Generating key practice points that enable optimal palliative care in acute hospitals: Results from the OPAL project's mid-point meta-inference

Abstract

Background

Internationally, the epidemiology of dying is changing with people dying at an older age from an expected death and with complex health care needs. An increasing proportion of people require hospital care with the need to strengthen the quality of this care well-articulated. Evidence about what enables optimal inpatient palliative care is available. Articulating how to enable this within clinical practice is required.

Objective

To investigate the domains of care that are most important to inpatients with palliative care needs and their families, to generate key practice points to inform optimal clinical care provision.

Design

A mid-point meta-inference of the Opal Project's data. The three phase Opal Project utilised a fully mixed sequential dominant design (Quan → QUAL). Phase 1 focused on ‘scoping the problem’ through a systematic review and meta-synthesis of important aspects of care for inpatients with palliative care needs and their families (studies 1a and 1b); Phase 2 focused on ‘understanding importance’ through a qualitative interview study (study 2); and a mid-point meta-inference of data obtained across Studies 1a, 1b and 2. Phase 3 included a global environmental scan (Study 3) and co-design workshop (Study 4) focused on understanding how to drive reform for Australian inpatient palliative care, based on outcomes from the mid-point meta-inference; and an end-point meta-inference to generate final recommendations.

Methods

Mid-point meta-inference of data obtained across Phases 1 and 2 involving: 1) verifying synthesis of data with palliative care consumers and clinical leaders; and 2) populating joint display tables to inform analysis and generate practice points.

Results

Three categories and 14 domains informing optimal inpatient palliative care were identified: 1) Person-centred care including respectful and compassionate care; effective communication and shared decision making; effective teamwork; enabling family involvement; and maintaining role, meaning and identity; 2) Expert care including excellence in physical care; impeccable assessment and care planning; effective symptom management; technical competence; patient safety; and supported access to senior clinicians; and 3) Optimal environment for care including patient and family focused structural factors; and cleanliness to support infection control. Data integration generated 68 practice points informing care provision.

Conclusions

Through a synthesis of patient and family perspectives about what is important for optimal inpatient palliative care, this study confirmed three categories of care, 14 domains of importance and 68 practice points. Importantly, these practice points guide clinical practice to enable each domain of care in practice.

Tweetable abstract

We know what patients with palliative care needs, and their families need for good care when they are in hospital. It is time to deliver care in line with these needs.

• •There are many people with palliative care needs receiving care within the hospital setting and not all receive optimal evidence-based care;
• •International research has documented what inpatients perceive to be most important in enabling optimal palliative care in the hospital setting;
• •No studies to date have synthesized all available data and presented these in actionable practice points to inform clinical care delivery.
• •The synthesis of perspectives from a large number of inpatients with palliative care needs (n = 1233) and their families (n = 3818) living in various high income countries about what enables optimal palliative care within the acute hospital setting;
• •The identification of what clinicians and ancillary workers can do in practice to contribute to optimal inpatient palliative care, from the patient and family perspective;
• •A robust foundation articulating patient and family identified need upon which future research, policy and practice change can build to see improvements in inpatient palliative care provision that are consumer-centerd.

Contribution of the paper.

What is already known about the topic

What this paper adds

Alt-text: Unlabelled box

1. Introduction

Globally, the epidemiology of dying is changing (Cherny et al., 2015; World Health Organization 2018). Whereas once people predominantly died suddenly, people living in high income countries now die at an older age from an expected death (Cherny et al., 2015). Living with advancing age and non-communicable disease leads to complex health care needs and mean that an increasing proportion of people require care in the hospital setting, even if most would prefer to be cared for at home for as long as possible. As a result, internationally a large proportion of hospital inpatients have palliative care needs (Pivodic et al., 2016).

Worldwide there have been calls to strengthen inpatient palliative care (Nevin et al., 2020; Heckel et al., 2020). The reasons for this are varied and complex (Nevin et al., 2020), but largely relate to: ineffective communication, too little input into decision making and poor symptom management (Coimín et al., 2019; Hillman, 2011; Garner et al., 2013). In part this may be related to the dominance of the biomedical model, with its focus on cure (Heckel et al., 2020; Garner et al., 2013; World Health Organization 2012) which has contributed to dying being viewed by clinicians as a ‘failure’ of care rather than as an opportunity to excel in a different kind of care (Teno et al., 2001). A focus on cure (Garner et al., 2013; World Health Organization 2012) serves the needs of people who require active treatment, but may make it harder for clinicians to move from a problem-solution orientated approach to a palliative approach to care that is person-centred and grounded in comfort and dignity (Hillman, 2011; Teno et al., 2001; The Lien Foundation 2010). Viewing death as a failure may inadvertently inhibit honest communication between clinicians, inpatients and/or families. Systemic issues relating to a lack of skilled clinicians, uncertainty in clinician roles, the complexity of multi-level organisations, and a lack of safety and quality indicators to measure performance and inform change are also noted barriers to the provision of quality inpatient palliative care (Nevin et al., 2020; Hillman, 2011; The Lien Foundation 2010; Shipman et al., 2008).

In order to drive improvements in hospital palliative care, understanding what patients with palliative care needs and their families identify to be important for optimal care is crucial. Significant data about what is important is available and has been for over 30 years (C Virdun et al., 2020; Virdun et al., 2015; Virdun et al., 2016; Robinson et al., 2018; Gott et al., 2019; Black et al., 2018). Domains of importance for optimal hospital-based palliative care, from the perspectives of inpatients and carers include: expert care (inclusive of physical care, symptom management and integrated care), optimal communication, respectful and compassionate care, valued family involvement in care planning and delivery, maintenance of self-identity for patients, environmental privacy for families, ensuring patient safety, supporting patient choices, preparing families for death and providing contact for families after a patient has died (Virdun et al., 2015; Virdun et al., 2016). How to enable this type of care in practice remains elusive (Virdun et al., 2016; Virdun et al., 2020; Virdun et al., 2018).

