Legislation: the Terminally Ill Adults (End of Life) bill

In response to changes in public attitude towards assisted dying, several governing bodies within the UK are now deliberating its introduction. The eligibility criteria for assisted dying within different UK jurisdictions are broadly similar, and so we have focused on the Terminally Ill Adults (End of Life) bill ² for England and Wales, as this will affect the most people if it becomes law. To qualify for assisted dying, a person must fulfil several basic requirements, such as being an adult who has resided in the jurisdiction for at least a year and is registered with a doctor. Then, to be deemed eligible, they must also satisfy several key criteria that stipulate that they express ‘a clear, settled and informed wish to end their own life’ that has been arrived at voluntarily, meaning that they have not been ‘coerced or pressured by any other person’. Further, the request can only be considered in the context of a terminal illness, and this is defined specifically as an ‘inevitably progressive illness or disease which cannot be reversed by treatment’ such that the person’s death ‘can be reasonably expected within six months’.

Importantly, the bill also specifies conditions under which a person will not be regarded as having a terminal illness and would therefore not qualify for assisted dying. For example, they are ‘not to be considered to be terminally ill only because they are a person with a … mental disorder’; in other words, if a mental illness is their sole underlying medical condition, it is, on its own, an insufficient reason for requesting assistance to end one’s life. Thus, the bill limits access to those dying from a terminal physical disease. In practice, this can usually be determined without much difficulty, as in most cases the person has a recognised condition such as cancer, the prognosis of which is reasonably well-known. By contrast, determining whether they have capacity is more of a challenge, as both the cognitive effects of cancer and its treatment can affect thinking and thereby compromise a person’s ability to make competent decisions. To determine whether this is the case, formal testing is necessary, and for this the bill relies on the Mental Capacity Act 2005 (MCA). ³

Mental capacity: the MCA and the Four Abilities (4A) model

Individuals value having significant say over medical decisions affecting their health. Indeed, patient autonomy and the ability to exert authority in this context is the basis of informed consent, an ethical practice that is now integral to good clinical management and research worldwide. However, not everyone has sufficient capacityFootnote ^a to provide informed consent, and accurately determining who can make appropriate decisions for themselves can be difficult, especially when significant risks are involved. This is because on the one hand, a person who lacks capacity may decide on a course of action that puts them in harm’s way, for instance, by electing to end one’s life by way of assisted dying; on the other hand, there is the risk of denying choice to someone who can justly decide for themselves and rightly wishes to exercise this freedom.

The bill stipulates that a person seeking help from a medical professional to facilitate their death must have ‘capacity to make a decision to end their own life’ and that this should be determined ‘in accordance with the Mental Capacity Act 2005’. Of note, the latter assumes that all individuals (aged 16 years and over) are capacitous – that is, they have capacity to begin with – and that when appraising capacity, all practical steps are taken to ensure the best interests of individuals are safeguarded by favouring options that are least restrictive of their rights and freedoms. Crucially, capacity is deemed to be specific to a particular decision, and its assessment must not be based on the wisdom of the outcome of that decision nor indeed the status of the individual per se, for instance, whether the person has a mental illness or not. Therefore, in conjunction with the Code of Practice, the MCA provides a deliberative framework that aims to both protect and empower those who may lack capacity because of mental impairment.

Elsewhere around the world, the 4A model is more commonly used than the MCA – especially in the USA, where it was first developed by Appelbaum & Grisso ^{Reference Grisso and Appelbaum4} – and components of the 4A model overlap considerably with those of the MCA, as they both rely on functional tests to appraise decision-making (Box 1). According to these schemas, having capacity means being able to understand and hold information that then informs the process of deciding and being able to communicate the outcome. However, formal capacity assessments based on these criteria have been criticised for not factoring in the potential influence of a person’s values and emotions on decision-making. The key concern is that although the four abilities are useful for assessing the cognitive processes within decision-making, they are nevertheless insufficient because they fail to capture individuals who intuitively should not be deemed as having capacity because they are unable to adequately look after their own interests. In other words, current capacity tests do not appear to account for the welfare of the person. To remedy this, Hawkins ^{Reference Hawkins5} proposes two requirements, which can be grafted on to the 4A model as additional conditions. Specifically, when considering a decision, first, ‘[t]here must be good evidence for thinking that the patient is making a serious, prudential mistake here and now’; and, second, ‘[t]he patient must be known to have a condition that, in turn, is known to make those who have it more likely than ordinary to make prudential mistakes’. A prudential mistake in this context is a decision that is destructive to oneself, and wrongly choosing to end one’s life would clearly qualify as a prudential mistake.

Box 1 Components and comparison of the Mental Capacity Act 2005 (MCA) and Four Abilities (4A) model)

The box shows the components of the two frameworks for assessing whether a person has decision-making capacity and lists three key differences. In the MCA, the person’s inability to make decisions is best assessed before determining whether any inabilities are the consequence of an impairment or disturbance of the mind or brain. The four abilities in the 4A model are also sometimes referred to as pillars, skills or principles.

These proposed refinements to any test of capacity are relevant to our discussion of assisted dying, because although the bill before Parliament attempts to avert the inclusion of patients with a mental illness, not everyone who develops a comorbid mental illness will necessarily be excluded – meaning that some patients may well undergo an assisted death despite having impaired capacity that has not been detected. The scenario in which this consequential concern is most likely to materialise is that of depression occurring in the context of terminal cancer. This is because cancer is often incurable and terminal, and depression is a common comorbid mental illness.

