Abstract
Background:
Advance directives enable patients to ensure that treatment decisions will be based on their autonomous will, even if they are incompetent at the time at which the treatment decision is taken. Although psychiatric advance directives are legally binding in Germany and their benefits are widely acknowledged, they are still infrequently used in German psychiatric practice.
Aims:
The aim of this study is to assess psychiatrists’ attitudes toward the use of advance statements in mental health care.
Methods:
A postal survey of psychiatrists in Germany was carried out to examine their views on advance statements in psychiatry. The survey addressed psychiatrists’ experiences of and attitudes toward different types of advance statements, including psychiatric advance statements written by patients without any specific assistance, and joint crisis plans ( ‘Behandlungsvereinbarungen’), where involvement of the clinical team is required. A total of 396 responded.
Results:
Results suggest that generally speaking, respondents held favorable views on joint crisis plans for mental health care. In all, 80.7% of participants agreed that more frequent use of joint crisis plans in clinical practice would be desirable. However, clinicians’ attitudes differ largely depending on the type of advance statement. Implications for the use of advance statements in psychiatry are discussed.
Conclusion:
The findings suggest that increasing the support structures available to train physicians and inform patients could lead to increased adoption of advance statements.
An advance statement is a legal document that contains a patient’s preferences for treatment in the event of impaired decision-making capacity. These documents enable patients to make sure that decisions will be based on their autonomous will, even if they are incapable of making or communicating a decision at the time of treatment. In the context of mental health care, these documents are primarily used by people with severe mental illnesses that involve fluctuating decision-making abilities, such as bipolar affective disorder or psychosis.
A variety of different names are used in the literature to describe these documents – a fact which has sometimes led to confusion. Both ‘advance statement’ and ‘advance directive’ are general terms encompassing all such documents. A more specific term is used to refer to advance statements written with the assistance of the treating psychiatrist. A ‘joint crisis plan’ (‘Behandlungsvereinbarung’) refers to an advance statement created after a conversation between the patient and the care team, and in which decisions are taken jointly with the treating psychiatrist (Atkinson, Garner, & Gilmour, 2004; Borbé, Jaeger, & Steinert, 2009; Henderson et al., 2004).
In general, disclosures to the patient are not a legal requirement in Germany in order for an advance statement to be lawful, and patients can and sometimes do write such a statement on their own without support of others.
Advance directives are generally seen as a way to increase the autonomy of patients. Research has also examined a number of additional potential advantages related to the use of advance statements in psychiatry. For instance, the wider use of advance statements could increase patients’ sense of empowerment (Kim et al., 2007) and lead to a more patient-oriented treatment culture (Wilder et al., 2013). Taken together, this would, in turn, reduce the need for coercive interventions (Jong et al., 2016).
These results suggest that using advance statements in psychiatry might also contribute to the aims of the United Nations Convention on the Rights of Persons with Disabilities (CRPD). Article 12(3) of the CRPD imposes an obligation on States Parties to provide appropriate measures for supported decision-making and advance statements may be useful tools to this end. In fact, the Committee specifically mentions advance statements as an important form of support for reconstructing the person’s will when communication is not possible in the General Comment on Article 12 (United Nations Committee on the Rights of Persons with Disabilities, 2014). As a result of the Committee’s position, implementing advance statements is necessary in order for psychiatrists to fulfill the mandate of Article 12. Providing opportunities for advance care planning is thus not merely optional but is becoming a legal requirement in wake of the demands of the CRPD.
Studies have shown that about three-quarters of psychiatric patients are interested in these tools (Srebnik, Russo, Sage, Peto, & Zick, 2003; Swanson, Swartz, Ferron, Elbogen, & van Dorn, 2006; Van Dorn, Swanson, & Swartz, 2009). However, while the advantages of the use of advance statements in psychiatry are widely acknowledged, the application of advance statements in German psychiatric practice remains low (Borbé, Jaeger, Borbé, & Steinert, 2012; Radenbach, Falkai, Weber-Reich, & Simon, 2014). This discrepancy between patients’ interest in and completion of advance statements has previously been reported in other countries, including the United States (Swanson et al., 2006), raising the question of why adoption is not higher.
