Why Mental Health Clinicians are Not Engaging in Advance Care Planning with Older People with Schizophrenia and Other Psychotic Illnesses

Abstract

Purpose

To explore mental health clinicians’ attitudes, experiences, and perceived barriers regarding Advance Care Planning (ACP) with older people (aged 55+) with schizophrenia/other psychotic illnesses.

Methods

Qualitative analysis of focus group interviews with multidisciplinary mental health clinicians from public mental health services in Sydney, Australia. A senior external clinician facilitated online focus groups exploring clinicians’ attitudes, experiences, and perceived barriers to ACP using a semi-structured interview guide. Transcripts were recorded and transcribed. N-VIVO was used to organise the data, which were subjected to reflexive thematic analysis grounded with an interpretive description framework.

Results

Fifteen mental health clinicians were recruited. Two overarching themes emerged from thematic analysis of focus group transcripts: (1) It is important, and I want to do it: and (2) But I do not do it because of the complexity. Subthemes in relation to this complexity included: (i) fear of harming; (ii) families and culture; (iii) systemic barriers; (iv) capacity and legal issues; (v) timing; (vi) lack of knowledge and training; (vii) neither prioritised nor embedded in practice.

Conclusion

These clinician-identified attitudes, experiences, and barriers to engagement in ACP with older people with psychotic illnesses highlight avenues of potential intervention to facilitate ACP in this cohort. Given the complexity of issues, clinicians need education and training in ACP combined with clear processes and policies to support practice. Clinician insights should be combined with the perspectives of older consumers with psychotic illnesses and their families to inform implementation of ACP.

Introduction

Advance Care Planning (ACP) is an essential part of care towards the end of life. It is an iterative process of discussing a person’s plans for future treatment and the effectiveness of medical interventions in the context of their values, wishes, and life goals.¹ ACP has been shown to have important benefits at the end of life for self-actualisation, quality of life, decreased unwanted medical intervention,² and even, in some settings, a trend towards survival.³ The optimisation of care towards the end of life is an international human right for all,^4,5 including people with serious mental illnesses such as schizophrenia. People with mental illness have poorer access to quality care^6–8 and worse healthcare outcomes^9,10 towards the end of life compared to the general population. These worse outcomes include reduced likelihood of accessing specialist care, including palliative care,^7,11 poorer symptom management, such as pain management,^6,7 and higher rates of cardiopulmonary resuscitation even when death from a life-threatening illness is expected.⁸

Despite the benefits of ACP being recognised since the 1970’s in the general population and in older people, ACP discussions are still rarely pursued with people with mental illness^9,12,13 and even less so with older people with mental illness.¹⁴ Clinicians have a vital role in identifying needs in terms of ACP discussions, initiating discussions, and supporting the collaboration between consumer, carer, and the healthcare team.^15,16 Existing studies exploring clinician perspectives towards ACP have mainly focused on patients with physical life-limiting illnesses such as chronic obstructive pulmonary disease,^17,18 chronic kidney disease,¹⁹ and heart failure.²⁰ Other studies have explored clinician perspectives regarding systems, service roles, and actual provision of end-of-life care for people with mental illness,^21,22 rather than the process of planning for future care.

While middle-aged people with mental illness are interested in ACP and able to consider and express preferences about end-of-life care when given hypothetical scenarios in a questionnaire,^23,24 clinician perspectives regarding ACP with people with mental illness such as schizophrenia are poorly understood. The particular ACP needs of older consumers are largely unknown, despite the obvious relevance to this population. As such, the reasons why this population remains under-served with regards to ACP remain unclear. The literature based on expert opinion would suggest that some of the reasons include clinician assumptions about impaired capacity in the context of cognitive difficulties or psychotic symptoms and concerns about causing distress,²⁵ not considering ACP as part of their role,^21,26 gaps in clinician skills,²⁷ and deficient knowledge about end-of-life law.²⁸ Further possible factors include ageism and mentalism (discrimination based on the presence of mental illness) which may foster nihilism and apathy regarding care of people with mental illness at the end-of-life.^5,29

These are speculative barriers to ACP. Studies specifically exploring mental health clinicians’ perspectives of ACP with older people with schizophrenia/other psychotic illnesses are essential, but currently lacking. There is a need to understand why, after over 50 years of ACP recognised as the gold standard practice for end-of-life care and various initiatives to promote such with people with mental illness,²⁷ there is still reluctance to adopt ACP in this cohort.

