Med Health Care and Philos (2014) 17:229–238 DOI 10.1007/s11019-013-9534-9

SCIENTIFIC CONTRIBUTION

Decision-making capacity should not be decisive in emergencies Dieneke Hubbeling

Published online: 27 December 2013 Ó Springer Science+Business Media Dordrecht 2013

Abstract Examples of patients with anorexia nervosa, depression or borderline personality disorder who have decision-making capacity as currently operationalized, but refuse treatment, are discussed. It appears counterintuitive to respect their treatment refusal because their wish seems to be fuelled by their illness and the consequences of their refusal of treatment are severe. Some proposed solutions have focused on broadening the criteria for decision-making capacity, either in general or for specific patient groups, but these adjustments might discriminate against particular groups of patients and render the process less transparent. Other solutions focus on preferences expressed when patients are not ill, but this information is often not available. The reason for such difficulties with assessing decision-making capacity is that the underlying psychological processes of normal decision-making are not well known and one cannot differentiate between unwise decisions caused by an illness or other factors. The proposed alternative, set out in this paper, is to allow compulsory treatment of patients with decision-making capacity in cases of an emergency, if the refusal is potentially life threatening, but only for a time-limited period. The argument is also made for investigating hindsight agreement, in particular after compulsory measures. Keywords Capacity
Decision-making
Depression
Anorexia nervosa
Personality disorder

D. Hubbeling (&) Wandsworth Crisis and Home Treatment Team, South West London and St. George’s Mental Health NHS Trust, 61 Glenburnie Road, London SW17 7DJ, UK e-mail: [email protected]

Introduction Patient autonomy is important, especially when people face life-changing decisions. For example, if a patient were to develop cancer in her leg, she might decide to have one of her limbs amputated, but she could also choose not to. Health professionals have to accept whatever the patient decides, unless she is not able to make such a decision. The concept of decision-making capacity is crucial in determining whether treatment refusals have to be respected by health professionals (Welie and Welie 2001). Patients can refuse treatment provided they have decision-making capacity, even if they are likely to die as a result of this treatment refusal; for example, Jehovah’s Witnesses may refuse a life-saving blood transfusion. However, patients only have a right to refuse treatment; they cannot demand specific treatments, even if they have decision-making capacity. Doctors, for example, do not have to provide treatments they consider futile (Sokol 2009). In particular, having decision-making capacity does not imply that patients can ask for euthanasia or help with suicide. This is only a possibility in countries where euthanasia or assisted suicide are legal. In many countries, the law makes a difference between respecting treatment refusal and offering help with suicide.1 For example, in England, according to the Suicide Act of 1961, assisting somebody to commit suicide is a criminal offence with a possible prison sentence of 14 years, but according to the Mental Capacity Act 2005 patients with decision-making capacity can refuse treatment, even if they die because of 1

It has been argued, for example by Rachels (1979), that there is no moral difference between killing and letting die, but others have disputed this (see for example Walton 1976), and an in-depth discussion is beyond the scope of this article.

123

230

this treatment refusal. Respecting the wishes of patients with decision-making capacity overrides the duty of doctors to save lives. Patients do not have to prove that they are capable of making certain decisions. It is a moral and legal assumption that patients have decision-making capacity regarding treatment options, unless health professionals demonstrate that they do not. Decision-making capacity should be assessed independently of the outcome of the decision and the diagnosis but is treatment specific. Although refusal of treatment might in practice trigger a formal capacity assessment, the ability to make a decision is independent of the particular decision the patient makes. Somebody has the decision-making capacity to consent to the amputation of her leg; this is independent of whether she gives permission for amputation or not. Criteria for decision-making capacity have to be similar for patients with different conditions: this is critical in order to avoid any possible discrimination against certain groups of patients, such as those with mental disorders. For example, giving somebody a diagnosis of Alzheimer’s disease might trigger a formal assessment of decision-making capacity in a similar way as somebody making a potentially harmful decision, but an individual with Alzheimer’s disease is not incapable of making decisions because of this diagnosis, even though in practice patients with dementia are more likely to lack decision-making capacity than patients with depression or schizophrenia (Vollmann et al. 2003). Decision-making capacity is treatment specific, because the implications of some treatments are easier to understand than others. Somebody might have the capacity to consent to taking a paracetamol tablet for headache, but not to give permission for the amputation of her leg. In summary, decision-making capacity as currently conceptualised depends upon the nature of the treatment in question, but does not depend upon the diagnosis or the outcome of the decision. The criteria for decision-making capacity are operationalized so as to make the assessment of decision-making capacity as transparent as possible. Judges, doctors, family, carers and possibly patients themselves have to be able to check (at least in principle) what the assessing health professional has taken into account when investigating decision-making capacity. Mainly based upon jurisprudence in the USA (Roth et al. 1977), Grisso and Appelbaum (1998) describe four elements of decision-making capacity: the ability to understand the relevant information; the ability to reason about treatment options; the ability to appreciate the situation and its consequences (i.e. the ability to realise what impact the decision will have for the patient herself); and the ability to express a choice. The basic elements of decision-making capacity as described by Grisso and Appelbaum are generally accepted in the Western world, but different jurisdictions emphasize different aspects of these

