Harmful Choices, the Case of C, and Decision-Making Competence

Abstract

In this paper, we make the case that a person who is considering or has already made a decision that appears seriously harmful to that person should in some cases be judged incapable of making that decision because of the harmfulness of the decision. We focus on the English case of C of 2015. C refused life-saving dialysis. The hospital wanted her declared incompetent to make this decision under the English Mental Capacity Act of 2005. The Judge argued that the consequences for a person’s welfare of their decision are irrelevant to the assessment of competence, a position labeled “internalism.” This aligns with an assessment of decision-making competence on a strictly cognitivist model. However, internalism misrepresents decision-making. The outcomes of decision-making processes should be part and parcel of judgments of decision-making competence, and in some cases are necessary for any judgment of incompetence to be made.

Keywords:

• Decision making
• law
• medicine
• philosophy
• professional-patient relationship

This article is referred to by:

An Externalist, Process-Based Approach to Supported Decision-Making

Preserve Patient Autonomy; Resist Expanding the Harm Principle to Override Decisions by Competent Patients

On the Relationship between Competence and Welfare

“Harmful” Choices and Subjectivity: Against an Externalist Approach to Capacity Assessments

Against Externalism: Maintaining Patient Autonomy and the Right to Refuse Medical Treatment

Against Externalism in Capacity Assessment—Why Apparently Harmful Treatment Refusals Should Not Be Decisive for Finding Patients Incompetent

When Limiting Liberty, Tread Carefully: Autonomous Free Choices Should Not Be Overruled Because of the Beliefs and Values of the Decider

The Value of Value in Decision-Making Competence Assessments

Determinations of Competence Ought Not to Be Primarily Grounded in Paternalistic Justifications regarding Welfare

Conceptual Compatibility and Transparency in Capacity Assessments

The Nature of Harm: A Wine-Dark Sea

We Don’t Offer What Can’t Be Chosen: Why Harmful Consequences Should Not Be “Decisive” in Assessing Decision-Making

Against Over-Protectionism: Riskier Decisions Require Clearer Evidence of Capacity But Don’t Call for Stricter Criteria

INTRODUCTION

This paper claims that there are ethical and conceptual reasons why doctors should take into account the serious harm of a patient’s choices in judging whether the patient is competent to make that choice. In short, we argue for an approach in which the welfare of patients is given weight in normative judgments about the competence of a person to make decisions. Our position reflects cautious protectionism. We hold that a person who is considering or has already made a decision that appears seriously harmful to that person should in some cases be judged incapable of making that decision because of the harmfulness of the decision.

We make two assumptions: first, that some people do not have sufficient capacity (and are therefore incompetent) to make decisions, and second, that a person who is unable to make an important or vital decision should have their decisional authority substituted. Both these assumptions have been contested, in the light of the discussion of the UNCRPD Article 12 (United Nations General Assembly 2006). The Committee of the Rights of Persons with Disabilities in its General Comment No 1 on Article 12 (Committee on the Rights of Persons with Disabilities 2014) interprets Article 12 to mean that no one lacks decision-making capacity provided sufficient support is offered to enable them to make decisions. The debate about whether substitute decision-making is ever permissible has followed (Byrnes et al. 2007; Ciavano 2014; Dawson 2015; Gooding 2015; Kohn and Blumenthal 2014; Mirfin-Veitch 2016). We hold that substitute decision-making is acceptable, but only after reasonable efforts have been made to support the person to make the decision for themselves.

In order to base the claims of this paper, in reality, we examine in some detail the case of C, considered in the English courts in 2015.¹ This contentious case has generated much discussion in medical and legal circles, but considerably less focus has been given to the ethical basis of the judgment. The ethical basis of the court judgment appears to be strongly related to internalism, an approach in which the possible harms resulting from decisions are not considered in the assessment of competence. In contrast, we take an externalist approach in which welfare judgments are part of judgments of competence. Internalism and externalism are defined further below

In preliminary support of our approach, we suggest that in cases such as C, the reality of decision-making is inconsistent with internalism. Rather, decision-making in many clinical contexts takes the form of supported decision-making, and it is on this background that competence judgments are made. A possible objection is that our approach is overly protectionist, by dint of being overly paternalist. However, there are numerous examples of paternalism in society, reflecting protectionism. As well as harm-reducing approaches to health and safety, we identify a general paternalism within professional health care, namely instrumental paternalism. In this, the health profession sets out the parameters of the care that they will provide, limiting the options that patients have in the name of acting upon evidence, in order to avoid unnecessary and serious harm.

The main thrust of our argument is to defend an externalist approach to competence judgments. We do this in three ways:

First, we show that within the standard ethical principles applied to decision making, which appear to support internalism, there is in fact a tension. This arises between the principle that welfare considerations are irrelevant to competence judgments, and the principle that each judgment should be decision specific.

Second, we suggest that normative judgments about the processes of decision-making are insufficient to provide a foundation for assessments of competence. This means that internalism, which focuses on decision-making processes does not provide a sufficient basis for judgments about competence.

Thirdly, having reviewed possible relationships between harm and competence judgments, we suggest a causal account of the relationship between decision-making processes and welfare which supports the externalist approach. In this approach, the harmful consequences of a decision would play a decisive role in the decision to assess the competence of the patient, and more importantly, in the assessment itself.

