Exploring the role that autonomy plays in achieving a good death: a case study

A good death

The concept of death is difficult to define, with definitions traditionally focusing upon the physical cessation of life (World Health Organization, 2009). The terms death and dying linguistically and conceptually may be viewed not only as the point at which the physical body ceases to function, but the stages leading up to it, as well as the perceived process after physical cessation. Komaromy (2009) thus argues that that any attempt to define a single social view of death is likely to fail, and that it must be considered both a physical and social event, with religious, cultural, racial, and historical factors influencing any individuals and communities definition.

Given that death is so hard to exclusively define, it is unsurprising that what constitutes a good death is also problematic to define. This is further complicated by the fact that there may be competing definitions of a good death, as individuals are rarely without family and friends, and as such a definition needs to be considered not solely from a person-centred perspective, but also a family-centred one also. Paramedics often find themselves dealing with those in the terminal phase as relatives and carers often become anxious, concerned, and unable to cope, leading to frequent paramedic attendance (Ingleton et al, 2009). Balancing the different definitions and expectations of various (often opposing) family members, while also considering the importance of service-user autonomy, is often challenging in terms of providing a good death.

Despite the fact that there remains no universally agreed definition of a good death and that attempts to define it are compounded by a lack of theoretical and empirical exploration of the relationship between EoLC pathways and conceptions of ‘good dying’ (Watts, 2012), multiple authors have described attributes that contribute to what they consider to be integral to the process, that paramedics should consider. Attributes include (Goldsteen et al, 2006; Department of Health (DH), 2008; Watts, 2012; Brady, 2013):

Although there is considerable variability in the definition of a good death, it is evident that many authors cite control and service-user autonomy as central factors. Given that paramedics often have very little time to glean much of the information surrounding service-user wishes and preferences, in situations which they would normally commence lifesaving interventions, it is unsurprising that complexities may arise. Thus an understanding of service-user autonomy, advanced planning, and mental capacity is integral to good decision making.

Autonomy

Autonomy is defined by the Care Quality Commission (2014), in the context of health care, as:

Autonomy as an ethical concept sits within four ethical principles posed by Beauchamp and Childress (2009), which are now considered the most widely accepted set of principles in medical ethics:

It has been suggested by authors such as Gillon (2003) that respect for autonomy is the most important ethical principle of all; however, while this principle has come to be seen as key to the provision of health and social care (Twomey, 2009), Beauchamp and Childress themselves do not advocate this stance, suggesting that all four principles are equally as important to each other in achieving good care (Lawrence, 2007).

Respecting autonomy is one of the key principles in EoLC that paramedics should be aware of, and can take many forms, such as:

By considering individual and family autonomy in EoLC, paramedics are themselves complying with the standards of proficiency set out by their registering body, in that that they are expected to ‘understand the need to respect and uphold the rights, dignity, values, and autonomy of services users’ (Health and Care Professions Council, 2014: 7).

However, with occasional poor knowledge of individuals' wishes, compounded by possible differing definitions and expectations of what is considered a good death, paramedics may encounter conflict, similar to that seen within the Case Study 1, and for which an understanding of the role autonomy plays in advanced care planning, and mental capacity, with the UK context of health care, is required.

Edward was a 72-year-old male who lived at home with his wife Elizabeth, and had been suffering from terminal metastatic malignant bronchial neoplasm for 2 years. Edward and his family knew it was unlikely he would live another year.

I met Edward within my role as a rapid response community paramedic, 1 week after he had been diagnosed with a chest infection by his own general practitioner (GP). Elizabeth and his two daughters had called 999 when Edward had started to hallucinate and have difficulty in breathing one night. I arrived and suspected Edward was suffering from sepsis, and was unlikely, given his current condition and poor observations, to recover and/or to benefit from attempts at life prolonging procedures and treatments, which in Edward's case would have been futile.

