted outcomes. Secondly, we also tend to underestimate our ability to adapt to negative events and habituate to positive ones.⁷⁰ As a result, individuals will often mis-predict their future preferences, and weigh relevant considerations differently depending on how far in the future a present decision is meant to apply. There is some correspondence between these findings and qualitative studies looking at the reasons why individuals do not see advance directives as useful,⁷¹ and in the experience with ACP and patient preferences in Singapore.⁷²

If we combine these insights based on history, culture, typical preferences for methods of medical decision-making and human psychology, we should question whether continued promotion of advance directives by improving the legal and supportive healthcare framework is the best way forward. Although there were discussions about amending the AMDA circa 2008, nothing came to bear on this.⁷³ Instead, in 2009, ACP was piloted in a Singapore tertiary hospital and subsequently scaled up to a national programme, “Living Matters”.⁷⁴ This approach to advance healthcare decision-making seeks to emphasise a relational, patient-centred process supported by trained healthcare professionals and trusted family members. It seeks to move away from the earlier, legally focussed, transactional process⁷⁵ towards a more open, inclusive and iterative communications process directed towards the same goal of enabling patients to retain control over their care once they lose decision-making capacity.⁷⁶ Ideally, when properly engaged, ACP will prepare patients and their proxy decision makers to make better in-the-moment medical decisions with their physicians on the basis of these communications, instead of advance decisions with incomplete information.⁷⁷

Even so, there will still be room for more flexible advance directives. Some patients have firmer preferences for end-of-life care, particularly when disease trajectories are relatively more predictable.⁷⁸ Family and social bonds may also be diminishing, with an increasing proportion of single adults and married persons without children.⁷⁹ This may result in a lack of adequate social support for effective proxy healthcare decision-making pursuant to an ACP programme. Differing social contexts suggest that the ACP model of decision-making may not suit all patients, some of whom may want more reliable means of ensuring respect for personal preferences in healthcare in the absence of trusted and willing proxy decision-makers. Studies elsewhere have revealed the existence of a stable minority of patients who want to retain control through the use of advance directives,⁸⁰ and the more informal ACP framework will need to be augmented to accommodate their preferred model of ACP.⁸¹

2.4 Conclusion

In summary, while Singapore was out of the gates very early as far as advance directives are concerned, AMDs have never really taken off in ACP and healthcare decision-making. For reasons that have to do with the social and policy climate at the time they were first introduced, prevailing cultural values and familial approaches to healthcare decision-making, and limitations on anticipatory decision-making in human psychology, the AMD initiative to jump start better respect for patient autonomy at the end of life has faltered. The current preferred Living Matters ACP programme seeks to learn from and avoid the difficulties experienced with AMDs, but it is argued that there remains a legitimate role for more flexible advance directives as the values, needs and expectations of the Singapore populace evolve.⁸²

3 South Korea’s End-of-Life Care Decisions Act: Law for Better End-of-Life Care

Ilhak Lee

3.1 Introduction

On 8 January 2016, the National Assembly of the Republic of Korea passed the Act on Hospice and Palliative Care and Decisions on Life-Sustaining Treatment for Patients at the End of Life (ELDA). The Act’s lengthy name arguably reflects the lengthy discussions in Korea about what constitutes acceptable end-of-life care. Although ELDA regulates end-of-life care in general, the legislators tried to narrow its potential application to the less-controversial condition of the “dying stage”. This chapter begins with a description of end-of-life care, especially of those features of such care for which remedies are needed, and then moves on to a discussion of the law and ethics of end-of-life care in Korea. My focus is on end-of-life care decision-making rather than palliative care. The implications of ELDA are also discussed.

3.2 Background and History

3.2.1 End-of-Life Decisions in the Courts

Whilst ELDA 2016 represents the most significant legal development concerning end-of-life care in Korea, a full understanding of the trajectory of legal developments in this area requires that we first look at two landmark court cases that preceded the law and played an important role in motivating legal reform and changes to professional practice: the Boramae Hospital case (1998) and Severance Hospital case (2008). The latter was a civil lawsuit brought against the hospital concerned to force it to forgo life-sustaining treatment (LST), and the former was a criminal case against surgeons who had discharged a patient at his wife’s request. The current legal regulation on end-of-life care decisions took its shape from the Supreme Court’s decisions in these two cases. It is important to note that both cases pertained to the forgoing of LST, not to euthanasia.

3.2.1.1 Boramae Hospital Case

This case dealt with the legitimacy of surrogate decision-making when the withdrawal of LST and discharge would mean the death of the patient.¹ The issue contested at trial was how far the duty of care extends when a family insists upon a particular treatment. The physicians at Boramae Hospital had discharged the patient concerned, whose treatment was deemed to be LST, at the request of his wife. She had requested that her husband be discharged for economic reasons, that is, the financial burden of continued hospitalisation. The final verdict arrived on 24 June 2004. The Supreme Court found the surgeons guilty and sentenced them to one year and six months’ imprisonment each. The court considered the surgeons to be guilty as accomplices to murder because they had (i) honoured the wife’s request, which could not be regarded as an authoritative proxy decision and (ii) not taken proper measures to prevent an anticipated harm.

This judgment rang alarm bells amongst Korean doctors, who became concerned about the decision-making process at the end of life. The judgment confirmed that doctors’ duty is this: treating a patient as an autonomous individual and considering his or her best interests under any circumstances.² However, confusion remained over whether every withdrawal of LST would result in liability or whether it was sufficient to consult all family members before any decision was made. Such confusion arose in part because there was no explicit written provision in Korean law. Further, most – but not all – decisions were made jointly by doctors and patients’ family members on the assumption that the latter were conferred with the power to consider what is best for the patient, even though, as noted, there was no explicit legal provision for this.

Despite the serious ethical discussions on end-of-life care following the Boramae Hospital case, no formal mechanisms such as advance directives (ADs), durable power of attorney or clinical ethics consultation were implemented. Accordingly, many professional bodies were quick to develop guidelines to realise the spirit of the Supreme Court’s decision. For example, the Korean Association of Medical Societies (KAMS) in 2002 developed guidelines on the forgoing of LST for dying patients that provide a clinical pathway resembling that of the American Medical Association.³ Although the KAMS guidelines were developed and modified to fit the Korean context, they did little to improve the practice of end-of-life care planning, primarily because they have little binding force: there is no duty for doctors to follow the KAMS guidelines. Accordingly, doctors wanted a secure legal basis for following a patient’s wishes, such as a clear exemption from liability. That desire stemmed from fear of the heavy penalty imposed by the Supreme Court in the Boramae Hospital case, not least the cancellation of the guilty doctors’ medical licences.

It was not only the medical community that was concerned by the case; there were also concerns amongst bioethics activists that this decision would lead to euthanasia being legislated. The controversy continued for several years until the Severance Hospital case in 2009, which triggered the start of the ELDA legislation process.

3.2.1.2 Severance Hospital Case

This case concerned a request made by the family of a patient, a woman in her late 70s who was in a persistent vegetative state, to discontinue the medical treatment she was receiving. The family filed a lawsuit against Severance Hospital, which had refused to stop the treatment. The trial process in the case was unusually quick, with the Supreme Court ruling in 2009 that the withdrawal of LST could proceed based on the patient’s presumed intent. The patient had once made a similar decision for her husband, and her preferences concerning LST were thus deemed by the court to be recognisable from her life attitudes. Notably, the Supreme Court imposed a limitation on the withdrawal of LST, ruling that “when it is recognised that a patient who has reached an irreversible stage of death exercises her right to self-determination on the ground of the constitutional rights of human dignity and right to pursue happiness, it is permissible to withdraw life-sustaining treatment”.⁴

The Severance Hospital case was important in the sense that it was the first case in which a Korean court recognised a patient’s right to refuse futile LST. Whilst the Supreme Court’s decision apparently recognised patients’ right to self-determination with respect to LST, we must take care not to interpret it as confirmation of the right to refuse life-saving treatment in general. There is a condition that must be fulfilled before a patient’s treatment preferences can be executed: “he/she (should) have reached an irreversible stage”. This condition leaves little room for interpretation. Nevertheless, the spirit of the court’s decision in this case is that a patient’s right to refuse treatment should be institutionalised by legislation on end-of-life medical decision-making. Hence, the legislative process began.

3.2.2 Consensus Building Process (2009–2012)

The period prior to ELDA’s passage saw the Government set up various consultation bodies to facilitate social consensus-building in relation to end-of-life decision-making issues.⁵

3.2.2.1 Ministry of Health and Welfare’s Council on the Institutionalisation of Forgoing Life-Sustaining Treatments (2009–2010)

In 2009, the Ministry of Health and Welfare responded to the Severance Hospital case by organising the Council on the Institutionalisation of Forgoing Life-Sustaining Treatments. The council released its report in 2010. The minimal consensus the council members reached, which subsequently served as the starting block for a public engagement process, was as follows. The permissibility of LST withdrawal was to be restricted to “terminal patients” and to certain treatments. Nonetheless, the report recommended documenting end-of-life care preferences to make them explicit and establishing a review committee and conflict resolution mechanism. The consensus report can be summarised as follows:

1. (i) The forgoing of LST is permissible only for terminal patients; patients in a persistent vegetative state are excluded unless they are in the terminal stage.

2. (ii) The only treatments that can be withdrawn are extraordinary LSTs such as cardiopulmonary resuscitation (CPR) and artificial ventilation; hydration and nutrition should not be withdrawn.

3. (iii) An “Advance Medical Directive (or Advance Medical Intention Documents)” should be written if a patient in the terminal stage wishes to express his or her preferences. Notably, an adult can write such a document after consulting with the attending physician following two weeks of mature deliberation.

4. (iv) A National Terminal Care Review Committee under the Ministry of Health and Welfare and Hospital Ethics Committees in individual healthcare institutes should be established to facilitate the decision-making process in the end-of-life care context.

The council’s 2010 report constituted baseline consensus amongst Koreans regardless of their attitudes towards the scope of end-of-life decisions. The National Bioethics Advisory Committee then joined the discussion by establishing a task force under the auspices of the National Bioethics Review Committee in 2013.

3.2.2.2 National Bioethics Review Committee’s Task Force for Decision-Making about Futile LST and Recommendations of the National Bioethics Advisory Committee (2012)

In 2013, the National Bioethics Review Committee, established under the Korean Bioethics and Safety Act 2005, launched the Task Force for Decision-Making about Futile Life-Sustaining Treatment with the aim of giving the council’s consensus concrete shape.⁶ The task force’s final report discussed six issues: basic principles, potential patients, medical interventions, identifying a patient’s preferences, social engagement and methods of institutionalisation.⁷ Although the National Bioethics Review Committee announced “agreed” recommendations for drafting an act, concerns remained over the potential abuse of end-of-life care decisions, and there were also demands for a more permissive law allowing patients to decide.

As for the basic principles of medical decision-making concerning LST, these were determined to be (i) a patient’s right to an informed decision, (ii) respect for self-determination, and (iii) the provision of hospice palliative care. The agreed recommendations limited the patients of concern to those in an irreversible condition with no hope of a cure and those with a rapidly aggravating condition that does not respond to treatment of the underlying disease. This narrow perspective can also be found in the treatments of concern, which are restricted to LSTs that require professional medical knowledge, skills and equipment, namely, extraordinary LST such as CPR, ventilator care, haemodialysis, anticancer chemotherapy and so forth. Patients can choose hospice palliative care. As for the method of identifying a patient’s preferences, the recommendations prioritised a patient’s explicit preferences over other means, such as a presumed will or surrogate decision-making. In the case of incompetency, documentation was recommended.

The task force’s recommendations recognised the importance of the cultural and socioeconomic environments and the role of collaboration for improvement, although such recognition appeared to be largely symbolic, with very few follow-up actions apart from the establishment of a National End-of-Life Day and Ceremonial Day and social media and other campaigns promoting the concept of self-determination. The National Bioethics Review Committee’s final recommendation was the institutionalisation of end-of-life care decisions through legislation.

3.2.3 Legislation

Before the successful legislative attempt in the 19th session of the National Assembly (2016), there had been seven bills drafted since 2006, namely, the Revision of the Medical Practice Act (2006), Hospice-Palliative Care Act (2008), Death with Dignity Act (2009), Law on the Right to a Natural Death at the End Stage of Life (2009) and three additional bills drafted during the 18th session and then again in the 19th. These draft bills, each of which had a different emphasis, were unsuccessful, perhaps because the public was unclear about the nature of futile treatment and there was insufficient societal pressure to engage with such issues more actively. It is noteworthy, however, that the Death with Dignity campaign subsequently took hold within Korean society.⁸ Owing to newspaper coverage of the Severance Hospital case and to the “dying well” campaign, the broader population gradually began to understand the importance of the right to self-determination and the possibility of dying free of any unnecessary medical interventions.

In 2015, a National Assembly member submitted a bill based on the recommendations of the National Bioethics Review Committee, and three other competing bills were also submitted. At a later stage of the legislative process, the bill on end-of-life decision-making and the bill on hospice palliative care were merged. In a sense, palliative care and end-of-life decision-making are closely related, and both had hitherto received less attention than they deserved. The new law came into force in July 2017, and its executive orders – a presidential decree and enforcement regulation – were legislated thereafter.⁹

As noted at the outset of this chapter, ELDA was passed on 8 January 2016. The Supreme Court had in the Severance Hospital case recommended legislation on the issues concerned in 2009, and thus the Act took many years to execute. ELDA’s goal is “to secure the patient’s best interest and to protect … human dignity and value by respecting self-determination” (Article 1 of ELDA). To achieve that goal, the Act is equipped with two mechanisms: the provision of hospice palliative care to terminal patients and clarification of the end-of-life care decision-making process. With respect to the former, ELDA declares the right of patients to hospice palliative care and the responsibility of healthcare providers and states to provide it. With respect to the latter, the legislation provides clinicians and patients with a decision-making framework, as well as the interpretation and application of relevant principles, which can potentially remove the confusion and settle the debate amongst Korean society.

3.3 Characteristics of ELDA¹⁰

3.3.1 Principles of End-of-Life Care and the State’s Responsibilities

3.3.1.1 Principles of End-of-Life Care

ELDA emphasises three principles in end-of-life care: human dignity and value, patients’ right to know and to self-determination and the duty of medical professionals to provide the best care and information and to respect a patient’s decision (Article 3 of ELDA). Whilst the language of the law has raised awareness of the concept of autonomy amongst patients and practitioners, the culture (and even the interpretation of the law) generally continues to place the best interests of the patient first. The tension between autonomy and best interests appears to have been dealt with by placing limits on autonomy and the scope of the right to refuse treatment.