CAPACITY, COMPETENCE, INTERNALISM, AND EXTERNALISM: CONCEPTUAL AND TERMINOLOGICAL ISSUES

It is important to clarify some of the central concepts and terms we shall use. In particular, we will introduce a distinction between capacity and competence, and onto this, we will map the distinction between internalism and externalism.

In many sources, it seems that the terms capacity and competence are interchangeable. But other than when we cite others who use them interchangeably, we shall use the two terms to refer to distinct ideas. Capacity will be used to refer to putatively measurable or assessable abilities such as understanding and reasoning, which can be taken to capture the processes by which decisions are made. Competence will be used to refer to the fact that these abilities are possessed by the person to a sufficient degree for them to be able to make a decision. Incompetence refers to the fact that a person has insufficient capacities. To illustrate the difference, length is a measurable property of, for example, a piece of wood. But it is another question whether that piece of wood is long enough for a particular purpose. Analogically, length is to capacity as long enough is to competence. In decision-making terms, to be competent is to have enough decision-making capacity for the purposes of particular decision-making.

Internalism we shall take to refer to the idea that decision-making competence is entirely a matter of a person’s internal abilities. Buller puts it thus:

According to the internalist, a determination of a person’s competence to make a decision is based on the person’s internal mental ability, that is to say, whether the person has the decision making capacity to understand the relevant information pertaining to a decision, and to be able to reason and deliberate. (Buller 2001)²

In defining internalism, Buller appeals to several aspects of decision-making including understanding, reasoning, and deliberating. Other accounts of the constituent capacities of decision-making also refer to these. The UK Mental Capacity Act of 2005, and the MacArthur Treatment Capacity Assessment (Appelbaum and Grisso 1995; Grisso and Appelbaum 1998) both include understanding and concepts related to reasoning, Culver and Gert’s account of capacity includes understanding (Culver and Gert 2009, 5 of 22). These capacities can all be seen as cognitive or (more broadly) mental, and are presumed to be assessable without reference to the outcome of the person’s decision-making, so that “the patient’s actual decision does not enter into the determination of competence” (Culver and Gert 2009, 5 of 22).

For the externalist, in contrast, “the determination of a person’s competence to make a particular decision is directly related to external factors, particularly risk” (Buller 2001, 102). Risk can be said to be external in two senses: it represents an assessment by others of the probability and severity of harm given a person’s choice, and it focuses on the content of decisions rather than on the cognitive processes applied (internally) to that content by the decision-maker.

Internalism, then, focuses on capacities, such as understanding, or more broadly on the processes of decision-making. In contrast, externalism additionally refers to such things as risk, which do not plausibly refer to capacities. Making a judgment about whether someone has sufficient capacity—i.e., is competent—is, as we argue, where the external things, such as risk, have a role.

KINGS HOSPITAL NHS FOUNDATION TRUST AND C & V

The case of C&V is relevant because it appears to adhere to internalism. The judge (MacDonald J) rejected the case made by the Trust that C was not competent to make a decision, agreeing with C and her family that she was competent, even though her choice would lead to her death.

In this section, we shall first describe the case, then draw out the principles upon which the judge in the case based his decision. We shall then assess the justification of using this case as a basis for a wider analysis of competence and the role of harm.

The Case of C

C was a woman of 50 at the time the case came to court. She was refusing dialysis, which she required because her kidney function was affected by an overdose of paracetamol. It was agreed that she required dialysis in order to live. She had taken the overdose because, broadly speaking, the life she wished to live seemed outside her grasp. She had been treated for breast cancer, but owing to various financial and interpersonal difficulties, felt that life lacked anything attractive. Although her doctors had assured her that dialysis would be temporary, she had come to the point where she no longer wished for life-sustaining treatment. The Trust sought a declaration under the Mental Capacity Act³ that C was not capable of making this decision (under s 15) and authority to provide medical care and treatment (under s 4 A and s 15).⁴ The basis of the application in relation to C’s competence centered on her ability to accept, use or weigh the information she had been given by her doctors. The case was heard by MacDonald J in the Court of Protection. The judge heard testimony from psychiatrists for the Trust that C was not able to accept, use and weigh the information given to her by her doctors because her thinking was rigid, black and white, and catastrophic. For example, he heard from Dr. R that he:

was concerned that C’s capacity to “weigh in the balance the risks and benefits of accepting/refusing treatment” was compromised by “very rigid thinking style and her distorted cognition such as black and white thinking and catastrophic thinking e.g. that she will forever require dialysis and will never recover to a stage where she can live an independent life again”.⁵

We take it that “capacity” here means what we mean by capacity, that is it refers to the degree of ability C had to do the weighing. Dr. R felt this ability was compromised to such a degree that C was not competent. But MacDonald J also heard psychiatric testimony that she was competent, testimony from her daughters that the basis of her decision to refuse dialysis was consistent with her past life and values, and testimony, which was material to his final decision, regarding the disclosures given to C about the likely success of dialysis. While at times the success of the dialysis was said to be assured, at other times it was not. In the light of this, the judge refused the Trust’s application re C’s competence, which meant the second application (for the authority to provide medical care) did not need to be considered.

Key Parameters in the Assessment of Competence

Assessment of patient competence is associated with several seemingly fixed ethical or quasi-ethical parameters. A useful summary is to be found in MacDonald J’s words from C:

• 26.… Thus capacity is required to be assessed in relation to the specific decision at the time the decision needs to be made and not to a person’s capacity to make decisions generally.