123

D. Hubbeling

and use slightly different formulations: for example, the English Mental Capacity Act 2005 reads ‘to use or weigh that information’, instead of ‘reasoning and appreciation’. Although in practice most assessments of decisionmaking capacity are uncontroversial, in some cases, respecting the treatment refusal of patients with decisionmaking capacity can be problematic, as a discussion of the examples of patients with anorexia nervosa, depression or borderline personality disorder will make clear. Some of these patients do endanger their lives by refusing treatment and yet they tend to have decision-making capacity as currently operationalized. The underlying reason for these difficulties with using decision-making capacity as a criterion is that we do not know enough about the psychological functions that underpin human decision-making. There are patients who make unwise decisions whose decision, one may well suspect, has been affected by their condition, even though they fulfil the criteria for decision-making capacity, as developed by Grisso and Appelbaum (1998). Given these difficulties, a number of authors have suggested amendments to the decision-making capacity criteria as developed by Grisso and Appelbaum (1998). Proposed solutions have focused on broadening the criteria for decision-making capacity, but this renders the process less transparent and, if the criteria are broadened for specific categories of patients, particular groups of patients such as mental health patients will be discriminated against. Another possibility is to follow the preferences expressed before the person became ill, but such information is often not available. This paper will firstly argue that, at least in emergencies, the fact that a decision seems ‘wrong’ could be considered a sufficient reason for not respecting unwise decisions with dangerous consequences, albeit only for a limited time period, and secondly that one should investigate retrospective agreement, especially after compulsory treatment.

Problematic cases: anorexia nervosa Patients with anorexia nervosa think that they are overweight. Even when they are underweight by objective standards, they still want to lose weight. They tend to do this by reducing food intake, increasing physical activity, using laxatives and/or inducing vomiting. Anorexia nervosa is a serious psychiatric condition and can result in an untimely death (Franko et al. 2013). Treatment consists of psychotherapy and specific measures, including feeding (sometimes via a nasogastric tube) to encourage weight gain; but this is often considered unacceptable by patients, and is sometimes refused altogether. Treatment refusal increases not only the risk of death but also that of other problems later in life, such as osteoporosis (Beumont et al.

Decision-making capacity

2006). The standard view is that these refusals should be respected if the patient has decision-making capacity: even though the consequences of this may include early death or problems with mobility in later life. The findings of the study by Tan et al. (2003) urge one to reconsider the application of decision-making capacity criteria in their current form. They investigated the decision-making capacity of patients with anorexia nervosa, through use of the MacCAT-T (Grisso et al. 1997), a formalised assessment of decision-making capacity, based on the criteria described by Grisso and Appelbaum (1998). They found that the patients did indeed have decisionmaking capacity; they tended to know that they had a health problem and that refusing treatment might well result in an early death but sometimes they still refused treatment. Tan et al. (2003) also conducted qualitative interviews. Here, it became apparent that the attitudes of the patients towards death and disability had changed: it was clear that patients might refuse treatment not because they want to die, but because death and disability seem less important than staying thin. Some patients acknowledged this themselves retrospectively, and claimed that it was justified that they were treated despite it being against their wishes at the time. Thus, the research identified a dilemma: namely, that there are at least some patients with decisionmaking capacity according to formal criteria whose treatment refusal might seem at odds with the right decision in medical terms, and who go on to agree in retrospect with their being treated against their wishes. Although patients with anorexia nervosa tend to have decision-making capacity, there are differences in their psychological functioning compared to normal controls; some of those differences persist even after they have gained weight. Empirical studies have found that the performance of patients with anorexia nervosa differs from normal controls in the Iowa Gambling task (Cavedini et al. 2004), and CatBat shift ratio (Holliday et al. 2005). Their attention seems to be impaired even when their weight has been restored (Bosanac et al. 2007). These differences in psychological functions might influence their decisions, even though they fulfil the criteria for decision-making capacity as currently operationalized. Therefore patients with anorexia nervosa can, while possessing decision-making capacity, refuse treatment, with potentially serious consequences for themselves. Tan et al. (2006) assert that the criteria for decision-making capacity need to be adjusted, and that the absence of pathological values (such as wanting to be thin) should be included in the requirements for decision-making capacity. But how is one to determine pathological values? Should pathological values be identified in a particular society, via clinical judgement, for people with a particular disorder or for a particular individual?

231

The suggestion of identifying pathological values at the level of a particular society or via clinical judgement should be abandoned. One of the ideas behind capacity assessments is self-determination, and another that patients should not be forced to live according to community values (Ha¨yry 1998), a complicated concept in any case. Furthermore, patients should not have to comply with the values of health care professionals, because medical paternalism needs to be rejected (Buchanan 1978). One could also claim that specifically patients suffering from anorexia nervosa should not have the pathological wish to be thin influencing their decisions and that the absence of pathological values should not be used as a general criterion. However, one of the underlying ideas behind introducing a decision-making capacity test is that the same criteria can be used for patients with different disorders, and that mental health patients in general or patients with anorexia nervosa in particular are not discriminated against (Richardson 2013). Another possibility would be to state that particular values could be pathological for a specific person. The values expressed should be consistent during somebody’s life. If somebody who always seemed to enjoy life and suddenly claims, after developing anorexia nervosa that she would rather die than gain weight, one could argue that she does not have decision-making capacity, because the values expressed are not consistent with those she is known to normally hold. In a later publication Tan et al. (2009 express the same idea somewhat differently and argue for the authenticity of values. They use an example of somebody giving all her money to charity. This could either happen because of her personality traits or because of mild mania. In the latter case, chances are that the individual will regret it once she is no longer manic. In such a case the decision to donate her money to charity was not authentic and, according the Tan et al. (2009), she did not have decision-making capacity. However, the problem with this position is that people should be able to change their values, and that they should also have the right to change them into values others might consider unwise. For example, one could become a Jehovah’s Witness as an adult and decide to live according to the standards of this particular religion and refuse blood transfusion. Responding to Tan et al. (2006), Grisso and Appelbaum (2006) argued that adjusting the criteria for capacity would not be necessary in the case of anorexia nervosa: they asserted that patients with anorexia nervosa do not fulfil the appreciation criterion. They argued that patients with anorexia do not really appreciate the situation they find themselves in and the risks involved, although they state that they know they might die when refusing treatment. A clear definition of appreciation is currently lacking (Saks and Behnke 1999), and Grisso and Appelbaum do not