The aim of this study was to qualitatively explore mental health clinicians’ attitudes, experiences, and perceived barriers to ACP with people with schizophrenia/other psychotic illnesses.

Materials and Methods

Study Design and Procedures

This qualitative study involved interviews with consumers (people aged 55+ with schizophrenia/other psychotic illness), their nominated carer, and mental health clinicians. This paper reports data from the focus groups with clinicians. Participants were recruited from public mental health services across three local health districts in Sydney, Australia. Mental health clinicians from any discipline (eg social workers, psychiatrists) were eligible to participate in the study if they were fully qualified healthcare professionals (ie, had completed their undergraduate degree and were not students) working with adults aged 55 or more in one of the mental health services of the study hospitals.

Recruitment occurred via e-mail invitations disseminated by team leaders of eligible mental health services and brief presentations in hospital grand rounds. All participants provided consent to participate, in writing or via e-consent, including to publication of their deidentified quotations. Focus groups of up to five clinicians were conducted via videoconferencing with the same facilitator (RB), an external senior clinician. Recruitment continued until data saturation was reached during iterative thematic analysis.

A semi-structured interview guide was used flexibly by the facilitator (RB) and is available on request. Topics discussed in the interview included clinicians’ experience, attitudes, and perceived barriers regarding ACP with consumers with schizophrenia and other psychotic illnesses. Clinicians were also asked about prior training regarding ACP with consumers. Each participant attended one focus group held over one session. There was no time limit set for focus groups. Focus group discussions were audio recorded and transcribed.

Data Analysis

Reflexive thematic analysis, as described by Braun and Clarke,^30–33 grounded within an interpretive description framework was used to analyse transcripts. Interpretive description was chosen as it keeps the generation of clinically- and practice-relevant knowledge central to the analysis. Rather than generating primarily theoretical contributions, this qualitative approach is designed to understand phenomena that yield practical clinical applications and can inform service delivery and improvement. As such, both informed questioning of participants exploring meanings and explanations and subsequent inductive analysis are focused on generating practically useful “descriptions” for informing clinical understanding.^34,35 As a practice-oriented approach developed for the health context, this was most appropriate for framing the thematic analysis and associated outputs to guide needs-oriented implementation of ACP.

Data were uploaded to NVivo 12. A primarily inductive approach to coding themes, acknowledging that meaning arises from the interaction between coder and text. In the first instance, analytic categories and predominantly semantic themes were generated line-by-line, to stay close to the voice and experience of participants. Guided by an interpretive description model, categories were initially created to respond to each specific research question (ie, clinician experiences, attitudes, and barriers to ACP). Semantic codes were coded derived as a means of grouping congruent experiences described in the data, as transcripts were read. AK was the primary coder with higher order coding undertaken by senior author (CP) and lead investigator (AW). Participants had the opportunity to provide feedback on the thematic analysis, with the aim of enhancing the transparency and reciprocity of the results, and leading to further refinement of the analysis.³⁶

As transcripts were read and predominantly semantic codes generated, the researcher reviewed the codes and considered potential hierarchical relationships between codes. These groupings were refined as transcripts were read and re-read. This process of constant comparison between codes and data allowed for themes and sub-themes to be generated. At this stage, the primary coderbegan to also code more latent themes (ie, ideas and conceptualisations underlying the words spoken by the interviewee) which captured the essence of the data and responded clearly to the research questions.

Reflexivity was considered. The researchers acknowledge having an influence over data analysis, which is shaped in part by their relevant personal and working history.^31,33 The primary coder (AK) is a clinical psychologist and qualitative researcher. The interviewer (RB) is a senior occupational therapist and provisional clinical psychologist working with older adults, including assisting with ACP. The project coordinator (YZ) is a Master of Clinical Neuropsychology student. CP is an old age and human rights psychiatrist and longstanding advocate for ACP. AW, is an old age psychiatrist working in a community-based older persons mental health team. The research team also included a specialist in palliative medicine who was involved in incorporating ACP into electronic medical records (MS), a retired nurse consultant in ACP (AM), a carer advocate for ACP with older people (CS), an academic geriatrician with clinical and research experience in ACP (VN), and a senior social worker and team leader of an older people's mental health service (DK). Perhaps, the most subjective influence on coding was the strong advocacy for ACP.

