Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

Legal regulations for advance directives exist in many Western countries. In Germany, for instance, a law on advance directives came into force in 2009. In day-to-day practice, such regulations provide binding guidance for all parties and greater legal certainty (Lipp 2009). Democratic legitimation ensures that regulations on advance directives do not merely serve the interests of particular groups or individuals. In connection with such regulations, important and often controversial problem areas need to be considered and discussed from an interdisciplinary perspective in the legislative process. These include the following ethical issues (Sass and Kielstein 2001; Nationaler Ethikrat 2005) addressed by the legislation in Germany (with similar regulations existing in other Western countries):

  • Authority

  • Scope

  • Requirements for validity

  • Implementation

The patient’s self-determined wishes expressed in an advance directive are binding and must be respected by the physician and other persons. The prerequisite is that the patient has written the directive voluntarily and with the capacity for self-determination, or competence to give consent. Furthermore, the treatment preferences for particular medical situations recorded in the advance directive must be as specific as possible and be applicable to the current situation. Advance directives are then more than merely an aid for physician decision-making or for determining the patient’s presumed wishes (authority). Advance directives are not restricted in scope, i.e. they are valid regardless of the type and stage of disease. There is, explicitly, no restriction of the scope of advance directives—as debated in the past—to the end of life, to an unfavourable prognosis or to terminal disease, etc. The requirements for validity specified in the legislation are that an advance directive is to be prepared in writing and signed by the patient, who is to be of legal age. In Germany (as in many other countries), a prior mandatory consultation with a physician or notarization is not required. The implementation of advance directives places the focus on the process of dialogue between the physician and the healthcare proxy or legal guardian. In clinical practice, decision-making is to be facilitated by professional communication among these parties. As in many other countries, the guardianship court is only to be involved in the event of disagreement among the above-mentioned parties in situations where there is a serious risk to the health or life of the patient.

The aim of the Law on Advance Directives is to enable autonomous individuals to record, in a binding form, self-determined decisions concerning future medical treatment in situations where they lack the capacity for self-determination. Many individuals thereby wish to avoid unnecessary suffering, which they may have experienced, for example, in previous episodes of illness or past hospital treatments, etc. Other motivations are to ease the burden for family members and attending physicians in making difficult treatment decisions in the event of the patient’s incapacity. Advance directives are binding if they contain information on treatment preferences applicable to the current health situation. The nature and extent of the treatments desired or refused should be indicated as specifically as possible. Because this is not always possible in advance, it is helpful to provide general information on one’s personal values, religious beliefs and attitudes towards life and quality of life. Some physicians object that previously expressed self-determined wishes can no longer be applicable for a patient who lacks the capacity for self-determination, since clinical experience shows that a patient’s wishes often change during the course of a disease. However, this objection is not valid: the purpose of the advance directive is precisely to enable individuals—at a time when they are capable of making treatment decisions—to give binding instructions for these medical situations in advance. The patient’s wishes specified in the advance directive then take precedence over the presumed “natural will” in the concrete medical situation. Moreover, it is difficult to imagine how the self-determined values expressed in the patient’s treatment choices could subsequently change at a time when the capacity for self-determination has been lost (Emanuel et al. 1994; Vollmann 2000a; Wittink et al. 2008). Accordingly, as long as the above-mentioned requirements for validity are met, a psychiatrist treating a patient with dementia who is incapable of giving consent must respect the self-determined wishes specified in the advance directive rather than current expressions of “natural will” (gestures, posture, mood, general health condition, etc.). In clinical practice, this can lead to difficult medical-ethical decision situations. Every individual should therefore understand the legal force of an advance directive and the associated consequences. An increase in patient self-determination always entails increased assumption of responsibility by patients. There is thus a need for individual information and counselling, which in practice has not been widely available to date owing to a lack of counselling programmes and expertise.

2 Advance Directives for Patients with Mental Disorders

Even though advance directives have primarily been discussed in relation to end-of-life treatment for patients with severe somatic diseases such as cancer (Sass and Kielstein 2001; Burchardi et al. 2005a, b; Lang-Welzenbach et al. 2005; Nationaler Ethikrat 2005; for overviews see Vollmann and Knöchel-Schiffer 1999 and Vollmann and Pfaff 2003), the scope of directives is not restricted either in Germany or in a number of other countries. This means that advance directives are valid regardless of the type and stage of disease, including all forms of mental illness (cf. Deutsche Gesellschaft für Psychiatrie 2009, 2010; Finzen 2009; for a legal perspective see Olzen et al. 2009).