• 28.… a person is not to be treated as unable to make a decision merely because he or she makes a decision that is unwise.

• 29.… the outcome of the decision made is not relevant to the question of whether the person taking the decision has capacity for the purposes of the Mental Capacity Act 2005 … .

(Note that in #29 “has capacity” means “is competent” in our terms.) Further describing #28 the Judge, citing an earlier judgment by Jackson J says:

The temptation to base a judgment of a person[‘]s capacity upon whether they seem to have made a good or bad decision, and in particular on whether they have accepted or rejected medical advice, is absolutely to be avoided. That would be to put the cart before the horse or, expressed another way, to allow the tail of welfare to wag the dog of capacity.⁶

(Jackson J uses the term capacity here to refer to what we refer to as competence.) While there are some differences between #28 and #29, they share the idea that welfare is irrelevant to competence judgments. Thus, MacDonald J has only two principles, the principle that welfare is not relevant to the assessment of competence (internalism), and the principle that assessment of competence must be decision specific.

Note that MacDonald J’s position does not imply that welfare is irrelevant to making decisions, and nor does internalism: Both externalists and internalists accept that people generally decide what to do at least in part based on their own assessment of the impact on their welfare of the various options before them, and tend to choose the one that promises the best outcomes. Thus, Jehovah’s Witnesses may choose to refuse a blood transfusion because the consequences of acceptance are spiritually disastrous, and spiritual matters weigh more heavily with them than worldly life (Watch Tower Bible and Tract Society of Pennsylvania 2012). Welfare as seen by the patient seems to be a primary foundation for decision-making. MacDonald J’s point is that the welfare implications of the decision are not relevant in determining whether someone is competent or not.

As noted in paragraph 29, the argument as to whether C was competent refers to the Mental Capacity Act of 2005. As already noted, this piece of legislation defines capacity (the abilities required to be capable of making a decision) in terms of the person’s internal thought processes. These capacities include the notion of weighing (“weigh in the balance”). This is a metaphor implying such things as making the information part of a decision (giving it a place in the scales if you like, even if the weight recorded is little or nothing), and evaluating and comparing various predicted outcomes described in the information given.

On the surface, it looks as though it would be difficult to justify externalism, and still hold on to the principles adumbrated by MacDonald J and a cognitivist model of decision-making. They seem designed to de-couple judgments of harm from judgments about decision-making. Externalism, on the other hand, includes a focus upon the outcomes of choices in assessing competence. But we shall argue that in the competence judgments, the principled ideas which set out the ethical foundations stand in a more nuanced relationship to one another than at first appears.

Significance of the Case

Any analysis of the relationship between harm and competence which focuses on a single case and/or judgment will have limitations. For example, a judgment, in any case, may be idiosyncratic or mistaken; and the case may not be relevant to jurisdictions other than its own. Furthermore, the relationship between legal analysis of competence and the legal and ethical principles which underlie this and actual practice may not be straightforward.

With respect to MacDonald J’s judgment, it has been referred to in subsequent cases, including his summation of the principles of the Mental Capacity Act of 2005, under which his judgment was given.⁷ Moreover, MacDonald J’s judgment refers directly or indirectly to important earlier judgments and discussions.⁸

With respect to the relevance of the judgment to US law and practice, a particular debate has arisen around the relationship between the MCA’s measures of capacity to which MacDonald J refers and those of the US. It has been argued that the US attribute of appreciation (part of the MacArthur assessment) has no equivalent in the MCA (Ryan, et al. 2019). However, Zhong et al. reply convincingly that the difference is largely verbal. The appreciation and rational manipulation requirements of the US test are broadly equivalent to the understand, retain, use and weigh elements of the UK test, and practice is not affected (Zhong, Sisti, and Karlawish 2019).⁹ Thus, though limitations remain, both the case and judgment are of general significance.

With respect to the relationship between practice and the legal and ethical analysis of competence, this suggests a potential further limitation to this analysis. Reference to the Mental Capacity Act of 2005 and to the MacArthur Treatment Capacity Assessment is to be found widely in professional guidelines and publications (Douglass 2017; McCulloch and Taylor 2018; National Institute for Health Care and Excellence 2018; Bourgeois et al. 2019; Barstow, Shahan, and Roberts 2018). These clearly aim to guide practice, but the extent to which they actually do may be limited. Amongst other things, there is evidence that few doctors use formal competence assessment tools (Young, Douglass, and Davison 2018; Jayes, Palmer, and Enderby 2017; Jayes et al. 2020).

Nonetheless, clarity about the legal and ethical expectations is important, as it is this that practice is judged against.

SOME DIFFICULT REALITIES OF DECISION MAKING IN MEDICAL CASES

In many instances, decision-making in medical cases presents a very different picture to the one implied by internalism. From an internalist perspective, the patient’s decision-making is abstracted from the interpersonal character of their situation, and the judgments about their competence are likewise abstracted and divorced from what the decision is actually about and what it might mean for the patient and others. C is a particularly interesting example, in this regard, because despite the standard internalist analysis of C, which is explicitly applied by the Judge and expounded by the standard ethical principlist model, the details of the case reveal a very much more involved picture than this abstraction allows, as is usually the case.