The study was approved by the Human Research Ethics Committee – Concord Repatriation General Hospital of the Sydney Local Health District (2023/ETH02283).

Results

Five focus groups each comprising two to five mental health clinicians (total N = 15) were conducted online (duration 42–50 minutes, mean 45.6 minutes) between January and May 2024. Demographic characteristics of the 15 participants, which included six different healthcare professions, are summarised in Table 1.

Table 1.

Demographic Characteristics of Participants

Characteristics	N (%)
Gender
Male	2 (13%)
Female	13 (87%)
Ethnicitya
Australian	5 (33%)
Australian/UK	2 (13%)
Australian/Italian	1 (6%)
Chinese	2 (13%)
Filipino/Chinese	1 (6%)
South African	1 (6%)
Unavailable	3 (20%)
Religiona
Atheist	1 (6%)
Buddhist	1 (6%)
Roman Catholic	3 (20%)
Protestant Christian	4 (27%)
Jewish Faith	1 (6%)
No religion	2 (13%)
Unavailable	3 (20%)
Mental health clinician discipline
Social worker	5 (33%)
Psychiatrist	4 (27%)
Nurse	3 (20%)
Clinical psychologist	1 (6%)
Clinical neuropsychologist	1 (6%)
Occupational therapist	1 (6%)
Prior training regarding ACP with consumers
Yes	13 (87%)
No	2 (13%)
Type of prior trainingb
Informal bedside teaching in clinical setting	8 (53%)
Formal course (live/interactive)	5 (33%)
Self-directed online learning (not interactive)	4 (27%)
Document/policy to read	4 (27%)

Notes: ^aSome participants did not provide information about ethnicity and religion. ^bSome participants had engaged in more than one form of training.

Thematic Analysis

Two overarching themes emerged, namely: (1) It is important, and I want to do it: and (2) But I do not do it because of the complexity. These two main themes with subthemes for the reasons for complexity are reported below with illustrative quotes.

1. It is important, and I want to do it

Clinicians valued ACP and identified their therapeutic relationship with consumers and expertise in mental health as beneficial.

I’d like to be better at it. I’d like to do it more routinely as part of my usual practices. (psychiatrist)

They were advocates for consumers and their families.

…. some of those patients might have difficulty articulating their wishes and decisions of the advance care planning but I guess knowing them for a period of time and knowing their illness cause and personality we might be able to sort of help them voice that better than maybe a GP who might just have like short reviews and …. do like focused questions around physical health issues…. And they may be more trusting of us. (psychiatrist)

For most people being able to have some autonomy or some ability to make decisions for their health is really important. (clinical psychologist)

[ACP discussion] relieves the burden on family as well because, if the decision has to be made, they know that they’re making a decision in keeping with the person’s wishes. (social worker)

2. But I do not do it because of the complexity:

…the complexity scares me and so I think it’s easier to just avoid it altogether. (psychiatrist)

Several themes emerged in relation to this complexity, elaborating the reasons for the reluctance to engage in ACP including: (i) fear of harming; (ii) families and culture; (iii) systemic barriers; (iv) capacity and legal issues; (v) timing; (vi) lack of knowledge and training, and (vii) neither prioritised nor embedded in practice.

(i) Fear of harming

Clinician anxiety and perceived potential harms for both the consumer and the therapeutic relationship were evident:

….are these conversations actually going to exacerbate their illness? (clinical psychologist)

One fear …. is that it might trigger some sort of delusion and get misinterpreted. Then you’re also somehow presenting, you’re somehow becoming a threat, and that that plays into the paranoid psychosis…. (psychiatrist)

You don’t want to engage a person in a process that could be …. distressing or that requires a level of capacity that they might not have. (psychiatrist)

There were assumptions that consumers do not want to talk about dying or may be offended by ACP discussions:

They automatically think that it’s about planning for them – they don’t want to think about dying. So it is a difficult conversation to bring up sometimes. (nurse)

You don’t want it to be perceived as ageism or, you know, that we’re assuming that they’re going to die soon…. (nurse)

(ii) Families and culture

Clinicians acknowledged the role of consumers’ families in decision-making towards the end of life, and the possibility of conflicting opinions.