In Germany, as elsewhere, advance directives have thus far played only a limited role in psychiatric practice (Haupt et al. 1999; Vollmann 2000a; Fritze and Sass 2003; Lauter and Helmchen 2006; Hansen et al. 2008). However, an online search using the terms “advance directive” and “psychiatry” turned up numerous websites and forums of interest groups for the mentally ill and people with experience of psychiatric treatment, as well as counselling programmes offered by associations, hospitals, lawyers, etc. Here, the points most frequently mentioned in relation to advance directives for people with mental disorders were as follows:

  • general refusal of treatment, e.g. with antipsychotics, electroconvulsive therapy (ECT), occupational therapy

  • specific refusal of treatment with high-potency antipsychotics

  • specific refusal of treatment with certain (other) psychotropic agents

  • specification of a maximum dose

  • specification of a maximum duration of treatment

  • specification of an antipsychotic-free treatment interval

  • approval of treatment under specific conditions (e.g. transfer to an open ward)

  • approval of treatment in specific situations (e.g. acute suicidality)

In the US, patients and healthcare providers have had many years of experience with advance directives in psychiatry (Appelbaum 1991, 2004; Swanson et al. 2006a, b; Swartz et al. 2006; Honberg 2010; National Resource Center on Psychiatric Advance Directives 2010). Empirical studies show that the content of advance directives most frequently concerns the following areas (Srebnik et al. 2005):

  • preferred medications, most often antidepressants and second-generation antipsychotics (listed by 81 % of study participants)

  • undesired medications, especially first-generation antipsychotics (64 %)

  • preferred alternatives to hospitalization (68 %)

  • methods of de-escalating psychiatric crises (89 %)

  • refusal of ECT (72 %)

In addition to the advance directive, 46 % of those surveyed had appointed a surrogate decision maker in case of mental illness. Fifty seven percent of the participants explicitly stated that they wished their directive to be irrevocable during periods of incapacity (Srebnik et al. 2005).

In contrast to the frequently expressed reservations and scepticism of psychiatrists towards advance directives, empirical studies from the US paint a more positive picture: for 95 % of advance directives, the patients’ treatment preferences were rated by psychiatrists as feasible, useful and consistent with practice standards (Srebnik et al. 2005). Thus, advance directives can play a constructive role in psychiatric practice and influence physicians’ treatment decisions in line with patients’ wishes (Wilder et al. 2007). Another study showed that successes in treatment can be achieved by educating and informing patients about psychiatric advance directives. Facilitated sessions increased the number of advance directives completed by patients. The specific information on treatment preferences given in the directives was rated by psychiatrists as consistent with standards of community practice (Swanson et al. 2006b).

From an ethical perspective, it can be noted that psychiatric advance directives are used in practice and are desired and respected by patients and psychiatrists alike. There is no fundamental ethical difference between the applicability of advance directives for mental as opposed to somatic disorders.

3 Capacity for Self-Determination

A requirement for the validity of advance directives is the patient’s capacity for self-determination at the time of writing. Since mental disorders can compromise the patient’s self-determination more frequently than somatic illnesses (Grisso and Appelbaum 1998; Vollmann et al. 2003, 2004), the assessment of capacity at the time when an advance directive is written merits special attention in psychiatric practice. Terms used in practice include “of sound mind” and “in full possession of one’s mental faculties”, as well as ability to manage one’s financial affairs, competence to consent or capacity for self-determination, capacity for judgement and decision-making, etc.; however, the definition of these terms often remains unclear.

Capacity for self-determination requires the fulfilment of the following criteria:

  • Ability to understand information

  • Ability to reason

  • Ability to appreciate a mental disorder

  • Ability to appreciate the possible benefits of treatment

  • Ability to make a decision and express a choice

The ability to understand information means that information given by a psychiatrist in an effort to obtain informed consent is understood by the patient. To prevent the patient from merely repeating what has been said without understanding it (which would be merely a test of memory rather than understanding), it is advisable to ask the patient to recapitulate the information in his or her own words. It is important that the information given is actively comprehended by the person concerned. The required ability to understand information applies not only to the communication of medical information by the psychiatrist, but also for all other areas.