While there are many examples of excellent palliative care within the hospital setting (Coimín et al., 2019; Gott et al., 2019; Robinson et al., 2015) delivering this level of excellence across all wards irrespective of location or population continues to be challenging. A recent qualitative study led by researchers in New Zealand identified examples of good care in the hospital setting in the last 3 months of life from the perspectives of bereaved relatives (Gott et al., 2019). This study concluded that adopting a bicultural relationship-centred compassionate care model to inform care provision, will enable improved practice (Gott et al., 2019). A quantitative descriptive post-bereavement study from Ireland has provided an example of how to use data from bereaved families to inform improvement efforts within a hospital environment, complemented by structured committees, with noted executive and quality improvement professional involvement (Coimín et al., 2019). A Netherlands led controlled before-and-after study examined the impact of ward based nurse champions on palliative care provision in the last 3 days of life and found no difference in quality of care experienced from the perspectives of bereaved relatives (Witkamp et al., 2016). Similarly, the implementation of a systematic approach for identification of patients deteriorating with uncertain recovery to support clinicians in their palliative care provision, the AMBER care bundle (Guy's and St Thomas' NHS Foundation Trust 2014), has shown mixed results when evaluated through a retrospective review of case notes, especially in relation to: improving identification of patients with palliative care needs (Etkind et al., 2015); improving care experience, particularly in relation to effective communication (Bristowe et al., 2015); and the complexities inherent in standardised package use across such varied clinical contexts where some wards may engage with this work regularly and others infrequently (Bristowe et al., 2018). Given the complexity both in relation to the care of those with palliative care needs and the different care settings within a hospital (ie. intensive, critical and high-dependency care units, ward and emergency departments), it could be argued that a standardised approach to care provision cannot succeed. Rather, tailoring to the local context will be important, albeit underpinned by evidence-based resources, tools and quality processes. Enabling the patient and family voice to remain central in informing all such improvement work is crucial to effect change that aligns with identified need (Virdun et al., 2020; Virdun et al., 2015; Virdun et al., 2016).

1.1. Rationale for the opal project

The Optimising care for People with palliative care needs, and their families, in the Australian hospitaL setting (the OPAL Project) refers to a doctoral program of research designed to improve understanding of how to enable optimal palliative care across all hospitals in Australia. This project focused on three key factors: (1) understanding the most important domains of care for inpatients with palliative care needs and their families; (2) identifying whether a suite of indicators to assist measurement of inpatient palliative care quality was available and aligned with consumer identified need; and (3) outlining key drivers for enabling improvement in palliative care provision within Australian hospitals. The complete OPAL project is published elsewhere as a doctoral thesis (Virdun, 2021). This manuscript presents the mid-point meta-inference of data informing what is most important for optimal care and what clinical practice factors contribute to the best possible experience of inpatient palliative care, from the perspectives of patients with palliative care needs, their families/carers (current and bereaved) and consumer representatives advocating for ideal palliative care provision.

1.1.1. Aim

To investigate the domains of care that are most important to inpatients with palliative care needs and their families, to generate key practice points to inform optimal clinical care provision.

2. Methods

2.1. Design and theoretical framework

A mid-point meta-inference within the three phase OPAL Project. In the fully mixed sequential dominant design Opal Project (Leech and Onwuegbuzie, 2009), each study was designed to prioritise the perspectives of inpatients with palliative care needs, and their families to enable person-centred care in practice for people with complex and chronic health care conditions. Informed by The Person-centred Practice Framework (2019 revision (McCormack and McCance, 2019)), the following key domains were considered within Project design: pre-requisites for person-centred practice; the care environment; person-centred processes; person centred outcomes (McCormack and McCance, 2016); and the macro context framing all such considerations (McCormack and McCance, 2019). Being sequential, each Phase of the OPAL Project informed the next with Phase 1 focused on ‘Scoping the problem’ and included a systematic review of published data describing what inpatients and their families perceive they require for optimal care (Studies 1a and b, (Virdun et al., 2015; Virdun et al., 2016)) Phase 2 focused on ‘Understanding importance’ and involved qualitative interviews with inpatients and their families to better understand domains of importance, implications for practice and relevance for the Australian population (Study 2) (Virdun et al., 2020). A mid-point meta-inference of the integrated data from Studies 1a, 1b and 2 was completed at this point to inform the direction of Phase 3 and shape Project conclusions. Phase 3 focused on ‘Driving reform’ through a global environmental scan (Study 3) (Virdun et al., 2018) and a co-design workshop (Study 4) (Virdun et al., 2020) with these studies designed to understand how to drive improvement that addressed the inpatient and family perspectives of what constitutes optimal inpatient palliative care. At the completion of the OPAL Project, an end-point meta-inference integrated data from all four studies to generate overall recommendations for system level reform founded upon that which matters most to people requiring this care (Virdun, 2021) Figure 1.

Fig. 1.

A mixed methods study: fully mixed sequential dominant design (Quan → QUAL) (Leech and Onwuegbuzie, 2009).

This paper reports the mid-point meta-inference of data obtained across Studies 1a, 1b and 2 to inform understanding of the domains of care that are most important to inpatients with palliative care needs and their families and how these can be achieved within clinical practice. This mid-point meta-inference involved: (1) verifying data synthesis with palliative care consumers and clinical leaders; and (2) populating joint display tables to inform the analysis (Creswell and Creswell, 2018; Guetterman et al., 2015).

2.1.1. Participants-verification of data synthesis

Palliative care consumers or an identified academic leader working in the field of palliative care research assisted verification of data synthesis. This study used the term ‘palliative care consumer’ to describe a person with palliative care experience (patient or family) supporting organisations in their work related to palliative care. This project worked with consumers from either the: University of Technology's Improving Palliative, Aged and Chronic Care through Clinical Research and Translation (IMPACCT) Centre's Consumer Advisory Group: https://www.uts.edu.au/impacct/consumers; or NSW Translational Cancer Research Network's Consumer Advisory Panel: http://www.tcrn.unsw.edu.au/groups/tcrn-consumer-advisory-panel.

2.1.2. Setting-verification of data synthesis

One workshop (3 h) was held with palliative care consumer representatives at The University of Technology Sydney, Australia. In addition to this, a face-to-face individual meeting (3 h) with an international academic leader was held.