You do sometimes need to involve the family because they’re the ones who make the decision in the end as well, and who have to state the decision to people. So sometimes it [ACP conversations] causes conflict…. family members don’t always agree. (social worker)

Sometimes the role of the family was confused:

If the family want to keep it [end of life care decisions] within the family, it’s not really your [the clinician’s] business. (nurse)

Cultural issues conferred an extra level of complexity.

Even to bring up the conversation with some cultures is probably a bit taboo and it’s a bit offensive. (nurse)

Knowing what is realistic [culturally] for them and what is part of their illness and just not writing off some of the things that they’re requesting. (nurse)

(iii) Systemic barriers

While being advocates for ACP, clinicians recognised systemic barriers such as ageism and mentalism:

I could clumsily embark on offering things that the physicians would just say, ‘no, because of your age we wouldn’t even offer that.’ (psychiatrist)

It almost feels like a very futile exercise, because [medical] teams come in say that they’re mentally unwell, they have psychosis… We’re not going to respect those wishes because they probably don’t have capacity, and sometimes family also have that view. (psychiatrist)

(iv) Capacity and legal issues

Both decision-making capacity and legal issues in relation to ACP with older people with a psychotic illness were identified as challenging and confusing for clinicians:

…[I have] anxiety about if people really have the capacity to be making those decisions. I know I certainly have some patients that will never really understand their illness or be well to the point that they really understand their health care. You know people that are in their 70s but they’ll perpetually be telling you that they’re 30 and they’re pregnant. (social worker)

It makes you worry if they do have the capacity to make those decisions. But I guess it’s not a legally binding document …. they can express their wishes on what they want and then it’s still up to the clinician at the time who’s got the care plan in hand. (social worker)

(v) Timing of ACP

Uncertainty regarding the timing of ACP conferred an extra layer of complexity both in relation to capacity and otherwise:

The other thing that scares me is that for someone with a psychotic illness, capacity can fluctuate …. so it’s hard. Do I do it every time I see them? Or do I do it once, and then [determine] that they maintain capacity? (psychiatrist)

I would see it as an awkward and almost irrelevant thing to bring up with somebody who’s just come in for an admission…. that can freak the patient out themselves and they might think that there’s an end-of-life issue occurring rather than, you know, this is just a general discussion. (clinical neuropsychologist)

Because I work mainly in an inpatient setting, or when people are referred because they’re acutely unwell and so they’re often psychotic. It doesn’t seem to be the right time to be trying to do advance care planning. (psychiatrist)

….we’re doing a lot of acute work. So, once a person is like maybe well enough to be making these kinds of big, considered decisions, …. we’re not looking after them anymore. (psychiatrist)

(vi) Lack of knowledge and training (also see Table 1)

Few clinicians had received formal training regarding ACP with people with mental illness and received their education informally on the job, using self-directed reading, team discussions, and by observing the practice of other clinicians (eg geriatric medicine trainees). It was noted that some Local Health Districts have an Advance Care Coordinator who provides education.

..... informal bedside teaching and sort of observing I guess when I was training what my seniors did in terms of advance care planning. (psychiatrist)

I think there is a HETI [online education] module that gets done but that’s pretty – not really the same as engaging personally. (nurse)

If you don’t do it very often, then you’re not very skilled at it. And so, then it comes out as very awkward when you do try to do it and that makes you not really wanna try again. (psychiatrist)

(vii) Neither embedded in practice nor prioritised

Clinicians observed a lack of procedural guidance, with ill-defined roles, responsibilities, and processes, causing abrogation of responsibility to other professionals or other settings:

…whose role is it to have these discussions? So I’m a neuropsychologist, maybe I would verge into the area that I believe is my role such as capacity, but my role is not to…. look at discussions around end of life, ….that would be a medical decision. (clinical neuropsychologist)

A number of consumers are in aged care facilities, so we think that’s [the facility’s] role to do that (nurse)

Is it our responsibility though? I mean, we have so many things that we need to tick off. Yeah, just another thing that we’re adding …we have to do the physical health, we have to do the Advance Care Directive, and our focus is the mental health, really. (social worker)

Clinicians did not see ACP as a priority in their work with older mental health consumers.