123

232

define the point at which they believe unwise decisions still fulfil the criterion of appreciation. The danger of their position is that one could argue that a patient did not sufficiently appreciate the information in the case of every unwise decision. The idea behind decision-making capacity is that patients can use their own values in making a decision; a broad interpretation of appreciation would block this. A similar argument can be made when looking at the terminology of the Mental Capacity Act 2005 in England. It might be asserted that patients with anorexia nervosa refusing treatment cannot use and weigh the information available to them, but with this wide interpretation everybody who makes an unwise decision can be considered not able to use and weigh the information (Richardson 2013). The distinction between unwise decisions made with and without decision-making capacity is not clear due to the broad interpretation of appreciation.

Problematic cases: depression Depression is a frequently occurring condition. Patients suffer from low mood, cannot enjoy activities which they used to enjoy, are often tearful, have no energy and sometimes even have suicidal thoughts. Many patients have a number of depressive episodes during their life, i.e. they get better and become ill again a couple of years later. The condition varies in severity. Mild cases might improve without specific treatment (Kirsch et al. 2008). Depression could sometimes even be regarded as a normal response in adverse or difficult circumstances (Horwitz and Wakefield 2007). However, the condition can also be very severe: so much so that victims refuse to eat and might die without medical intervention. Sometimes patients with severe depression refuse to be treated. Silberfeld and Checkland (1999) discuss an example of an individual with severe depression who refused to eat and drink and, although he understood that electro-convulsive treatment might improve his condition, still refused it. He did seem to have the capacity to decide; but intuitively, it was difficult to accept his refusal because of the severe consequences of this choice. Owen et al. (2013) conducted a qualitative study into experiences of depressed patients. Their main finding was that severely depressed patients have a different experience of time and a different idea of their future. This can influence the decisions which they make while their reasoning and appreciation, as measured by the MacCAT-T (Grisso et al. 1997), and their using and weighing information remain within normal limits. There are also differences in psychological functioning between depressed patients and controls: for example, in executive functioning (Snyder 2013) and reasoning (Channon and Baker 1994). In their study of reasoning in

123

D. Hubbeling

depression, Channon and Baker (1994) did not test reasoning according to the MacCAT-T (Grisso et al. 1997), which implies that it is quite possible that somebody with depression differs in reasoning from normal controls while still possessing decision-making capacity, according to the MacCAT-T. It could be argued here that, as elaborated upon by Silberfeld and Checkland (1999), the patient in their example did not fully appreciate the situation he was in and, although he understood the information provided, he could not apply it to himself. But the general argument against a wide interpretation of appreciation applies here as well: namely, that every unwise decision can become a failure of appreciation and the idea behind introducing decision-making capacity was that patients would be able to make unwise decisions. Although not emphasised by Silberfeld and Checkland (1999), one could also note that the patient whom they described had pathological values, analogous to the argument of Tan et al. (2006) regarding anorexia nervosa patients. The patient in Silberfeld and Checkfeld’s example might now prefer death or think that he does not deserve treatment. Yet with no consensus in Western societies regarding pathological values, some are of the opinion that wanting to die is always a sign of some mental disorder, though others disagree (Ginn et al. 2011), and ‘rational suicide’ has been defended in the philosophical literature (Hewitt 2010). In countries where euthanasia and/or physician assisted suicide are legalised, patients have to have decision-making capacity for their request to be granted (Schroevers et al. 1998). Patients with a terminal illness might well be depressed (Ganzini et al. 2008) and although help with suicide and/or euthanasia will not be allowed when somebody does not have decision-making capacity, it is a concern that somebody might request it because of their depression (Hendin and Foley 2008), while still having decision-making capacity according to current criteria. Rudnick (2002) discussed a similar example as Silberfeld and Checkland (1999) of a severely depressed patient with decision-making capacity according current criteria who refused treatment. He asserted that in cases of fluctuating conditions, such as depression, treatment refusal of patients with decision-making capacity should only be respected if their refusal is consistent during periods with and without illness. If there is no evidence of this consistency in preferences, one should consider treating them against their wishes if it seems in their best interests; in other words if there is a favourable risk–benefit ratio. Any such individual should also be encouraged to make an advanced directive for future episodes. Rudnick does not discuss the idea that, although somebody might have different episodes of depression, the clinical presentation and recommended treatment can sometimes be very different;

Decision-making capacity

but one could easily expand his argument by stating that different clinical presentations should be treated as new episodes, and that treatment against somebody’s wishes is indeed possible with a different clinical presentation. Although Rudnick refers to past treatment preferences and Tan et al. (2006) refer to pathological values as an additional criterion, if one interprets the suggestion of Tan et al. exclusively at a personal level—namely as having consistent values over a certain period of time—their view has many similarities with Rudnick’s. An implication of the view of Tan et al. is that if people radically change their values, such as by becoming a Jehovah’s Witness, this would be by definition pathological, if one requires personal values to be consistent. Extrapolating from Rudnick’s proposal one could argue that by becoming a Jehovah’s Witness, one also changed ones preferences regarding blood transfusion. Although Rudnick’s (2002) solution has advantages over introducing a broad concept of pathological values or expanding the appreciation concept, it will be difficult to apply it in practice. Very often, there will not be enough information available with which to investigate whether the refusal of treatment has been a consistent and permanent wish outside the illness episode itself. However, Rudnick is clear that if there is limited information available, health professionals can override treatment refusal if it seems to be in the patient’s best interests.