3.3.1.2 State’s Responsibility to Improve Quality of End-of-Life Care

There have long been concerns about the poor quality of death amongst Koreans, with most commentary suggesting palliative care as an alternative. However, palliative care had remained only an ideal, not everyday practice, for several reasons, not least the fear that recommending hospice care constituted a recognition of failure by physicians and the abandonment of patients.¹¹ What was notable for Korean legislation was that the opponents of ELDA argued that good hospice care should precede end-of-life decision legislation.¹² It was partly as a result of that argument that the legislative strategy of merging the hospice palliative care bill with ELDA surfaced in the last phase of parliamentary review.

ELDA provides a justification for the state to operate nationwide hospice palliative care (which had in fact been part of the state’s responsibility since 2005, as defined by the National Cancer Control Act). It mandates that the state establish a discussion body (the National Hospice and Palliative Care Committee) and submit and implement a national plan (General Plan for Hospice and the Provision and Withdrawal of Life-Sustaining Treatment) (Article 8 of ELDA). The new legislation authorises the Ministry of Health and Welfare to plan and implement programmes. To a certain extent, ELDA can be seen as an extension of hospice palliative care to disease categories other than cancer, as well as a clarification of the Government’s authority to set standards for and support hospice palliative care providers.

3.3.2 Scope of the Application

Article 2 of ELDA provides definitions and defines the subjects of application. Interestingly, the law distinguishes the terminal stage of disease from the dying stage and terminal patients from dying patients. A “terminal patient” is defined in Article 2 as

a patient who has been diagnosed as expected to die within a few months [by] the doctor in charge and one medical specialist in the relevant field in accordance with the procedures and guidelines prescribed by Ordinance of the Ministry of Health and Welfare, because there is no possibility of a fundamental recovery, and the symptoms [will] gradually worsen despite proactive treatment.

He or she is also regarded as a potential hospice palliative care beneficiary, but one who is unable to elect the termination of LST. The article originally listed four diseases as terminal illnesses: cancer, AIDS, chronic obstructive pulmonary disease and chronic liver cirrhosis. However, other diseases were added by ordinance of the Ministry of Health and Welfare following the 2018 revision. Patients at the dying stage, or in the “end-of-life process”, are defined as in “a state of imminent death, in which there is no possibility of revitalisation or recovery despite treatment, and [whose] symptoms [are] worsen[ing] rapidly” (Article 2 of ELDA).

Article 2 of ELDA also originally limited the potential LSTs that could be foregone to four specific medical interventions: CPR, haemodialysis, anticancer chemotherapy and artificial ventilation care. This narrow definition was designed as a safeguard against the possible hasty withdrawal of LST but was criticised for ignoring clinical realities. It was thus subsequently revised to include (i) extracorporeal life support, (ii) transfusions, (iii) the infusion of hypertensors and (iv) any procedures medically assessed by the physician to be withheld or withdrawn in the patient’s best interests. It can thus be seen that the legislative approach moved away from narrow definitions towards leaving the determination of whether a given treatment constitutes a futile LST to clinical judgement. As for how such judgement should be exercised, there is published professional guidance that suggests, for example, how a patient’s vital signs should be interpreted. This approach demonstrates an understanding that not everything can be specified by law, and hence that the application of the law should be left to professional guidelines.

3.3.3 Institutionalisation of End-of-Life Care Decisions

According to ELDA, the state can establish several institutions to support the making and executing of end-of-life decisions on LST. Three official institutions have accordingly been established: the National Agency for the Management of Life-Sustaining Treatment (established under Article 9), the Agency for the Registration of Advance Statements for Life-Sustaining Treatment (ARAS) (established under Article 11) and Institutional Ethics Committees (IECs) (established under Article 14).

The National Agency operates the national end-of-life infrastructure, provides certified education and handles public relations. It works as a national AD registry, identifying and confirming LST plans (LSTPs) and ADs, and is accountable for producing and managing databases of ADs for LST, which cover both AD and LSTP registries. The National Agency also provides research and statistical information on end-of-life decision-making and the execution thereof and is responsible for granting licences to institutions such as hospitals to become registering institutions.

ARAS oversees the private and public institutions that are the focal point of end-of-life decision-making services for the public. These institutions, which can be healthcare institutions, public health centres or qualified non-governmental organisations, provide information to patients (see Section 3.4.2 for further details) and transfer the documents that patients create.

Finally, IECs have similar functions to clinical ethics committees. Their various functions include (i) deliberation (including the review of consulting requests from patients and physicians and requests to change the attending physician) and the review of ADs (although they do not decide on them); (ii) the education of professionals; and (iii) reporting and referral to the National Agency. Physicians may also consult an IEC on a particular prognosis or to seek legal clarification. Whilst IEC recommendations carry some weight, they are not technically binding. In light of the IECs’ wide range of functions, they are expected to become the focal point of reporting and communication with the National Agency.

3.4 Application of ELDA

ELDA provides for two processes relating to end-of-life decision-making, which will be referred to here as LSTPs and Advance Statements for LST (ASLSTs). The Act requires that a patient be in the “dying stage” before his or her decision regarding LST under either an LSTP or ASLST can be executed. Although a patient’s decision on LST under a physician’s order for LST (Physician Order for Life-Sustaining Treatment, POLST) is binding, decisions under a valid ASLST represent only the patient’s (explicit) preferences and are not binding until and unless the ASLST is verified. The two processes are discussed in further detail next.

3.4.1 Life-Sustaining Treatment Plans

According to Article 10 of ELDA, an end-of-life care plan takes the form of an LSTP. Under this regime, a doctor may supply the patient with the information required to prepare a plan to terminate LST or, at the patient’s request, prepare an LST plan for a terminal patient at a medical institution. Although the process can be initiated by a patient or the doctor in charge, physicians are seen as playing the leading role.

The process of preparing an LSTP under ELDA is akin to advance care planning, but is limited to LST during a terminal patient’s last days of life. The patient may in his or her LSTP specify matters relating to the termination of LST, as well as matters concerning the use of hospice care (Article 10(4) of ELDA). As safeguards, Article 10 of ELDA prescribes formalities for LSTPs and provides requirements for the registration of POLSTs at a medical institution, record-keeping and notification of the National Agency.

3.4.2 Advance Statement for Life-Sustaining Treatment

In Korea, an ASLST can be viewed only as evidence of a person’s preferences regarding LST until and unless it has been verified, as discussed in further detaillater. At the point of verification, it becomes binding on the doctor. An ASLST is a document similar to a living will. It is written when a person is relatively healthy and independent. It contains the person’s decisions regarding (i) the termination of LST and (ii) the use of hospice care where necessary (Article 12 of ELDA). There are no legal requirements in relation to either witnessing or how the person’s mental capacity is to be assessed. ELDA does, however, mandate that personnel from the registering institution meet with the person to confirm that he or she understands the following matters before the ASLST can be considered valid.

1. i. Matters concerning the methods of implementing LST and making a decision to terminate such treatment.

2. ii. Matters concerning selecting and using hospice care.

3. iii. Matters concerning the validity and invalidity of ASLSTs.

4. iv. Matters concerning the preparation, registration, keeping and notification of ASLSTs.

5. v. Matters concerning amending and withdrawing ASLSTs and subsequent measures.

6. vi. Other matters prescribed by ordinance of the Ministry of Health and Welfare.

The definition of registering institutions is construed narrowly under Article 11(1) of ELDA as a safeguard, limiting them to regional healthcare institutions under the Regional Public Health Act; medical institutions; public institutions under the Act on the Management of Public Institutions; and non-profit corporations registered with the Government that have the proper qualifications and trained personnel to provide services related to ASLSTs. Although registering institutions must be licensed, their personnel need not be medically or legally trained and may be representatives of the medical institution concerned or even volunteers. As they may have only a basic understanding of ASLSTs, it is not uncommon for the assessment of whether a person is mentally capable of making an ASLST to be left to the institution’s nurses or volunteers. After the ASLST is prepared, it has to be registered and kept at the registering institution (Article 12(4) of ELDA), and the National Agency must be notified of the registration (Article 12(5) of ELDA). Licensed healthcare professionals will then be able to access this ASLST when a decision regarding the use of LST or hospice care needs to be made.

As noted, although a registered ASLST is valid under ELDA, it can serve only as evidence regarding the patient’s LST preferences. For an ASLST to be binding, the doctor must verify the patient’s intentions with him or her when the ASLST is retrieved for use (Article 15 of ELDA). When the patient is in the “end-of-life process” but is still of sound mind, the requirement to verify the patient’s intentions does not pose much of a hurdle. In situations where the patient becomes incapacitated and is unable to verify the ASLST, however, the doctor needs to obtain the agreement of a second doctor (Article 17 of ELDA), and during this confirmation the doctors typically have a discussion with the family about the authenticity of the document. An additional complexity arises when a patient in the “dying stage” becomes incapacitated and has no ASLST. The current law does not provide for the designation of a proxy for LST-related decision-making, although it is worded such that the patient’s family members can bear witness to his or her LST preferences, which means that they are effectively able to make LST-related decisions on behalf of the patient in the event of incapacity. In such a situation, the doctor will consult with family members, and if they unanimously bear witness to a consistent expression of the patient’s preference, then that preference will be regarded as the patient’s will (in other words, his or her presumed will), and the doctor will need to follow it (Article 17, Clause 2 of ELDA).

3.4.3 LSTPs versus ASLSTs

There are thus two legal mechanisms by which individuals can indicate their preferences in relation to LST under ELDA, with LSTPs having priority over ASLSTs. That priority explains much of the confusion surrounding the force and application of ASLSTs. If an LSTP is made, any ASLST made prior to it becomes invalid. There are also many more hurdles to overcome before an ASLST can be made and/or implemented, including confirmation of the six matters listed previously and verification of the patient’s intentions (where the patient is capacitous), which reflects a rather reserved attitude towards ASLSTs.

There have been attempts to give more weight to ASLSTs, but there seems to be no appeal in legislation. In fact, the original objective was for LSTPs to support the implementation of ASLSTs, with an AD being a small but important component of giving a patient full control over a key decision within the larger process of planning his or her end-of-life care. However, that objective appears to have been lost in the legislative process. ASLSTs have yet to move past being an overarching statement of patient preferences that assists physicians in deciding on the use of LST. We will likely have to wait for another legislative opportunity for ASLSTs to reach their full potential. In the meantime, it is likely that they will remain marginalised or perhaps become even further marginalised given the reality of decision-making by patients and families in the Korean context, as we will see in the following section.

3.5 End-of-Life Decision-Making in Practice

3.5.1 Low Utilisation Rate

Equipping patients with the power to decide what end-of-life care, LST in particular, best suits them is closely tied to the general well-being of the population. Such end-of-life decision-making mechanisms as LSTPs and ASLSTs enable patients to communicate their preferences regarding the use of LST and hospice care, allowing sufficient planning for a death with dignity and reducing the need for futile LST. However, as of January 2020, 85,076 LST decisions (LSTD) had been implemented, with the number of decisions made through family statements or agreements (55,775) nearly twice as high as the number made using an LSTP or ASLST (29,301). The LSTD system thus appears to be based on family decisions, which may or may not be based on patients’ preferences or wishes, rather than on direct records of those preferences/wishes, and less than 20 per cent of dying people seem to be benefiting from the legislation. One potential reason is the age and health status of those who currently make an AD, and the situation should improve as awareness of AD use grows and more healthcare professionals gain experience and knowledge of using ADs for end-of-life decision-making. Other reasons for the low utilisation rate of LSTPs and ASLSTs may be linked to the sociocultural context of Korea, which is explored in further detail next.

3.5.2 Sociocultural Context

The two key sociocultural factors that play an important role in end-of-life decision-making in Korea are the role of the family and the role of Confucian values.

In relation to first factor, as in other Asian countries, the life of the individual in Korea is strongly tied to the family. Individuals live for the family and by its support. The dignity of the individual is seen to come from the family, especially from generational relations. Accordingly, decisions regarding end-of-life care tend to be taken collectively. A study of end-of-life communication amongst elderly Koreans found that less than 20 per cent of patients prefer making treatment decisions alone,¹³ with over 90 per cent of cancer patients and their caregivers preferring family involvement in treatment decision-making.¹⁴ Family members do not perceive their involvement as infringing the patient’s autonomy, but rather as a means to show support and alleviate the patient’s burden.¹⁵ There is thus less emphasis on the individual in end-of-life decision-making in Korea than in the West: patients generally do not express their preferences for a given treatment, and the family can be expected to take an active role¹⁶ or even decide on the patient’s behalf when it comes to life/death decisions.^{17, 18}

In relation to the second factor, the current underutilisation of LSTPs and ASLSTs reflects the ethos of a traditionally Confucian society such as Korea,¹⁹ particularly with respect to the elderly. Confucian values dictate that the care of elderly parents is the duty of children,²⁰ and the elderly are, in turn, often dependent – financially and physically – on their children for healthcare and are expected to follow the decision of the family, particularly with respect to decisions pertaining to such value-laden issues as birth, marriage and death. The concept of Hyo (meaning filial piety) is also a cardinal Confucian virtue that plays a role in end-of-life decision-making. For example, it is common for children to feel obliged to continue LST for their family members because they equate providing medical treatment with being filial.²¹ Furthermore, filial piety requires children to keep their parents comfortable for as long as possible. As many Koreans consider knowledge of a poor prognosis to be painful and even unbearable, information relating to their condition is often concealed from patients until the very end of their illness trajectory.²² Given that research demonstrates a strong association between a patient’s understanding of his or her condition and his or her preference for making an AD,²³ such concealment of information likely contributes to the low utilisation rate of LSTPs and ASLSTs.

3.6 The Way Forward

In considering the way forward for Korea, we need to address some of the concerns with the current legislation.

3.6.1 Limited Autonomy

Although patients have opportunities to express their preferences to the people who care for them, such as their doctors, family members and other caregivers, the current law may limit their ability to exercise their autonomy, for example, by viewing an ASLST only as evidence of a person’s preferences rather than a binding statement and permitting surrogate decision-making by family members when the person is incapacitated. The result could well be too much emphasis on the family’s opinion at the expense of the patient’s. In addition to the risk of family members misinterpreting patients’ preferences, there is also the possibility of family members putting forth statements contrary to the ASLST if they do not agree with it and the patient is incapacitated. Revising ELDA in such a way that a valid ASLST always trumps the family’s decision-making authority would be one way to ensure that patients’ autonomy is strengthened in incapacitation scenarios.

In addition, the scope of ELDA’s application is overly narrow. For example, it applies only after a patient’s condition has been confirmed by two physicians to be irreversible and not responsive to treatment. Moreover, there are just four types of LSTs that can be withdrawn, namely, CPR, haemodialysis, the administration of anticancer drugs and mechanical ventilation (Article 2 of ELDA). These restrictions leave little space for patients’ preferences to be afforded much significance.