C was hospitalized for a number of weeks, and her carers, friends, and family had many contacts with her. For example, MacDonald J records an occasion when she was visited by her father and a friend:

Following interventions from her father and a friend on 29 September, C changed her mind and consented to treatment. It is clear from the medical records that C appears, with the assistance of her father and her friend, to have undertaken an exercise of using or weighing information as it is recorded that an hour was spent talking to C about her grave medical condition, her chances of recovery, and her prognosis for the future.¹⁰

It was in this context of constant interactions between C, her family, and carers that the doctors began to fear that C was making decisions that did not reflect the advice she was receiving and to doubt her competence.

In addition, in their assessment of the patient’s competence, the doctors in C and in other cases bring a significant focus on “the good.” Their concerns reflect not only the focus on autonomy but also—in the traditional principlist analysis—on beneficence (doing good) and non-maleficence (avoiding harm) (Beauchamp and Childress 2019). The doctors’ concern to do good is commonly seen as being in potential conflict with respect for autonomy, though Cohen offers a more nuanced typology of the relationship between the two principles (Cohen 2019) and Eastman and Hope argue that there needs to be a balance between the two concerns in making competence judgments (Eastman and Hope 1988). Where doctors resort to the courts it is often in the name of the patient’s wellbeing, as in the case of C. This focus on welfare is given scope—where doctor and patient agree—in a form of a trusting contract. Philosophically this represents epistemic trust in a practical sense. This contract recognizes the role the doctor is fulfilling, enabling “the good.” Whilst the doctor accepts responsibility for care, they trust the patient will allow them to fulfill their role (Fonagy and Allison 2014). This reflects elements of supported decision-making.

It is interesting, in this connection, to note that MacDonald J’s judgment records clear instances where C shows her awareness of this beneficent role of the doctor. She states that “They [the doctors] are doing their best to do everything they can for me and unfortunately that is not what I want” and “I know they need to save lives. But I’ve chosen a different route.”¹¹

As a matter of fact, then, the case of C, as with many others, does not fit well with an internalist account. However, from an ethical viewpoint, it may be feared that a particular picture of C’s capacities and competence might arise in the course of the doctors’ close engagement with C, her family and friends, and other health professionals and that they might then be unable to subject that picture to objective assessment. Ethically, protection from this could justify the abstraction of decision-making processes from the context in which C found herself.

However, abstracting judgments of people’s competence from the realities in which decisions are actually made, risks prejudging the question of whether harm judgments are separable from competence judgments. Since externalism focuses on the harm of decisions, reflecting the ethical concern with non-maleficence and beneficence, and holds that this focus is intrinsic to making competence judgments, this prejudgment of decision-making is ethically as well as conceptually conflicted.

EXISTING ELEMENTS OF PATERNALISM

One potential ethical objection to externalism is its paternalism. In the case of C the doctors in judging C incompetent act paternalistically, substituting their ideas, or perhaps society’s ideas, of “the good” for the ideas of their patient. This replacement of the individual’s own idea of what is good by the ideas of others is at the heart of paternalism. In accusing doctors of paternalism, those who support internalism usually have as their primary focus “ends paternalism.” In the medical context, ends paternalism is generally rejected, as the patient is assumed to know their own wider goals and preferences and to make decisions with reference to these. In the case of C, the judge heard from a member of C’s immediate family that C’s decision was “not only fully thought through, but also entirely in keeping with both her (unusual) value system and her (unusual) personality.”¹²

MacDonald J acknowledged the existence of alternative sets of values and beliefs in his conclusion:

The decision C has reached to refuse dialysis can be characterised as an unwise one. That C considers that the prospect of growing old, the fear of living with fewer material possessions and the fear that she has lost, and will not regain, ‘her sparkle’ outweighs a prognosis that signals continued life will alarm and possibly horrify many … .¹³

However, MacDonald J clearly believed that it would be unacceptable to substitute any of the expectations, reasons or values, of others, for the reasons and values of C herself. This constitutes a clear rejection of ends paternalism.

It should be noted, however, that this substitution of what is good or bad for a person is not always held to be a bad thing, even in liberal democratic contexts. For example, paternalistic laws such as those making wearing a seat belt compulsory are widely enacted, including in the US (at both federal and state-level) and in the UK (Wikipedia n.d). In a pandemic, laws restricting travel or social interaction, are partly paternalistic: that is to say, they are aimed in part at the good of each individual. The person is coerced by these laws to take their own safety into account, even if their personal values or preferences make liberty, convenience, bravado, fashion, or some other consideration more important.¹⁴

In addition, ends paternalism is not the only form of paternalism. Within clinical medicine, instrumental or means paternalism is well established. The means to recover health offered to people in many societies are those which have been developed and established by those regarded as having special knowledge and skills. In the developed West, and many other parts of the world, these are the professional medical specialists and the supporting research and commercial institutions (e.g., universities, drug companies). Patients are not expected to know that much about the treatments offered, nor to receive or follow advice from outside the socially legitimate channels. In short, the choice of the means has been handed over almost entirely to people other than the patients to decide upon.

Of course, this does not mean that the patient has to accept the offer or offers of treatment. However, instrumental paternalism places a clear limit on what the patient can demand of her carers, and on the range of options offered. Means paternalism represents a judgment founded on the balance of harms and benefits as assessed by health professionals.