As a result, some research has investigated potential barriers to the implementation of advance statements (Nicaise, Lorant, & Dubois, 2013; Shields, Pathare, van der Ham, & Bunders, 2014), such as limited knowledge about these tools and critical opinions by psychiatrists (Van Dorn et al., 2006). Because psychiatrists are a key participant in the creation of advance statements, their endorsement is an important precondition for the successful implementation of such documents, and it is important to understand how they view these documents. What, then, do psychiatrists in Germany think about the use of advance statements in mental health care? And which barriers, from the psychiatrists’ point of view, prevent the further implementation of advance statements within psychiatric treatment? The present study examines the views on psychiatrists on different types of advance statements.
Methods
Sample and procedure
The survey was distributed to psychiatrists who are licensed to teach doctors during their residency training in psychiatry. In the summer of 2016, 1,164 psychiatrists with this license (called a ‘Weiterbildungsbefugnis’) were identified via the 17 medical associations in Germany and sent the survey in form of a cover letter, a survey with a cover sheet, and a pre-addressed return envelope. Data collection occurred between July 2016 and November 2016. In all, 229 questionnaires were returned after the first letter. Those who did not return a questionnaire received another questionnaire and a reminder. After the reminder, of the 1.164 questionnaires sent out, 396 psychiatrists responded, representing a response rate of 34.0%. In the cover letter, it was made clear that all respondents would be treated anonymously. The survey has been approved by the research ethics committee of the medical faculty of the Westfälische-Wilhems University Münster, Germany (Reg No. 2016-153-f-S).
Design of questionnaire
The survey consisted of 50 questions divided into four sections. The first two sections concerned joint crisis plans. A joint crisis plan is a specific subtype of advance statements involving a discussion between the patient and care team. While a care team may be expected to enter into a joint crisis plan despite not fully endorsing all aspects thereof, the care team need not agree to treatment decisions in the joint crisis plan which they do not support at all. The care team thus has the right to not sign a joint crisis plan; however, under such circumstances, the patient would still be able to write an advance statement without consultation and signature of the care team. In this context, it is also important to bear in mind that the power dynamics within these discussions between psychiatric staff members and patients are asymmetrical. It can’t be ruled out that the way the discussion about the joint crisis plans proceeds may well have a decisive impact on the content of the joint crisis plan.
The first section asked psychiatrists about their overall attitude toward joint crisis plans, their potential benefits, and potential barriers to their use. The second section asked psychiatrists about their disposition toward specific provisions within joint crisis plans and specific use cases. The third section compared attitudes among different models of advance statements, and the final section collected demographic information. Each section is described in greater detail below.
In the first section, participants were asked questions about their overall attitudes toward joint crisis plans. They were presented with four statements about their attitude toward joint crisis plans in general, six potential benefits or reasons for using joint crisis plans, and nine potential barriers to use. Participants were asked to rank all of the statements on a five-point Likert-type scale from strongly agree to strongly disagree.
The second section consisted of statements suggesting different matters that could potentially be covered by a joint crisis plan and asked participants to rate the usefulness of joint crisis plans in each case. The four response possibilities were ‘very useful’, ‘somewhat useful’, ‘mostly not useful’, and ‘not useful at all’. Participants rated the usefulness of instructions about methods of de-escalating crises, medication preferences, refusal of certain medications, preferences on the use of electroconvulsive therapy (ECT), and affirmative consent to hospitalization in the event of a crisis. This section also listed different psychiatric disorders and asked psychiatrists in which cases they considered joint crisis plans to be useful. Schizophrenia, bipolar affective disorder, depression, manic episodes, addiction, and emotionally unstable personality disorder were all listed.