3.6.2 ELDA’s Inability to Reflect Societal Changes

ELDA adopts a narrow definition of “family members”, requiring them to be of the same bloodline (lineal ascendants, descendants and siblings) as the patient with the single exception of the patient’s spouse. In reality, not only are unregistered partnerships on the rise in Korea, but indirect family members (e.g. nieces and nephews) are also increasingly becoming patients’ primary or sole caregivers. As discussed previously, family members have a role in bearing witness to the patient’s preferences regarding LST, and, under the current legislation, that role is limited to family members as narrowly defined in ELDA to the exclusion of other caregivers who might be better able to offer evidence of those preferences. Furthermore, recent studies have shown a decrease in the younger generation’s perceived duty to take care of elderly parents.²⁴ This shift in perceived responsibility and levels of involvement may also affect the feasibility of verifying a patient’s preferences through their family members. To adapt to Korea’s changing needs, it is worth considering whether ASLSTs could be implemented without further verification, particularly given that there are already procedures in place to confirm a patient’s understanding prior to the making of an ASLST.

In terms of the way forward, patient empowerment should be the first priority. The current legislative framework permits too large a role for the family in end-of-life decision-making. This is not to say that the family should not be engaged in the process at all, but rather that enabling patients to exercise their right to know and have their preferences respected should come first. From a legislative point of view, ELDA currently leaves many matters relating to ASLST validation open ended, and the legislative framework would thus benefit from a more systematic, well-planned validation process for ADs.²⁵

3.7 Conclusion

ELDA is the result of experience and a painstaking consensus-building process, although the law is narrow in application and areas for improvement remain. What Korean society has agreed is that every dying patient deserves respect and humane care and that such care is something that society should provide. As with other social institutions, however, ELDA and related social institutions need further improvement, particularly with regard to the scope and application of ADs. Accordingly, continued discussion of ELDA and the implementation of end-of-life decision-making tools is necessary. Respectful communication and collaboration amongst all relevant stakeholders will be necessary to balance the opinions of the public with empowering patients in the end-of-life decision-making process.

4 The Law and Practice of Advance Directives in Taiwan

Daniel Fu-Chang Tsai

4.1 Introduction

Achieving a good death (善終) is regarded as one of the Five Blessings (五福) of life in traditional Chinese culture. The introduction of modern Western medicine and the practice of cardiopulmonary resuscitation (CPR) and other forms of life-sustaining treatment (LST), however, made a good death difficult to achieve for many of those accepting end-of-life care in hospital settings. Through three decades of effort to promote hospice and palliative care, Taiwan has developed comprehensive and pioneering legislative frameworks to govern advance directives (ADs) and end-of-life decision-making more generally. In 2015, Taiwan was ranked number six in the world and number one in Asia in terms of “quality of death” by The Economist. Despite these advances, however, the uptake of ADs in Taiwan remains low, which may be the result of insufficient understanding and awareness, as well as sociocultural factors such as an emphasis on the role of the family in medical decision-making. This chapter begins with an introduction to the legal regulation of ADs in Taiwan, followed by a discussion of the available data on the practice of ADs and an exploration of the possible reasons for their low uptake. The chapter concludes with several suggestions for the way forward.

4.2 The Legal Regulation of Advance Directives in Taiwan

The regulation of ADs in Taiwan is governed primarily by two legislative frameworks, the 2000 Hospice Palliative Care Act (HPCA) and the 2019 Patient Right to Autonomy Act (PRAA). No court decisions to date have independently recognised the right of terminally ill patients to withhold or withdraw LST because the use of court decisions to resolve end-of-life care disputes is inefficient under the continental law system of Taiwan, nor is it an acceptable practice in medicine.

4.2.1 Legislative Background

Although the first hospice (安寧病房, or palliative care ward) in Taiwan was established in 1990 in a Christian hospital (MacKay Memorial Hospital),¹ and many academic medical centres have begun to establish palliative medicine wards, only a small portion of terminally ill patients are able to receive hospice care owing to the limited service volume and such care being an unfamiliar concept. Because the “doctor’s duty to rescue” was emphasised in the earlier version of the Medical Care Act², many terminally ill patients received painful and burdensome CPR and other forms of LST in their final days in hospital. There was no definitive regulation or protocol concerning the withholding or withdrawal of LST from terminally ill patients prior to the HPCA’s enactment in 2000.

In response to the increasingly urgent need for Taiwan to establish clear legal procedures in relation to end-of-life care, a joint effort by healthcare professionals, social and religious leaders, legislators, and government officials was launched in 1993, with the HPCA finally being passed in 2000 after years of discussion and negotiation.³

4.2.2 Hospice Palliative Care Act⁴

The goal of the HPCA is to ensure that patients who are terminally ill have the freedom to choose hospice palliative care and refuse LST,⁵ as demonstrated by Article 1, which states that the Act’s aim is to respect the wishes of terminally ill patients in relation to medical treatment and to protect their rights. A terminally ill patient is defined in Article 3 of the HPCA as an individual (i) who suffers from serious injury or illness, (ii) who has been diagnosed by a physician as incurable and (iii) for whom there is medical evidence showing that the inevitable prognosis is death within the near future.

Whilst the term “advance directive” is not used in the HPCA, Article 4 establishes a mechanism called the “letter of intent” by which a terminally ill patient can make a decision about whether he or she wishes to receive hospice palliative care or LST. To make a letter of intent, an individual needs to be a capacitous adult (Article 5).⁶ The letter, which should state the individual’s wishes regarding the receipt of hospice palliative care or LST, should also contain his or her personal information and the date on which the letter was written (Article 4). The signing of the letter of intent should be witnessed by two individuals with full capacity. However, neither healthcare workers who are affiliated with medical institutions that provide hospice palliative care nor the individuals who will carry out the individual’s wishes with respect to LST are permitted to act as witnesses to avoid conflicts of obligations or roles (Article 4). Individuals can withdraw their wishes in writing at any time, either on their own or via their medical surrogate agent (Article 6), who can be appointed as part of the letter of intent (Article 5).

Once the letter of intent has been signed, the relevant competent authority (as defined in the HPCA) will note the letter’s existence in the individual’s national health insurance certificate, and the letter will be stored as an electronic file in the central competent authority’s database. If the contents of the letter of intent are inconsistent with the written wishes specified by the individual during the consent process prior to clinical treatment, the contemporaneous written consent, or refusal of consent, will prevail. In terms of application, where there is a signed letter of intent refusing consent to CPR or other forms of LST, these interventions cannot be administered as long as two physicians have confirmed that the patient is terminally ill (Article 7). If CPR or other LST has already begun, and the foregoing conditions have been satisfied, Article 7(5) allows such treatment to be terminated or withdrawn.

Where there is no letter of intent, and the patient is unable to express his or her wishes clearly, Article 7(3) of the HPCA provides that consent can be given instead by his or her closest relative, or by the physician in the patient’s medical best interests, although it is explicitly stated that neither of these decisions can contradict the patient’s previously expressed wishes. Article 7(4) provides a list of “closest relatives”, which are defined broadly to include third-degree relatives and first-degree relatives by marriage. In the event of disagreement between a patient’s closest relatives, the order in which the relatives are listed in Article 7(4) determines their priority. Article 7 defines the essential elements and practice of ADs for the withdrawal/withholding of CPR/LST from terminally ill patients. It also indicates the priority ranking of decision types: first, the patient’s personally signed AD/letter of intent; second, an AD/letter of intent signed by the patient’s surrogate or closest relative; and, third, a doctor’s medical order when the patient has no close relatives. In the third condition, the doctor’s decision must be made in the best interests of the patient after consultation with a palliative care specialist. None of these surrogate decisions should be contrary to a patient’s formerly expressed wishes.

The HPCA stipulates strict penalties for the violation of Article 7, which provides for the circumstances in which CPR or LST should be withheld. Article 10 states that physicians who violate Article 7 will be fined a sum between NT60,000 and NT300,000, and will either be suspended from practice for a period of between one month and one year or have their licence revoked completely. In terms of individuals who choose to end the lives of their family members without following the procedures laid down by the HPCA, court decisions suggest that such actions will be considered homicide.

4.2.3 Concerns with the HPCA

After the HPCA’s passage in 2000, clinical practice relating to end-of-life care in Taiwan changed gradually but remarkably. Both patients and doctors began to embrace the concept of respecting patients’ autonomous decisions at the end of life and to recognise that refusing CPR and LST was acceptable as long as the relevant procedures and documentation required by the HPCA were followed.

However, the HPCA has a number of limitations. First, it has been argued that the Act’s protection of patient autonomy is insufficient. For example, Article 8 requires that physicians inform terminally ill patients or their family members of the patient’s condition, hospice palliative care treatment and LST choices. Article 8 further stresses that patients are to be informed if they explicitly express the desire to be informed of their condition. Although this stipulation, on the one hand, protects a patient’s legal right to know about his or her own diagnosis, on the other hand, it also seemingly permits the traditional practice of information (bad news) being shared only with a patient’s family members but not the patient. Owing to such a vague zone of who can or should be informed of a diagnosis, some have argued that the individual’s right to informed consent is not fully protected under the HPCA.⁷ However, such criticism lacks understanding of the fact that medical practices must progress over time with the evolution of social conceptions of patient autonomy. In a family-oriented model of end-of-life care practice such as that which prevails traditionally in Taiwan, the patient’s right to informed choices is made clear and affirmed through efforts to advance legal frameworks.

Second, the HPCA applies only to terminally ill patients. It does not cover individuals who cannot be described as having a terminal illness, such as those in a permanent vegetative state, those with severe dementia, those in an irreversible coma or those with motor neuron impairment arising from such rare diseases as amyotrophic lateral sclerosis. In response to this concern, representatives of rare disease patient groups and Legislator Yu-Shin Yang of the Legislative Yuan (the unicameral legislature of Taiwan) have launched efforts to meet the needs of these groups by advocating for a more fundamental solution through a legislative framework aimed at protecting a patient’s right of self-determination.

4.2.4 Patient Right to Autonomy Act

The PRAA was passed in December 2015, announced in January 2016 and became effective in January 2019. Its aims, as stated in Article 1, are to respect patient autonomy in the healthcare context, to safeguard patients’ right to a good death and to promote a harmonious physician-patient relationship.

The scope of the PRAA is wider than that of the HPCA. It no longer limits the withdrawal or withholding of CPR and LST to cases of terminal illness assessed by two specialist physicians. The LST category has also been broadened to include such interventions as artificial nutrition and hydration and antibiotics where they are being used to sustain life. Moreover, the PRAA expands the clinical conditions in which CPR and other forms of LST can be withdrawn or withheld beyond terminal illness: (1) irreversible coma; (2) a permanent vegetative state; (3) severe dementia; or (4) other disease conditions announced by the central competent authority (as defined in the PRAA) that meet the requirements that the condition or suffering is unbearable, the disease is incurable and no other appropriate treatment options are available given the medical standards at the time of the disease’s occurrence. The patient’s condition must also be diagnosed by two physicians in relevant specialties and confirmed through at least two consultations convened by the palliative care team (Article 14).

In terms of decision-making, the PRAA provides for an “advance medical decision” (預立醫療決定), defined as “a prior written and signed statement expressing the willingness of a person to accept or refuse life-sustaining treatment, artificial nutrition and hydration or other types of medical care, and their wishes relating to a good death, when they have been diagnosed with specific clinical conditions” (Article 3). As noted previously, the scope of this AD is wider than that of the letter of intent under the HPCA, which covers only decision-making with respect to hospice palliative care and LST.

The PRAA also provides more comprehensive protection for a patient’s right to informed consent. Article 4 states that patients have a right to be informed of their diagnosis and treatment options, as well as the potential effectiveness and risks of those options, and the right to choose and make decisions regarding treatment options. Moreover, a patient’s legal representative, spouse, relatives, healthcare agents and other closely related people (defined in the PRAA as “parties concerned”) are not to prevent a medical institution or physician from acting on the patient’s treatment decisions. This measure arguably provides patients with a rights-based form of resistance against the common phenomenon of family-based decision-making in Chinese culture, discussed further subsequently, and physicians with a way to preserve their authority to carry out a patient’s decision free of family influence. Accordingly, there is much greater emphasis in the PRAA on both enhancing and protecting patients’ autonomous decision-making, as expressed through an AD and other legislative requirements.

This is not to say that the PRAA does not envisage a role for the family in the end-of-life process, and indeed there are implicit tensions between the value of maximising respect for patient autonomy and the value of involving the family in identifying patients’ best interests within some of the PRAA’s articles. For instance, after specifying the list of items about which a medical institution or physician must inform the patient (in other words, the scope of the physician’s “duty to inform”), Article 5 states that “in the absence of clear objections from the patient, the parties concerned may also be informed”. Thus, unless the patient explicitly objects, his or her physician is permitted to share this information with his or her family. It is noteworthy that instead of requiring consent from the patient before such information is shared, the presumption is that the family is to be informed. Therefore, it is important to recognise that this requirement could be interpreted such that it does not mandate the physician to initiate a discussion with the patient about sharing information with his or her family during which the patient may or may not express an objection.

Another notable feature of the PRAA is the way in which it embeds the process of making an AD in a wider legal requirement to instigate advance care planning (ACP; 預立醫療照護諮商). The procedures for establishing an “advance decision” through ACP consultation must fulfil the following requirements: (i) a medical institution has provided consultation on ACP to the declarant and affixed its seal on his or her advance decision; (ii) the advance decision must be notarised by a notary public or witnessed by two or more persons with full disposing capacity; and (iii) the advance decision must be registered in the declarant’s national health insurance IC card. Furthermore, ACP must take place with the participation of at least one relative of a first or second degree of affinity (Article 9).

The Taiwanese Government has not yet taken any steps to cover ACP through the national health insurance system, which means that ACP participants currently need to pay a small out-of-pocket fee, generally in the region of US$80–150, depending on the standard set by the individual hospital. Some have thus argued that financial incentives may be necessary to increase the public’s willingness to embark on ACP consultation.⁸ This author argues, however, that because ACP is resource-demanding and time consuming (as with any counsel of a psychologist or social worker), if the patient in question lacks motivation and understanding of the value and importance of ACP, then a fee-waiver policy would simply place an unnecessary burden on ACP consultants and waste scarce resources.

Other components of Article 9 are also worth noting. For example, it is expressly stated in Article 9(3) that at least one relative of a first or second degree of affinity must participate in the ACP process, which is in line with the emphasis on family-oriented decision-making in the Taiwanese context.⁹ The rationale for such a legal design is that when tough end-of-life choices need to be made, family members must often act as patients’ surrogate decision-makers. The purpose of requiring close relatives to participate in the procedures of ACP consultation and AD signing is to build understanding of and consensus on a patient’s informed choices among family members to ensure that when the time comes to execute the AD there will be less disagreement and conflict.