Edward had in place advanced directives in the forms of a ‘do not attempt resuscitation’ order (DNAR), as well as a treatment escalation plan (TEP), which stipulated that if a time came that he was not able to make a decision for himself, he would not want any: intravenous intervention, including antibiotics, artificial nutrition and or hydration, or invasive urinary catheterisation. Most importantly, however, he did not wish under any circumstances to be taken to any hospital, however wanted to be moved to a local hospice he had previously received respite care in, and made as comfortable as possible, until he died with his family by his side. Edward, under no circumstances wanted to die at home, in part due to his wife Elizabeth's love of their family home, and his eldest daughters plan to live there once Elizabeth herself had died, and how his death there would have tainted these feelings.

Unfortunately, upon calling the hospice they had no beds and neighbouring hospices were unable to accept service users not registered under a local GP and/or not on their waiting lists, despite stressing his advanced directive wishes. Community hospitals were also considered but who, already under pressure from norovirus ward closures, were only accepting discharges from acute hospitals, leaving the only choice of an accident and emergency (A&E) department, which in this case was an entirely inappropriate place for a dying person.

Edward's somewhat unexpected deterioration had meant that there had been little preparation for hospice care, which alongside local capacity pressures, and commissioning policy in terms of GP practices and hospice admissions, meant that despite his clear and pre-prescribed intentions, stipulated within his TEP, a box of palliative care drugs was delivered, and between myself and an out-of-hours GP, Edward was made comfortable at the home he wished not to die in.

Edward died some hours later, in a home he did not wish to die in, albeit with his family by his side, comfortable, pain free, and without any invasive intravenous or urinary interventions.

Advanced planning

Autonomy for Edward in Case Study 1, in terms of attempting to ensure a good death, has been explored through the use of both his ‘do not actively resuscitate’ (DNAR) and his treatment escalation plan (TEP) forms, aiming in his case, for example, to ensure avoidance of invasive procedures, and the ability to die pain free with his family close by. Travis et al (2001) states that the fundamental aim of an expression such as advanced directives (ADs) is to provide a means for an individual to continue to exercise autonomy at the end of life, where due to lack of capacity, they may be unable to do so. Such ADs reflect society's desire to respect personal autonomy and allow for better deaths, better service-user experience, and fewer hospital admissions (Mullick et al, 2013).

ADs also reflect the need for family and friends to be aware of their loved one's wishes, and to reduce undue stress at having to make decisions that may have not been in line with their wishes. Johnston (2007) highlights that multiple studies have shown that neither health professionals nor surrogate decision makers accurately reflect what people actually want, with surrogates tending to err in favour of giving treatment which someone would not have chosen, and physicians being more likely to withhold treatment that the person said they would have wanted. Such disparity or confusion in treatment views highlights the validity of ADs as a means by which autonomy can be respected in achieving a good death, especially in situations where paramedics have been called, and are often looked upon to make important decisions in a short space of time, with no previous knowledge of the serviceuser or family.

Despite these advantages, ADs have been described by multiple authors as a way of helping individuals influence particular decisions about their future care (National Care Association, 2007; Twomey, 2009; Levi and Green, 2010). It is the term influence that this author finds challenging in relation to autonomy. Take for example the Office for National Statistics (2013), who state that only 50% of those who express a preference to die at home actually do, raising the question that if we, as clinicians, are unable to categorically guarantee all of the wishes expressed within ADs, are they tokenistic in nature, and if so, should individual expressions of autonomy just be used as a guide, and not a rule, to achieving a good-death?

In Edward's case, the TEP was used as a guide, and not a rule, as this was not possible. Paramedics will find themselves in similar situations, especially given the unpredictability and uncertainty surrounding the trajectory of death and deterioration. Mere completion of ADs does not necessarily improve EoLC, as in Edward's case and many others, they cannot be enacted completely. Rather, paramedics need to pay attention to their intent and not only their specific instructions (Rosenberg et al, 2013), in order to guide and inform their decisions about how best a service user would have wanted to die, a position which some might argue achieves a form of good death, and acts in the best interests of the service user and family.