The third section differentiated between different types of advance statements based on the degree to which health professionals are involved in the process of their creation. Participants had previously rated their agreement with the statement ‘it is desirable that joint crisis plans are used more frequently in clinical practice’ with the explanation that a joint crisis plan is a specific type of advance statement. In this third section, they were asked to use a Likert-type scale to rate their agreement with three variants of the same statement, each describing an advance statement involving a different amount of physician input. These statements were as follows: ‘It would be desirable that advance statements are used more frequently if the patient has written the advance statement alone’ (1) ‘[…] after a disclosure by the treating psychiatrist’, (2) ‘[…] after a disclosure by another physician, e.g., the general practitioner’ and (3) ‘[…] if there was no disclosure by a physician’.
The demographic section included questions about age, gender, the current work setting (clinic vs. private practice), professional position, knowledge about advance statements in general, and practical experience with different types of advance statements.
Data analysis
Descriptive statistics and correlations were calculated for relevant variables. Data were entered and analyzed using IBM SPSS Statistics for Windows (Version 24). Statistical analyses are mostly descriptive. Differences between groups of respondents were analyzed using chi-square and Fisher’s exact test. p-values were regarded as exploratory, not confirmatory. Not all participants answered every single question, so the number of participants who answered the questions varies from question to question.
Results
Demographic data
Data about demographic information were missing for 20 of the 396 participants. Of the remaining 376 participants, 101 were women (26.9%). In all, 182 (48.4%) of the participants were between 50 and 59 years old, 109 (29.0%) were younger than 50 years and 85 (22.6%) were older than 59 years old. A total of 285 (76.0%) of the participants were working in the clinic, among which 166 (44.3%) were senior consultants (‘Chefärzte’). In all, 90 (24.0%) of the respondents were working in the ambulatory sector.
Knowledge and practical experience with psychiatric advance statements
The majority of respondents reported familiarity with advance statements. In response to the question, ‘How much knowledge do you have about advance statements in general?’ slightly more than a third (37.7%, n = 142) reported ‘deeper knowledge’, about half the sample (53.3%, n = 201) reported that they had ‘basic knowledge’, and only few (9.0%, n = 34), stated that they had ‘no knowledge’ at all.
The survey also asked participants about their practical experience specifically with joint crisis plans. Only about one quarter (23.1%, n = 87), however, reported having ‘a lot of experience’, about half the sample (58.9%, n = 222) reported having ‘some experience’, and about one fifth (18.0%, n = 68) reported having ‘no experience’. The numerical results are shown in Table 1.
Reported knowledge about advance statements and experience with joint crisis plans.
General opinions about joint crisis plans
In general, respondents held favorable views on joint crisis plans. In all, 80.7% (n = 318) agreed with the statement that ‘it is desirable that joint crisis plans be used more often in psychiatry’. In all, 74.1% (n = 292) also agreed that ‘physicians should advocate the use of joint crisis plans in psychiatry’. Furthermore, a large majority (82.7%, n = 326) agreed with the statement ‘from a clinical-practical point of view, it is feasible to use joint crisis plans more often’. There was also high agreement among participants (87.8%, n = 346) with the statement ‘more staff training regarding joint crisis plans is needed’. Figure 1 shows the exact agreement percentages for each question.
General opinions about joint crisis plans.
A chi-square analysis comparing the answers of those who had more experience against those who had less experience showed that doctors with more practical experience were more likely to be favorable toward joint crisis plans (11,236, df = 2, p = .003). Of the psychiatrists who indicated having ‘a lot of experience’ with joint crisis plans, 88.2% (n = 75) agreed that their use is desirable, while 84.1% (n = 185) of those having ‘some experience’ and 67.2% (n = 42) of those having ‘no experience’ agreed to this statement.
In all, 70 psychiatrists (17.7%) did not agree to the statement ‘it is desirable that joint crisis plans be used more often in psychiatry’. Among these, however, 31 (44.3%) nevertheless agreed that, from a clinical-practical point of view, it is feasible to use joint crisis plans more often and 41 (58.5%) agreed that more staff training is needed.