2.1.3. Recruitment-verification of data synthesis

Palliative care consumers were sent information about the data verification meeting by email and invited to attend. Identified key palliative care academic leaders were emailed and asked whether they would be available for a meeting. Response and attendance were voluntary with attendance noted as consent to participation.

2.1.4. Workshop-verification of data synthesis

A modified nominal group process (Wainwright et al., 2014; de Wolf-Linder et al., 2019) was used. A nominal group process is a highly structured method for generating ideas and solutions within groups, with the aim of generating recommendations for best practice (Wainwright et al., 2014; de Wolf-Linder et al., 2019). This process is comprised of a number of steps that can generate both quantitative and qualitative data (Wainwright et al., 2014). This study's modified group process followed the following steps: a detailed introductory statement by the research team framed the context for the research; participants were asked to generate ideas individually; group discussion about generated ideas; group clarification of issues; generation of key priority areas for further analysis; and workshop conclusion (Wainwright et al., 2014). Participants were asked to work independently before coming together as a group to provide consensus opinion in relation to the workshop objective (verification of data representation). Each participant was provided with data outlining key elements of optimal inpatient palliative care informed by Studies 1a, 1b and 2. Firstly, participants were asked to independently categorize the key elements of importance for optimal care, into provided domains. Where they felt a point did not fit, this was placed into a domain titled ‘other’. All individual responses were collected for analysis. Secondly, the group were brought together for an overall discussion about the data and its impact on clinical practice to inform the development of key practice points. Field notes of this discussion were taken by two members of the research team to inform analysis.

2.1.5. Data analysis

Data analysis built on each prior study (1a, 1b and 2) through an integrated analysis (Bradley et al., 2007) approach. This used the domains from the initial quantitative data set (Study 1a) to inform the coding structures for subsequent qualitative Studies (Studies 1b and 2). To build on this, three key processes were used (Green et al., 2007):

• 1Verification of the concepts described within the domains by palliative care consumers (workshop). Where three or more consumers agreed with the coding within one specific domain, this was felt to be verified. Where verification was not found, the research team further reviewed and discussed this key point and/or domain to identify how increased clarity could be obtained and how to best describe this in terms of clinical practice delivery.
• 2Careful analysis of workshop outcomes (verification work and field notes review) and the relationships within and across domains and how these can inform clinical practice (research team and key research leader).
• 3Meta-inference of all data using joint display tables.

2.1.6. Ethics approval

Ethical approval was granted by St Vincent's Hospital Sydney, Human Research Ethics Committee Ref. No. 2019/ETH03307 in October 2019.

2.1.7. Positioning of the researchers

The first author (CV) is a registered nurse with long-term clinical experience in inpatient specialist palliative care provision. The supervision team for this research program are experienced researchers with nursing (JP and PD), medical (KL) and allied health (TL) expertise in palliative care research.

3. Results

3.1. Part 1-Workshop: verifying the data synthesis with palliative care consumers

The consumer representatives (n = 6) who participated in the three-hour November 2019 workshop were provided with a summary of domains of importance (n = 12) generated from the Study 1a, 1b and 2 patient data. Family data were not included at this initial stage, to prioritise the patient voice. The summary of the patient data included 12 patient-identified domains, which contained a total of 131 key points considered essential for enabling optimal inpatient palliative care (Appendix 1). Consumer representatives were asked to work individually with the 131 key points and allocate each one into the domain they felt they best aligned with, to verify the mapping completed by the research team. Results from this individual mapping revealed strong group consensus (100%) regarding four domains, majority group consensus (> 50%) regarding 8 domains and minority group consensus (< 50%) regarding four of the 12 domains (Table 1).

Table 1.

Consumer agreement with coding to domains completed by research team.

Inpatient identified domains of importance (N = 12)	Number of key points in each domain	Degree of consumer agreement* on the key points (N = 131)	Suggested alternative domains
Addressing nutritional needs	5	5/5–100%	Not applicable
Comfort provided by easy access to medical and nursing specialists	2	2/2–100%	Not applicable
Family involvement in care provision	4	4/4–100%	Not applicable
Minimising burden	1	1/1–100%	Not applicable
Adequate environment for care	13	12/13–92%	Maintenance of patient safety
Maintenance of patient safety	5	4/5–80%	Not applicable
Maintenance of sense of self	17	12/17–71%	Expert care
Respectful and compassionate care	18	11/18–61%	Effective communication and shared decision making; and Maintenance of sense of self / self-identity
Effective communication and shared decision making	22	11/22–50%	Expert care; and Respectful and compassionate care
Expert care (good physical care, symptom management, integrated care and other)	35	17/35–49%	Effective communication and shared decision making; Maintenance of sense of self / self-identity; Respectful and compassionate care; and Trust and confidence in clinicians
Financial affairs	4	1/4–25%	Family involvement in care provision; other - not coded consistently
Trust and confidence in clinicians	5	1/5–20%	Effective communication and shared decision making; and Respectful and compassionate care

^⁎Consumer agreement with mapping completed by the research team

The key points (n = 131) articulating what patients stated enabled optimal care within each domain (n = 12) were perceived to be comprehensive and a good overview of areas of importance. The degree to which participants verified the classification of key points into domains was variable, suggesting that further synthesis was needed to enhance clarity and reduce duplication. Group discussion revealed that the taxonomy of domains (n = 12) was perceived to be unclear, with significant overlap in concepts, especially in relation to what key components of care informed effective communication, expert care, trust and confidence in clinicians, and what was understood to inform the financial affairs domain. Participants made a distinction between Person-centred care (encapsulating for example communication, compassion, respect) and Expert task-orientated clinical care, and felt this distinction was not reflected in the current domains. They recommended that Person-centred and Expert care be two separate categories to reflect the type of care that is required for optimal inpatient palliative care provision. Based on this advice, the 12 domains and 131 key points were reviewed before undergoing further synthesis and consideration of the viability of adopting these recommended changes.