Having understood the information given, the patient must integrate it into the context of his or her life and system of values in order to be able to make a self-determined decision. The patient must assess what this medical information means for his or her life planning, preferences and values. The advantages and disadvantages of the proposed treatment options need to be weighed up and assessed with respect to the consequences and alternatives. It is precisely this individual assessment process which enables self-determined decision-making (reasoning).

Patients with the capacity for self-determination are also able to recognize that they have a mental disorder (appreciation of the disorder) and that there are treatment options which could help (appreciation of treatment benefits). It is explicitly not required that the patient must personally accept a medical diagnosis (e.g. schizophrenia) or psychiatric conception, or use this terminology him/herself. Rather, for the assessment of capacity, what is important is that patients recognize that they are affected by a mental disorder and acknowledge potential treatment options. Finally, patients must be able to make a decision and express a choice (Grisso and Appelbaum 1998; Vollmann 2000b).

In the international psychiatric literature, the MacArthur Competence Assessment Tool (MacCAT) is considered to be the gold standard for assessing decisional capacity (Grisso and Appelbaum 1998; German translation in Vollmann 2008). Tools are available for the areas of treatment (MacCAT-T), clinical research (MacCAT-CR) and criminal adjudication (MacCAT-CA). MacCAT-T formed the basis for the development of the Decisional Competence Assessment Tool for Psychiatric Advance Directives (DCAT-PAD). This tool assesses patients’ basic grasp of advance directives, understanding of content, and reasoning abilities (Elbogen et al. 2007). Specifically, it evaluates their ability to:

  • understand the key components of psychiatric advance directives (PADs)

  • appreciate whether or not PADs would be relevant to them and their treatment

  • reason about how PADs would affect their lives

  • choose whether they would want to fill out a PAD

  • understand the pros and cons of hospital treatment

  • appreciate whether hospitalization may be a relevant option for them

  • reason about how hospital treatment would affect their lives

  • choose whether they would want to be hospitalized if they became ill

Professional assessment and documentation of the capacity for self-determination when an advance directive is written of great significance for the ethical and legal validity of the patient’s wishes recorded therein. For unlike mere expressions of the patient’s natural will, a self-determined expression of wishes imposes an ethical and legal obligation on third parties (e.g. physician, family members) to respect the patient’s wishes, even when these run counter to the physician’s advice or a medical indication for treatment.

4 Problems with Assessments of Capacity

In medical ethics, the principle of self-determination obliges physicians to obtain a patient’s consent prior to medical treatment (Beauchamp and Childress 1994). A condition for valid informed consent is the patient’s competence, which may be impaired, particularly in patients with mental problems (Helmchen and Lauter 1995; Koch et al. 1996; Helmchen 1998; Vollmann 2000b).

In clinical practice, physicians generally presume competence on the patient’s part. But if they have occasion to assess competence, they will generally proceed using their own subjective judgement and clinical experience and have difficulty applying standards suggested in the literature (McKinnon et al. 1989; Markson et al. 1994). Physicians also often make inconsistent evaluations of competence in a given case (Marson et al. 1997). Considering the ethical and legal significance of competence and the desire for physicians’ evaluations to be transparent and reliable, various objective testing procedures have been developed and applied in clinical trials in recent years (Janofsky et al. 1992; Bean et al. 1994; Marson et al. 1995; Kitamura et al. 1998; for an overview see Bauer and Vollmann 2002).

In studies using structured test interviews, a considerable proportion (40–75 %) of acutely psychotic and schizophrenic patients had no competence to consent (Grossman and Summers 1980; Appelbaum et al. 1981; Hoffman and Srinivasan 1992; Grisso and Appelbaum 1995). Patients with depression, however, displayed fewer impairments in empirical studies (Grisso and Appelbaum 1995; Appelbaum et al. 1999). Among patients with dementia, the proportion of non-competent persons increased both with increasing severity of disease and with increasing stringency of the legal standards applied (Marson et al. 1995; Kim et al. 2001). A significant correlation between competence and multiple cognitive functions has been described in demented patients (Marson et al. 1996).

In only a few studies has competence been directly compared among various diagnostic groups using the same test instrument. One such study, the MacArthur Treatment Competence Study (Grisso and Appelbaum 1995), compared schizophrenic, depressed and medical patients using instruments related to the following four legal standards:

  1. 1.

    ability to understand information relevant to the decision about treatment;

  2. 2.

    ability to manipulate the information rationally (or reason about it) in a manner that allows one to make comparisons and weigh options;

  3. 3.

    ability to appreciate the significance for one’s own situation of the information disclosed about the illness and possible treatments; and

  4. 4.

    ability to express a choice.