DECISION-MAKING PROCESS NORMS AND THE ROLE OF HARM

A degree of paternalism is then well established. But when it comes to individual adult patients’ treatment decisions, it is generally agreed that paternalism is justified only when the patient lacks the ability to make her own decisions. Moreover, the internalist position aims to separate concerns about protecting the patient from judgments of competence. Accordingly, paternalist instincts are restricted to motivating concerns about competence. They should not, MacDonald J suggests, be part of a judgment about competence.

In contrast, the externalist approach is expressed in setting a higher bar for competence assessment where harmful outcomes may ensue. This approach is generally preferred to one in which choices, competent or otherwise, may be overruled on the basis of the harm they may produce. The preference to adjust the bar for competence, rather than the bar for intervention in a person’s life, is a very commonly taken position following on from the seminal work of Buchanan and Brock (1989; see also Brock 1991). For a person to be regarded as competent to make a choice which will—or can be reasonably expected to—lead to harm to the person, that person will have to demonstrate greater decision-making capacity, than that same person would need to, in order to be regarded as competent to make a choice which will—or can reasonably be expected to—lead to little or no harm. In the traditional example: you have to have greater decision-making capacity (i.e., be more able to make decisions) to be adjudged competent if the choice is life or death and you want to choose death, then you have to have if the choice is tea or coffee and you want to choose tea.

This section proceeds as follows. First, it will be argued that the fundamental ethical and legal principles set out by MacDonald J are potentially inconsistent and that this is revealed in his judgment.

Second, it will be argued that judgments about incompetence, such as those offered in C&V require reference to the harm of the decisions made. Without this reference, there is no basis for assessment of any particular level of decision-making capacity as sufficient for competence. To illustrate this claim, we shall suggest that thought processes very like those alleged of C can be described in others, who are competent decision-makers. The example we shall use is that of climate activist Greta Thunberg. Both C and Greta Thunberg have similar decision-making processes (viz. rigid, black and white, and catastrophic). Yet Greta Thunberg is rightly regarded as competent to make the decisions she does about focusing her life on climate activism. It follows that there is no equivalence between the level of decision-making capacity each reveals (which is broadly the same) and their respective levels of competence. Reference to the harm of decisions is the reason C is incompetent, but Greta Thunberg is competent.

Third, it will be suggested that the correct analysis of the role of harmful outcomes in competence judgments involves recognizing that capacity judgments, in general, are not in many cases a sufficient basis for competence judgments. We will suggest that an explanatory connection can exist between decision-making processes and harmful choices. Proceeding on the judgment that a choice is severely harmful to the person and that some reasonable alternative was available to choose, an investigation of the level of the person’s decision-making capacity should aim at finding if there was any form of non-normative cognition or decision-making process causing the person to make the choice. A judgment about their competence involves an assessment of whether the level of decision-making capacity reaches that required to make such a harmful choice.

Inconsistency in the Ethical Principles

MacDonald J, reflecting the standard ethical expectations in relation to competence, insisted that welfare was irrelevant to decision-making assessment and that every assessment of decision-making competence must be decision specific. There is however a potential tension between these two claims, which implicitly emerges in MacDonald J’s judgment.

In C the judge indicated that judgments must not be about competence to make decisions in general. This implies that a person could be competent to make one decision but not another. This is a widely accepted and indeed almost self-evident belief, especially if we consider decisions made in quite different domains: say surgery and engineering. There is little reason to think that a surgeon would make competent engineering decisions, for example (see Culver and Gert 2009). What we need to know, as Buller (2001) rightly insisted, is what distinguishes the specific decisions from one another in the way required to assert that a person is competent to make a specific decision (or type of decision) a but not competent to make a specific decision (or type of decision) b.

In C, MacDonald J felt C was competent to make the choice she made: presumably this is a judgment about her competence to make the specific decision (or to choose between the particular options) she was faced with. The judge did not appear to allow what was at stake with regard to C’s welfare to influence his decision toward not finding her competent. That is to say, he found she was competent despite the fact that her specific decision to refuse dialysis was (accurately)¹⁵ predicted to have the consequence of death in a short time. But, despite the judge’s own principles, when considering the specifics of the case, he may have allowed welfare considerations to influence his finding that she was competent.

Consistent with an internalist viewpoint, the judge noted that the rosy future sometimes promised by her doctors was not as rosy when looked at from C’s own point of view. From her view, whether or not the dialysis was medically successful it would not enable her to “attain her former lifestyle” nor meet “her desire not to get old and lose her appearance.”¹⁶ The judge acknowledged that C’s subjective view might not reflect the views of others.

C’s decision is certainly one that does not accord with the expectations of many in society. Indeed, others in society may consider C’s decision to be unreasonable, illogical or even immoral within the context of the sanctity accorded to life by society in general.¹⁷

He went on “None of this however is evidence of a lack of capacity.”¹⁸ He resisted the idea of substituting a different set of values for C’s. What mattered in his judgment was that she had used her values—however strange they might seem—to make her decision, weighing the information she was given in the scales which those values provided.

However, the judge adds an interesting further set of reflections on the specifics of the case.