4.2.5 Differences between the HPCA and PRAA

Whilst the HPCA and PRAA both concern a patient’s advance expression of his or her wishes in relation to end-of-life care, there is a significant difference between them in terms of physician liability. As discussed previously, the HPCA includes strict penalties for physicians who violate end-of-life procedures. The PRAA, in contrast, contains no such punitive measures, which were removed during the legislative process. The Ministry of Health and Welfare has explained that “since patient autonomy involves ethics, professional judgement, personal belief and other variables, using penalty or compulsory means to force physicians, patients or families to comply with the law would be non-beneficial to the promotion of such legislation”.¹⁰

One of the difficulties that arose during the legislative process concerned physicians’ unwillingness or refusal to withhold or withdraw LST and/or artificial hydration or nutrition from non-terminally ill patients due to conscientious objection.¹¹ Indeed, Article 14 allows very wide space for professional autonomy in that it permits a physician to refuse to implement a patient’s AD on the basis of his or her “professional expertise or wishes”, a wider scope of opt-out for a professional than reliance on his or her personal values or beliefs. In such a case, the physician is required only to inform the patient or parties concerned (Article 14), and the patient may then turn to other healthcare professionals who are able to assist.

Article 14 further protects physicians from liability by stating that they will not be subject to criminal or administrative liability if they terminate, withdraw or withhold LST and/or artificial nutrition and hydration in accordance with the article and that they will bear no responsibility to compensate for any damage incurred unless intentional or grossly negligent conduct is involved and the action is in violation of the patient’s AD. These detailed, comprehensive guarantees of protection from liability for physicians, which resulted from negotiation during the legislative process, are arguably a way to reduce the pressure and burden involved in forgoing LST and/or artificial nutrition and hydration and to provide reassurances to mitigate the tensions that may arise.

4.3 Professional Guidelines Relating to Advance Directives in Taiwan

Whilst the HPCA and PRAA have been promoted and endorsed by many professional associations in Taiwan,¹² unlike their Western counterparts few of these associations have issued professional guidelines concerning end-of-life care, although the Taiwan Academy of Hospice Palliative Medicine, whose members are primarily physicians practising in the field of palliative medicine, issued a position statement expressing a firm objection to euthanasia and physician-assisted suicide in 2017.¹³ The National Health Research Institute also published a “White Paper for Palliative Care in Taiwan” in 2019¹⁴ and a “Policy Statement and Action Plan for Palliative Care in Taiwan”¹⁵ in 2020. However, although both relate to the future development and promulgation of a comprehensive national plan for advancing citizens’ rights and dignity with high-quality end-of-life care, as well as the expansion of ACP within hospitals, neither make any mention of ADs. Thus, there is likely to remain much uncertainty in practice about how the basic legal provisions should be implemented and managed in professional healthcare settings.

4.4 The Practice and Sociocultural Context of Advance Directives in Taiwan

This section now considers the practice of ADs in Taiwan in light of the implementation of the aforementioned frameworks, as well as the sociocultural factors affecting such practice.

As of March 2022, of Taiwan’s population of 23 million, approximately 800,000 people (3.4 per cent) had completed a letter of intent under the HPCA, and only 33,000 (0.14 per cent) had completed an AD under the PRAA.¹⁶ It is understandable that the completion rate of ADs would be low, as the PRAA has been in force only since 2018. However, the reasons for the low uptake are still worth exploring, given the pioneering nature of this legislation. Two possible such reasons, namely, insufficient awareness and/or understanding of ADs and the sociocultural emphasis on family-based decision-making, are explored next.

4.4.1 Insufficient Awareness and/or Understanding

There are data suggesting that there is insufficient awareness and understanding of the concepts of ADs and palliative care among the general public. In 2016, Lin et al. reported that the majority of the elderly individuals and their family members whom they had surveyed in the intensive care unit of a medical centre in central Taiwan had an insufficient understanding of ADs.¹⁷ In a more recent study conducted in 2021 that compared the differences in AD acceptance between millennials and baby boomers, Liang et al. found that, despite the Ministry of Health and Welfare of Taiwan announcing measures associated with the PRAA to assist the general public in understanding ADs, participants from both generations were still unfamiliar with ADs and who was able to accept palliative care.¹⁸

The data from these studies align with commentary suggesting that the general public has misconceptions concerning the nature of advance decisions to refuse LST such as CPR and of hospice care. Among the misconceptions are that signing up for either effectively means that they have “given up” or that they are giving up all treatment.¹⁹ Liang et al. similarly found the low willingness among their participants to sign a letter of intent under the HPCA to be “associated with their erroneous perception that signing the intent [letter] means giving up all medical treatment”.²⁰

In relation to medical professionals, the data are slightly less clear. Preliminary data from Chen et al. suggest that the PRAA has had a significantly positive impact on the knowledge of healthcare practitioners, with concepts such as “informed consent” and “decision-making directives” becoming more widely accepted in 2018 (as compared with 2013).²¹ In a study conducted among healthcare professionals in the field of nephrology, Chiu et al. found that when asked about their knowledge of laws related to ADs, there were three particular areas in which the respondents achieved relatively low scores: (i) understanding of the scope of the HPCA following its revision; (ii) understanding of the matters that one should be aware of when intervening with an AD; and (iii) the difference between an AD and a do-not-resuscitate order.²² Whilst the latter study does not provide us with sufficient information to properly interpret its results,²³ taken together the findings of the two studies seem to suggest that although efforts to promote the PRAA appear to have had a positive effect on healthcare professionals in Taiwan, there are still specific areas in which their knowledge and understanding remain weak and require further training.

4.4.2 Emphasis on the Role of the Family

In addition to insufficient awareness and understanding of ADs, particularly among the general public, the low uptake of ADs may also be explained by certain sociocultural characteristics of Taiwan. In addition to the taboo associated with death, which results in the avoidance of conversations about end-of-life care,²⁴ there is a significant emphasis on the role of the family in the medical decision-making context. Medical decisions in Taiwan are commonly seen as family decisions that should be made collectively between the family and the patient rather than decisions that patients should make on their own.²⁵ In extreme cases (which were not uncommon as recently as one or two decades ago), the family takes complete responsibility for medical decision-making, with patients denied any opportunity to participate. Presumably, however, this type of collusion between physician and family will be made much more difficult by the regulatory requirement to inform patients of their diagnosis, treatment options and right to make autonomous decisions, as enshrined in the PRAA.

In addition, because of the strong influence of the Confucian doctrine of filial piety, which results in adult children remaining subordinate to their parents and being expected to provide care without question, in some instances adult children in Taiwanese society do not proactively engage in discussions about ADs with their parents or other elders.²⁶

4.4.3 The Way Forward

What might be the way forward for Taiwan in light of these possible challenges to the acceptance and uptake of ADs? In terms of improving awareness and understanding, there are data suggesting that implementing ACP programmes containing education on palliative care and ADs has the potential to improve knowledge of and attitudes towards ADs in Taiwan.²⁷ The lack of focused good practice guidelines emanating from within the healthcare profession on how to translate basic legal requirements into a complex, skilled and multidisciplinary process of making advance care plans and advance decisions across different health and community care settings should also be addressed.

Additionally, the 2018 study of Chu et al. on the factors associated with the completion of ADs found that patients were more likely to complete them if social workers participated in the ACP meeting required by the PRAA.²⁸ The authors observed that social workers provide support in multiple ways, including offering patients more information about their rights, identifying the major decision-makers among family members and offering emotional support. Because of such support, they suggested that social workers have an important role to play in the ACP process. In light of these findings, the development and implementation of ACP programmes that provide education on ADs, as well as further research on the proper role of social workers in the ACP process, may be steps that can be taken to improve the uptake of ADs in Taiwan.

A thornier issue may be the cultural influences on the acceptance of ADs. For example, both the traditional emphasis on family-based decision-making in the medical context and the expectation that children will take care of their parents in their old age without question may make it difficult for families to support an individual-oriented form of decision-making through the practice of ADs. However, from the author’s experience and observations in clinical practice and ethics consultation, such traditional concepts and practices have changed gradually but remarkably in the past two decades, during which Taiwan has also undergone tremendous transformation in terms of political freedom and a democratic system.²⁹ The young and even middle-aged adults who have grown up in a rapidly developing democratic society that emphasises personal freedoms and individual rights are likely to have little difficulty embracing patients’ right to autonomy in medical decision-making. The aforementioned cultural influences and traditional practices are increasingly being transformed into an endorsement of patient autonomy and self-determination, or at least “shared decision-making”, as more desirable standards. Portrayals of Chinese culture as characterised by “family-determinism” rather than “individual-determinism” are imprecise, dichotomised, unfeasible and inapplicable to the current Taiwanese context.³⁰

The AD approach in the PRAA is family oriented to some extent. As discussed previously, whilst the Act protects individuals’ right to informed consent and to treatment in accordance with their wishes without familial interference, it also contains provisions that presume that information about a patient’s condition should be shared with family members unless otherwise explicitly indicated and, more importantly, that require a family member to attend the ACP meeting in order for the individual to be able to make an AD.

Hence, the PRAA preserves key features of family-based decision-making while advocating for respect for individual autonomy, and should arguably be promoted as such to emphasise its alignment with certain traditional cultural elements. In this sense, describing the Act in terms of a basic requirement to protect a person’s autonomy rights fails to do justice to the complex and multiple values that intersect within its regulatory requirements and that need to be negotiated by practitioners, patients and family members in practice. The Asian Delphi Taskforce for Advance Care Planning has done initial work on developing good practice in light of these tensions, in particular through its publication of the 2019 Taipei Declaration on Advance Care Planning, which recommends tasks for each role in ACP, taking into account Asian cultural influences.³¹ Of particular note here is (i) recognising that family members and other caregivers have needs that should be attended to as part of the patient’s care; (ii) expecting family members to participate actively in ACP discussions; and (iii) co-opting family members in the ACP process to remind, help or accompany the patient in the process of sharing goals and preferences.

4.5 Conclusion

Over the past two decades, Taiwan has developed comprehensive legislative frameworks for ADs and end-of-life decision-making that have expanded over time. It is particularly notable that Taiwan’s recent approach has been to embed ADs within a wider regulatory push towards introducing and improving ACP more broadly. The legislation now in place emphasises the importance of the fundamental ethical principle of respect for patient autonomy while simultaneously aiming to enable patients’ family members to retain a central role in medical decision-making in ways that are likely to give rise to tensions in practice.

Despite the enactment of the two pieces of legislation discussed herein, however, AD uptake remains low in Taiwan. Further research needs to be conducted to learn more about the practice of ADs in Taiwan, but the existing data suggest that greater effort needs to be devoted to increasing the awareness and understanding of the general public (and healthcare professionals to a lesser extent) about ADs and end-of-life care. Whilst certain cultural factors may hinder the acceptance of ACP and ADs, efforts to implement them in a culturally sensitive manner, such as the PRAA provisions discussed here do, as well as the recommendations of the Asian Delphi Taskforce for Advance Care Planning in the 2019 Taipei Declaration on Advance Care Planning, are important steps towards aligning high-quality end-of-life care and decision-making in ways that endorse the conflicting ethical values that play out within the healthcare context in Taiwan.

5 Advance Directives: The Thai Context

Thitinant Tengaumnuay

Recognising the principle that an individual’s autonomy and dignity are to be respected, the law of Thailand provides the right for patients to make advance decisions concerning their future medical treatment. Advance directives (AD) are used by terminally ill patients to specify their wishes in the event that life-saving medical interventions would be considered futile, merely prolong death, or cause unnecessary suffering. Thailand is among the growing number of countries globally to have enshrined within its laws a well-regulated framework for the use of ADs. However, there is no single template or concept of the format of an AD, an absence that has the potential to undermine the effective application of ADs in practice. Furthermore, within Thai culture, clear tensions exist between the principle of individual autonomy and the importance and influence of the family in determining the extent to which their relatives’ wishes expressed within an AD are upheld. This chapter explores these tensions between competing desires and interests and how they complicate the effective implementation of ADs.

The chapter begins with an outline of the legal regulations guiding the use of ADs in Thailand (Section 5.1). Section 5.2 explores the challenges posed to medical professionals involved in the implementation of ADs, and considers sociocultural values such as familism, which enables the patient’s family to exercise influence in the context of ADs. The discussion in Section 5.3 identifies some particular problems of ADs in Thailand and concludes that effective implementation requires more than legal backing; it requires the introduction and implementation of formal procedures such as a system for registering ADs. A registration system would help to ensure that the legal framework for ADs, implemented in response to the recognition of the right of terminally ill individuals to autonomy as they approach death, and to be enabled to die with their dignity intact, can be achieved in practice.

5.1 The Legal Framework: Background, Scope and Conditions of ADs in Thailand

5.1.1 Background of Thai Laws on ADs

Since 2003, the notion that individuals may aspire to achieving a ‘good death’, aligned with the principle of a patient’s right to refuse medical treatment, has generated robust debate between medical professionals, legal scholars and practitioners. Before the introduction of Universal Health Coverage (UHC)¹ in 2002, the high costs of medical treatment imposed considerable financial burdens on most Thai people. The ability to access medical and health services was therefore regarded as a matter of “luck”²; few people therefore even considered the idea that in the event of a diagnosis of a terminal illness, they might wish to refuse medical treatment, or that such treatment might prolong the dying process and cause the terminally ill patient unnecessary pain and suffering. Unlike some other countries where health and human rights groups have been advocating for patients’ rights – especially the right to refuse medical treatment – most Thai people are unfamiliar with the principle of the right of patients to self-determination. Thailand’s Constitution recognises the principle of individual autonomy and the rights of human dignity, yet, the notion that some people might wish to refuse medical treatment has been characterised by some people – including some within the legal profession – of representing a threat to public order and morals; refusal of treatment was therefore void under Thai law.