Patients with scores below defined limits were categorized as impaired in that standard. The majority of patients in all diagnostic groups performed adequately in all standards. However, patients with schizophrenia—as a group—had significantly more deficits than the other two diagnostic groups, and this group displayed the largest proportion of patients with impairment. A hierarchical order of the standards was not detected. For each patient group, a combination of standards resulted in an increase in the proportion of impaired patients, with impairment being defined as poor performance on any of the standards. Since the instruments used were too cumbersome for application in day-to-day clinical practice, the authors developed a short form with a high degree of inter-rater reliability, the MacArthur Competence Assessment Tool-Treatment (MacCAT-T) (Grisso et al. 1997; Grisso and Appelbaum 1998).

Using different standards of competence, substantial differences were found among patients in various diagnostic groups (Vollmann et al. 2003). Studies showed that schizophrenic patients had greater impairments than depressed patients in the standards relevant for competence to consent to treatment (Bean et al. 1994; Grisso and Appelbaum 1995). Patients with dementia showed even greater impairments than schizophrenic or depressed patients. For each standard, a significantly greater proportion of patients with dementia was impaired than in the other two diagnostic groups (Vollmann et al. 2003). Despite the substantial proportion of impaired patients, many patients displayed no impairments in the standards used; thus, the diagnosis is not suitable as the sole indicator of incompetence in individual cases.

The paradoxical situation observed among many schizophrenic patients, who generally appreciated the treatment benefit despite reduced appreciation of the disorder, reminds us of the “double accounting” among patients with schizophrenic psychoses, familiar from clinical practice. The conspicuously high proportion of such patients with full appreciation of treatment benefit may possibly be explained by a selection effect, since patients with a negative attitude towards treatment frequently also refuse to participate in clinical studies.

For a thorough assessment of competence in practice, all standards should be examined, since deficits in just one standard can call overall competence into question. However, the more thoroughly competence is examined (i.e. the more standards are used in assessing it), the more patients will be evaluated as incompetent (Vollmann et al. 2003). The selection and combination of standards depend on previous value judgements. This influences both the content requirements and the threshold for defining incompetence to consent.

Compared to the patients categorized as unable to consent by clinical assessment, the proportion of patients impaired in at least one standard of the MacCAT-T is significantly higher. The discrepancy between objective testing methods and clinical assessment in evaluating competence has been described in the literature (Rutman and Silberfeld 1997; Vollmann et al. 2003). The question arises of whether clinical assessment or objective testing methods are more suitable for satisfying the ethical demand to respect patients’ self-determination and also to protect the well-being of patients with impaired competence (and safeguard them against serious and dangerous consequences of incompetent decisions). In the light of the discrepancy found between clinical assessment and objective testing, we must decide whether the risks of decision substitution for possibly competent patients (objective testing) are greater than the risks of possibly incompetent patients making their own decisions (clinical assessment) (Vollmann 2000c).

The MacCAT-T showed good applicability in clinical practice. Most patients evaluate the interview as positive since it gave them the opportunity to discuss their illness and possible forms of therapy at length with a physician. In our experience, patients have great communication needs in this regard. However, some patients complained after the interview that it required a lot of concentration and was difficult at times. For instance, one patient with her first manifestation of schizophrenia said that the test was too much for her on the day of admission (Vollmann et al. 2003). This patient’s statement illustrates the practical significance of informed consent as an “educational process” (Roth 1983) and evaluation of competence as an “evolving process influenced by therapeutic interventions” (Mahler and Perry 1988). Reduced abilities related to competence in schizophrenic patients can be improved by educational interventions (Carpenter et al. 2000). More intensive education strategies tailored to individual patients are of great importance in this regard.

There are difficulties administering the MacCAT-T to patients with severe dementia who have major cognitive impairments. This raises the question of whether, as critics suggest (Elliot 1997; Charland 1998; Welie 2001), the criteria for objective testing methods are too strictly or too one-sidedly oriented towards cognitive functions, while ignoring evaluation of major emotional factors. The identification of criteria used for decision-making by people with cognitive impairments (e.g. emotional, social-context-specific and biographical), and the integration of these criteria into the evaluation of competence, may be starting points for further research (Breden and Vollmann 2004). In view of demographic developments—with an increasing proportion of older people in the population, and hence also of patients suffering from dementia—such research approaches are of considerable practical significance.