3.1.1. Post workshop data synthesis and consensus: verifying the data with palliative care research leaders

The consumer representative workshop outcomes reported above guided the next stage of the analysis. The suggested changes along with the 12 patient reported domains were initially reconsidered by the research team before being discussed with an international palliative care and research academic expert with measurement expertise. During this three-hour meeting, held in November 2019, the team considered the domains of importance and their degree of alignment within the suggested new overarching categories of ‘Person-centred care’ and ‘Expert care’. The team also considered the synthesis and renaming of the domains to enhance clarity of meaning. Whilst consensus with the consumer-recommended changes was secured, the research team identified that the integrated data also highlighted the need to consider the environmental needs of inpatients with palliative care needs. At the end of this process were two outcomes: 1) three overarching categories were agreed to, namely: 1) Person-centred care; 2) Expert care; and 3) Optimal environment for care; and 2) synthesis and renaming of inpatient-described domains of importance for optimal palliative care, with a final set of 14 domains noted (ReferFig. 2).

Fig. 2.

Categories (n = 3) and domains (n = 14) of importance for optimal inpatient palliative care.

Having confirmed these three categories and their 14 related domains, the next stage of the mid-point meta-inference sought to: integrate all data inclusive of inpatient and family participants; and develop key practice points for potential use by hospital managers and clinicians to strengthen inpatient palliative care.

3.2. Part 2-mid-point meta-inference

The first step of the mid-point meta-inference was to integrate all data inclusive of both patient and family-identified domains of importance from the systematic review and meta-synthesis (Studies 1a and 1b) and the qualitative study (Study 2), in a series of joint display tables (Appendix 2). Collectively, these data reflect the perceptions and desire for change over the past 30 years from a large number of inpatients (n = 1233) and their families (n = 3818) living in various high income countries. (Virdun et al., 2020; Virdun et al., 2015; Virdun et al., 2016) Integrating these data facilitated identification of 15 domains of importance, with an additional three domains evident from family data alone (Table 2).

Table 2.

Domains of importance for optimal inpatient palliative care derived from meta-inference of data from Studies 1a, 1b and 2.

Domains of importance (n = 15)	Identified within patient data (n = 12)	Identified within family data (n = 13)
Effective communication and shared decision making	✓	✓
Expert care	✓	✓
Respectful and compassionate care	✓	✓
Trust and confidence in clinicians	✓	✓
Family involvement in care provision	✓	✓
Adequate environment for care	✓	✓
Financial affairs	✓	✓
Maintenance of sense of self / self-identity*	✓	X
Minimising burden*	✓	X
Maintenance of patient safety and prevention of harm;	✓	✓
Preparation for death*	X	✓
Duty of care extending to the family after patient death*	X	✓
Enabling patient choice at the end of life*	X	✓
Nutrition	✓	✓
Access to senior clinicians	✓	✓

^⁎Domain unique to either patient or family data only;.

The mid-point meta inference confirmed that:

• •Person-centred care; Expert care; and Optimal environment for care (identified during the consumer representative and research leader verification process) were relevant and appropriate categories; and
• •The newly described domains (n = 14) reflected all patient and family data, with data alignment fitting well within each of the three categories.

This process also facilitated identification of key practice points (n = 68) as illustrated in the joint display tables (Appendix 2) and described in the next section.

3.2.1. Person-centred care

The five domains identified as being important within the category of Person-centred care are: (1) respectful and compassionate care; (2) effective communication and shared decision making; (3) effective teamwork; (4) enabling family involvement; and 5) maintaining role, meaning and identity. The type of care provision that will enhance the person-centred care experience for inpatients and their families in each of these identified domains is described in the next section, along with a series of key practice points (Table 3) that are considered as being important to guide clinical practice, policy and research in each domain.

Table 3.

Key practice points (n = 68) aligned to each noted domain of importance.

