Mental health law and mental capacity: Dogmaand evidence

Mental capacity and empirical research

In order to make autonomous decisions a person must have mental capacity. The definition and characteristics of mental capacity have been the subject of intensifying debate. Mental capacity represents a junction between ethics, law, medicine and social science. It also represents a potent embodiment of the gap between physical and mental healthcare. In British law there are crucial differences between mental health and mental capacity legislation despite both having a similar function – laying out the extent to which clinicians and other agencies can act when patients are deemed to be unable to make decisions for themselves.

In England, existing and planned mental health legislation does not take mental capacity into account. Despite the advice of the Government's expert committee (Expert Committee, 1999), and further pressure from the Joint Scrutiny Committee, the Government has remained steadfast in its view that mental capacity (or impaired decision making) has only a very limited place in mental health law (the only place it appears in the Mental Health Bill is in relation to electroconvulsive therapy, which, provided it is not required in an emergency – will not be given to a patient with capacity who refuses the treatment). English mental health law uses a “status approach” to define when compulsion is acceptable – i.e., where mental disorder is present and of a “nature and degree” where the health or safety of the patient, or safety of others, is at risk. Having established that, mental health practitioners have wide scope to detain and treat the patient as they see fit, without necessarily taking into account the patient's previous wishes, current views, or fitness to make specific treatment choices.

The Mental Capacity Act 2005, which will shortly be implemented in England and Wales, has five principles, including the assumption that people have capacity; the right of individuals to make decisions which others might consider unwise; and the crucial point that any decision made on behalf of a person lacking capacity must be made with his or her best interests in mind. The Act uses a “functional” approach in which capacity can only be said to be impaired in relation to a specific decision, implying that the same individual may have capacity to make some, but not other, choices. The Act recognizes the importance of prior statements relating to care (including treatment) which, provided they are properly formulated, are legally binding. The assessment of best interests places a considerable responsibility on those making a decision to divine the patient's prior wishes, or his or her likely wishes given known attitudes. Best interests do not simply equate with what the clinical team thinks should be done.

These differences have led some commentators to suggest that mental health legislation is discriminatory (Campell & Heginbotham, 1991; Szmukler & Holloway, 1998). Patients subject to mental health legislation do not have the same safeguards as those treated under mental capacity legislation. For example, the presumption that a patient has capacity is, in an era of homicide enquiries and assertive outreach, turned upside down in patients with mental disorders. An argument put by the government is that mental capacity based legislation would prevent the use of preventive orders in patients thought likely to relapse due to non-compliance. Although psychiatrists making decisions about the use of compulsion seem to use covert tests of best interests (Peay, 2003), they are under no obligation to do so. Instead all is focussed on responding to risks which may be difficult to quantify. Despite the importance of this debate, it has been remarkably little informed by empirical research. What can the available research evidence tell us, and what questions can future research answer?

Mental capacity assessments can be made in a reliable manner

Mental capacity is difficult to define, and although we may feel we know what it means, it is hard to convey that meaning. Clearly this causes problems (although there are, of course, similarly vexed problems which exist in the definition of mental disorder used in status based approaches). If mental capacity is too difficult to use reliably, one might expect that if two psychiatrists using current legal definitions interviewed the same patient they would struggle to reach a consistent view. However various studies (Bellhouse, Holland, Clare, Gunn, & Watson, 2003b; Cairns et al., 2005b; Wong, Clare, Holland, Watson, & Gunn, 2000) suggest this is not the case. When psychiatric inpatients were interviewed by two psychiatrists within 2 days of each other, they were able to agree 91% of the time on the presence or absence of mental capacity to make a treatment decision relating to use of medication (Cairns et al., 2005b). This translated into a kappa value of 0.82, indicating excellent inter-rater reliability (Landis & Koch, 1977). Such reliability is considerably better than that seen for many apparently “objective” tests in physical medicine, such as cardiologists reading an exercise ECG, or radiologists interpreting a mammogram (Sackett, Richardson, Rosenberg, & Haynes, 1997).