Controversy surrounds the question of whether objective testing methods can replace clinical assessment of competence. A major problem lies in setting the cut-offs in the tests. In many studies, these are set on the basis of statistical considerations. Studies comparing objective testing methods with other evaluation methods (clinical assessment, forensic-psychiatric study) range from complete agreement (Janofsky et al. 1992) to significant discrepancies (Rutman and Silberfeld 1997). In our opinion, cut-offs have the primary purpose of allowing statistical comparison of groups and illuminating deficits in examined standards; a categorical decision as to whether competence is present or absent cannot be made solely on the basis of the test results. We view the MacCAT-T as a suitable instrument for detecting deficits in patients’ decision-making abilities in a concrete case; it should be followed by a thorough clinical evaluation that also includes non-cognitive aspects.

5 Ethical Conclusions

Advance directives are valid for the treatment of all illnesses, regardless of whether they are somatic or mental. The considerations presented here refer exclusively to the treatment of patients with mental disorders. The special cases of court-ordered commitment to inpatient treatment and coercive measures for individuals posing a danger to others cannot be discussed further here (cf. Olzen et al. 2009).

The prerequisite for the validity of an advance directive is the patient’s capacity at the time when it is written. Since patients with mental disorders may have lost the capacity for self-determination, special attention must be paid to the assessment and documentation of this matter. This requires a clear ethical and legal definition of capacity for self-determination or competence to give consent. Apart from the question of capacity, important points are how closely the situation envisaged in the advance directive corresponds to the current medical situation and the specificity of the treatment preferences expressed. Since, in the nature of the case, the criterion of a close fit between the advance directive and future situations cannot always be met, it is helpful to include more fundamental information on values, the patient’s biography, etc. It is further recommended to combine the advance directive with the designation of an informed and trusted individual via a durable power of attorney for healthcare.

Neither the severity of disease nor a conflict between the self-determined wishes and a medical indication for treatment is sufficient to override the wishes expressed in a valid advance directive. In the event of a conflict, the patient’s self-determined, closely fitting wishes prevail over specialist medical advice. The physician’s duty to provide medical care, a medical indication or a psychiatric patient’s fundamental right to receive medical treatment cannot be taken to constitute a duty to treat, outweighing the patient’s self-determined wishes. This also applies to life-threatening diseases, as shown by the extensive discussion in medical ethics and medical law with regard to somatic illnesses.

These clinical conflicts and ethical dilemmas have been clearly resolved by legislators in favour of the patient’s self-determined wishes. This is demonstrated by examples and experience in somatic medicine, such as the fundamental refusal of blood transfusions on religious grounds by patients who are Jehovah’s Witnesses, even in life-threatening situations. By analogy, this also applies to life-threatening situations in psychiatry, such as the self-determined refusal of ECT as a medically indicated and life-saving measure in pernicious catatonia. The prerequisite for the legal force of the patient’s wishes, however, is that they are self-determined in accordance with clearly defined criteria, as opposed to mere expressions of natural will. This means that the patient has been adequately informed about his or her health situation, treatment options (nature, purpose, advantages and disadvantages, alternatives) and prognosis. If the patient’s guardian and the attending physician disagree in their interpretation of the advance directive, the guardianship court must be called on to make a decision if there is a serious risk to the patient’s life or health.

The more authentically the patient’s self-determined wishes are expressed, the more helpful and convincing an advance directive will be. This applies particularly to the expression of values which run counter to a medical indication. Timely and appropriate physician-patient communication on treatment options and patient self-determination is in practice often an important prerequisite for achieving this goal.

The above-mentioned empirical studies and practical experience with psychiatric advance directives in the US show that shared decision-making between patient and psychiatrist on treatment matters can be achieved in clinical practice. Here, the instrument of the advance directive should be understood as an integral part of the joint process of communication and decision-making between patient and physician within the framework of a longer-term treatment plan. Furthermore, the frequency and quality of advance directives can be increased through targeted educational measures in the psychiatric hospital. This should not be forgotten in the debate on advance directives, which often focuses on the few cases where conflicts arise (Vollmann 2010; Zeug 2010). For, particularly in the treatment of mental disorders, the patient’s individual personality, lifestyle and values play a key role. The utilization of advance directives in psychiatry presents an opportunity to improve psychiatric care through timely communication with patients and ascertainment of their self-determined wishes.