Problematic cases: borderline personality disorder Patients with personality disorder have on-going difficulties in many areas of their lives—be they related to work, family or friends—though the nature of these difficulties varies with different types of personality disorder. Patients with borderline personality are impulsive: their mood fluctuates rapidly in response to minor triggers; they have problems with their identity, and often lack a sense of continuity in themselves. Given these rapid, impulsive responses to relatively minor triggers, overdoses and other acts of self-harm are quite common: patients with borderline personality disorder often attend accident and emergency departments. A patient with borderline personality disorder drank antifreeze and refused renal dialysis which would probably have saved her life. She had written a letter stating that she did not want to be treated if she were to drink antifreeze. When this patient arrived at the accident and emergency department, she pointed at this letter. Health professionals thought she had decision-making capacity, so she was not treated and died (David et al. 2010). In the legal procedures that followed, the coroner decided that her letter was not a legally binding advanced directive, but that she had decision-making capacity according to the Mental Capacity Act

233

2005 at the time that she was refusing treatment, and this implied that the decision not to force her to have renal dialysis was justified. Qualitative research into experiences of patients with borderline personality disorder found that patients with this condition often fear losing control of their emotions and feel safer if they can experience at least some control (Fallon 2003), and this implies that patients with borderline personality may wish to make decisions in order to exert control over their lives, rather than because they really want the outcome of their decision itself. Although many patients with borderline personality disorder seem to have decisionmaking capacity, research findings suggest abnormal frontal lobe function (Bazanis et al. 2002); superior performance on a theory of mind test compared with ‘normal controls’ in at least in one study (Franzen et al. 2011); and altered use of feedback in decision-making (Schuermann et al. 2011): thus there are differences in various psychological functions between patients with borderline personality disorder and controls. It could be argued that this patient did not fulfil the criteria for decision-making capacity because she was unable to appreciate her condition but, similarly to patients with anorexia nervosa and depression, it can again be posited that for every unwise decision, there is a failure of appreciation. Pathological values are also difficult as criteria, because preferring to die would then always be pathological, and there is no consensus about this (Ginn et al. 2011). This patient had a letter stating that she did not want to be treated and, although it was not considered to be a legally binding advanced directive in this particular case, it does make clear that applying the solution proposed by Rudnick (2002) for depression can be very difficult in practice for patients with rapidly fluctuating conditions. One can never be certain whether a patient was really well when writing down his or her preferences and therefore whether the preferences are consistent during periods of being well and being ill; moreover, psychological research reveals differences in psychological functioning with normal controls, so even with decision-making capacity the outcome of their decision might well have been influenced by the disease process.

Problematic cases: summary Thus, in all the cases detailed so far—those regarding anorexia nervosa, depression and borderline personality disorder—there are patients with decision-making capacity according to Grisso and Appelbaum’s (1998) criteria who refuse potentially life-saving treatment. The actual decision they made seemed to be influenced by their disease

123

234

process, yet they had decision-making capacity as currently operationalized. One could argue that the treatments such as feeding via a nasogastric tube, electro-convulsive treatment and renal dialysis are physical treatments for patients with mental health problems, and, especially in the last case of renal dialysis, only the effects of the overdose are treated and nothing is done about treating the underlying mental disorder. However, the difference between mental and physical disorders is culturally determined, and it has been disputed whether such a difference can be constructed anyway (see for example Pickard 2009 for a negative answer). But even if there is such a difference, the point of introducing decision-making capacity is that one should not discriminate against certain groups of patients and that the same functionally defined criteria for capacity should be used for patients irrespective of diagnosis or type of disorder. There are practical issues about treatment without consent. Some treatments can be offered without the cooperation of the patient but this is not possible with talking treatments. So far, a number of amendments to Grisso and Appelbaum’s (1998) operationalization of decision-making capacity have been discussed which were all inspired by problems with a specific patient category; Tan et al. suggested introducing pathological values to solve the problems with decision-making capacity assessment in anorexia nervosa patients; other authors (Grisso and Appelbaum 2006; Silberfeld and Checkland 1999) wanted to expand the appreciation concept to solve the problems with anorexia nervosa patients and depressed patients respectively; and Rudnick (2002) emphasized the importance of coherent wishes during periods of being well and periods of being ill to solve problems with depressed patients. These proposals might solve the problem in some cases, but in case of a wide interpretation of pathological values and appreciation there is no clear boundary between unwise decisions and appreciation or pathological values, and, when trying to apply a personal-level interpretation of pathological values or coherent wishes, the relevant information will often not be available; rapidly fluctuating conditions remain a problem and it is not always clear whether people were really well when informing others of their preferences. A number of other authors have also suggested amendments to the current operationalization of decisionmaking capacity in general. Breden and Vollmann (2004), for example, argued for an extension of the MacCAT-T approach with other factors, using personal construct theory as a guidance, while Banner (2012) advocates taking all beliefs, values and emotions into account. Banner is quite explicit in stating that decision-making capacity criteria cannot be fully operationalized, and that clinical judgement will always play an important part. It is possible

123

D. Hubbeling

to expand the decision-making criteria and/or accept clinical judgement. However, if one takes more factors into account the process becomes more time-consuming, which is likely to be a disadvantage (although not necessarily an insurmountable one), and if clinical judgement is a major factor, assessing decision-making capacity becomes less transparent. It is easier to grasp which factors the assessing health professional has taken into account with the Grisso and Appelbaum (1998) criteria than if it was a matter of clinical judgement; differences of opinion about decisionmaking capacity will be more difficult to solve. Thus, it is very difficult in practice to operationalize the concept of decision-making capacity if one wants to retain all the requirements: namely transparency of the assessment-process, independence of diagnosis and outcome of the decision and appropriateness for the nature of the treatment. The reason for this difficulty is that the Grisso and Appelbaum (1998) criteria do not map onto specific psychological functions. Available research findings suggest that decisions are often not made by consciously weighing all the available information.