The issue of patients’ right to refuse life-saving medical treatment was explicitly raised in 2003, when the Ministry of Public Health submitted a legal question to the Office of the Council of the State (OCS), a body which functions as a legal consultant for state agencies.³ The question concerned a case involving a member of the Jehovah’s Witnesses faith community, who made an AD stipulating that they wished to refuse blood transfusions as part of any life-saving treatment. Medical professionals were undecided as to whether they should follow such a directive, or should prevail in providing the blood transfusion treatment to forestall any criminal liabilities. Responding to the concern of medical professionals, the OCS suggested that in cases where blood transfusion was a critical element of the life-saving treatment, doctors must provide the blood transfusion, even though doing so went against the patient’s expressed will. Medical professionals are dutifully and legally bound to save a patient’s life, so to follow the patient’s will, and to therefore cease to provide life-saving treatment, would be construed as having breached Thai law, and hence, they might be criminally liable for failure to provide life-saving treatment.⁴ The OCS argued that implementation of the patient’s will – here specifically, allowing the patient to refuse treatment – could potentially ruin the existing positive relationships between doctors and patients. More so, to implement a patient’s AD to withhold life-saving treatment would represent an act of non-compliance with medical ethics, which stipulate that doctors must act to save life. And finally, it was argued that for doctors to observe patients’ right of autonomy would also conflict with the concept of the good Samaritan, a humanistic principle which holds that a person should help others – including strangers – who are in danger. Several doctors disagreed with the OCS’ legal advice, arguing instead that allowing the implementation of a patient’s AD in no way represented a threat to public order and morals, but should instead be respected as the exercise of a patient’s right to autonomy.⁵

While not binding, the legal opinion of the OCS is the starting point for debates over which patients’ rights require legal protection. This legal issue was of particular concern to a number of public health professionals,⁶ presumably because of the serious implications for their possible criminal liabilities. The debate led to much discussion about, and the drafting of specific legislation on ADs, in 2004, one year after the OCS had rendered its legal opinion. Notably, the drafting committee included representatives of the medical professions.⁷ The subsequent introduction of the National Health Act in 2007 incorporates ideas that emerged from the drafting process, and Article 12 explicitly gives recognition to a patient’s right to self-determination in the context of ADs. The details of the AD regime were further clarified in a ministerial regulation of 2010.⁸

5.1.2 Scope and Conditions of ADs under Thai Laws

Article 12 of the National Health Act 2007 states that a person has the right to make an AD setting out their desire “to refuse any futile health services which simply prolong death or cause suffering”. Public health professions are exempted from any liabilities arising from their actions taken in accordance with fulfilling the patient’s expressed will. Article 12 seeks to preserve the human dignity of a patient, as enshrined in the Constitution’s specific mandate for the protection of patient autonomy over their lives and bodies. It gives patients the right to refuse medical treatment and to have a “good death”, rather than to accept life-prolonging treatment involving the use of medical technologies that would likely cause further pain and suffering but which would ultimately prove futile.⁹ The scope of an AD is therefore limited to ensuring that a patient has some control over their dying and can experience a “good death”.¹⁰ In other words, an AD explicitly sets out the patient’s refusal of medical treatment. An AD will only be implemented if (i) the patient is in a terminally ill stage; or (ii) the patient has an incurable disease for which medical treatment will merely prolong an inevitable death, including a persistent vegetative state.¹¹ In cases where the patient remains conscious and retains the capacity to communicate, the medical professional responsible for the treatment of such a patient must ask for the patient’s confirmation of their wish as expressed within the AD, prior to implementing their will.¹² It must be noted that Article 12 does not allow for either “mercy” killing or active euthanasia, but is applicable solely within the context of a strictly defined “end-of-life” scenario. Its use is limited to the right to refuse to accept treatment when the disease has progressed to the stage where treatment is futile and/or causes further suffering.

Thai legislation does not prescribe a specific formal format for an AD; Article 12 merely requires a written statement to be made while the individual is conscious, of sound mental capacity and has full understanding of their actions and the consequences. There is no requirement that medical or legal personnel should be involved or present; a person can make an AD, alone, without having to consult public health professionals. The practical difficulties involved in establishing whether an AD was made when its maker possessed the mental capacity to make such a momentous decision is an issue to be discussed in a later section, but in theory or practice, Thai laws do not require an assessment of mental capacity at the time of writing the AD. Ministerial regulations merely provide guidelines on the information to be included in an AD:¹³

• The personal biodata information of the person who makes the AD, namely name and surname, age, national identification number, and address or contact number.

• Date the AD was made.

• Name(s), surname(s), and national identification number(s) of a witness/witnesses, and their relationship to the person making the AD.

• The specific medical services or treatment that the patient does not wish to receive.

• In cases where individuals have requested another person to write or type the AD on their behalf, the surname and national identification number of that person should be stated.

• The signatures or fingerprints of the person who makes the AD and that of the witness/witnesses.

It must be noted that the ministerial regulations require only that the intentions of the patient making the AD are clearly stated; other requirements, such as stating the name of the witness and their signature, are solely intended as guidance on how to make a clear AD. It is an illustrative, rather than an exhaustive list, and as such, the AD is valid whether it provides less – or more – information than the AD guidance suggests. It stipulates only that the AD be in the form of a written statement and clearly indicates the individual’s will.

Since the clarity of intention of an AD is required, in order to ensure the precise understanding of their intention, a patient can specify the name of a person who is able to clarify their intention as stated in the AD. Generally, the patient chooses a family member with whom they share close relationship, and who has been (unofficially) informed by the patient of their will.¹⁴ The option to enlist the help of a trusted and familiar person is particularly helpful in cases where there is some confusion or ambiguity, and the person who made the AD is no longer in a state where they are able to explain their will. As stated previously, it is imperative that a patient who makes an AD put their signature and national identification on the document,¹⁵ so as to enable the later identification of such persons. In some cases, while the patient might believe that the terms used in the AD are unambiguous and clear, it is possible that the language or terms employed might be confusing to the reader. In such a situation, it is envisaged that the named person would be able to clearly and correctly articulate and explain the patient’s will to the doctor. The inclusion of this named person therefore serves to ensure that the AD conforms with the requirement for clarity; the named person is not authorised to make any decisions on behalf of the patient, and they play no part in proving the validity of an AD, their sole purpose being to help clarify the patient’s will.

The state does not require the registration of ADs with state officials; the ministerial regulation suggests only that a patient give their AD to public health professionals as soon as possible, when receiving medical treatment.¹⁶ Article 12 exempts medical professionals from criminal liabilities for carrying out the patient’s will as set out in an AD. However, neither Article 12 nor the ministerial regulation imposes sanctions for non-compliance with an AD. In other words, if a doctor does not follow the patient’s request for termination or withholding of medical treatment, they will not face any particular liabilities under the laws regulating ADs. This then raises questions about the actual effectiveness of ADs in Thailand, as will be further discussed in Section 5.3.

In terms of implementation, practical problems can also result from the informality of the template of an AD. While the laws require the clarity of an AD, the inclusion of some important information – albeit, limited – to support the validity of an AD, such as the identification and signature of a witness, is merely a recommendation, and not mandatory. Flexibility may facilitate and perhaps encourage more people to make ADs, but an AD that does not contain sufficient information to support its validity can be the source of problems that might inhibit its implementation. When identification of witnesses is not required, the possibility arises that some people might choose to simply forgo a witness, an issue to be further discussed in Section 5.3.

The provision of palliative care, however, must continue even if medical treatment is refused.¹⁷ There is no definition of “palliative care” within Thai laws; instead, Thai healthcare services employ the definition provided by World Health Organisation (WHO) to clarify the particular healthcare provision and duties of health professions for patients.¹⁸ The obligation for healthcare professions to provide palliative care against the patient’s expressed will raises the question of how to balance the tensions between maintaining respect for individual autonomy and enabling the achievement of a good death. The concept of “patients’ rights” is conventionally based on the notion of individuals having decision-making autonomy over their own lives, in accordance with their own values. These tensions have long been the subject of debate; what, if any, are the limits to patient autonomy, and how can this be squared with medical knowledge and judgement on the best course of action (or inaction) for a patient?¹⁹ Thailand has grappled with these tensions after the patients’ right to self-determination was recognised in Article 12 of the National Health Act, in a form of ADs. Some medical professionals were concerned about their possible criminal liability under this provision; it is they, after all, who generally shoulder responsibility for their patients’ healthcare, and the decision on whether to continue or terminate medical treatment should, they reasoned, be made by physicians, rather than by patients.

In 2011, a group of physicians brought a case to court, claiming that Thai legal regulations on ADs were against public order and morals, since such rules impose burdens on medical professionals who carried the weighty responsibility of deciding whether a patient had reached the terminal stage, before implementing an AD. They must also decide what forms and nature of treatment constitute palliative care. However, the law allows medical professionals to avoid any possibility of being held criminally liable if they are acting in accordance with the patients’ advance decisions. An AD was hence alleged to enable a patient to reject the evidence-based knowledge, expertise and decisions of medical professionals. The court decided that the legal rules on ADs did not impose any significant burdens on medical professionals, for under Thai law, the right to the use of an AD was not to be construed as allowing active euthanasia or mercy killing; it was instead to be interpreted as the patient’s right to refuse treatment in order to die a peaceful and dignified “good death”; medical professionals merely had to respect the patient’s decisions. The Thai legal rules on ADs are thus lawful and constitutional.²⁰

Superficially, the court’s decision in this case appeared to emphasise the value of patient autonomy over medical judgement, by prioritising respect for the patient’s decision rather than privileging the doctors’ judgements of the best course of action and treatment for the patient. However, Article 12 does not leave all decision-making power with the patient. The mandatory provision of palliative care under the ministerial regulation demonstrates continued state intervention to a certain extent, in particular in relation to the imposing of normative values about what constitutes a good death. Moreover, this insistence on continued palliative care overrides the patient’s will. In practice however, such state intervention might not be harmful to most patients; should a patient decide to refuse what they consider to be ultimately futile treatment,²¹ the palliative care regime usually coheres with the patient’s desire for less suffering.

5.1.3 Further Development of the AD Regime

Apart from establishing specific legal rules for ADs in the form of the National Health Act and the ministerial regulations, the National Health Commission Office (NHCO) has published a manual (generally referred to as the NHCO Guide) which provides guidance for health service providers on dealing with ADs. Despite its 2007 introduction in the provisions of the National Health Act, the concept of an AD is not well understood among health professionals. The aforementioned court case brought in 2011 provides a salient and instructive example of the limited knowledge of some public health professionals as to the proper scope of palliative care.²² It also revealed their prominent anxieties about the possibility of facing criminal liability should they fail in their professional and ethical duty to provide medical treatment. The claimants in this case (whom it will be recalled were medical professionals) had misinterpreted the regulations and believed that an AD enabled a patient to request a “mercy killing”. The basis of their claim indicates the paucity of their knowledge and understanding about ADs, as well as a degree of uncertainty about the correct meaning of some terms (such as palliative treatment and futile treatment that simply prolongs death) contained within the ministerial regulations. Acknowledging this problem, the NHCO responded with efforts to enhance understanding of ADs, including the publication of a web-based manual which provides clarity to the concept of an AD, the conditions framing their use and implementation, as well as clarifying the definition and processes of palliative care.²³

The NHCO’s guidance manual also suggests the proper practice of public health professionals when faced with a situation where a patient has written an AD. When informed of the AD, public health professionals must first observe and confirm the mental capability of the patient to ensure that they had created the document while fully conscious of its intended outcome and were of sound mental capacity at the time of writing. It must be noted that this action constitutes guidance rather than an actual directive mandating a public health professional to conduct a preliminary assessment of the patient’s mental capabilities. Such an approach is inherently problematic, however, for the mental capability of a patient at the time they provide the AD to the doctor generally does not always equate to the patient’s mental state at the time of crafting their AD. It is possible that while the AD was made when the patient was conscious and cognisant of their intentions, by the time the AD was submitted to the relevant medical professional, the patient no longer had capacity. Another possible scenario is where a patient did not have mental capacity when the AD was made, but subsequently regained their mental capacity at the time of submitting the AD. In such cases, the medical professional might face difficulty in verifying the patient’s capability at the time of making the AD, and therefore not be able to implement the AD.

After the preliminary assessment of the patient’s capacity, the patient’s will as indicated in an AD must be noted in their medical record. The physician then makes a copy of the AD and returns the original AD to the patient. Should the patient be later moved to another public health facility, another copy of the AD should be made and sent with the patient to the new public health facility.²⁴ The NHCO also raises public awareness of ADs through the publication of information leaflets, guidance manuals, and samples of ADs, which can be easily found on an online website which provides specific information on ADs under the supervision of the NHCO.²⁵

5.1.4 Child ADs

Does a minor have the autonomy to make an AD of their own? This troubling question has been the subject of much debate, since Article 12 does not specify any minimum age as one of the conditions of an AD. Some lawyers have referred to the Declaration of Patient’s Rights²⁶ in seeking answers to this question. This declaration confirms parental rights to exercise a patient’s rights on behalf of their minor children below eighteen years of age, thus leading some scholars to argue that parents are therefore empowered to write an AD for their child.²⁷ Meanwhile, the NHCO guidance mentioned previously states that a person below eighteen years of age can make an AD with the permission of their parents.²⁸ The difference between the declaration and the NHCO guidance matters, in that in the former, parents may make an AD on behalf of their child, while in the latter interpretation, a child patient has the right to make an AD, but only with parental permission. In other words, the former assumes the patient’s will, while the latter directly reflects the patient’s will.

The matter of children’s capacity and right to make an AD remains undecided by the courts, and in fact, has not generated much debate, presumably because children are not usually expected to have fatal illnesses in their youth; it is rare indeed to find an AD made by or on behalf of a child patient. Nevertheless, it is still of theoretical interest that a minor might be interpreted as having the capability to make an AD, albeit with parental permission. Conventionally, Thai laws recognise the incapacity of a minor to make a juristic act, and prevents a minor from doing as such for the minor’s own benefit, though there are some exceptions including some personal acts that a minor must decide their own, and in which no other persons are permitted to intervene, or to pre-empt such decision, such as certifying a child’s birth.²⁹ A minor is allowed to make a will when they reach the age of fifteen years.³⁰ Any will made before that, even if it lasts until a minor becomes sui juris, are deemed void.³¹ Parents cannot make a will on behalf of their child, nor can they give permission to a child below fifteen years of age to write their own will. An AD – usually called a living will – is comparable to a typical will. The limitations placed on the juristic acts a minor is able to make are justified on the grounds that they are intended to protect a minor’s benefits, and to safeguard them from those who might seek to take advantage of their young age. While the law allows a minor of fifteen years of age to make an enforceable will, there is currently no minimum age for the same minor to make an AD (even if parental permission is required), which could disadvantage rather than benefit a minor. What a “good death” means to a person is considerably subjective and personal, and the idea that parents can make an AD on behalf of their child expands parental powers over a child, which is not coherent with other legal provisions that preserve some rights solely for the individual, rather than the entire family. Even the other interpretation that imposes no age minimum but instead allows a minor to themselves make an AD raises doubts about the appropriateness of allowing a child to make such a grave and consequential life-concerning decision. Although the family relationship is of great importance within Thai culture, some legal rights are nevertheless personal, and should be reserved for one’s own decision. I explore further the tension between the principle of autonomy and the role of family in implementing an AD in Section 5.2.