I don’t know if there’s a policy. I don’t know what the standard practice is. (occupational therapist)

In our service advance care planning I just don’t think is really on the radar…. particularly with people presenting with psychosis is it’s not the first cab off the rank …. You’re thinking about treating the psychosis. (social worker)

There needs to be a cultural shift. It’s like talking about suicide. Any mental health clinician is now quite good at [talking about suicide], because everyone does it and we’re comfortable doing it. Whereas with something like advance care planning, you know, if no one else in your department does it, then it’s very easy to just go along with it and conspire and not do it. Unless there is some cultural shift, and shift in attitude, and getting used to the conversations and not being afraid of it, then I think it might be quite challenging to do it routinely. (psychiatrist)

It’s not a priority…. when people are discharged from the ward, there’s so many things that need to be followed up…. and then maybe a few months down the track, when it is more appropriate to have that conversation, for whatever reason, it doesn’t even enter my mind to discuss it. (social worker)

Clinicians identified not having time for meaningful in-depth discussions about ACP:

Really giving them the time and the space to explore that [ACP], maybe think out loud…. You can’t just do this in a five-minute conversation. (social worker)

Who’s got the time? …. for some people, it’s not gonna be a decision that they make and then might come back with questions, and they may want more information about something. They wanna change it…. (social worker)

There was also lack of clarity about where and how to document ACP discussions:

I presume there would be formal templates for this stuff? I’ve got no idea where that stuff is stored. I know that on EMR [Electronic Medical Record] there’s something about advance care planning but it just is a tick box about whether a document exists. (social worker)

So if you fill one out with your GP like does that go on something in My Health Records so… you don’t have to cart it around with you to show like a physical copy? … I don’t know. (occupational therapist)

Adding to the general uncertainty, was the lack of guidelines regarding which age group to target for ACP. No age group seemed appropriate:

I don’t think it’s a conversation that you have with older people…. or people who are unwell…I am aware that people with mental illness, schizophrenia, have a much lower life expectancy so you might want to adjust to include that, but I don’t think it’s that relevant with our younger population. (occupational therapist)

Discussion

To the best of our knowledge, this is the first qualitative study examining the perspectives of mental health clinicians regarding advance care planning (ACP) with older people with schizophrenia and other psychotic illnesses. Two overarching themes emerged, namely: ACP is important, and I want to do it; but I do not do it because of complexities conferred by: (i) fear of harming; (ii) families and culture; (iii) systemic barriers; (iv) capacity and legal issues; (v) timing; (vi) lack of knowledge and training; and (vii) neither prioritised nor embedded in practice. Capturing the voices of a diverse range of clinicians was a major strength of the study given the consensus that ACP is a multidisciplinary clinical task that requires multidisciplinary input and effort.³⁷ Interpretation of the data was supported by opportunities for participant feedback on the initial analysis.

Mental health clinicians indicated that they would like to be involved in ACP with their consumers, citing their relationship with older consumers as of particular benefit. The importance of clinicians supporting people with serious mental illness to make end-of-life decisions has been recognised in policy and guidelines.²¹ However, not all mental health clinicians see ACP as their role. Some clinicians in the present study did not consider ACP as “mental health work”, echoed in other studies,^21,26 an important target for intervention.

Despite valuing ACP, mental health clinicians in our study identified many reasons why they are not presently involved. Clinicians expressed fear of harming consumers through ACP, leading to avoidance. In the specific setting of end-of-life care, others have similarly reported clinician fears of causing distress, emotional instability, or exacerbating the underlying mental illness.^21,22 However, the literature suggests that when these discussions have occurred people with serious mental illness (albeit in mid-life rather than older age) have been receptive, able to participate and not adversely affected.^22,24,38 In contrast with assumptions of some clinicians in the present study that people with mental illness do not want to talk about ACP, previous work has revealed that while they may not routinely think about or start discussions about end-of-life care, when engaged on the topic they can express preferences.^24,39

Another identified barrier to ACP was clinician uncertainty navigating capacity assessment in people with psychosis. Difficulties untangling chronic psychotic symptoms from preferences regarding end-of-life care, fluctuation in capacity, and concerns about the legal and health implications of ACP all contributed to clinician anxiety about engaging in ACP and its timing. In related contexts such as end-of-life care, clinicians have similarly expressed concern that symptoms of mental illness may influence understanding, and revealed blanket assumptions of lack of capacity in people with serious mental illness leading to avoiding ACP.^9,22 This is despite common law presumption of capacity for all adults regardless of diagnosis⁴⁰ and policy and guidance to the contrary.²¹