Mental capacity assessments have construct validity, up to a point

Of course inter-rater reliability is not the same thing as validity, and in mental capacity there is no true “Gold standard” criterion, but good inter-rater reliability is at least reassures us that the concept can be applied in a consistent manner. A variety of tools have been devised in order to assess mental capacity. The best known is the MacArthur Competence Assessment Tool – Treatment (MacCAT-T) (Grisso, Applebaum, & Hill-Fotouhi, 1997). This is a semi-structured interview in which the patient is questioned about his or her condition and its treatment, and asked to make a treatment choice. The interviewer is able to disclose information, and depending upon the patient's responses, rates four underlying dimensions: understanding, appreciation, reasoning and expressing a choice. Although it is not designed to give a binary capacity rating, it can be used effectively with a clinical interview to guide such a decision. The MacCAT-T and other capacity measures have been used in a various clinical populations and a number of predictable associations are seen. Diagnostic groups who tend to perform least well are those with either severe cognitive disorders or psychosis, whereas individuals with depression typically perform somewhat better, although incapacity is also common (Vollmann, Bauer, Danker-Hopfe, & Helmchen, 2003). There is also reasonable evidence that the more severe the psychopathology, the more severely capacity is affected (Billick, Della Bella, & Burgert, 1997; Grisso & Appelbaum, 1995; Raymont et al., 2004).

Mental incapacity in some diagnostic groups is more difficult to define

The impairments in decision making present in psychotic illness or cognitive impairment are often obvious. When rating transcripts of MacCAT-T interviews one suspects that most psychiatrists would be comfortable that the degree of thought disorder in psychosis; or failure to retain simple information in the cognitively impaired; are compelling reasons for rating the individual as lacking capacity. However there are other instances where the assessments seem to miss more subtle impairments of decision making. Jacinta Tan, in a series of elegant papers, has described the problems of decision making in adolescents with anorexia nervosa (Tan, Hope, & Stewart, 2003a, b, c). These young women performed remarkably well on the MacCAT-T, but had a set of beliefs and attitudes which seem beyond reason, and were deeply embedded parts of the individual's sense of identity. Similar problems might arise in individuals with personality disorders and this has been one of the Government's reasons for avoiding mental capacity based mental health legislation.

Thus the view has been expressed that our current understandings of mental capacity are too restrictive and “cognitive” to be able to encapsulate the range of mental disorders which are currently under the “protection” of mental health legislation (Berghmans, Dickenson, & Meulen, 2004; Tan & Fegert, 2004). Fulford argued that psychopathology in mental disorder leading to mental incapacity is so radically different from that causing mental incapacity in physical disorders that it was difficult to bridge the two (Fulford, 1998). However, the common law cases which have, to a very large extent, set out the principles on which the Mental Capacity Act 2005 are based, have been dominated by treatment refusal for life-threatening physical illness in patients with mental disorders. Thus the impact of delusions (in the case of Re C) and phobias (in the case of Re MB) on the decision making processes about physical disease have set the agenda. Since these and other seminal cases, the Law Commission and the Lord Chancellor's Office have drafted discussion documents which have ultimately led to the new Act (Law Commission, 1995; Lord Chancellor, 1999). These proposals and the new Act itself have used a deliberately broad diagnostic threshold which could, in extreme cases, even include individuals with powerful emotional responses to life situations or severe physical symptoms such as pain. As Fulford acknowledged, mental health legislation has also used a broad definition of mental disorder but placed certain safeguards to ensure a narrower use. This does not seem especially different to the approach developed in mental capacity legislation.