5.2 Practice, Value, Commitments and Sociocultural Influences in Thailand

With an individual’s autonomy as the normative foundation of an AD, Article 12 allows the patient to make an AD by themselves, without having to obtain consent from or to discuss with other family members. A medical professional does not have to ask the patient’s family for permission before implementing an AD. Even where the patient is unconscious, the only duty of a medical professional towards the patient’s family is to explain the current stage of their family member’s illness, and the medical process for implementing the patient’s AD.³² In cases where a patient has not made an AD, or where there are ambiguities within their AD, but the patient has not specified a named individual able to clarify their intention, it is not possible to simply appoint family members to make an AD or to make decisions about palliative treatment and/or any futile treatment that simply prolongs death for the patient. If the patient’s will as stated within the AD lacks clarity, it cannot then be implemented; the patient’s right to refuse treatment cannot be given by proxy to any other person, even to a close family member. The right to make an AD is preserved solely for the individual,³³ and as such stands in stark contrast to traditional Thai values of collectivism and paternalism.

Like many other Asian countries, Thai society embraces the concept of collectivism rather than individualism.³⁴ The closeness of familial relationships makes it possible for an individual’s everyday decisions to be influenced by their family. That paternalism has been embedded within Thai culture for a long time³⁵ is demonstrated by the traditional convention that children should respect and accept the advice of seniors.³⁶ This value is so deeply entrenched that Thai people allow “elites” or experts whom they believe to possess greater knowledge to make personal decisions on their behalf.³⁷ In the case of ADs, this would suggest that Thai individuals would be discouraged from making ADs, but would instead be inclined or persuaded to defer the decision to specialist and expert medical professionals. The perceptions of the patient’s family can also influence whether the patient will make an AD, and the terms of that directive. This part explores perceptions about the “good death” in the Thai context, and discusses the tension between the principle of autonomy and the role of family in implementing an AD.

Long before Thai laws adopted the concept of a “good death” and recognised a patient’s right to refuse medical treatment, public opinion emphasised the duty of medical professionals to treat patients, despite the fact that an AD had been made stipulating the patient’s refusal of futile treatment. A 1984 survey revealed that a majority of the public at that time considered a doctor who terminated life-saving treatment according to a patient request made in advance was guilty of committing a crime; 71.8% of those surveyed regarded termination of treatment as murder while 24.3% thought that the doctor should be held liable for having committed a minor offence; only 3.4% believed that the doctor should be free of any criminal liability.³⁸ The later public survey in 1995 indicates the changing perspectives, as respondents indicated more positive responses towards a “good death” - the termination of treatment requested in advance by a patient was not considered a crime by the majority (67.4%) of respondents.³⁹

Although both studies discussed previously were conducted well before the passing of the National Health Act, they show how substantively public opinion can change from one extreme to another within the course of a decade. With growing understanding that some medical treatment is futile and cannot cure the patient’s illness but will merely prolong their death, and thus cause more suffering, the termination of treatment has become more acceptable to the Thai public. Apart from the concepts of collectivism and paternalism embedded in Thai society, Thai culture and norms also draw heavily from Buddhism,⁴⁰ which prompts adherents faced with the prospect of death, to always recall the uncertainty of life, and to learn to let go of the materiality of life. The concept of a “good death” is consonant with most Thai beliefs that in cases where no treatment can cure the particular illness, a person should die peacefully rather than be forced to undergo futile treatment that merely prolongs their suffering.⁴¹ The concept of good death and the recognition of the patient’s right to self-determination have their origins in Western countries, which generally rests on cultures of individualism, and which promote awareness of citizens’ rights and freedoms. As discussed previously, Thai culture fosters paternalism: most people tend to believe in the superiority of expert knowledge and unquestioningly obey seniority, norms that conflict with the principle of individual autonomy, which supports an individual’s right to make an advance decision concerning their medical treatment, especially where it leads to death. Buddhism’s strong focus on the truth of impermanence and the contemplation of death, however, may have contributed to acceptance of the view that individuals should be permitted to plan ahead to enable them to achieve a “good death”. It is thus the influence of Buddhism that led Thai laws to adopt Western ideas of the “good death” and ADs.

More recently, empirical studies conducted in 2017 and which explored patients’ views of ADs found that only 1.4% of those surveyed held negative perceptions of their use, although it was also found that most participants were unaware of their legal right to make an AD until they were informed about this during the study. Nearly half (42.9%) of participants had limited knowledge about ADs, while 38.6% had ‘moderate’ knowledge, and only 18.6% were highly knowledgeable.⁴² Interestingly, after being informed of their right to do so,⁴³ over half (57.1%) of all participants decided to make an AD, suggesting that general attitudes towards ADs continue to be positive.

While Buddhism emphasises the recognition of a life cycle in the sense that death surely follows life, no human can avoid death, and should therefore be prepared to die peacefully, a philosophical approach that supports the concept of the “good death”, Buddhists also enshrine some of the wider social conventions, such as the norm of gratitude towards their parents. That children should express gratitude to their parents is a deeply rooted norm in Thai society;⁴⁴ sons or daughters who hurt or kill their parents receive extremely severe public condemnation. This norm instils obedience to the parent, alongside gratitude, and can act as a restraint on some patients’ children, who may refuse to allow medical professionals to terminate their parent’s treatment, despite the presence of an AD. Even though the patients’ families may fully understand the futility of treatment, some are afraid of attracting social condemnation for not having made every effort to keep their parents alive.⁴⁵ As noted before, Article 12 of the National Health Act enables a patient to make an AD and exempts medical professionals from criminal liability for any acts undertaken in implementing the AD. Yet, as also noted, this legal provision does not impose liabilities for acting against the patient’s will where the patient’s family decide that their family member should continue medical treatment. Medical professionals do not have legal standing to petition for the judicial enforcement of an AD. In practice, resistance from the patient’s family can obstruct the implementation of an AD, as most medical professionals wish to avoid a suit brought by the patient’s family.⁴⁶ To date, the laws have not been able to provide a satisfactory solution out of the impasse when there are conflicts between the desires of the patient’s family and that of the patient as expressed in the AD.

This causes a fundamental practical difficulty in the implementation of ADs. As mentioned previously, the principle of individual autonomy provides the theoretical grounding for ADs, and the individual’s will regarding their medical treatment is considered personal; there are absolutely no legal measures permitting family intervention. However, resistance from the patient’s family can potentially induce hesitancy among some doctors, causing them to delay or abstain from implementing their patient’s AD. This problem suggests that the concept of individual autonomy has not been adopted well in Thailand’s legal framework for ADs. Further regulatory development is needed to ensure that ADs will be actually implemented, and that the patient’s family cannot influence the physician to go against the wishes of their patient’s AD. One possible solution is to adopt the ministerial regulation that proposes that individuals should identify an advocate to act on their behalf, in the event that there is a need for clarity about the patient’s desire as expressed within their AD. Since the advocate will presumably share a close relationship with the patient, and is cognisant of, and understands the patient’s rationale for choosing to refuse medical treatment, selecting a family member as advocate can be an interesting approach that facilitates mutual understanding between the patient and the family, and helps to diminish the family’s resistance to the AD. Another possible solution is to mandate registration of ADs, as will be discussed in the following section.

5.3 Would a Flexible Format in the Design of ADs Support or Discourage Their Use in Practice?

Thai legal rules on ADs as embodied namely within Article 12 of the National Health Act and the ministerial regulations, do not require a strict format for ADs. Compared with other jurisdictions where specific legal rules set out mandatory requirements for ADs to be regarded as legally binding documents, the Thai state imposes few requirements, making the creation of ADs a more straightforward and uncomplicated task than demanded by neighbouring states. Singapore, for instance, requires that ADs assume a fixed format, while South Korea and Taiwan insist that ADs must be registered with national authorities. Thai laws do not require ADs to have a fixed template – a simple written statement will suffice. There is no mandate for ADs to be registered, and no involvement of national authorities or public health professions is required for an AD to be deemed valid under Thai laws. This flexibility in the making of ADs might appear to be supportive of patients’ right to exercise self-determination in the matter of their death. However, fewer requirements do not always ensure fewer difficulties in implementation. The fact that Thai laws merely set out a few basic guidelines for an AD can, on one hand, benefit a patient in that they can easily make an AD, but on the other hand, medical professionals can face difficulties relying on an AD. Such challenges may reduce the effectiveness of ADs in actual application. Three problems can potentially undermine Thailand’s flexible rules for ADs, which are (1) uncertainty about its validity, (2) unawareness of its existence, and (3) conflicts with the patient’s family.

5.3.1 Uncertainty about the Validity of an AD

As we have seen, under Thai law, a written statement by a mentally capable individual is sufficient to confirm the validity of an AD. However, the laws do not indicate how the individual’s capacity is to be assessed. The NHCO guidance simply suggests that health professionals should conduct preliminary assessments to confirm the consciousness and mental state of the patient at the time the physician is presented with an AD.⁴⁷ Although the ministerial regulations suggest that an AD should be witnessed, this is not a mandatory requirement, and leaves the possibility that a patient lacking full mental capacity could proceed without a witness. This flexibility can raise concerns for medical professionals, for example, in cases where a patient allows the family to give the AD to the physician on their behalf. In such cases, the physician cannot be certain that the AD was actually made by the patient, or whether the statement was made when the patient was fully conscious and aware of the implications of their decision. In addition, the NHCO’s suggestion that the capacity of a patient be assessed when the AD is received poses yet another problem; as discussed previously, mental capability at the time the AD is produced does not necessarily mean that the patient was mentally capable at the time of the writing of the AD.

A comparable case with an AD is the ‘last will’ of an individual. There are five legal formats that the last will can assume in order to be enforceable under Thai laws.⁴⁸ One such format is a document wholly written and signed by the testator.⁴⁹ This format does not require witnesses or registration, and therefore can be compared with an AD. However, the application of this form of the last will does not cause the same concern as an AD, because in cases where there is any doubt surrounding its validity, a challenge can be brought to the court. In contrast, the physician’s decision-making on whether to rely on an AD is shaped by the limited time in which the case may be brought to the court.⁵⁰ In other words, the physician has to make a nearly immediate judgement on the authenticity and validity of an AD. This difficult situation can worsen if there is conflict within the patient’s family over the reliance on an AD.

The requirement that an individual must register their AD with national authorities might – to some extent – resolve these problems. Registration does not only address the issue of medical hesitation and prevent a family’s objection to an AD, it also provides an opportunity for authorities to examine and validate (or otherwise) the AD. It is also possible for the law to require that a physician certify that an individual who makes an AD is capable of understanding and communicating their will at the time of making the AD. This examination or verification can prevent ambiguity or lack of clarity that might cause problems in its later application. Medical professionals can feel more confident of the validity of an AD and will be less likely to hesitate before implementing their patient’s directive. The requirements for either registration or verification from a physician can also help to protect the enforceability of the patient’s will, ensuring that the principle of the patient’s autonomy is pragmatically secured.

5.3.2 Unawareness of the Existence of an AD

In some circumstances, a patient will not have had the opportunity to inform others that they had made an AD. Since Thai laws do not require registration, there is no record of an AD unless the patient informs the physician or other health personnel. It is possible therefore for others to be unaware of the existence of an AD, in which case, the patient’s exercise of their right to make an AD fails to achieve its purpose. In some jurisdictions such as Taiwan,⁵¹ an AD is required to be registered and recorded in the data memory of an individual’s national health insurance card; this can help to provide information immediately in the event the patient is admitted to a healthcare institution.

5.3.3 Conflicts with the Patient’s Family

As already mentioned, the cultural norm that children express gratitude to their parents is deeply ingrained within Thai society. This can leave medical professions with a problem whereby the patient’s family is unwilling to allow physicians to terminate their family member’s treatment despite the declared wishes of the AD. Although reliance on an AD is not conditional on permission from the patient’s family, most medical professionals would prefer not to have to act against the family’s wishes. Article 12 exempts doctors from criminal liability arising from acting in accordance with an AD, but it cannot shield medical professionals from threats of legal actions from dissatisfied families. Most physicians do not wish to undergo the exhausting judicial process, and tend therefore to accede to the wishes of the patient’s family, rather than observe the patient’s desires as stated within the AD.⁵²

Growing recognition of the notion of a “good death” and the changing perception of Thai society towards ADs can decrease resistance from the patient’s family. Apart from relying on raising awareness via education, increasing the formalities associated with ADs can also demonstrate respect for patients’ wishes. Introducing mandatory registration can signal formal recognition of an AD and help raise its significance above the views of the patient’s family, which will, in turn, result in greater confidence among medical professionals to act in accordance with their patients’ ADs. Selecting a family member who can act on behalf of the patient and help to clarify any ambiguities of the AD may also reduce familial resistance to the AD.

5.4 Conclusion

With specific legal rules for ADs, namely Article 12 of the National Health Act and the ministerial regulation, the Thai government has legislated a formal legal framework for ADs. Empirical studies indicate the changing perceptions of Thai society towards the understanding and acceptance of the concept of a “good death”, and of terminally ill patients’ right to autonomy. Current positive responses towards ADs will likely lead to increased use of ADs in the future. However, the lack of formalities in relation to the making of ADs under Thai laws can obstruct the application of ADs in practice, as medical professions might encounter difficulties in relying on ADs, namely uncertainty about their validity, unawareness of their existence, and conflicts with the patients’ family. This chapter suggests the requirement for registration as an approach to relieve such difficulties, and to ensure that the aim of an AD to protect the patient’s right to self-determination and preserve human dignity can be achieved in practice.

6 Advance Directives in India: Seeking the Individual within the Community

Kelly Amal Dhru & Ravindra B. Ghooi

पुनरिप जननं पुनरिप मरणं,

पुनरिप जननी जठरे शयनम्।

इह संसारे बहु दुस्तारे,

कृपयाऽपारे पािह मुरारे।।

(Born again, dead again, born again to stay in the mother’s womb. It is indeed hard to cross this boundless ocean of samsara. Oh Lord! Redeem me through thy mercy.)

– “Bhaja Govindam”, Adi Sankara, Eighth Century

6.1 Introduction

The legal regime surrounding advance directives (ADs) and living wills in India took a long journey to reach where it is, and yet the journey remains incomplete. Legal recognition of end-of-life ADs in India is usually seen as closely tied to questions of the legalisation of passive euthanasia and decriminalisation of suicide attempts. Much of the literature has overemphasised the connection between euthanasia and ADs, particularly because they are conflated with do-not-resuscitate orders and enforcement of the right to refuse treatment that ADs may contain.¹ This chapter explores legal developments in relation to end-of-life care and ADs in India, with the aim of decoupling ADs from the euthanasia discourse and focusing on the former. We lay out the current guidelines on ADs laid down by the Supreme Court of India, followed by a consideration of ADs in the psychiatric context, and compare some of the crucial elements of each, including the role of the family and the consultation process.