How do people actually make decisions? Although humans intuitively believe that they make decisions by reasoning with the available evidence, empirical psychological research suggests otherwise. Nisbett and Wilson (1977) asked people attending a shop which nylon stockings they preferred and why. Study participants came up with various reasons such as nicer texture, etc., yet in actual fact all the stockings were similar, and people simply preferred those placed on the right hand side. Study participants did not know that all the stockings were the same, and were not aware that the reasons they provided for their choice were post hoc rationalisations. One could wonder whether this study is really applicable to difficult medical decisions: Nisbett and Wilson (1977) asked for reasons for one item having been chosen over the other when they were, in fact, all similar. In healthcare there will be a difference between the various options available for a patient. However, given the importance of providing reasons in current decision-making capacity assessments, it is an important empirical finding that post hoc rationalisation can occur and that people do not report this when giving their reasons, so the reasons people give during an assessment of their decision-making capacity might well be post hoc rationalisations. Dijksterhuis et al. (2006) developed the ‘deliberationwithout-attention’ hypothesis. They asked people to make simple or complex choices between cars: the cars in the simple condition were described by four attributes, while those in the complex condition by 12. Participants knew

Decision-making capacity

they had to name the best car after 4 min. In the unconscious thought condition, they had to do a different task for 4 min, and in the conscious thought condition, they were left to think about the various options for 4 min. There was no statistically significant difference for the simple choice between cars; but for the complex choice, participants in the unconscious thought condition performed better. The underlying theory behind this experiment is that conscious thinking has limited capacity and cannot take all alternatives into account at the same time. These findings suggest that it is not always necessary to consciously weigh up the various arguments to make good decisions. It could also be considered that in Dijksterhuis et al. (2006), the decision was not akin to those which patients normally have to make, because there was a time constraint of 4 min. There will be more time available in making difficult decisions about medical treatment, and patients can use more aides (such as writing things down) in order to increase their capacity for conscious thinking. However, it remains a very important finding that sometimes, people actually make better decisions without thinking about the issues consciously. The study findings from Dijksterhuis et al. indicate that encouraging people to consciously weigh the advantages and disadvantages of certain decisions might not always lead to the best possible decision and make the idea that consciously weighing the information should be a criterion for proper decision-making controversial. Both the study by Nisbett and Wilson (1977) and the study by Dijksterhuis et al. (2006) suggest that one should not use reasoning as solid evidence for proper decisionmaking. Other empirical studies have shown that the situation is even more complicated: experiencing disgust influences moral judgement (Wheatley and Haidt 2005), and people can have strong moral views without being able to give reasons (Haidt 2001), manipulation of mood in healthy subjects can influence executive functions (Mitchell and Phillips 2007), and executive functions are important for planning and decision making (Grimes et al. 2000). People will make different decisions when environmental circumstances are different; this is why estate agents advice prospective home sellers to bake something nice when potential buyers are viewing their property. Being influenced by mood or a nice smell is not something decision-making capacity criteria should prevent from happening, but it becomes a problem when we attempt to determine exactly when people are so influenced by their mood that they can no longer decide for themselves. This is particularly pertinent in cases where one cannot take the outcome of the decision into account, as there does not seem to be a sharp dividing line between mild and severe depression (Varga 2013). It suggests that one cannot

235

clearly disentangle unwise decisions caused by the disease process and unwise decisions due to other causes. The fact that we do not know enough about the psychological processes underlying decision-making has influences at different levels, not only at the patient level but also at the level of the assessing health professional. The post hoc rationalisation described by Nisbett and Wilson (1977) might well occur in health professionals, ethicists and legal professionals. Patients might be influenced by various environmental biases such as the smell in the room when making their decision, but this is probably also the case for assessing mental health professionals. The limited available research suggests that both philosophers (Schwitzgebel and Cushman 2012) and doctors (Sabin et al. 2012) are not immune to at least some of these biases. Some suggestions have been made to reduce the influence of biases (Albisser Schleger et al. 2011), but it has not been shown empirically that they will be effective (Cain et al. 2005). The fact that assessing mental health professionals can be influenced by various biases when conducting their capacity assessment is an argument for making the process as transparent as possible, so that one can check what the professional has been doing. All these findings from psychological research suggest that one cannot base criteria for decision-making capacity upon objective standards of normal decision-making (Welie and Welie 2001). The current operationalization does not reflect how people actually make decisions and this is problematic at two levels: at the level of establishing whether somebody has decision-making capacity and at the level of the assessing health professional, because he or she might be influenced by various biases.