Mental capacity

The concept of mental capacity for decision making is intimately connected to that of personal autonomy and advanced planning, and is reflected within the Mental Capacity Act 2005 (c.9), which includes that:

In Edward's case, for example, he had made certain decisions and expressed certain preferences, which some may consider unwise, in terms of refusing invasive life-preserving treatments. However, that such a decision and expression of personal autonomy is protected within legislation arguably means that Edward is able to ensure he will experience a good death, highlighting the role autonomy plays in achieving this aim and the legislative and professional responsibilities paramedics have.

It is widely accepted both in terms of ethicality and legality that when one lacks mental capacity, they also lack autonomy, and thus treatment and care should be provided in line with their best interests (Freyenhagen, 2009). Chapter 9, paragraph 4, subsection 1–7 of the Mental Capacity Act 2005 (c.9), for example, sets out how best to determine the best interest of an individual should they lack capacity, and does so in a way that fundamentally aims to protect an individual's autonomy, requiring the decision maker to:

Such requirements mirror some attributes of a good death, and acting in the best interests of an incapacitated individual, if done so in accordance with section 4 of the Mental Capacity Act 2005 (c.9), ensures autonomy is considered and protected where practicably possible, and arguably increases the possibility of ensuring a good death.

What is important to note is that if a service user does not have ADs, paramedics are still able to act in a similar way to Edward's case, by using the guide within the Mental Capacity Act 2005 (c.9). Thus paramedics can start, stop, or not even start life-saving treatments at all, without TEPs or DNARs, if it is considered to be within the service users best interests, and cannot be held liable if they have complied with chapter 9, paragraph 4, subsection 1–7 of the Mental Capacity Act 2005 (c.9) (Medical Protection Society, 2012).

While using ADs as a guide or determining a service users best interests, may contribute to a good death, some argue that if the wishes of an individual are not carried out exactly as they are prescribed, then this is said to be acting paternalistically, in so far as their pre-prescribed decisions are being overridden, ignored, or substituted.

It is recognised that much conflict surrounds the concept of paternalism and utilisation of ADs (Ekore et al, 2012). However, Twomey (2009) points out that if someone is being treated paternalistically, their decisions are being substituted by someone else, and thus paternalism will immediately fail to respect someone's autonomy, which is disconcerting considering it is considered integral to achieving a good death. This is not to say that paternalism is not justified in some circumstances (Buchanan, 2008), and further raises the questions as to whether individual autonomy should be considered merely as a guide in the achievement of a good death, especially within the challenging context of emergency and unscheduled care.

Mann and Cornock (2007) further highlighted conflict between an individual's ‘best interests’ concerning a good-death and autonomy, by questioning whether the autonomy of service users who lack capacity can be respected while also upholding another equally important ethical principle of beneficence. Beneficent actions are those taken to improve an individual's situation or to help prevent or remove possible harms, such as in Edward's case, where he was not moved to a busy, noisy, and uncomfortable A&E department to die, rather his autonomous decisions were respected, where practicable, and used as a guide to ensure the best death possible.

This author is confident that paramedics would be questioned and critiqued if they were to take an end stage service user in their terminal phase into an emergency department to die, if their wishes were otherwise, and rightfully so. However, this is compared against the possible questioning they may also receive for not respecting and enacting all of a service users pre-prescribed wishes, as set out in their ADs, if the situation (such as Edward's) would not allow.

This highlights the conflict between respecting service-user autonomy and providing a good death. However, if paramedics can be seen to have, where practicable, acted in the best interests of their service user, as set out within the Mental Capacity Act 2005 (c.9), which includes using ADs as a guide, then they cannot be readily criticised. Paramedics face these challenges in what is a fluidic and multifaceted area of health care, within an ever changing social context, which does not always allow for complete service-user autonomy.

Death within a social context

Empowering individuals to be and remain autonomous is challenging in regards to a good death when EoLC provisions are not unlimited and are subject to local and national availability restrictions and commissioning limitations (Foley and Komaromy, 2009), as seen in Edward's case specifically.