Mental incapacity is common in some clinical populations

Although the first principle of the Mental Capacity Act is that individuals making decisions should be presumed to have capacity, studies in both psychiatric and acute medical settings indicate that rates of mental incapacity are high. At one level this represents a category error – the legal approach is to presume capacity exists, and so to use an epidemiological approach to uncover incapacity is inherently misguided. Nonetheless mapping the extent of incapacity in clinical populations provides potentially useful information. In consecutive patients admitted to psychiatric hospitals between one quarter and one half of all patients lack capacity to make treatment decisions (Bean, Nishisato, Rector, & Glancy, 1994; Bellhouse, Holland, Clare, Gunn, & Watson, 2003a; Billick, Naylor, Majeske, Burgert, & Davis, 1996; Cairns et al., 2005a). Clearly the exact figure will depend upon case-mix and clinical service under study. Given the stigma of psychiatric disorders, an important message from such research is that even in this severe group, the majority of people with psychiatric disorders are capable of making valid treatment choices. What is more surprising is that a remarkably similar proportion of acutely ill medical patients lack capacity. For example, our study of consecutive patients admitted acutely to a general hospital found that approximately one third of the interviewed sample had impairments which interfered with their ability to make decisions (Raymont et al., 2004).

Mental incapacity in patients is often not recognized by clinicians

Despite such high rates of incapacity, clinicians – especially those working outside mental health settings – only recognize a very small proportion of their case load who have decision making difficulties (Raymont et al., 2004). Does this matter? Clearly many patients simply acquiesce to treatment and are happy to place their trust in doctors. There is a risk that too legalistic an approach might erode the trust that forms the basis of the doctor-patient relationship (O'Neill, 2002). However there may be circumstances where the assumption of capacity works against such patients' interests – where for example end-of-life decisions are being made, or nursing home placement is considered. Failing to identify that an acquiescent patient lacks mental capacity may allow more scope for such individuals being coerced into decisions which may not – in the sense of the Act at least – be in their best interests. Common law rulings (Re T: Adult: refusal of medical treatment (1992)) have recognized that the more serious the consequences of a decision, the greater the level of capacity required to make it. Taken to an extreme this implies that the presumption of mental capacity should be relaxed as decisions become more serious. It is noteworthy that the current proposals to legalise physician assisted suicide (Assisted Dying for the Terminally Ill Bill, 2004) do not make such a presumption, and if enacted would force those processing the patient's request for assisted suicide to assess mental capacity.

Mental incapacity and use of the mental health act overlap, but not completely

If capacity-based legislation existed, what proportion of patients currently detained would be free to leave? Data here are very sparse, but suggest a figure of approximately 15 – 20% (Bellhouse et al., 2003a; Cairns et al., 2005a). Much more needs to be known about the characteristics of this group – it might be, for example, that they have fluctuating mental capacity – having been compulsorily admitted to hospital in a crisis, their mental state has settled by the time when the researcher performs an assessment. The available evidence also shows that this group feels especially coerced (Cairns et al., 2005a). The other side of the problem is the extent to which people assenting to admission but lacking capacity are admitted “informally”, a problem which has gained more urgency since the European Court's ruling on the Bournewood case (R v. Bournewood Community and Mental Health NHS Trust, ex parte L (1999)). Such admissions will be open to greater scrutiny and represented approximately one quarter of informal admissions in our study (Cairns et al., 2005a).

Conclusions

Results from observational studies are beginning to illuminate some of the key areas of current legal debate. Further, such research may provide useful data on other topics: the way in which clinicians define and use capacity tests; how clinicians and other agencies judge best interests; the likely impact of advance directives in both mental and physical illness; and the extent to which perceived risk of violence and mental capacity overlap in the same individuals. Research in this area also has its limits. Hard cases have dominated the declarations which courts have previously made on mental capacity to consent to medical treatments. Such cases will always remain the exception, and will never be captured by empirical research. Ultimately there is nothing infallible about the process of assessing mental capacity. It is a framework for thinking about decision making, which may be influenced by a wide range of additional contextual factors. Its principal advantage is that it forces those making the assessment to give precise reasons for over-riding the wishes of a patient.