Possible solution In practice, there is sometimes disagreement between patients and health professionals, but it is rare for this to end up in court. The best solution is often to postpone the decision (if at all possible), because despite having decision-making capacity, patients might change their mind. It can sometimes be helpful to involve family and carers to support the patient when making a decision, although involving the family while respecting patient’s privacy can be difficult in practice (Landeweer et al. 2011). Also, involving others is not always acceptable for the patient. When discussing the patient with borderline personality disorder who drank antifreeze, Richardson (2013) hypothesised that this patient would not have agreed to have her decision reconsidered in the context of her family and social relationships. It also would not have been a realistic possibility given the narrow time-frame for a possible intervention in that particular case.

123

236

When the condition is life threatening, postponement is not really an option. Rudnick (2002) has argued in favour of treating patients with capacity during their first episode (probably—although not explicitly—stated during their first episode with similar presentation), even when they do not themselves consent, as long as the treatment seems to be in their best interest and assuming that there is no advanced directive. His position is very similar to the treatment of some medical emergencies such as intravenous administration of glucose in hypoglycaemia in insulin-dependent diabetes, whereby people receive treatment, even without their consent, unless there is a valid advanced directive. Tan et al.’s (2006) suggestion of using pathological values could lead to a similar outcome in most cases, provided one interprets pathological values at a personal level. The idea of overriding the wishes of patients with decision-making capacity as currently operationalized, something Rudnick (2002) and Tan et al. (2006) support, is not new. It is already happening in some medical emergencies and, for example, in England the Mental Health Act 2007 allows compulsory treatment of psychiatric patients with capacity: however this is controversial, because it could easily be construed as a form of discrimination against patients with a mental disorder (Holloway and Szmukler 2003). However, if treatment refusals in both patients with mental and physical health problems are overridden, there would no longer be discrimination; but to do this each and every time when treatment seems to be in patients’ best interests, if there is no information about past preferences available, would clearly reduce patients’ freedom to make decisions others considered unwise. A possible solution would be to make the criteria for treating patients with decision-making capacity against their wishes more stringent than ‘if this serves the best interests of the patient’ (Rudnick 2002, p. 153), and only treat patients if there is a danger of immediate death or severe disability. Having a high threshold, namely immediate danger for the patient, would respect a patient’s autonomy more than simply acting when treatment seems to be in their best interests. However, in order to fully protect personal autonomy, one should only allow this to take place for a limited time-period: namely, the period in which improvement can be expected, i.e. a couple of months at most. Health professionals would have to demonstrate immediate risk, and treating people against their consent would still be a fairly transparent process, more transparent than with some of the other solutions that have been suggested, such as using a broad concept of appreciation, pathological values or using clinical judgement. There would also not be any discrimination against particular groups, because one would override the wishes of patients with capacity for both patients with physical and mental disorders.

123

D. Hubbeling

In deciding to treat people (with or without their consent) when they are at immediate risk of death or severe disablement, one does give up the idea behind assessing decision-making capacity that decisions made in the right way should always be respected, independent of outcome. The reason for this is that we do not know enough about the psychological processes underpinning normal decision making to establish normal decision-making. The proposed solution would solve some of the problems of anorexia nervosa patients, depressed patients and patients with personality disorder; but it might well be that a different decision will be made in some other situations, such as in the case of an 86-year-old woman living on her own who seemed physically unwell, and who was not treated by emergency services, because she had decision-making capacity (Naess et al. 2001) and might well have died as a consequence of her treatment refusal. Treating somebody in such a condition against her wishes would be defensible with the framework currently proposed. However, the proposed solution is a temporary one and in future we should develop criteria which are more in line with developments in psychological research, if possible. Tan et al. (2006) stress the importance of retrospective agreement. It is important to investigate if patients who have been treated against their will in emergencies are in agreement with hindsight. Hindsight agreement does not prove that one has done the right thing; however, if there is no hindsight agreement, one has to change one’s practice. Probably, most Jehovah’s Witnesses would not be in agreement after they were given a blood transfusion without consent and, if this is confirmed empirically, this seems to be a valid reason not to force them to have blood transfusions in emergencies. The retrospective agreement should also be investigated for people whose decision was respected, because although it may have seemed unwise, it may not have been immediately life threatening. There is very limited data available about retrospective agreement (Owen et al. 2009) and this needs to be studied empirically, as it is likely that some people will agree that their overall well-being has increased after compulsory treatment or disagree that their refusal was respected, even though they had capacity.

Conclusion Examples have been discussed of patients with anorexia nervosa, depression and borderline personality disorder who appeared to make unwise decisions due to their disease process while having decision-making capacity as currently operationalized. Proposals have been made to adjust the criteria for decision-making capacity as developed by Grisso and Appelbaum (1998), but it is impossible for decision-making capacity criteria to be transparent,

Decision-making capacity

independent of actual decisions and diagnosis, and treatment specific. Most alternative solutions make the process less transparent, and some of these discriminate against patients with certain disorders. Given that very little is known about decision-making processes, it will be difficult to devise criteria which really reflect decision-making at a psychological level. Therefore, in this article a different solution is proposed, namely allowing the treatment of patients with capacity for a limited time period, if the consequences of not treating are severe (i.e. death or disability). Health professionals are required to demonstrate lack of decision-making capacity according to the Grisso and Appelbaum (1998) criteria, or some variant (such as the Mental Capacity Act 2005 criteria); or they have to demonstrate a high risk of death or disability, if the patient is not treated. This would have allowed health professionals to treat the borderline patient who drank antifreeze, somebody suffering from depression who is refusing to eat or drink, and those with anorexia nervosa who are dangerously underweight. Further research is necessary, not only into psychological decision-making but also into hindsight agreement. Hindsight agreement on its own cannot justify compulsory treatment, but if there is no hindsight agreement, one should consider changing one’s practice in similar situations.