Category	Domain	Key practice points
Person-centred care	Respectful and compassionate care	1. Clinicians who provide care in a compassionate, empathetic, personalised, kind, friendly, supportive and willing way, who are professional and appear happy to be at work 2. Clinicians who are honest, trustworthy, knowledgeable, able to anticipate needs, and confident, efficient, attentive, responsive and mindful in their care delivery 3. Clinicians who connected to their particular situation and circumstances, were unhurried, non-judgmental and acknowledged them as a unique human being 4. Preservation of dignity described in relation to physical care, consideration of quality of life, and through the tone of communication used (particularly noting the need for use of names and not terms such as ‘darling’, and not talking down to the patient or family member) 5. Use of humour throughout an admission to assist a patient to cope and have a laugh 6. Feeling welcomed and deserving of a hospital admission, and where they may have experienced multiple admissions, to be acknowledged by clinicians and ancillary staff, assisting a feeling of homeliness 7. Kind and empathetic kitchen staff and diet aides who return to talk with patients about dietary choices if they are not available on their first visit (asleep, at a procedure, in the bathroom etc.) 8. Provision of good food at the right temperature, in line with preferences, that aides in wellbeing and a sense of comfort (E.g. a hot cup of tea or coffee in the morning) as well as timely assistance with accessing food items on the tray when unable to manage independently 9. Gentleness and kindness from ancillary staff, particularly noting the impact of cleaners, catering staff and those working in hospital cafes 10. Checking in with family members when their loved one is dying (prognosis of days only), even when there are no clinical tasks to be completed, to show kindness, support and care and to understand if they have any concerns at this time* 11. Talking with patients when unconscious and after death, in relation to their care provision*
	Effective communication and shared decision making	1. Communication that is honest, clear, uses layman's language, demonstrates understanding of the current situation and is delivered in a compassionate, connected and mindful way (including sitting at eye level, near to the person, and managing hearing impairments effectively) 2. Communication that enables a shared understanding and feels unpressured by time 3. Communication that works with a patient's/ family's own knowledge and expertise about their condition 4. Listening to the patient and family member and using this information to guide clinical assessments and care planning, especially in relation to symptoms, being mindful of the fact patients and family members might want to meet separately with clinicians at times 5. Communication that maintains positivity and hope within the context of their illness 6. Communication that focuses on the bigger picture in relation to their care, not just the immediate day-to-day issues 7. Accurate documentation of communications held with patients, families and across teams and departments 8. Difficult conversations should be led by a senior clinician and held directly with patients, as appropriate 9. Acknowledgment and support for the requirements within cross-cultural communication to prevent misunderstanding and negativity across patient, family and clinician groups 10. Regular updates throughout an admission about the patient's current condition and plan of care 11. Consistent messages across teams and departments, with a lead clinician to pull complex information together and a key contact who can address any noted concerns 12. Clinicians that explain what is likely to happen with the patient's physical condition over time, to enable planning 13. When the patient is close to death, clinicians being explicit about the fact that the patient may die within the next few days/a week, to enable important conversations and preparations as much as possible* 14. When the patient is close to death, clinicians checking with family members about whether they are wanting explicit information about the physical changes that occur as a patient nears death (e.g., breathing and circulatory changes) * 15. Being engaged in care planning and supported in making decisions, in line with personal needs to be able to live as well as possible 16. Discussion of patient choices with family members supported informally or more formally through advance care planning processes, advance care directives, and nomination of preferred decision makers
	Effective teamwork	1. To see clinicians working together in relation to the patient's care, both within the ward environment and across different teams and/or departments, and also with generalist care providers or lead clinicians from other hospitals 2. To ensure complex care needs are accurately handed over in depth between shifts to enable good care and prevent continual repeating of needs by patients, and where possible to roster the same nurse to a patient to support continuity in care 3. To enable access to multidisciplinary expertise to help patients be as well as they can be, with specific noting of the need for increased access to physiotherapy support, and a key clinician to support inpatients with dementia 4. Efficient discharge planning and management with specific noting of timely provision of paperwork, accurate medication management, with this information also provided for the GP, physical supports such as a wheelchair and/or wardsperson to assist with getting to the car, and integration with generalist care supports as required 5. Effective integration with specialist palliative care services in relation to when they are consulted, how their service is described to families (to enable understanding of their expertise and role) and how their advice is integrated into care planning
	Enabling family involvement	1. Family inclusion to advocate for optimal care and help patients understand complex information provision, given they are so unwell and their cognition can be affected by illness and/or various medications 2. Family inclusion to provide comfort, emotional healing and ongoing sharing as part of a long-term partnership 3. Supportive hospital processes to enable family members to fully participate, including:  • understanding the timing of medical ward rounds (to prevent having to sit all day waiting)  • supporting families to be part of the team, respecting their knowledge of their loved one's care needs (enhancing patient safety) and inviting them to participate in care if they are wanting to  • enabling access for visits at any time  • comfort when staying overnight (to enable rest) and consideration of the family member's wellbeing 4. A clear process for communicating with a senior member of the medical team, in charge of the care plan 5. Opportunity for a family conference to guide care planning at complex points of care 6. Explicit information and guidance on admission about:  • how to best manage parking costs  • availability of Wi-Fi  • supported accommodation options for those who are from rural locations  • access to carer's payments  • access to subsidies for clinical equipment required for transition home (E.g. Incontinence pads) 7. Supportive care and processes at the time of an inpatient's death (respectful care of the body; not feeling rushed; timely and accurate death certification completion) * 8. Supportive information provision (verbal and written) about processes of care after death (death certification, removal of body) * 9. A follow-up contact (call or email), conducted in a compassionate and kind manner, by a clinician connected to their loved one's care to check in with the family and provide information about options for counselling or other supports (with routine bereavement letters from a separate service noted to be less supportive and feeling somewhat tokenistic)*
	Maintaining role, meaning and identity	1. To be supported and encouraged to maintain independence and live as well as possible, within the context of life limiting illness, minimising feelings of being powerless and burdensome# 2. To isolate their illness from life and engage in meaningful activity on a day-to-day basis to assist in wellbeing, even whilst in hospital, suggesting the following would assist in this: access to reading materials (papers/magazines) or puzzles/games, access to information about what is available across the hospital and encouragement for inpatients to get out of bed and engage in an activity# 3. To feel a sense of control and avoid becoming institutionalised noting the importance of access to technology to support social connections, beauty treatments for those who require it (waxing and hair for example) # and the need to maintain a focus on work where able
Expert care	Excellence in physical care	1. Excellence in physical care that is responsive and provided in a compassionate, empathetic and willing way 2. Attention to the inpatient's nutrition to maintain weight throughout an admission and assist in management of chronic conditions (E.g. diabetes) and overall recovery/wellbeing
	Impeccable assessment and care planning	1. That the clinician is knowledgeable about their specific health care needs within the context of their broader health issues, enabling them to accurately assess the current situation and related care requirements to achieve a good clinical outcome 2. That the clinical team pays attention to a person's holistic health care needs, inclusive of their physical, social, emotional and spiritual wellbeing 3. Identification of when a patient is in their last days of life and ensuring their comfort *
	Effective symptom management	1. Regular assessment and rapid management (pharmacological and non-pharmacological) of any symptoms causing discomfort or distress, especially in relation to pain and breathlessness 2. Responsive attention to any forms of distress identified for a patient who is imminently dying, with medication orders available for nurses to act immediately (not have to wait for medical review) *
	Technical competence	1. That their clinician is an expert in procedures such as cannulation (insertion and care) or management of a pleural effusion 2. Ability to discuss and appropriately deactivate an implanted cardiac device in a timely way *
	Patient safety	1. To feel their care is safe and well-managed without them needing to continually have oversight of the care given the patient is so unwell 2. Accuracy in medication prescription and administration, with particular noting of translation of complex medication regimes in the home setting to the hospital setting (and vice versa), accuracy in timing of administration and managing an error in the computer system once entered
	Supported access to senior clinicians	1. Access to senior medical and nursing clinicians with high levels of expertise (email / mobile phone numbers) enabling prompt attention to any noted concerns, answers to questions, time for support and planning, comfort in relation to ongoing support on discharge home and continuity of care
Optimal environment	Structural factors - patient	1. That the bed area is restful to enable sleep and recovery, including the need for quiet, appropriate lighting and temperature# 2. Access to a window to enable a connection to the day/night cycle, to feel connected to the world, to feel warmth from sunlight and to prevent feeling claustrophobic 3. Prevention of frequent bed changes, as patients try to develop a sense of belonging within their space and feel disorientated when moved regularly (the loss of personal items when moved was also noted) # 4. To consider comfort when waiting in the emergency department, through the provision of a bed to wait in (rather than a chair only), given they are so unwell 5. To consider practical supports such as emergency access to toiletries for unplanned admissions, and more practical audio-visual controls for the TV (not to be handheld if possible) # 6. Consideration of supports for people with breathlessness, including the provision of fans, cooling air and adequate ventilation in bathrooms specifically to assist with showering# 7. To consider the implications of a shared room, with positives noted in relation to the company and helpfulness of fellow patients but challenges noted in relation to being disturbed by noise, having to tolerate different smells/odours, feeling quite confronted by other people's illnesses, discomfort with mixed gender wards and lack of privacy for both care and important conversations
	Structural factors - family	1. Access to seating and preferably a place to lie at night, alongside the patient's bed, as well as simple kitchen items (water, ice, microwave) 2. A dedicated space or room to enable private conversations to be held, a break from caring or for quality family time away from the bed-space (inclusive of the patient) 3. Access to affordable meals onsite (a suggestion of subsidised hospital cafeteria access for this population) 4. A private room for someone who is very close to dying to enable privacy for meaningful conversations and space for families to be present *
	Cleanliness to support infection control	1. Cleanliness, specifically in relation to infection control and managing the unique requirements of those who are immunosuppressed, in a timely and efficient way