Kapp (2001) states that while respecting service users' autonomy is a worthy goal, it is frequently more costly because it requires more individualised care and treatment, especially in regards to TEPs. Thus there again appears to be conflict in empowering individuals and families to express their autonomy, if their wishes are unable to be enacted due to fiscal rationing or lack of availability. Baily (2011) argues that it is important to speak to the limits of service users' autonomy, and to emphasise that having a cost-conscious standard of care does not mean having no choice, rather it just means not having unlimited choice, and so autonomous expressions of treatment preference are important to ensuring such choices are enacted where practicable. Baily's (2011) point is highlighted in Edward's story, in that most, albeit not all, of his choices were enacted, thus autonomy did play a role in achieving a good death in part.

Paramedics may think that such commissioning decisions sit outside their role of face-to-face care; however, they do not. Paramedics will find themselves having to make decisions in the best interests of a service user and family, but also within the confines of what is available at the time. Again, if paramedics are seen to be complying with the best interest of a service user, and were possible using ADs as a guide to this determination, then they cannot be readily criticised. This supports the assertion that autonomous expressions should, and at times can only, be used as a guide in achieving a good death, and not as a rule.

The limitations of autonomy in achieving a good death are partly as a result of the increased medicalisation of death. Medicalisation is the process by which all conditions and problems come to be defined and treated as medical ones. Legitimised in a legal context, and for which there are perceived preventative and curative options, this process, which many have argued is unhelpful in terms of death and dying, has historically been viewed as reducing individual capacity, increasing service users' dependency, and dehumanising death and dying (Conway, 2010). Medicalisation sees death not as a normal life process (Van-Brummen and Griffiths, 2013), but within the context of commissioning of services, eligibility, and rationing in the face of fiscal constraint.

Paramedics can only respect individual autonomy in terms of a ‘good death’ so far as there are suitable resources available to enact the decisions made by an individual, and so they find themselves having to attempt perhaps not what the individual and family considered the best death, but a good enough death.

A ‘good enough’ death

Levi and Green (2010) describe a criticism underlying the overwhelming emphasis on the respect for service-user autonomy, explaining that some call it misguided, and argue that when an individual is at their most vulnerable, the primary responsibility should be to protect them from harm, and not to engage in quixotic quests to discern their exact, authentic wishes. This view which heavily contrasts traditional theories, is arguably applicable to paramedic practice, given the challenges paramedics sometimes face in respecting autonomy, determining best interests, and all within a limited social context. This view is somewhat mirrored by a small amount of authors (McNamara, 1997; 2004; Masson, 2002; Molan, 2012), who describe the concept of a good enough death.

As seen in Edward's case there are times within clinical practice that the ideology of service-user autonomy cannot always be realised in terms of achieving what the individual and family considered exactly to be a good death (McNamara, 2004). A negotiation process needs to take place in which relativism and pragmatism coincide in line with ADs, and an individual's and family's best interests, to form a perception of what an acceptable or good enough death might be (Masson, 2002), and enacted upon.

This author believes that this concept of a good enough death is often used, perhaps unknowingly by paramedics already, to attempt to negotiate between the service user's and family wishes, and the constraints and limitations they experience in everyday practice. Such an approach, which does not entail the lowering of professional aspirations due to the use of language of contingency (McNamara, 1997), takes into consideration the points covered within this educational article, such as autonomy, advanced planning, and capacity, and uses them as a guide to achieve a good enough death. If done so with the service users best interest at heart, it makes it difficult for paramedics to be readily criticised in what is an emotive and increasingly complex area of clinical practice.

Conclusions

This paper has reviewed what the provision of a good death might entail, and has explained some of the theories, complexities, and nuances of autonomy that paramedics may face in practice, in an attempt to increase the reader's understanding and confidence of this area. Respect of individual and family autonomy remains integral to the provision of a good death, but as seen in Edward's case, at times can only be used as a guide to provide what is known as a good enough death. Limitations and conflicts involved in attempting to respect individual autonomy in terms of tokenism, paternalism, and the medicalisation of death, highlight the multiple challenges paramedics may face. However, if what is done is in line with the individual and family's best interests, then paramedics are ethically, professionally, and legislatively compliant, and such efforts engender and epitomise the view that:

Key Points