References Albisser Schleger, H., N.R. Oehninger, and S. Reiter-Theil. 2011. Avoiding bias in medical ethical decision-making. Lessons to be learnt from psychology research. Medicine, Health Care and Philosophy 14(2): 155–162. Banner, N.F. 2012. Unreasonable reasons: normative judgements in the assessment of mental capacity. Journal of Evaluation in Clinical Practice 18: 1038–1044. Bazanis, E., R.D. Rogers, J.H. Dowson, P. Taylor, C. Meux, C. Staley, and B.J. Sahakian. 2002. Neurocognitive deficits in decision-making and planning of patients with DSM-III-R borderline personality disorder. Psychological Medicine 32: 1395–1405. Beumont, P.J., S. Zipfel, M.J. Seibel, B. Lo, C. Kasperk, and W. Herzog. 2006. Osteoporosis in eating disorders: A follow-up study of patients with anorexia and bulimia nervosa. The Journal of Clinical Endocrinology and Metabolism 86(11): 5227–5233. Bosanac, P., S. Kurlender, L. Stojanovska, K. Hallam, T. Norman, C. McGrath, and J. Olver. 2007. Neuropsychological study of underweight and ‘‘weight-recovered’’ anorexia nervosa compared with bulimia nervosa and normal controls. International Journal of Eating Disorders 40: 613–621. Breden, T.M., and J. Vollmann. 2004. The cognitive based approach of capacity assessment in psychiatry: A philosophical critique of the MacCAT-T. Health Care Analysis 12(4): 273–283. Buchanan, A. 1978. Medical paternalism. Philosophy & Public Affairs 7(4): 370–390. Cain, D.M., G. Loewenstein, and D.A. Moore. 2005. The dirt on coming clean: Perverse effects of disclosing conflicts of interest. The Journal of Legal Studies 34(1): 1–25.

237 Cavedini, P., T. Bassi, A. Ubbiali, A. Casolari, S. Giordani, C. Zorzi, and L. Bellodi. 2004. Neuropsychological investigation of decisionmaking in anorexia nervosa. Psychiatry Research 127: 259–266. Channon, S., and J. Baker. 1994. Reasoning strategies in depression: Effects of depressed task mood on a syllogism. Personality and Individual Differences 17(5): 707–711. David, A.S., M. Hotopf, P. Moran, G. Owen, G. Szmukler, and G. Richardson. 2010. Mentally disordered or lacking capacity? Lessons for managing serious deliberate self harm. British Medical Journal 341: 587–591. Dijksterhuis, A., M.W. Bos, L.F. Nordgren, and R.B. Van Baaren. 2006. On making the right choice: the deliberation-withoutattention effect. Science 311(5763): 1005–1007. Fallon, P. 2003. Travelling through the system: the lived experience of people with borderline personality disorder in contact with psychiatric services. Journal of Psychiatric and Mental Health Nursing 10: 393–400. Franko, D.L., A. Keshaviah, K.T. Eddy, M. Krishna, M.C. Davis, P.K. Keel, and D.B. Herzog. 2013. A longitudinal investigation of mortality in anorexia nervosa and bulimia nervosa. American Journal of Psychiatry 170(8): 917–925. Franzen, N., M. Hagenhoff, N. Baer, A. Schmidt, D. Mier, G. Sammer, and S. Lis. 2011. Superior ‘‘theory of mind’’ in borderline personality disorder: An analysis of interaction behavior in a virtual trust game. Psychiatry Research 187(1–2): 224–233. Ganzini, L., E.R. Goy, and S.K. Dobscha. 2008. Prevalence of depression and anxiety in patients requesting physicians’ aid in dying: cross sectional survey. BMJ 337: a1682. doi:10.1136/bmj. a1682. Ginn, S., A. Price, L. Rayner, G.S. Owen, R.D. Hayes, M. Hotopf, and W. Lee. 2011. Senior doctors’ opinions of rational suicide. Journal of Medical Ethics 37: 723–726. Grimes, A.L., L.B. McCullough, M.E. Kunik, V. Molinari, and R.H. Workman. 2000. Informed consent and neuroanatomic correlates of intentionality and voluntariness among psychiatric patients. Psychiatric Services 51(12): 1561–1567. Grisso, T., and P.S. Appelbaum. 1998. Assessing competence to consent to treatment. New York, Oxford: Oxford University Press. Grisso, Thomas., and P.S. Appelbaum. 2006. Appreciating anorexia: Decisional capacity and the role of values. Philosophy, Psychiatry and Psychology 4(4): 293–297. Grisso, Thomas., P.S. Appelbaum, and C. Hill-Fotouhi. 1997. The MacCAT-T: A clinical tool to assess patients’ capacities to make treatment decisions. Psychiatric Services 48: 1415–1419. Haidt, J. 2001. The emotional dog and its rational tail: A social intuitionist approach to moral judgment. Psychological Review 108(4): 814–834. Ha¨yry, H. 1998. Should the decisions of ethics committees be based on community values? Medicine. Health Care and Philosophy 1: 57–60. Hendin, H., and K. Foley. 2008. Physician-assisted suicide in oregon: A medical perspective. Michigan Law Review 106(8): 1613–1639. Hewitt, J. 2010. Rational suicide: philosophical perspectives on schizophrenia. Medicine, Health Care and Philosophy 13: 25–31. Holliday, J., K. Tchanturia, S. Landau, D. Collier, and J. Treasure. 2005. Is impaired set-shifting an endophenotype of anorexia nervosa? American Journal of Psychiatry 162(4): 2269–2275. Holloway, F., and G. Szmukler. 2003. Involuntary psychiatric treatment: Capacity should be central to decision making. Journal of Mental Health 12(5): 443–447. Horwitz, A.V., and J.C. Wakefield. 2007. The loss of sadness. New York, Oxford: Oxford University Press, Inc. Kirsch, I., B.J. Deacon, T.B. Huedo-Medina, A. Scoboria, T.J. Moore, and B.T. Johnson. 2008. Initial severity and antidepressant