Perceived barriers to use of joint crisis plans and negative experience
Table 2 shows the percent of respondents who agreed with statements about potential barriers, showing both the responses of all participants and of those participants who indicated having a lot of experience with joint crisis plans (‘experienced psychiatrists’). The mentioned barriers included both fundamental limitations and challenges (like the difficulty of predicting future medical situations and making joint crisis plans precise enough) and procedural difficulties (such as increased administrative burdens).
Reported perceived barriers to use of joint crisis plans in psychiatric treatment.
A large majority of psychiatrists (91.8%, n = 362) agreed with the statement ‘patients might not update their joint crisis plans often enough’. About two thirds of participants agreed with the statements ‘joint crisis plans might not be concrete enough’ (62.8%, n = 248) and ‘there is lack of trained physicians that help patients to write joint crisis plans’ (69.8%, n = 276).
Regarding practical barriers such as lack of time or extra paperwork required, about half of the psychiatrists agreed to these statements (53.2%, n = 210; 53.0%, n = 209, respectively). In the group of psychiatrists who indicated having a lot of experience with joint crisis plans, however, fewer agreed to each of these statements (40.7%, n = 35; 41.4%, n = 36, respectively).
On the other hand, it is noteworthy that the majority of participants disagreed with the statement ‘patients are usually not interested in joint crisis plans’ (56.6%, n = 223), while an unusually large proportion reported that they did not know (10.7%, or n = 42, compared with approximately 1% – 4% on other questions). A majority also disagreed with the statement ‘joint crisis plans are problematic because they might restrict therapeutic options of physicians’ (75.7%, n = 299).
Content of joint crisis plans
The participants strongly agreed that it would be useful for joint crisis plans to contain medication preferences and methods of de-escalation. 93.4% (n = 369) agreed that naming preferred medications would be useful, while 85.7% (n = 339) agreed that the inclusion of methods of de-escalation would be useful. 84.3% (n = 332) also agreed that it would be useful if patients would give prior authorization to be treated in the hospital in case of a crisis.
Fewer participants, but still a majority (79.4%, n = 313), stated that the refusal of certain antipsychotics would be useful and 72.1% (n = 284) agreed that a statement regarding ECT would be useful. Furthermore, doctors were especially critical of the idea of patients including a maximum dose of medication in their joint crisis plans. Only 38.7% (n = 157) agreed that this would be a useful inclusion, while 59.4% (n = 234) stated that including a maximum dose of medication would not be useful.
Views on different advance statements
It is noteworthy that the more the advance statement was described as an instrument written by the patient without the involvement of the attending physician, the more psychiatrists were critical. Figure 2 shows the exact level of agreement with a relevant statement. In all, 80.7% of respondents (n = 318) agreed with the general statement that ‘it is desirable that joint crisis plans are used more often in psychiatry’. Slightly fewer (73.7%, n = 291) agreed that ‘it would also be desirable that advance statements are used more frequently if the patient has written the advance statement alone after a disclosure by the treating psychiatrist’. When the disclosure would be conducted by another physician, such as the general practitioner, however, only about one third (35.5%, n = 140) agreed. Few psychiatrists (11.4%, n = 45) believed it would be desirable for advance statements to be used more frequently if no disclosure took place at all. More than 80% of psychiatrists disagreed (n = 325) with more than 50% (n = 207) of psychiatrists disagreeing strongly that more frequent use of advance statement that are done without the consultation of a physician would be desirable.
Views on different advance statements.
Discussion
Discussion and interpretation of results
Knowledge and general opinions
Our results indicate that most psychiatrists in Germany are familiar with advance statements. Most respondents indicated, however, that they possess only theoretical familiarity and lack hands-on experience. This result supports the finding of other studies that while health care professionals are increasingly aware of this topic, the actual use of advance statements is not yet common in clinical practice in Germany (Borbé et al., 2012; Radenbach et al., 2014).