However, I am not satisfied that the medical records bear out the assertion that C was, in fact, receiving uniformly positive and reassuring information concerning her prognosis (most especially in relation to the likelihood of her being able to live a life without dialysis).¹⁹

Summarizing he notes:

it is not in my judgment accurate to characterise the prognosis C was being given as consistently positive. Her more categoric responses in respect of her prognosis must in my judgment be seen in this context when determining whether they are probative of an inability to use and weigh her prognosis in her decision making.²⁰

It might be recalled here that one of the elements of reduced capacity alleged of C was her categorical (black and white) thinking. The categoric responses are those in which she seemed simply to deny that dialysis would help her in any way. It was these responses that particularly concerned the psychiatrists who judged her incompetent. But the judge notes that these categoric responses must be understood in the light of the fact that C was not always told the same thing about the likely success of dialysis, and that it was when she was given less positive predictions that she tended to deny that dialysis would be of any good to her. The implication seems to be, that if we want to know whether her categorical denials show a sufficiently reduced capacity (or ability) to use and weigh the prognosis she was given (and hence incompetence to decide), we need to understand that she was as often given reason to doubt its efficacy as to expect it to work fully. MacDonald J concludes that “in the context of the information given to C regarding her prognosis … these fears on the part of C cannot be considered irrational.” ²¹

This suggests the judge thought that the assessment of C’s ability to use and weigh information should be relative to the information she actually had about its likely benefit. When she had the information that the dialysis was less likely to benefit her, and was more likely to end up harming her, the Judge seems to say, this makes her decision to refuse dialysis more understandable (rational). Thus, while her decision to refuse dialysis was still no doubt (from an outsider’s view) the worst of the two options, the alternative was not necessarily so starkly better that the decision was difficult to understand (irrational). It was partly in the light of this explicitly externalist perspective that the Judge declared C competent.

The judgment thus leaves open the possibility that C’s competence might have been assessed differently if the objective difference between the benefit of dialysis she had described to her and her assessment of the value of dialysis had been starker, but she had persisted in refusing the treatment. This is not to claim that the Judge would have made a different decision based on other facts about the relative harms of the options before C. But he seems to offer himself reasons that would have supported a judgment of incompetence had the specifics of the case, particularly with respect to the comparative harm of C’s choice, been different.

Potentially, this could be analyzed as a failing of MacDonald J’s reasoning. However, it can be seen to follow quite logically from his concern to judge the case with proper attention to the facts specific and relevant to it. MacDonald J’s approach reveals a tension in his explicit commitment to internalism, and this emerges in his reasoning because the tension is present and inherent in the two principles he adumbrates.

The Relationship of Harm and Incompetence

In the previous section, it was argued that the requirement for competence assessment to be both unrelated to welfare and decision specific is potentially inconsistent. In this section, we argue that judgments of competence cannot rely wholly on internal elements of decision-making. In the light of this, harm presents itself as a means to supplementing judgments about internal decision-making processes (capacities), in order to make competent judgments.

Recall that an internalist position holds that a judgment of competence has to be made about elements of the process of decision making in the absence of judgments about its outcomes for patient welfare. For example, in the case of C, the Trust’s psychiatrists, concerned about C’s ability to use and weigh information, described her lack of capacity in terms of two things. First, her failure to believe certain pieces of information (specifically, that dialysis would give her kidneys a chance to recover to functioning); and second, her thinking process which was described as rigid, black-and-white, and catastrophic.²² Implicitly, the psychiatrists endorsed an account of decision-making capacity measured on a scale where rigid, black & white, and catastrophic thinking were unacceptable poles. While these scales are not explicitly designated, they might be thought of as scales of flexibility, nuance, and moderation. On this scale, C was assessed to be at a point below the level of competence (or so the psychiatrists believed). Her thinking was too inflexible, too stark, and too immoderate to be competent.

This implicit basis of assessment represents a series of related evaluations. It implicitly values true belief; greater flexibility and nuance of thought; and a moderate assessment of outcomes. Internalism cannot avoid some measure of evaluation of this sort if it is to allow that anyone is ever decision-making incompetent (or has low levels of decision-making capacity). Analyses of this sort, responding in various different ways, and in various different contexts, to the highly cognitivist focus of much of the literature, are to be found in Banner (2012), Charland (1998, 2001), and Tan et al. (2007).

But the problem for internalism is that such evaluative features of decision-making capacity, though they may set up reasonable norms, do not necessarily support a case in and of themselves for a judgment of insufficient decision-making capacity (i.e., of incompetence). It is not necessarily the case that a rigid, black-and-white, catastrophic decision-making process is inherently flawed thinking, even if it might be unusual and might not match up against the norms of good decision-making in many cases.

An example might be found in the reasoning of climate activist Greta Thunberg. Thunberg has been diagnosed with Asperger’s Syndrome. She has been reported as saying that the disorder helps her see things in “black and white” and she has consistently identified the world’s choices as extremely limited: “You can’t be a little bit sustainable. Either you’re sustainable, or not sustainable” (O’Malley 2019). It could be argued that Thunberg fails to present reality accurately, for example, in insisting that sustainability is a stark binary when there are in fact degrees of sustainability available.²³ The question is whether, even if this is true and it is true also that she thereby demonstrates some reduced level of decision-making capacity, this is a sufficient reduction to validate the further judgment that she is not competent. A perfectly reasonable response is that her decisions—including a life-changing focus on climate activism—seems altogether rational, in the light of what she is drawing attention to, viz. the potential disaster of climate change for the current children of the earth. Given her sense of this impending threat of harm, a reasoning style involving black and white, and catastrophist thinking, has a strong claim to be competent, however, removed from the usual norms.