An important thread connecting the legalisation of euthanasia with ADs is the role of individual autonomy. In the landmark case of Common Cause (A Registered Society) v. Union of India² in March 2018, which saw the legal sanction of passive euthanasia and legal recognition of ADs, as well as the creation of the aforementioned current guidelines on ADs, the judges explored the value of patient autonomy. In his judgment in the Puttaswamy v. Union of India decision, which was quoted in Common Cause, Justice Chandrachud said:

Life is precious in itself. But life is worth living because of the freedoms, which enable each individual to live life as it should be lived. The best decisions on how life should be lived are entrusted to the individual. They are continuously shaped by the social milieu in which individuals exist. The duty of the State is to safeguard the ability to take decisions – the autonomy of the individual – and not to dictate those decisions.³

There remain in the background, however, concerns over the social and religious relevance of the notion of autonomy. Whilst the value of individual autonomy may be quite central to Western culture, the cultural context has often been neglected in discussions about ADs in India. For example, the family plays an important role in India’s sociocultural context, with respect to decision-making in particular, but there is also a presumption by the courts that the patient has to be protected from the family. We argue in this chapter that the role of doctors and the courts can be seen as protecting patients from their families. The chapter also briefly addresses the questions surrounding autonomy in the religious and medico-cultural context.⁴

In terms of the practice of and attitudes towards ADs, there is very little understanding, although a survey we conducted on the level and extent of knowledge about ADs amongst educated persons with and without a medical background provides some empirical data on the level of AD awareness in India. The last section of the chapter discusses this survey, highlighting that there is a need for greater awareness of ADs. We conclude with reflections on the problems with the current AD regime and recommendations for the way forward.

6.2 The Legal Background: De Lege Lata

6.2.1 The Legal Understanding of Life and Death

To understand the legalisation of ADs, one must first understand the Indian legal approach to life and death. Article 21 of the Constitution of India guarantees the right to life: “No person shall be deprived of his life or personal liberty except by a procedure established by law”.⁵ Following the case of Maneka Gandhi v. Union of India,⁶ Article 21 of the Constitution of India has been given a very expansive interpretation by the Supreme Court of India to embody dignity⁷ and a meaningful, complete life worth living. Since then, Article 21 of the Constitution of India has also been the anchor of many welfare rights.

In the context of patients suffering from terminal illness, the pertinent question is whether the right to life includes the right to die and to die with dignity. The legalisation of end-of-life ADs has been by extension of the right to self-determination involved in the refusal of treatment and withdrawal of life support in this context. As the legal situation currently stands, ADs are seen as a mechanism to ensure death with dignity, as they involve self-determination. What the situation thus demands is translation of the abstract right to death with dignity into concrete regulatory processes to ensure realisation of that right. Accordingly, it is important to understand, as part of the AD context, the journey that led to the legalisation of death with dignity in the form of passive euthanasia.

6.2.2 The Aruna Shanbaug Case (2011)

The battle surrounding the legalisation of euthanasia took centre stage and garnered nationwide attention with the Aruna Ramchandra Shanbaug v. Union of India case.⁸ The case concerned the brutal rape and assault of a nurse at KEM Hospital Mumbai by a ward boy, which left her in a persistent vegetative state. The Supreme Court held that passive euthanasia (understood here as the withdrawal of life-sustaining treatment) could be legalised only through an act of Parliament. The two-judge bench of the court made a distinction between active and passive, and voluntary and involuntary, euthanasia. The procedure it devised required that a high court be approached directly concerning every single case in which life-sustaining treatment was to be withdrawn.

Just as in the Airedale National Health Service Trust v. Bland case,⁹ Aruna Shanbaug had made no express provision for an AD. In Aruna Shanbaug, a reference to ADs was made in the report submitted to the court by medical experts, who also stated that they had no information concerning the expressed views or wishes of Aruna Shanbaug.¹⁰ The amicus curiae in the case also brought up the link between self-determination, used interchangeably with informed consent, and ADs.¹¹ The court laid down guidelines on the withdrawal of life support, which were to remain in force until Parliament enacted a law. Under those guidelines, the decision to withdraw life support should be taken in the best interests of the patient if she is unable to make a decision for herself.¹² Such a decision could be taken by a person or body of persons acting as “the next friend” and could also be taken by the patient’s doctors. However, the decision required the approval of the high court concerned.¹³ The Supreme Court in this case limited itself to acceptance that there had been no expression of wishes without going into the question of the legality of ADs per se.

6.2.3 Law Commission of India Reports (2006 and 2012)

The next milestone in the development of end-of-life AD legislation was two Law Commission of India reports, specifically the 196th report published in March 2006,¹⁴ which discussed the withholding of life support measures from terminally ill patients, and the 241st report published in August 2012¹⁵ after the Aruna Shanbaug decision.

The 196th report recommended allowing passive euthanasia for both competent and incompetent patients. However, despite acknowledging the recognition of ADs in other countries, the report resisted recognising their legal validity, citing “complex legal and factual issues”, such as subsequent changes in circumstances, or changes in technology, and problems of proof in the event of a subsequent withdrawal.¹⁶ The Law Commission, in the 196th report, seems to have been greatly influenced by Justice Munby’s considerations in HE v. Hospital NHS and Another,¹⁷ which centred on the efficacy of ADs (in this case in the context of blood transfusions) and a test of their applicability being a matter of proof.¹⁸ The commission seems to have been concerned about the problems of proof in oral ADs, particularly given the level of illiteracy in India. For policy reasons, it thus recommended that ADs not be legalised in India, fearing the huge amount of litigation stemming from conflicting evidence, even if the ADs were in writing.¹⁹

The 241st report of the Law Commission upheld the findings of the previous report. Interestingly, some of the questions concerning the subsequent withdrawal of consent, for example, which are found in contemporary discussions of ADs, were cited as reasons for not accepting their validity. In light of these findings, Section 11 of the draft bill arising from the two reports (i.e. the 2016 draft Treatment of Terminally-Ill Patients (Protection of Patients and Medical Practitioners) Bill) considered ADs to be invalid.

6.2.4 The Common Cause Decision (2018)

The Common Cause case, which was referred to the Supreme Court with a request that it declare that the right to life includes the right to die with dignity and legalise ADs, dealt more generally with the end-of-life context in India. The court also considered the question of legalising ADs in the case. In light of international developments, in India too there had been repeated requests for the legalisation of euthanasia, many of which ended up on the doorstep of the Supreme Court. Advocacy groups had been writing about euthanasia and ADs and recommending legal sanction for the latter.²⁰ Everything changed in March 2018, however, with the court’s ruling in Common Cause, which was heard by a five-judge bench comprising Justices Dipak Misra Chief Justice of India (CJI), Khanwilkar, Sikri, Chandrachud and Bhushan.

In the Common Cause decision, the court upheld the arguably problematic distinction between active and passive euthanasia²¹ and legalised passive euthanasia, which it understood as the withholding or withdrawal of life support, not as physician-assisted suicide. In relation to ADs, Justice Sikri stated that the fear of misuse cannot be grounds for a blanket ban on ADs.²² The Court laid down a set of detailed guidelines, discussed later, which are the law in place until the Indian Parliament enacts legislation. The guidelines provide for a rather long procedure in the event that the executor of an AD develops a terminal or incurable illness.

In this case, we find at least a prima facie reiteration of the right to decisional autonomy in the five-judge bench decision, particularly Justice Chandrachud’s judgment.²³ Even the CJI, just before giving the guidelines for ADs, quoted John Rawls from his book Political Liberalism, wherein he emphasises choice, holding that between best interests and decisional autonomy, it is decisional autonomy that triumphs.²⁴ It could be said that the legal milieu for the recognition of ADs was established by the Supreme Court’s increasing focus on individual decisional autonomy, as is highly visible in the recognition of the right to privacy in 2017 by a unanimous decision of a nine-judge bench of the Supreme Court in Puttaswamy.

There has also been acknowledgement of the role that the right to privacy played in the court’s reasoning in the Common Cause case. Justice Sikri, for instance, when formulating the questions concerned states that “with [the] right to privacy now a fundamental right under Article 21 (right to life), the principle of self-determination in India stands on a higher footing than before”.²⁵ In the Puttaswamy decision, which was quoted in Common Cause, Justice Chelameshwar said, “An individual’s right to refuse the life-prolonging medical treatment or terminate life is another freedom which falls within the zone of the right to privacy”.²⁶ Hence, the domain of privacy has been extended to include end-of-life decisions.

6.3 Current Legal Position on Advance Directives in India

The key guidelines pertaining to ADs²⁷ come from paragraphs 191–4 of the Common Cause decision, described previously, under the judgment of the CJI and Justice Khanwilkar, which was reiterated by the rest of the bench:²⁸

1. (a) The format: Under the guidelines, an AD can be executed only by an adult of sound mind, who is in a position to understand and communicate the purpose and consequences of the directive. The Court decision does not specify how soundness of mind is to be assessed, although soundness of mind in Indian law is a presumption under the Indian Contract Act 1872.²⁹

   • Reference is also made to “informed consent” given without any undue influence or constraint. The AD must be clear, unambiguous, with specific terms and instructions, and in writing. It can be withdrawn at any time, and the withdrawal must also be in writing. The format requirement is that the AD has to be signed by the executor (who is an adult of a sound and healthy state of mind) and two (independent) witnesses and counter-signed by a judicial magistrate first class (JMFC). It is unclear what exactly is meant by “independent”, but it presumably means someone with no vested interest in the property of the person making the AD.

   • The JMFC is a post established under Section 11 of the Criminal Procedure Code 1973.³⁰ The JMFC’s role is to preserve the document in both paper and digital form and send a copy to the Municipal Corporation or local Panchayat (local village council³¹) for the appointment of a custodian of the document. The JMFC also needs to inform the executor’s family of the directive, which risks compromising the executor’s autonomy. Finally, the JMFC is required to maintain a copy of the AD in his or her office, with another copy forwarded to the relevant district court. The amount of responsibility placed on the JMFC is concerning given the huge backlog of cases in the courts.

   • The signature of the JMFC is likely to constitute a major hurdle given that the number of judicial officers of this level is very small in any given city. For example, in the city of Pune, there are only 19 such officials for a population of six million. There is also a huge backlog of cases with the lower judiciary in India, a concern that does not seem to bother the Supreme Court.

2. (b) When ADs apply: Under the Supreme Court guidelines, an AD can be given effect only in the event that the executor becomes terminally ill and is undergoing prolonged medical treatment with no hope of recovery or cure of his or her ailment or is surviving on life support.

3. (c) The process if there is an AD: The physician, when made aware of the existence of an AD, must check its authenticity with the JMFC and then inform the hospital, which in turn will form a hospital medical board comprising the head of the treating department and at least three experts from the fields of general medicine, cardiology, neurology, nephrology, psychiatry or oncology with good standing in the medical profession and at least 20 years’ experience.

   • The experts visit the patient and form a preliminary opinion on whether to execute the directive. If it is to be executed, the JMFC is to be informed, and he or she in turn appoints a medical board consisting of the Chief District Medical Officer as chairperson and three expert doctors (who are not members of the hospital’s medical board). If this board agrees with the hospital board that the instructions in the AD may be carried out, it will approach the JMFC, who will visit the patient and “after examining all aspects” authorise the implementation of the board’s decision.

   • If the JMFC-appointed medical board refuses implementation, the executor of the AD, his or her family members, or even the treating doctor or hospital staff may approach the High Court. The two-judge bench of that Court can, if it wishes, form another committee with expert doctors (each again with at least 20 years’ experience), although the High Court is to keep in mind “the best interest of the patient”.

4. (d) No AD: In the case of no AD (or an ambiguous AD), the procedure to be followed is this: after consultation with the patient’s family, which must give its consent in writing, the hospital’s medical board is consulted for a primary opinion, and the JMFC-appointed medical board for a secondary opinion.

This is obviously a highly complex and rather bureaucratic procedure, with too many opinions required to be obtained before life support can be withdrawn. There is also no separate procedure for emergency situations. The power to interpret and give effect to an AD lies largely with the physician and the various boards. There is little or no representation of persons other than doctors in the expert committees. It is likely that in future cases,³² the Supreme Court will clarify the procedure, as well as the convoluted nature of the existing guidelines. In the meantime, however, the value of self-determination that the Common Cause decision ostensibly serves is taken away by the lengthy procedure laid out in the guidelines given by the Supreme Court, which remain the law of the land until the Indian Parliament enacts legislation concerning ADs.

Interestingly, in a different context, India has experienced much less controversy in enacting AD legislation. We now turn to the psychiatric advance directive (PAD) regime in India.

6.3.1 PADs under the Mental Healthcare Act 2017

The recognition of PADs under Section 5 of the Mental Healthcare Act 2017 provides analogical strength in favour of the legalisation of end-of-life ADs. For example, Justice Sikri uses PADs in arguing against the Law Commission’s concerns that ADs might be misused, relying on the safeguards in the PAD regime as an example of the safeguards that could be implemented to prevent abuse.³³

Under the procedure for PADs, the power to determine the manner of the directive’s implementation is delegated to the Rules and Regulations of the Mental Healthcare Act 2017, of which Part II deals with the same. An AD format is given in Form CR-A of the Mental Healthcare Rules 2017, made under the Mental Healthcare Act 2017. The AD is to be signed by two witnesses, and is to be registered with the Mental Health Review Board of the jurisdiction in which the person resides. There is also an obligation to make the AD available online within 14 days. According to Form CR-A, a person can appoint two or more nominated representatives in order of precedence to make decisions about mental healthcare treatment in the event of incapacity. In some regards, this stipulation renders a PAD more akin to a power of attorney than to an AD. There is also a need for certification by a medical professional that the capacity to make decisions exists at the time the PAD is made.

Whilst there appears to be no apparent basis for distinguishing between end-of-life ADs and PADs on the issue of capacity, there is an established procedure for assessing capacity under the Mental Healthcare Rules 2017, unlike under the aforementioned guidelines laid down by Common Cause. The determination of the capacity to consent for patients with mental illness is to be made by the expert committee appointed by the Central Mental Health Authority under Section 81 of the Mental Healthcare Act 2017.³⁴

Although PADs were used to provide analogical support for the legalisation of end-of-life ADs in that decision, as discussed previously, there remain crucial differences between the procedures and powers concerning PADs and end-of-life ADs. For example, compared with an end-of-life AD, a PAD confers a duty upon the medical caregiver to follow the directive.³⁵ Less discretion thus lies with the medical practitioner. The AD format is also specified by the Mental Healthcare Rules 2017, which constitute a much more detailed legislative scheme.³⁶

The media attention surrounding euthanasia and ADs has been greater than that surrounding PADs. Because the legalisation of PADs generated little controversy, discussion or resistance in India, they may well have been easier than ADs to legislate. The law under which they were legalised, the Mental Healthcare Act 2017, received cross-party support, as it recognised the legal right to mental healthcare.³⁷ There is also a greater role for the family under the PAD regime, which will be discussed in the next section.