Normal font = data from patients and families.

^#= data from patients only; italics = data from family members only (current and bereaved); * = data from bereaved family members and specific to the care in the last days of life and at the time of death.

3.2.1.1. Respectful and compassionate care

Respectful and compassionate care for inpatients resonates with many areas noted as important for the non-palliative inpatient population, including the need for compassion, empathy, kindness, personalised care, clinicians being happy and willing, as well as clinicians who are knowledgeable, efficient and connected to the unique needs of each inpatient. (Håkansson et al., 2019) Additional requirements in relation to respectful and compassionate care for those with palliative care needs relate to care provision when someone is imminently dying, noting the importance of clinicians checking in with family members, even when there is no clinical task to be completed, and talking with patients about care provision even when unconscious or after death. Furthermore, consideration of the comforts provided from access to food and beverages is important, with kind and empathetic kitchen staff enabling the provision of good food, at the right temperature and, in line with preferences, supportive.

3.2.1.2. Effective communication and shared decision making

Effective communication and shared decision making for inpatients also reflects the expressed needs of the non-palliative inpatient population, (Håkansson et al., 2019) noting the need for honest, mindful, informed and clear communication that enables a shared understanding and feels unpressured by time. However, those with palliative care needs also require clinicians to acknowledge and work with their expertise in relation to managing their own condition and they have also stated that it is important for difficult conversations to be led by a senior clinician. Furthermore, additional requirements when a person is imminently dying are noted including, the need for: clinicians to be explicit about the fact the patient may die within the next few days/a week, to enable important conversations and preparations as much as possible; and for clinicians to ask family members whether they are wanting explicit information about the physical changes that occur as a patient nears death (E.g. breathing and circulatory changes). The need for clinicians to support patient choices either through informal discussions or formal advance care planning processes is also noted as important.

3.2.1.3. Effective teamwork

Effective teamwork evidenced through care integration within teams and across departments resonated again with expressed need from the non-palliative inpatient population. (Håkansson et al., 2019; Australian Commission on Safety and Quality in Health Care 2018) The additional area described by families of those with palliative care needs was that of a greater need for effective integration with specialist palliative care services in relation to timing of consultation, how this service is described to family members and then how their advice is integrated into care planning.

3.2.1.4. Enabling family involvement

Enabling family involvement is valued to ensure inpatients’ needs are understood, to advocate for optimal care and promote emotional wellbeing for inpatients with chronic and complex healthcare needs. (Mackie et al., 2019; Kelley et al., 2019) Patients specifically articulate the importance of family involvement to assist understanding of complex information whilst so unwell and potentially with affected cognition related to illness or medication provision. Families of people with palliative care needs also specifically noted wanting to connect with clinicians in a way that respects their knowledge, enabling them to feel part of the care team. They value a clear process for communicating with a senior member of the team able to make an impact on care planning and provision, as well as opportunities for family conferences to guide planning at complex points of care. Supportive care and processes at the time of an inpatient's death, such as respectful care of the body, adequate information provision, not being rushed from the ward and timely and accurate death certification are appreciated. Finally, a follow-up contact (call or email) by a clinician connected to their loved one's care to check in and provide information about options for counselling or other supports is valued.

3.2.1.5. Maintaining role, meaning and identity

Maintaining role, meaning and identity is articulated as important by inpatients. Across multiple studies and for over two decades, patients have noted this as important for the hospital setting, but families have not mentioned this at all - the only domain of importance where this is the case. Inpatients want to be supported to live as well as they can, within the context of their illness, to isolate their illness as much as possible and to engage in meaningful activities on a day-to-day basis. They also emphasise their need to retain a sense of control and to avoid becoming institutionalized whilst an inpatient. The need to connect with each patient and value their individuality is well noted in policy and research across healthcare; (Håkansson et al., 2019; Picker Institute Europe 2019; National Safety and Quality Health Service Standards (Version 2)) however, this need is accentuated for people with palliative care needs, where the notion of maintaining one's role and identity is considered especially important in the context of declining health and functioning.

3.2.2. Expert care

The Expert care category for inpatients includes six domains of importance: excellence in physical care; impeccable assessment and care planning; effective symptom management; technical competence; patient safety; and supported access to senior clinicians. Aspects of care provision that enhance care experience are noted below in relation to each identified domain, followed by the key practice points (Table 3) to guide clinical practice, policy and research.

3.2.2.6. Excellence in physical care

Excellence in physical care, with specific attention to the maintenance of personal hygiene, responsive attention and care provided in a compassionate, empathetic and willing way is important. In addition to this, attention to a patient's nutrition is important for this population who noted their need to access a diet that enhances wellbeing, maintains weight and assists management of chronic conditions.