Comorbid mental illness: depression in the context of terminal cancer

In the proposed bill, having depression alone will be insufficient to gain eligibility for assisted dying, but given that approximately one in four cancer patients have depression, and that their odds of being depressed are more than five times those of the general population ^{Reference Hartung, Brähler, Faller, Härter, Hinz and Johansen6} and further, it is common for terminally ill patients with comorbid depression to experience a decline in both their cognitive abilities and capacity, it is likely that a significant number of depressed patients will be assisted to die. Hence, the question of having decision-making capacity is critical. However, the precise impact of comorbid depression on capacity can be difficult to gauge: first, because having cancer can alter a person’s thinking, as can its treatment; and, second, because the effect of depression on cognition and capacity varies.

Further, cancer-related and chemotherapy-induced cognitive impairment, known colloquially as ‘brain fog’ and ‘chemo-brain’, respectively, affect 30–70% of cancer patients and typically present as fatigue and difficulties with concentration and memory, symptoms that overlap with those of depression. Which patients will experience these symptoms is difficult to predict. Similarly, the effects of comorbid depression also vary considerably from one individual to the next, and within the same individual over time. Overall, the main cognitive effects of comorbid depression include the development of negative and pessimistic biases, as well as cognitive rigidity, in which decision-making becomes inflexible and increasingly dichotomous such that it limits the person’s ability to reason and formulate alternatives to solve problems. These deficits compromise capacity in nearly one-third of depressed patients, ^{Reference Owen, Szmukler, Richardson, David, Raymont and Freyenhagen7} particularly limiting ‘appreciation’, the ability to apply information, including the consequences of one’s decision for oneself or one’s own situation. This impairment, along with the effects of depression on attention, memory, and speed at which thoughts are processed – all of which can interfere with decision-making and do so variably – casts suspicion on the authenticity of any choice a person makes when they are depressed. In other words, and put more simply, depression can disrupt decision-making capacity, and the extent to which it does this and the pattern of impairment it produces vary. This means that it is difficult to predict how capacity will be impaired, and this can only be known once the individual has been assessed. At the same time, it is important to note that although depression can ‘corrupt’ capacity, a significant proportion of patients with comorbid depression are relatively unaffected in this regard and for all intents and purposes manage to retain their capacity for decision-making. However, in the context of assisted dying, this seems counter-intuitive, as the decision to end one’s life when depressed seems like a serious prudential mistake. This is because suicidal thinking is a symptom of depression, and the desire to end one’s life when depressed is thought to be a feature of the illness rather than a well-considered decision, hence the intuition of allowing a depressed person to kill themselves – let alone assisting them to do so – being both morally and medically wrong. This lends support to the view that a welfare criterion should be added to any test of capacity applied to this unique circumstance, even though this will increase the time it takes to assess capacity, and do so at a time when time is arguably at its most precious.

What we propose: three suggestions for consideration

The assessment of capacity is a morally weighty responsibility, because declaring that a person lacks capacity deprives them of the right to make decisions concerning their health and well-being. At the same time, failing to identify a compromise in their capacity risks their making a decision that is inauthentic and has serious consequences. In assisted dying, the stakes could not be higher, and so proper procedures are necessary with stringent safeguards.

As we have shown, the bill as is stands will allow terminally ill patients to access assisted dying if they are judged to have capacity. However, many patients requesting assisted dying for a terminal illness will have a comorbid psychiatric disorder that is most likely to be depression. Therefore, excluding patients who only have a psychiatric condition, as specified in the bill at present, will not capture patients who develop a comorbid mental illness, and some of these patients may then have their death hastened with medical assistance when in fact they do not wish to die. To remedy this, we propose that the following suggestions be given careful consideration.

First, all patients who request assisted dying should undergo a comprehensive psychiatric evaluation and be subject to a robust assessment of their capacity. For this, we propose that a new test of capacity should be developed. This should retain the key components of the MCA and 4A model but include additional considerations that capture those elements of a person’s life that also significantly impinge upon their decision-making abilities, for example, assessing their ability to appraise their best interests and give consideration to their welfare, and gauging their suffering and willingness to endure hardship. ^{Reference Holmes, Lange, Stewart, White, Willmott and Dooley8}

Second, whatever form these assessments take, they must be performed by healthcare professionals with sufficient psychiatric knowledge and experience; ideally, they should be performed at several time points and on multiple occasions and involve as far as possible immediate family members and significant others, as well as carers and friends. Further, all relevant past psychiatric and psychological history should be sought, and, where practicable, any doctors who are familiar with the patient’s management should be consulted and their perspectives carefully considered. At the same time, it is important to explore when the person’s opinions on assisted dying were formulated, to determine whether their decision to end their life is a longstanding and settled view or one that is newly formed and has not yet been given sufficient consideration. This approach is akin to the determination of advance directives in which an earlier well-considered view is used to navigate decisions when the person no longer has capacity because of a condition that has deprived them of their cognitive faculties.

In addition to these suggestions as to what to assess and how, our third suggestion is that alternatives to assisted dying should always remain an option and be actively explored even when the illness is terminal. Further, throughout the process, and especially in personal interactions, the person’s suffering should be acknowledged, and they should be offered support and realistic hope.

Presently, the bill is in the process of being debated in Parliament, and it is possible that it will undergo revision and be modified. However, if it passes unchanged, clinical instances in which a person has comorbid depression will inevitably complicate the assessment of capacity, and the difficulties this creates will need to be addressed. These problems will be difficult to resolve when provisioning assisted dying, and so they should be addressed now, while the medicolegal and clinical details are still being formulated and before implementation of any legislation.

For all these reasons, we argue that if the bill passes and becomes law, further research must be the next step, ^{Reference Malhi and Hofstede9} and ideally this should occur before assisted dying becomes accessible. Crucially, until we can properly assess capacity in the context of assisted dying, we should perhaps reconsider whether we want this option in its present form or, indeed, whether we want it at all.