It follows, given it is not sufficient in Thunberg’s case merely to evaluate her cognitive processes to support a judgment of incompetence, that the same will be true of C. Indeed, in general, it is not enough to point out that a decision-making process is non-normative in some way in order to show incompetence. There must be some other judgment at work alongside these.

In general, this should not be a surprise. It is difficult to see what other means there are of judging that a particular process of decision-making, which seems unusual, is in fact not merely unusual but also incompetent. Judgments about the mechanisms of a person’s way of making decisions can only tell us if the decision-making has matched normative standards applied to decision-making processes in the abstract. They cannot also tell us if the decision-making processes are appropriate in the circumstances. This judgment, about whether a decision-making process is appropriate has to assess the circumstances. A primary element of the assessment of circumstances in many medical cases relates to the welfare implications of choices made in those circumstances.

However, this raises a problem: the form of this relationship between the outcome of decisions and competence seems difficult to model. In the next sections, three options are considered, then rejected, and an alternative (causal) model is suggested.

The Relationship of Harm to Competence Judgements

At first sight, there are three ways in which welfare threatening decision-making and some sort of decision-making incompetence could be related.

First, the relationship might be constitutive, poor decision-making being part of what is meant by lack of competence. Second, the relationship might be one of some systematic connections between riskiness of choice and decision-making incompetence. Or third, the relationship might be indicative—that is to say, poor decision-making might act as an alert that there may be decision-making incompetence. We shall advance the fourth possibility, that the choice for a harmful outcome, where it is explained by a non-normative decision-making process, is what enables the judgment that the non-normativity represents incompetence.²⁴

The first (constitutive) approach is that a poor decision would be sufficient to reveal an incompetent decision-making process. A version of the constitutive approach is alleged (by Wicclair 1991) to be part of the position taken up by Buchanan and Brock (1989). Brock (1991) writes:

We [Buchanan and Brock] appear to hold that justified paternalism requires an independent finding of incompetence, he argues, but in reality make an assessment of whether paternalism is justified a part of the determination of competence. It becomes, in effect, merely true by definition and so an empty tautology that patients cannot have their competent treatment choices set aside for their own good. Whenever others believe that patients’ treatment choices can be justifiably interfered with for their own good, those choices can simply be labeled as incompetent. (Brock 1991, 108–109)

Brock admits the force of Wicclair’s objection, though he denies it applies to the position he and Buchanan espouse. In effect, Wicclair’s point is that this approach makes the judgment that someone is incompetent a consequence of the judgments that they have chosen a harmful outcome, or, in other words, judgments of harmfulness are constitutive of judgments of incompetence. We agree with Wicclair (and Brock) that this view of the relationship between harm judgments and competence judgments is untenable.

The second possibility is that there is a systematic but non-constitutive relationship between the harmfulness of a decision and decision-making capacity level and hence competence. As Buller (2001) suggests, intuitively, the ability to undertake a task is linked to its complexity and difficulty. A task is more of a challenge to undertake, and therefore requires higher levels of ability, if it is more complex or difficult. A decision-making task requires greater capacities if the decision involves complexity or difficulty of understanding or reasoning. The difficulty and complexity presented by a decision-making task may provide a basis for judging whether a person’s decision-making abilities are equal to the task.

There is no doubt that many profound or risky decisions are challenging in this way. For example, it could be argued that riskiness is constituted by a complex interaction of likelihood of outcome (which may be assigned some form of probability) and severity of outcome (where severity may be described on a range of dimensions, such as physical pain, psychological pain, ability to function in society, and so on). This is reflected in the use of subjective expected utility in combination with decision trees in decision analysis (Detsky et al. 1997; Naglie et al. 1997). This account of decision-making seems to have some truth about it. But it doesn’t help the case that needs to be made. A risky option (high likelihood × severely harmful multidimensional outcome) doesn’t seem necessary to be more complex than (say) a low-risk option (low likelihood × mildly harmful multidimensional outcome) or even a beneficial option (likelihood × beneficial multidimensional outcome).

Likewise, in C, the Trust based its judgment that C was incompetent on her rigid, black and white, and catastrophic cognitions. But this description could be applied in any number of different cases, with very diverse levels of consequent predicted harmfulness.

On the surface of it, then, Buller’s arguments provide a basis for rejecting externalism. The externalist relates the consequences for a person’s welfare to their decision-making capacity, but Buller suggests this relationship is imaginary.

This leaves the third option, which has been labeled an indicative relationship between harm and competence judgments. An indicative relationship is one in which a specific poor decision might occasion questions about decision-making processes. This is in fact the relationship between harmful decisions and decision-making incompetence which internalism accepts. For example, the internalist accepts that a doctor may seek further information about a person’s capacity to make decisions if they make poor decisions. Likewise, it seems to be rather like Buchanan and Brock’s process focus.

We argued that the competence evaluation should address the process of the patient’s decision making and how well the patient’s understanding and reasoning leads to a decision that promotes the patient’s own aims and values. (Brock 1991, 109)

The connection which Brock describes is one in which a decision-making process (understanding and reasoning) “leads to” a decision. We shall focus on situations where decision-making processes lead to—or cause as we shall say—risks to the patient’s wellbeing.

An Explanatory Account

We accept that a harmful choice can indeed play an indicative role, occasioning doctors to look more deeply at the decision-making processes. However, we propose that in many cases, in order to judge that the decision-making process failed to reach the level of competence, one has to take into account the harm that the choice would cause.