3.2.2.7. Impeccable assessment and care planning

Impeccable assessment and care planning, highlighting the need for clinical teams to pay attention to a person's holistic health care needs, inclusive of their physical, social, emotional and spiritual wellbeing, are vital. In addition, inpatients with palliative care needs want clinicians to be knowledgeable about specific health care needs within the context of broader health issues, enabling an accurate assessment of the current situation and related care requirements to achieve a good clinical outcome. That is, inpatients and their families seek care that is not purely focused on one component of care or one organ system only. Finally, families articulated the importance of identifying when an inpatient is in their last days of life and ensuring their comfort.

3.2.2.8. Effective symptom management

Effective symptom management, including the need for regular assessment and rapid management (pharmacological and non-pharmacological) of any symptoms causing discomfort or distress, especially in relation to pain and breathlessness, is important. Responsive attention to symptom distress was highlighted and families specifically noted the need for rapid management of distress for an imminently dying person, with medication orders available for nurses to act immediately.

3.2.2.9. Technical competence

Technical competence in relation to specific procedures such as cannulation and pleural effusion drainage is valued. Distress was caused by multiple cannulation attempts. In addition, the need to proactively discuss and appropriately deactivate an implanted cardiac device for an imminently dying person is important from the perspective of families.

3.2.2.10. Patient safety

Patient safety with respect to care being well-managed without the need for continual oversight by the patient themselves, given the unwell status of this population, is important. Medication management was also discussed as an area of critical importance and of great concern for patients and families alike in relation to accuracy in medication prescription and administration.

3.2.2.11. Supported access to senior clinicians

Supported access to senior clinicians with high levels of expertise to enable prompt attention to noted concerns, answers to questions, time for support and planning, comfort in relation to ongoing support on discharge home and continuity of care is valued and supportive.

3.2.3. Optimal environment for care

Environmental factors also impact on inpatient and family experience of palliative care and feelings of wellbeing, (Robinson et al., 2018) hence being represented as a third category of importance for optimal care. This category includes three domains of importance: patient focused structural factors; family focused structural factors; and cleanliness to support infection control. For inpatients, an optimal environment aligns with many of the person-centred design principles recommended for hospital environments, including: the built environment takes a holistic approach to health and wellbeing; the built environment is welcoming and feels safe; spaces are designed to facilitate patient and family engagement; design of spaces promotes partnership and equality between patients and care teams; physical barriers are minimized to promote communication; and the environment empowers patients and feels like ‘their own’. (Australian Commission on Safety and Quality in Health Care, 2018) Aspects of the inpatient environment that enhance care experience for people with palliative care needs, are noted below in relation to each identified domain, followed by the key practice points (Table 3) to guide clinical practice, policy and research.

3.2.3.12. Patient focused structural factors

Specific attention to providing a restful environment is desired, inclusive of minimal noise, soft lighting and comfortable temperatures. Access to a window is important to enable a connection to the day/night cycle. It also enables inpatients to feel connected to the world, feel warmth from sunlight, and helps to prevent feeling claustrophobic. The preference for a shared or private room varied, suggesting that each inpatient, where possible, ought to be provided with choice. This discussion would be helpful even if choice is not available, as understanding the key concerns for either option could enable clinicians to tailor care accordingly – for example, where someone is particularly concerned by smells, use of an odour neutraliser in the room could be helpful. Discomfort with the use of mixed gender wards was a frequent concern for both inpatients and their families, who felt vulnerable and questioned the appropriateness of such a policy. Preventing frequent bed changes is important for inpatients who try to develop a sense of belonging to their bed area. People living with breathlessness spoke of their need for greater environmental supports for their symptom control, such as the provision of fans, cooling air and adequate ventilation in bathrooms specifically. Patients also spoke of practical supports that enhanced their care experience, such as the provision of a bed whilst waiting in the Emergency Department, given how unwell they are, and the opportunity to access emergency supplies of toiletries if needed.

3.2.3.13. Family focused structural factors

The provision of simple comforts, such as seating for family members and preferably a place to lie down at night, alongside the patient's bed, was important to ensure rest and wellbeing whilst providing care on a day-to-day basis. Families described how supporting people with palliative care needs can be exhausting, and the profound benefits comfortable sleep provides. In addition, simple kitchen items (water, ice, microwave) are helpful to enable family members to provide care for their loved one, as well as food and drinks for themselves. Dedicated space to enable private conversations, a break from caring or for quality family time away from the bed-space (inclusive of the inpatient) were also highlighted as supportive. In addition, access to affordable meals onsite was said to be important, with the suggestion that subsidised hospital cafeteria food be made accessible for families. Finally, bereaved family members noted the fact a private room is essential for someone who is very close to dying to enable privacy for meaningful conversations and quality time.

3.2.3.14. Cleanliness to support infection control

Cleanliness to support infection control was emphasised as important, specifically in relation to feeling safe in the bed and ward area. Patients especially described their need to limit exposure to infections if they are immunosuppressed Table 3.

4. Discussion

This mid-point meta-inference has confirmed three categories of care that, when considered from the perspectives of inpatients and their families, would ensure optimal palliative care within hospital settings, namely: ‘Person-centred care’; ‘Expert care’; and ‘Optimal environment for care’. The meta-inferences also revealed specific domains (n = 14) and practice points (n = 68) that when implemented would potentially strengthen and improve inpatient palliative care delivery. To be effective, all of the categories and domains will need to be considered in their entirety. That is, a focus on excellence in physical care alone will not enable optimal inpatient palliative care. This meta-inference is reflective of the views of a large cohort of inpatients living in high income countries and is a robust contribution from which to move forward in informing practice evaluation and subsequent improvement.

Given the studies included in the systematic review and meta-synthesis (Studies 1a and 1b) were published up until April 2015, (Virdun et al., 2015; Virdun et al., 2016) an updated search was completed in November 2020, with three additional studies identified (Robinson et al., 2018; Gott et al., 2019; Black et al., 2018). These three studies confirmed data already outlined within this meta-inference. However, there was one additional component for consideration from a more recent New Zealand study (Gott et al., 2019) centring on the need to consider ‘traditions and culture’. This emphasis is consistent with New Zealand's bicultural society and reflected in its national policy and practices. The perspectives of people from minority cultures and Indigenous perspectives are neglected in work to date and this would be a valuable focus for future research.