123

238 benefits: A meta-analysis of data submitted to the food and drug administration. PLoS Medicine 5(2): e45. Landeweer, E., T.A. Abma, L. Dauwerse, and G.A.M. Widdershoven. 2011. Triad collaboration in psychiatry: privavy and confidentiality revisited. International Journal of Feminist Approaches to Bioethics 4(1): 121–139. Mitchell, R.L.C., and L.H. Phillips. 2007. The psychological, neurochemical and functional neuroanatomical mediators of the effects of positive and negative mood on executive functions. Neuropsychologia 45: 617–629. Naess, A.-C., R. Foerde, and P.A. Steen. 2001. Patient autonomy in emergency medicine. Medicine, Health Care and Philosophy 4: 71–77. Nisbett, R.E., and T.D. Wilson. 1977. Telling more than we can know: Verbal reports on mental processes. Psychological Review 84(3): 231–259. Owen, G.S., A.S. David, P. Hayward, G. Richardson, G. Szmukler, and M. Hotopf. 2009. Retrospective views of psychiatric inpatients regaining mental capacity. British Journal of Psychiatry 195(5): 403–407. Owen, G. S., Freyenhagen, F., Hotopf, M., and Martin, W. (2013). Temporal inabilities and decision-making capacity in depression. Phenomenology and Cognitive Sciences, in press. Pickard, H. (2009). Mental illness is indeed a myth. In L. Bortolotti and M. Broome (Eds.), Psychiatry as Cognitive Neuroscience: Philosophical Perspectives (pp. 83–102). Oxford: Oxford University Press. Rachels, J. 1979. Killing and starving to death. Philosophy 54(208): 159–171. Richardson, G. 2013. Mental capacity in the shadow of suicide: What can the law do? International Journal of Law in Context 9(1): 87–105. Roth, L.H., A. Meisel, and C.W. Lidz. 1977. Tests of competency to consent to treatment. American Journal of Psychiatry 134(3): 279–284. Rudnick, A. 2002. Depression and competence to refuse psychiatric treatment. Journal of Medical Ethics 28: 151–155. Sabin, J.A., M. Marini, and B.A. Nosek. 2012. Implicit and explicit anti-fat bias among a large sample of medical doctors by BMI, race/ethnicity and gender. PLoS ONE 7(11): 1–7. doi:10.1371/ journal.pone.0048448. Saks, E.R., and S.H. Behnke. 1999. Competency to decide on treatment and research: MacArthur and beyond. Journal of Contemporary Legal Issues 103: 103–130.

123

D. Hubbeling Schroevers, R.A., F.P. Asmus, and W. Van Tilburg. 1998. Physicianassisted suicide in psychiatry: Developments in the Netherlands. Psychiatric Services 49(11): 1475–1480. Schuermann, B., N. Kathmann, C. Stiglmayr, B. Renneberg, and T. Endrass. 2011. Impaired decision making and feedback evaluation in borderline personality disorder. Psychological Medicine 41: 1917–1927. Schwitzgebel, E., and F. Cushman. 2012. Expertise in moral reasoning? Order effects on moral judgment in professional philosophers and non-philosophers. Mind and Language 27(2): 135–153. Silberfeld, M., and D. Checkland. 1999. Faulty judgment, expert opinion, and decision-making capacity. Theoretical Medicine and Bioethics 20: 377–393. Snyder, H. R. (2013). Major depressive disorder is associated with broad impairments on neuropsychological measures of executive function: A meta-analysis and review. Psychological Bulletin 139(1): 81–132. Sokol, D.K. 2009. The slipperiness of futility. BMJ 338: 1418. Tan, J.O.A., T. Hope, A. Stewart, and R. Fitzpatrick. 2003. Competence to refuse treatment in anorexia nervosa. International Journal of Law and Psychiatry 26: 697–707. Tan, J.O.A., A. Stewart, R. Fitzpatrick, and T. Hope. 2006. Competence to make treatment decisions in anorexia nervosa: Thinking processes and values. Philosophy, Psychiatry and Psychology 13(4): 267–282. Tan, J.O.A., A. Stewart, and T. Hope. 2009. Decision-making as a broader concept. Philosophy, Psychiatry and Psychology 16(4): 345–349. Varga, S. 2013. From melancholia to depression: Ideas on a possible continuity. Philosophy, Psychiatry and Psychology 20(2): 141–155. Vollmann, J., A. Bauer, H. Danker-Hopfe, and H. Helmchen. 2003. Competence of mentally ill patients: A comparative empirical study. Psychological Medicine 33(8): 1463–1471. Walton, D. 1976. Active and passive euthanasia. Ethics 86(4): 343–349. Welie, J.V.M., and S.P.K. Welie. 2001. Patient decision making competence: Outlines of a conceptual analysis. Medicine, Health Care and Philosophy 4: 127–138. Wheatley, T., and J. Haidt. 2005. Hypnotic disgust makes moral judgments more severe. Psychological Science 16(10): 780–784.