Lack of knowledge, training, and practical skills emerged as wide-ranging barriers to ACP. Inadequate training may lead to feelings of anxiety and clinician avoidance, negatively impacting the quality of care provided to people with mental illness.^22,41,42 This was echoed by clinicians in our study who described their fears about ACP, often stemming from lack of knowledge and skills, resulting in avoidant behaviours. A qualitative study of mental health and palliative care clinicians exploring barriers to end-of-life care provision for people living with serious mental illness similarly identified lack of confidence in areas such as navigating end-of-life care health systems and services and assessing capacity to make end-of-life decisions, leading to avoidance of such conversations.²² The literature suggests educational approaches such as in-services, core education within professional disciplines, and having end of life and mental health clinicians learning from each other.²¹

Clinicians identified changes needed at a systemic level to support their involvement in ACP with older adults with psychotic illnesses. This included having clear and easily accessible means for documentation and data sharing, including leveraging electronic medical records. Although there is a facility for documenting ACP discussions in the medical record system of this Australian jurisdiction, participants appeared unfamiliar with it. A digital tool supporting advance care planning in palliative care, “Coordinate my Care”, has been used by people with severe mental illness (the latter comprising 1.2% of all records).⁴³ Relatively few care plans in this sample (20%) included information about consumer preferences for physical and mental health treatment, instead clinician statements about the patient’s capacity or presentation were documented, and many lacked the patient’s voice.⁴³ The advantages of having up-to-date, accessible, digital data on clinical treatment plans and patient preferences for care were noted, including facilitating information sharing across services.⁴³ The accessibility and transferability of ACP documents across care settings with centralised electronic records has previously been emphasised,² goals increasingly within reach in contemporary health settings.

A related theme which emerged from clinicians was how to incorporate ACP as a routine part of practice. Concerns were raised about ACP not being a core business in mental health, having other priorities, inadequate time to properly explore ACP wishes, and lack of protocols or procedures to support implementation. The general lack of guidance and strategy regarding end-of-life care for people with serious mental illness has been previously highlighted.²² In a non-mental health setting, lack of time and competing clinical priorities were similarly identified as impeding implementation of an approach to improve end-of-life decision-making and treatment in hospitalised older adults (aged 75+) in the Appropriate Care and Treatment (InterACT) study.⁴⁴ The intervention involved notifications to treating clinicians that their patient was at-risk of short-term death, with the aim to trigger a range of ACP-related responses. In this study, informed by nudge theory, site champions (senior staff members) promoted the intervention, supported the participation of clinicians, and provided tailored feedback, alongside electronic notifications and prompts to consider the intervention.⁴⁴ With changes in treating teams over the course of hospitalisation, clinicians identified uncertainty about whose role it was to initiate end-of-life conversations,⁴⁴ a question also raised by mental health clinicians in the present study. Although multipronged in approach, and generally supported by clinicians, the InterACT was evaluated as unsuccessful, presumed to be due to issues with leadership support, cultural context, and implementation.⁴⁴

Potential Solutions

Acknowledging the important role of mental health clinicians facilitating access to physical health care for their consumers,⁴⁵ and as stated earlier, solutions for ACP should be informed by multidisciplinary clinicians, not single disciplines.³⁷ In Table 2, we use the barriers identified by our participants in the thematic analysis of focus group transcripts to craft practical solutions. Many of the barriers to ACP could be addressed through targeted clinician education specific to this population of mental health consumers, with a practical, skills-based component to improve knowledge and confidence such as through demonstration videos, role-play, supervision, and mentorship. Further strategies include educating and partnering with families in the process of ACP with their relative.

Table 2.