It might be argued that this is not in effect any different from the indicative approach, in which while an investigation of a patient’s competence might be undertaken because of a harmful decision, the actual judgment of competence is based independently upon features of the patient’s decision making processes. That these decision-making processes caused—or would have caused—harm is not relevant to this judgment of competence.

But consider again the case of C. As has been argued already, the claim that her thinking was “rigid, black and white, and catastrophic” is insufficient to establish that she was incompetent, even if it shows some reduced level of capacity. We can find situations in which the same features of thought would be regarded as appropriate. So, the decision-making process alone doesn’t seem to be the decisive factor. What makes the reduced capacity within the process reason for a judgment of incompetence is the decisions it gives rise to.

This connects harm to competence judgments in a reasonable way. It provides the basis needed to turn a judgment about the unusual nature of a person’s decision-making processes into a judgment about the person’s competence.

CONCLUSION

The accepted approach to decision-making competence is internalist, as illustrated in MacDonald J’s principles in the case of C. Adherence to internalism, and the cognitivist basis of competence assessment together support the idea that judgments of harm are not relevant to judgments of incompetence. However, externalism better reflects the realities of decision-making and competence judgments. Its focus on the harm that decisions may bring upon people reflects a reasonable protective concern for people’s wellbeing. In addition, externalism provides a means of moving from an assessment of capacity to a judgment of competence. On its own, in many cases, internalism cannot provide this. Harm resulting from decision-making is both ethically and logically relevant to competence judgments.

Notes

1 Kings College Hospital NHS Foundation Trust and C&V [2015] EWCOP 80 IN THE COURT OF PROTECTION 30/11/2015 Case No: COP 1278226. This case will be referred to as “C.”

2 Buller bases his account on Wilks (1997, 101–102).

3 Mental Capacity Act of 2005, c.9

4 C at 5 and 6.

5 C at 43(ii).

6 Heart of England NHS Foundation Trust v JB [2014] EWHC 342 (COP) at [7].

7 See for example: Newcastle City Council v TP and FW [EWCOP B4]; London Borough Of Brent and (1) NB, (2) SA, (3) AD, (4) MB, and (5) SB [2017] EWCOP 34; R (on the application of Conway)—and—The Secretary Of State For Justice—and—Humanists UK (1) Not Dead Yet (Uk) (2) CNK Alliance Ltd (3) [2018] EWCA Civ 1431; A Local Authority—and—AB and SB (by her litigation friend, the Official Solicitor) [2020] EWCOP 32.

8 MacDonald J’s judgment refers to other judgements making similar points prior to his (e.g. to the Jackson judgement) (Heart of England NHS Foundation Trust v JB [2014] EWHC 342 (COP)) Jackson @23 takes us even further back: “The sequence in subsection (1) [of Section 3 MCA—where the 4 requirements are set out] has its origins in the 1991 Law Commission Consultation Paper No. 129, "Mentally Incapacitated Adults and Decision-Making." This approach was adopted in the influential decision of Thorpe J in In Re C (Adult: Refusal of Treatment) [1994] 1 WLR 290 at 295”. Thorpe J noted in his judgment: “I consider helpful Dr. Eastman's analysis of the decision-making process into three stages: first, comprehending and retaining treatment information, second, believing it and, third, weighing it in the balance to arrive at choice. The Law Commission has proposed a similar approach in paragraph 2.20 of Law Commission Consultation Paper No. 129, “Mentally Incapacitated Adults and Decision-Making.” Dr. Eastman, however, is an externalist (see Eastman and Hope 1988).

9 “Fortunately for medical practitioners who must navigate the difficult waters of comparative jurisprudence, the underlying concepts remain essentially the same” (Zhong et al. 2019, 502).

10 C at 74.

11 C at 87.

12 C at 63.

13 C at 97.

14 Examples are The Health Protection (Coronavirus, Restrictions) (England) Regulations 2020 (SI 350) (https://en.wikipedia.org/wiki/Health_Protection_(Coronavirus,_Restrictions)_(England)_Regulations_2020 accessed 14 June 2020); Coronavirus Act 2020 (https://en.wikipedia.org/wiki/Coronavirus_Act_2020 accessed 14 June 2020); No. 202 Executive Order Declaring a Disaster Emergency in the State of New York (https://www.governor.ny.gov/news/no-202-declaring-disaster-emergency-state-new-york accessed 13 June 2020) and Executive Order No. 202.6: Continuing Temporary Suspension and Modification of Laws Relating to the Disaster Emergency (https://www.governor.ny.gov/news/no-2026-continuing-temporary-suspension-and-modification-laws-relating-disaster-emergency accessed 13 June 2020).

15 Indeed, by the time MacDonald J’s judgment was made available in full (30th November 2015), C had already died. The hearing on which MacDonald J’s judgment was based took place on 13 November 2015, C died on 28th November.

16 C at 89.

17 C at 97.

18 C at 97. By “a lack of capacity” MacDonald J means “incompetent” in our terms.

19 C at 83.

20 C at 85.

21 C at 90. Our italics.

22 C at 43 (ii).

23 We owe this point to one of the anonymous reviewers of the paper.

24 It is worth noting briefly here that there are situations where a person’s capacities are so lowered, that a harmful decision is not required to make the judgment that the person lack competence. For this reason, it would be mistaken to propose that externalist considerations are required for any and every judgment of competence, and so we do not make this claim.