It thus seems that whilst ADs in the psychiatric and end-of-life contexts both aim to provide individuals with the opportunity to exercise autonomy in relation to decisions about medical treatment if and when they lose capacity, the respective regimes are very different, although the differences (in relation to the assessment of mental capacity or enforceability, for example) do not appear to be based on principle.

6.4 Advance Directives in Context

6.4.1 The Role of the Family

Illness in India, scholars have argued, is a family matter, and hence the role of the family is the most crucial for healthcare decision-making.³⁸ The role of the family is central in the epics that scholars often cite as creating the landscape for Indian ethics.³⁹ In an interaction between the Indian legal landscape through personal laws and Hindu religious-cultural practices, the law itself allows for the fictional unit of the “Hindu Undivided Family”, particularly in the context of the ownership of family property and inheritance under the Hindu Succession Act 1956.⁴⁰ Under this system, the eldest male of the family, that is, the father, husband or eldest son, has traditionally been considered to be the Karta or “manager” (or, by extension, the decision-maker). The definition of Karta has now been extended to include females.⁴¹ Family lies at the heart of Indian society, and yet the treatment that ADs afford it is bittersweet. Under the Supreme Court guidelines, doctors have a greater role to play than family members. Nevertheless, the JMFC is to inform the family of the person making the AD that such a document has been made. How is one to interpret this apparent discrepancy?

Given the country’s huge young population and cultural norms against the abandonment of one’s parents in their old age, one would expect the family to play a crucial part in this context in India. Yet, the legal battles around who gets to make a decision on behalf of the patient in the case of ADs have centred on the roles of the courts and doctors. In light of the importance of the family in other law and ethics contexts, as demonstrated in the way decisions are to be made by the head of the family, and legal recognition of the Hindu Undivided Family, one can only wonder at the lack of any role for the family in the end-of-life context.

It is possible that the situation arises from concerns over the potential for abuse in that context. Such concerns are apparent in court decisions pertaining to ADs, as well as in the aforementioned Law Commission reports. Hence, one can argue that the main reason for the resistance is the potential misuse of ADs by family members. In light of this scepticism of the role of the family, the state or third-party agencies perceive themselves as having a crucial role in preventing such abuse and preserving the autonomy of the individual, even though that perception may be unwarranted given the lack of empirical evidence on such abuse.

Scepticism about the role of the family is also reflected in the Aruna Shanbaug decision. The hospital and her caretakers were considered to be the next of kin rather than Aruna Shanbaug’s family or even Pinky Virani, the humanistically intentioned person who filed the case. It was because of the decision of the staff and nurses at KEM hospital who had looked after Aruna Shanbaug in a persistent vegetative state for 37 years that she was kept on life support.⁴²

In the Aruna Shanbaug decision, the role of the family in giving consent on behalf of the patient was also resisted. Justice Katju, for instance, said the following:

The question arises, as to who should give consent for withdrawal of life support. This is an extremely important question in India because of the unfortunate low level of ethical standards to which our society has descended, its raw and widespread commercialisation, and the rampant corruption, and hence, the Court has to be very cautious that unscrupulous persons who wish to inherit the property of someone may not get him eliminated by some crooked method.⁴³

Justice Katju discussed the appointment of a guardian in the Aruna Shanbaug case. He cited chapter IV of the 196th Law Commission Report, which considers the family and relatives to be just one consideration, placing power with the High Court as parens patriae. As with the High Court’s authority to make decisions regarding the withdrawal of life support, there is strong jurisprudence on the role of the state, particularly that of the High Court as parens patriae. Somehow, the Supreme Court under Common Cause and Law Commission of India under the reports seem to view themselves as the protectors of individual patients from their own families. Under the new procedure resulting from the Common Cause case, the focus seems to have shifted from the role of the High Court to that of medical practitioners and the JMFC, with the family still placed in a secondary position. The sentiment remains similar: to protect the individual from his or her family.

Interestingly, concern over the potential for abuse is less evident in the PAD context, in which there appears to be a larger role for the family. In the PAD regime, for example, a relative or the caregiver of the person concerned can make an application to the hospital board to review, alter, modify or cancel the AD.⁴⁴ The greater role for the family in this regime may be the result of dialogue with stakeholders, who expressed scepticism about PADs. The Standing Committee Report on the Mental Healthcare Bill, which resulted in the Mental Healthcare Act 2017, attributes PADs to the Convention on the Rights of Persons with Disabilities, to which India has been a signatory since 2007.⁴⁵ In the report,⁴⁶ which, as noted, resulted from dialogue with stakeholders, there is scepticism regarding the creation of PADs, as well as resistance on cultural grounds, namely, that they may be in conflict with the role of the family in decision-making in India.

Whilst there has been no similar dialogue with stakeholders in the end-of-life context, and it is therefore unclear whether a similar concern would be raised, the foregoing discussion of the importance of the family in Indian culture and other legal contexts suggests that despite court scepticism about that role, there may be cultural reasons for the family to be given a larger role in the context of end-of-life ADs.

6.4.2 Autonomy, Pain and Desire: The Role of Religions

Law operates in the context of its society, which, in the case of India, brings together the peculiarities of that society owing to its highly diverse, pluralistic nature. In fact, doubts have been expressed in the past concerning whether a unified “Hindu” bioethics can exist given the plurality of Indian culture. The management of patients at the end-of-life is also influenced by religious and cultural beliefs, although there appears to be little discussion of personal will and desire in the context of end-of-life treatment.⁴⁷

For a significant number of people belonging to the Hindu, Buddhist and Jain faiths, amongst others, there is the concept of an ongoing cycle of birth, death and rebirth. In these traditions, societal existence, or Samsara, sees its outer boundaries in the pursuit of the attainment of Moksha or Nirvana. According to this view, death is not the end, and there are many ideas pertaining to renouncing all material desires, and eventually the body itself.

Interestingly, the judgment of the CJI and Justice Khanwilkar in Common Cause holds that although Buddhism, Jainism and Hinduism are against euthanasia in the strict sense, the concept of the “good death” in each of these religions can be seen as reflecting the concept of dying with dignity.⁴⁸ More broadly, death can be seen within Indian religious tradition and philosophy as a union with the Divine, the greater Self.⁴⁹ In the Hindu epic of The Mahabharata, Bhisma, who had decided to remain celibate owing to various circumstances, was given in return the boon of Iccha-mrityu, or “the death of desires”, meaning that he could choose the time of his death “as per [his] desire”. This example demonstrates that a good death is considered the highest of rewards.

What about alleviation of the suffering of the Self? As per the famous bhajan Bhaja Govindam quoted at the beginning of this chapter, whilst the individual is to embrace pain, it is left to God to save the individual from pain and suffering. In Hindu and Buddhist philosophy, suffering is considered to be an eternal truth about life, to the extent that it is one of the noble truths of Buddhism. Hence, the desire to avoid pain is not celebrated. Pain is to be embraced and may depend on past-life deterministic forces such as karma, and freeing oneself of pain is not sufficient justification for bringing about death. Accordingly, the rejection of life-sustaining treatment to alleviate pain and suffering, which will in turn bring about the person’s death, would not be seen as in line with these religious beliefs.⁵⁰

6.4.3 The Medico-cultural Context

In ancient India, physicians took an oath after completion of their studies. One of the commitments they made was that they would not try to prolong life in end-of-life situations or resort to what is known today as futile treatment.⁵¹ In the West, however, medicine evolved on different principles. The original Hippocratic oath placed two important conditions on the practice of medicine. The first was to disallow euthanasia, and the second was to never conduct an abortion, with no exceptions.⁵² In India, the Indian medical tradition was replaced by an interpretation of the Hippocratic formulation of doctors’ duties in ways that stress the need to act paternalistically towards patients. Doctors in India, who may even have been favourably inclined towards self-determination at the end-of-life, were taught that life has to be preserved, at all costs if necessary.

The greater role of doctors and medical paternalism also have roots in Ayurvedic systems, where great faith is reposed in the medical community despite the issues of medical malpractice and institutional problems concerning hospital infrastructure. One can see a reflection of Ayurvedic ideas in doctors being seen as a rebirth of Dhanvantari, the lord of medicine. In Ayurvedic systems, there were codes of ethics, which can be found in the Sushruta Samhita and Charaka Samhita. The Charaka Samhita, for example, makes reference to autonomy and consent, although preference is afforded the physician in the case of, for instance, withholding the truth from the patient in his or her best interests, as opposed to full disclosure in order for him or her to make an informed decision.⁵³ Although medical paternalism and a larger role for doctors may in some jurisdictions result in greater reluctance to accept ADs (as individuals would presumably expect to defer to doctors about decisions at the end of life), this has not been the case in India. Instead, preference for the role of the physician in overseeing and enacting a bureaucratic process of reviewing the place and applicability of an AD has interestingly been accepted as a key component of the AD regime laid down by the Supreme Court, wherein medical practitioners have the greatest say in decision-making. This situation may well result from the fear of abuse by family members, as discussed previously, but is also consistent with the Indian medico-cultural context.

6.5 Survey Mapping the Level of AD Awareness

Autonomy, if it is to be realised, presupposes the knowledge required for an informed decision. Despite the legal permissions given by the Supreme Court, the common person in India remains unsure about the implications of AD legislation and the Common Cause ruling.

The question we come to now is this: to what extent is there awareness of the law as it currently stands? One way to understand perceptions of ADs is to conduct a survey amongst a broad swathe of people. To the best of our knowledge, no such survey had previously been undertaken in India since the Common Cause decision.⁵⁴

6.5.1 Demographics

We conducted an online survey using Google Forms. The survey collected information from individuals who were internet users and had obtained at least a graduate degree in any field. Our respondents do not represent the Indian population in any way; rather the population we selected was a highly educated, elite group of people.⁵⁵ Our final sample of 411 responses reflected a healthy mix of male and female respondents from the fields of medicine, science, the arts, engineering and law, with a mean 18.16 years of work experience. The responses were sorted by gender, qualifications and field of education for detailed analysis using simple statistical measures.

6.5.2 The Findings

In relation to their awareness of ADs, only 115 (27.9 per cent) of the respondents reported having some knowledge of ADs, meaning that 296 (72.1 per cent) had no such knowledge. Further questions about ADs were therefore not directed towards those with no knowledge of the subject matter. Amongst those who knew about such directives, 100 reported having thought about making them, but only 21 (18.2 per cent) had actually done so. Taking all respondents as a whole, only 5.1 per cent had prepared an AD.

We analysed the knowledge of respondents from the medical profession separately. Of the 147 medical professionals, 92 (62.6 per cent) had heard of ADs, whereas 55 (37.4 per cent) had not, and only nine (7.8 per cent) had actually made one. One question that we asked all respondents was whether, if they were unable to voice their preference about medical treatment, they would like to nominate someone as their proxy. We received a response from 226 respondents. Most (133, or 58.8 per cent) said they would want their spouse to act as their proxy. Seventy-one (31.4 per cent) chose their parents, and 18 chose family or their children. Only three said they would like a doctor to make decisions on their behalf if they were unable to do so. It is alarming that the doctors in our sample were only slightly better informed about ADs than our other respondents; roughly half were clueless.

Several findings emerge from these data. First, despite all of the benefits of ADs, not all are inclined to make them. Second, considerable effort is required to make ADs acceptable. Our survey revealed that although most of our respondents wanted autonomy over their medical treatment and the way in which they were treated, very few had an understanding of ADs, their legality or the process of making them.

The inference that can be drawn from our survey is that the likelihood of a well-educated Indian making an AD is rather slim. An added worry is that there appears to be little knowledge of ADs amongst medical professionals. Thus, the likelihood of a patient who has prepared an AD being treated by a medical practitioner with knowledge of ADs is remote. This situation, added to the fact that responsibility lies with the medical practitioner, makes AD implementation a challenge. There is thus an urgent need for a campaign by governmental and non-governmental organisations to spread awareness of ADs, their advantages and the need for them. The Indian courts seek legal grounding in other common law jurisdictions, and yet ground-level adoption is often neglected. Whilst the governments of other countries have put huge effort into popularising ADs, no such effort has been seen in India. Without knowledge of ADs, they will remain an unrealised right.

6.6 Conclusion

India has come a long way in jurisprudence concerning end-of-life care, culminating in the current guidelines on ADs laid down by the Supreme Court of India, which remain the law of the land. There remain, however, various concerns about this regime. The first is in relation to the lengthy, bureaucratic and unrealistic procedure that the court has laid down for the creation of an AD. Whilst at first glance, the court’s decision in Common Cause seems to have furthered decisional autonomy, the complex procedures it put in place may actually serve to disempower patients, reducing their control over decisions about their care and discouraging them from undergoing the AD process.

The second concern relates to the institutional dimension surrounding the legalisation of ADs. The Common Cause decision, a result of India’s activist judiciary, raises questions about the separation of powers by showering rights on people without any democratic dialogue. Implementation is also a challenge, especially when new legal requirements are introduced without careful consideration of what is practically feasible in a healthcare context. Because of the different workings of the legislature and the judiciary, the idealism of the judiciary seldom translates into ground-level practices. The Health Ministry, for instance, has yet to hold a consultation meeting with stakeholders concerning the procedure for ADs.⁵⁶

A third concern relates to how the law sits uncomfortably with the social reality of India, which is communitarian and family-based. Interestingly, despite the importance of the family being recognised in the PAD context, the courts have expressed concerns about the potential for abuse by family members in the end-of-life context. Such concerns have not, however, translated into an endorsement of the person-centric approach that ADs encapsulate. Instead, a key role for healthcare professionals is envisaged, and looks to have been generally accepted.

In terms of the way forward, our survey demonstrates a significant lack of awareness of ADs, which needs to be remedied. Concerns over the lack of AD knowledge have also been expressed in the context of PADs.⁵⁷ One of the potential solutions has been to set up legal aid centres to spread awareness and afford access to vulnerable populations.⁵⁸ Such centres may be a way to educate the public about ADs, although the bureaucratic procedures will need to be streamlined if such directives are to be widely taken up by the population. At the hospital level, ethics committees could be empowered to keep a check on futile treatments and even make decisions where necessary.