Dispositionally, psychiatrists broadly support the use and implementation of advance statements. This is an important finding, as it contradicts earlier studies that have suggested that critical attitudes among psychiatrists might be a major barrier to implementation (Van Dorn et al., 2006). This attitude varies strongly, however, depending on the exact type of advance statements, with psychiatrists indicating greater favorability for joint crisis plans when compared with other forms of advance statements involving less input from the care team.
Barriers/negative experiences
Psychiatrists indicated that they perceive multiple barriers that threaten to hamper implementation. Many psychiatrists agreed with statements citing fundamental challenges of the use of advance statements, such as difficulty predicting future medical situations and challenges related to the renewal and specification of advance statements. More than 90% of psychiatrists agreed that it must be feared patients would not update joint crisis plans frequently enough. When advance directives first became legally binding in Germany in 2009, a controversial amendment to require mandatory updates was discussed, but eventually discarded. It seems, however, that many psychiatrists perceive there to be a risk of advance statements becoming outdated.
Fewer psychiatrists were concerned about procedural barriers to implementation, such as lack of time or extra paperwork. This result is consistent with earlier studies (Van Dorn et al., 2006; Wilder et al., 2013). Moreover, there was a negative correlation between hands-on experience with joint crisis plans and concern about these barriers. This suggests that while there may be practical barriers to implementation, inexperienced health professionals overestimate their impact.
In contrast to previous research, most psychiatrists indicated that they were aware that patients are interested in joint crisis plans. This result contrasts with a study in the United States, where approximately 61% of clinicians were found to incorrectly believe that patients would not want to complete advance statements (Wilder et al., 2013).
Content of joint crisis plans
More psychiatrists were in favor of patients naming preferred medication than of patients refusing certain medications. In addition to being in favor of patients proactively indicating preferred medication, psychiatrists also supported patients affirmatively giving their consent or prior authorization to being treated in case of a crisis. This is linked to the idea of the so-called ‘Ulysses contract’, or self-binding advance directive, in which patients commit themselves to treatment during future episodes of illness. These irrevocable ‘opt-in models’ of advance statements may be especially interesting for people with bipolar affective disorder (Gergel & Owen, 2015). In the German legal framework, however, the legal binding force of irrevocable opt-in directives is questionable (Henking & Bruns, 2014). Prior research has shown that a substantial number of patients are interested in irrevocable advance statements or ‘Ulysses contracts’ (Srebnik et al., 2005). The fact that psychiatrists seem to embrace features of these documents as well suggests that it is pressing to resolve both ethical and legal issues regarding these tools. That the use of joint crisis plans was ranked as useful for people with bipolar affect disorders by almost 90% of participants underscores this conclusion.
Attitudes on different types of advance statements
The survey results show that psychiatrists’ attitudes on advance statements vary depending on the level of professional input. Marginally more psychiatrists have a positive view of joint crisis plans (in which the patient and treating psychiatrist create the plan together) than advance statements created by the patient alone following a disclosure from the treating psychiatrist. Fewer than half as many respondents, however, believed that advance statements written after a disclosure from another medical professional were desirable, and even fewer supported advance statements with no disclosure at all. Put another way, the survey demonstrates that physicians typically do not believe that the care team must be involved in the creation of the advance statement, but they do believe that a disclosure from that team is necessary. There are several potential explanations for this result.
First, psychiatrists may worry about the communicative efficacy of advance statements written without the help of the treating psychiatrists. These statements may not provide sufficient or accurate information, and, without their prior involvement, psychiatrists may fear that they were completed without appropriate reflection or discussion. As a result, it may be difficult to interpret and act upon such statements. In other words, insofar as advance statements are a way for patients to communicate with their doctors, statements written without the involvement of those doctors may be prone to error, misinterpretation, or other difficulties. This interpretation is in line with results from a similar study from the United States., in which approximately 63% of psychiatrists indicated that ‘lack of quality information in the document’ is a barrier to implementation (Van Dorn et al., 2006).