Solutions for a Cultural Shift in ACP

Barrier	Solution
Fear of harming	Multi-disciplinary, person-centred risk analysis of timing and appropriateness of ACP with each consumer, including the consumer’s nominated family/carer
Families and culture	Education of and partnership with families to work on ACP with their relative and clinicians over time Develop resources for families: eg Empowered Project (2020). End of Life Video. Capacity Australia https://empoweredproject.org.au/dementia-video/
Systemic barriers	Systemic education and awareness raising regarding ageism and mentalism Accessibility and visibility of ACPs ● Standard recording in the medical record ● Give copies to GPs, residential aged care facilities, disability services (where applicable), and other related care providers
Capacity and legal issues	Education on capacity (see O’Neill and Peisah, 2021) specific to international jurisdiction (for example in Australia; https://end-of-life.qut.edu.au) Clinician and consumer education on supported decision-making: (https://cdpc.sydney.edu.au/research/planning-decision-making-and-risk/supported-decision-making/)
Timing	Multi-disciplinary, person-centred risk analysis of timing
Lack of knowledge and training	Clinician training ACP practice and communication skills (demonstration videos, role-play, and practice opportunities); documentation, End-of-life law and capacity issues; Patient advocacy and human rights, Augmented by mentorship (shadowing experienced clinicians), supervision, reflective practice and debriefing opportunities.
	Family Education about ACP, palliative care, psychotic illnesses. See also Empowered Project (2020). End of Life Video. Capacity Australia. https://empoweredproject.org.au/dementia-video/
Neither prioritised nor embedded in practice.	Systemic shift Support from management/leaders in the mental health service New rules: ACP is part of care Embed ACP within care plans ACP is routinely brought up in ward rounds Iterative discussion with community-based consumers over a number of visits (ie break it up into a series of short discussions over time) Use consultation model involving multidisciplinary team (eg discussion with the psychiatrist if there are questions of capacity/complexity) Appoint champions Audit and address existing document/template use Processes ACPs recorded in the medical record Printed educational materials about ACP in this population bespoke for the target audience (clinicians, consumers, and carers/families) Information on display in wards and take home leaflets Routine and consistent documentation on EMR shared with GP

While a couple of mental health clinicians raised the possibility of referring their consumers to an (external) expert in ACP, the possibility of partnering more broadly with palliative care services and sharing respective knowledge, skills, and pertinent clinical information to overcome barriers to ACP were not raised. Other studies exploring clinician perspectives of end-of-life care in people with serious mental illness have highlighted the need for partnership between mental health and palliative care services.^9,22,38 The two services may learn from each other through formal training and collaboration during clinical encounters to ensure access and support for people with mental illness to achieve optimal outcomes at the end of life; breaking down silos of care.²²

Strengths and Limitations

This study recruited a broad range of multidisciplinary healthcare professionals, reflective of the composition of mental health teams. Given participation in the study was voluntary and participants self-selected,⁴⁶ it is likely that clinicians already interested in ACP took part, as suggested by the majority of participants reporting some form of ACP training. Thus, we may not have captured the full breadth of clinician perspectives, particularly those who do not want to be involved or do not see ACP as part of their role, potentially limiting generalisability. However, the range of responses indicated uncertainty and lack of knowledge was present even in those interested in the topic and those who had received some training. To generate data, focus groups rely on facilitated discussion which may be dependent upon the skill and impartiality of the facilitator.⁴⁷ The strength of the study was having an experienced trained facilitator, herself a clinician, with advanced knowledge and practice with ACP but who did not work with participants. Most of the study investigators had clinical and/or research experience in ACP, which could be considered a potential bias on data analysis and interpretation. However, this was counted by having three coders analyse the data and consideration of reflexivity. The study was conducted in three urban mental health services, so results may not be generalisable to ACP in rural and regional communities where specialist resources may be limited.

Although this study focused on clinicians working with older adult consumers with schizophrenia or other psychotic illnesses specifically, it makes intuitive sense that many of the barriers discussed are also pertinent to people with other serious mental illnesses.³⁸

Conclusion

We have captured clinician voices regarding ACP. This study provides valuable insights that can be used to inform the development of mental health clinician training and practical implementation of ACP and policies to support ACP with older people with psychotic illnesses. The next phase of this study examines the perspectives of older people living with a psychotic illness and their carers, which are crucial to understand, but hitherto largely absent from research in this field.^22,48 These data from all key stakeholders will be triangulated to guide evidence-based implementation of ACP in this population, which should then be evaluated and refined accordingly.

Funding Statement

This work was supported by the Moyira Elizabeth Vine Fund for Research into Schizophrenia Program 2023, administered by the Faculty of Medicine and Health, University of Sydney, Australia.

Ethics Statement

Ethical approval was granted by the Human Research Ethics Committee- Concord Repatriation General Hospital of the Sydney Local Health District (2023/ETH02283).

Disclosure

The authors report no conflicts of interest in this work.