There is considerable evidence outlining that person-centred care contributes to improved outcomes and experiences for both inpatients and their families (McCormack and McCance, 2016; Australian Commission on Safety and Quality in Health Care, 2018; Picker Institute Europe, 2019; National Safety and Quality Health Service Standards (Version 2), 2017). This study has identified that ‘Person-centred care’ and ‘Expert care’ are two separate but important categories of care with this distinction at odds with traditional person-centred frameworks, which tend to encompass both clinical competence and technical skills (McCormack and McCance, 2016; Australian Commission on Safety and Quality in Health Care, 2018; Picker Institute Europe, 2019; National Safety and Quality Health Service Standards (Version 2), 2017; Picker Institute Europe, 2017; Nolan et al., 2004; Harding et al., 2015). While The Person-centred Practice Framework, (McCormack and McCance, 2019) notes professional competence as a prerequisite to enable person-centred outcomes, it does not give it the emphasis that palliative care consumers believe good care demands. This change in thinking is also consistent with a recent review of the key attributes of high-performing person-centred healthcare organisations undertaken by the Australian Commission on Safety and Quality in Health Care, which identified the two key influencers on experience of care were the way a person is treated as a person and the way they are treated for their condition. (Australian Commission on Safety and Quality in Health Care, 2018) This distinction speaks to the importance of differentiating between the person and the technical capabilities of the clinician and confirms the consumer representative argument presented in this study.

Importantly, there is considerable cross-over in relation to what is important to the non-palliative inpatient population in relation to person-centred care (Håkansson et al., 2019; Australian Commission on Safety and Quality in Health Care, 2018; Picker Institute Europe, 2019; National Safety and Quality Health Service Standards (Version 2), 2017; Picker Institute Europe, 2017; Nolan et al., 2004) and what is important for those with palliative care needs. The need for respect, compassion, effective communication, shared decision making, holistic care, engagement and coordinated care is well stated across both populations. (Virdun et al., 2020; Virdun et al., 2015; Virdun et al., 2016; Håkansson et al., 2019) Maintaining role, meaning and identity is not specifically discussed within the broader discourse about person-centred care, (Håkansson et al., 2019; Australian Commission on Safety and Quality in Health Care, 2018; Picker Institute Europe, 2019; Picker Institute Europe, 2017) but it is noted within literature focused on working with older people. (Nolan et al., 2004) This suggests that maintaining role, meaning and identity may be of specific importance to people living with life-limiting illnesses. Important areas of care articulated by inpatients with palliative care needs that are not evident in current person-centred care discourse include enabling family involvement, valuing of time with senior clinicians and the importance of food and nutrition. Furthermore, there are specific requirements that are supportive for people who are imminently dying, and their families, that are not evident in the broader person-centred care literature (Håkansson et al., 2019). These include the care and communication required when a person is close to dying and immediately following their death. Understanding such distinctions is important when considering evidence to inform indicators of quality that may be accessible from non-palliative inpatient population data, to drive reform. That is, there may well be data currently available to inform aspects of care noted to be important across all inpatients (palliative and non-palliative), such as respectful care, compassionate care and/or effective communication. However, consideration ought to be given to areas noted to be important specifically to those with palliative care needs (e.g. enabling family involvement, food and nutrition, and access to senior clinicians). A different process is also likely to be needed to obtain data from people who have recently experienced a death in hospital of a close family member.

Given the busy hospital environments caring for diverse populations, the number and frequency of indicators used for quality assurance processes is critical. Therefore, driving improvements in inpatient palliative care will benefit from a review of possible use of: existing data sources; (Maetens et al., 2016) prioritisation of tools that assess patient-identified areas of importance; (Steinhauser et al., 2002) tools that can be implemented into routine clinical practice (Currow et al., 2015; Hearn and Higginson, 1997) and which preference patient-reported outcome and experience data (Hearn and Higginson, 1997; Antunes et al., 2014) The insights generated by this meta-inference can be used to inform and influence the use of current data and available indicators with the most potential of informing improvements, as well as identifying where new indicators are required.

4.1. Strengths and limitations

The strength of this study is the meta-inference of data from different studies identifying the key domains of importance for optimal inpatient palliative care from the perspectives of patients with palliative care needs and their families. These data have been collected in a systematic and rigorous way (representing thousands of perspectives) and were mixed using joint display tables informing a robust base to work from when designing changes to enable improvements. The key limitation in relation to this is the lack of cultural diversity reflected in the outcomes from this work. Taking these domains to diverse populations and Indigenous populations is important as this work progresses to ensure their views and needs are represented in future improvement initiatives.

5. Conclusion

This mixed methods meta-inference has identified and confirmed three categories of care that enable optimal inpatient palliative care, namely: (1) Person-centred care; (2) Expert care and (3) Optimal environment for care. Within these three overarching categories are 14 domains of importance for inpatients with palliative care needs and their families. Many components within each domain are also likely to be important for other inpatient populations. However, there are some unique requirements for inpatients with palliative care needs that warrant careful consideration, particularly in relation to the inpatient who is imminently dying. Furthermore, this meta-inference provides a clear set of practice points aligned to each domain and category to inform optimal inpatient palliative care. Understanding how to drive reform to enable this care, within the busy and varied environments of hospitals caring for diverse populations across metropolitan and rural contexts, is an important next step.

Relevance to clinical practice

This mid-point meta-inference study reports two key outcomes of relevance to clinical practice. Firstly, a synthesis of patient and family perspectives about what is important for optimal inpatient palliative care ensuring all research and improvement work is founded upon that which matters most to those who require this care. Secondly, specific guidance about how to enable each domain of care in practice is provided through the articulation of key practice points. This enables clinical and research leaders to work with each domain and their related practice points and consider how they can enable care in line with these within the busy and varied contexts of acute hospital settings across the globe.

Data availability statement

All data informing this mid-point meta-inference study are freely available either via published manuscripts or within a doctoral thesis with these thesis available via an online repository. All referencing provided throughout the manuscript connects the reader to available source data.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.