Second, psychiatrists may worry about the ability of their patients to write an advance statement without their guidance. Patients may not fully understand the scope of the issues involved or reflect adequately without input from qualified professionals. Advance statements written without a full understanding of their stakes may not fully or accurately reflect a patient’s desires were they to have been better informed.
A third, and related, explanation is that psychiatrists may fear that, without a consultation, it will not be clear whether the patient was unambiguously capable of making treatment decisions when the statement was written.
Finally, psychiatrists may simply believe that their expert training enables them to make better health judgments than their patients, and that they should therefore be involved in treatment decisions to avoid poor outcomes for the patient. The fact that psychiatrists are critical of advance statements written solely by the patient might therefore also be explained by paternalistic attitudes of psychiatrists. Physicians sometimes believe that their duty to ensure the patient’s best interests justifies overriding the wishes and preferences of the patient when these run counter to his or her best interests. Paternalism has played a significant role throughout the history of psychiatry, yet this has been regarded as problematic in recent times as values of autonomy and the right to self-determination have been more fully embraced (Beauchamp & Childress, 2013). Nevertheless, our study seems to reveal that paternalistic attitudes are still dominant in parts of psychiatry today.
The view that supporting patients in the creation of advance statements is important is not new, with some even going so far as to insist that this be a legal requirement for their validity. This issue was debated when advance statements were first legalized in Germany, with an amendment to the law creating a requirement for a consultation (a ‘Beratungspflicht’) being introduced and ultimately defeated. Proponents of this requirement argue that decisions made by poorly informed individuals cannot be regarded as autonomous, so a consultation or disclosure is necessary to create a useful advance statement, and should therefore be mandated in order for the advance directive to have legally binding force. Skeptics point to the commonly held principle that, in a liberal society, the decisions of a person able to consent should not be overruled except in cases where they infringe upon the rights of others. Under these constraints, western society typically errs in favor of allowing individuals to make whatever personal decisions they want, using whatever decision-making process they deem appropriate. Nevertheless, even opponents of a requirement typically acknowledge the value of the availability of a consultation when desired by a patient and hold the view that these consultations should be easily accessible for patients.
Limitations
The most important limitation of this study is a low response rate, which may have led to a selection bias. In particular, it is likely that respondents with an interest in – and/or greater familiarity with – this topic may have preferentially replied. Second, the results of this study may not be generalizable because the participants were mostly senior consultants and physicians in leading positions, and it is unclear how other demographics might respond to similar questions. Finally, there may have also been a social desirability bias in the survey. Participants were asked to put their names on a cover sheet, which was discarded before the results were analyzed, and this may have led to a perceived lack of anonymity that could have influenced the results.
Conclusion
Our survey indicates broad support for the implementation of joint crisis plans among psychiatrists. There was overwhelming agreement with the theoretical benefits of joint crisis plans, and those with hands-on experience reported that experience to be positive. Nevertheless, psychiatrists were skeptical of other forms of advance statements involving less psychiatric input and perceive practical barriers to the implementation of advance statements in general, although our findings also suggest that practical experience ameliorates how severe psychiatrists judge these barriers to be.
There are several major implications of these findings. First, our results suggest that the attitudes of psychiatrists are not a barrier to implementing advance statements when they are created in conjunction with the treating psychiatrist. Second, because increased practical experience with advance statements is correlated with a decrease in perceived problems, increased training may result in more widespread adoption. Finally, the results of this study indicate that an overwhelming majority of psychiatrists are skeptical of the usefulness of advance statements written without the consultation of a physician, despite overwhelming support for advance statements in general. If the support of psychiatrists is taken as an important precondition for the implementation of advance statements, then these results strongly militate in favor of increasing the availability of such consultations. Taken together, these findings suggest that increasing the availability of support structures, such as training for practitioners and disclosures for patients, is likely to enable an increase in the adoption of advance statements.
