A Scoping Review of the Barriers and Facilitators to Clinician Engagement in Hospital-Based Palliative Care in Australian Hospitals

Abstract

Background: Current research has shown that inpatient palliative care (PC) services are under-utilized, poorly integrated, and frequently introduced too late during inpatient hospital stays. The aim of this study was to identify a comprehensive list of multi-disciplinary facilitators and barriers to inpatient PC in Australian hospitals through a scoping literature review. Methods: This review identified articles published since 2000 from 3 electronic databases (CINAHL Plus, MEDLINE and Embase), which included discussion of collaboration among non-palliative care clinicians and palliative care professionals in Australian hospitals. We used an inductive approach to identifying key domains of barriers and facilitators. Results: Thirty-four articles met inclusion criteria following full text review. Barriers and facilitators were categorized in 7 domains: (1) Patient concerns, (2) Family concerns, (3) Clinician knowledge, education, and experience in palliative care, (4) Recognition and acceptance of prognosis when a patient was dying or needing end of life treatment, (5) Reconciliation of individual and professional values around PC, (6) Clinician access to resources for PC in the hospital, and (7) Communication between the PC team and ward clinicians. Each domain included potentially substantial barriers to PC delivery and practice. However, given the small sample sizes and specialized settings of many included studies, it was difficult to draw conclusions on the relative significance of different barriers across hospitals. Conclusion: This review identified a number of barriers and facilitators across studies. Subsequent research needs to more comprehensively compare factors impacting PC use in order to improve implementation of PC across hospital settings.

Introduction

Palliative care (PC) is delivered in both hospital and outpatient settings to provide an integrated approach to quality of life for patients with life-limiting illnesses and their families.^1-3

PC delivery has changed over time, particularly in how hospital-based PC services operate. Initially, in-hospital PC services were aimed towards patients with cancer and terminal care, but increasingly, there is a focus on providing equitable access to those with non-malignant conditions.^4-6 There is also a paradigm shift from equating palliative care with end-of-life care, towards provision of PC concurrently with disease-targeted therapies, ⁷ as PC has proven effective at preventing and relieving suffering through early assessment and symptom management. ⁸ As demand for PC in the hospital setting has increased, many hospitals employ PC staff to provide care either through specialist consults or in designated PC units. Despite these changes, hospitals continue to find it challenging to integrate inpatient PC with curative care or disease-targeted treatment. ⁹

Early access to PC is also critical for reducing burden on healthcare systems, particularly as the population ages and faces increased burden from chronic diseases. ² Additionally, early access to PC has been shown to decrease healthcare expenditures by reducing medically unnecessary hospital presentations and investigations, the length of inpatient admissions, and the utilization of intensive care services. ¹

Despite this, early integration of PC in a patient’s disease journey is seldom seen in practice.^9-11 Even when PC services are readily available, physicians delay referral despite patients meeting referral criteria. ¹² PC is frequently poorly integrated and called for too late during an illness, and though this issue has long been recognized, it remains a persistent challenge.^10,11,13 To improve the timely provision of PC, it is necessary to characterize barriers and facilitators to its early integration. Much of the contemporary work has been undertaken within the American context. In contrast to the American model of health care, Australia adopts a hybrid model whereby the Government provides basic-level insurance (universal healthcare) known as Medicare, and individuals can choose to supplement this by purchasing additional private health insurance. Medicare provides palliative care and inpatient or hospice care in a public facility free of charge. However, patients may also utilize private insurance to cover additional or alternative private services and may incur out-of-pocket costs or ‘gap-fees’ for certain medicines or specialist appointments that are not fully covered by Medicare. ¹⁴ Factors impacting on the provision and timing of PC in an Australian context have been minimally explored.

Aim and Objectives

The aim of this study was to identify multi-disciplinary facilitators and barriers to inpatient PC in Australian hospitals through a scoping literature review.

Methods

This scoping review was undertaken in accordance with the methods described by Arksey and O’Malley ¹⁵ and Levac et al ¹⁶ and reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement. Consistent with these methods, thematic domains identifying key types of barriers and facilitators were generated through exploration of the data to help generate hypotheses regarding which key factors are most likely to impact PC provision in hospitals.

Search Methodology

We searched 3 databases (CINAHL Plus, MEDLINE and Embase) from 2000 up to April 2024 to focus on more current models of palliative care. The search strategy used medical subject headings (MeSH) as well as key terms related to ‘inpatient,’ ‘palliative care,’ ‘Australia’ (inclusive of state names as well) and ‘hospital.’ The search strategy as implemented in CINAHL is contained in our supplementary materials. Titles, abstracts, and full texts were independently screened by 2 researchers against the inclusion and exclusion criteria. We pre-specified our methods in an unpublished protocol.

Study Selection

Eligible articles included those (1) written in English, (2) published from January 2000 through April 2024, (3) focused on the hospital setting and based in Australia, (4) limited to an adult population (18+ years), (5) related to palliative care in hospital, (6) referring to collaboration with non-palliative care clinicians, and (7) including palliative care professionals. Studies were excluded for the following reasons: (1) in home-based hospice or sub-acute settings; (2) focused solely on patient outcomes; (3) focused on paediatric care; (4) did not include any palliative care expert or team involvement (for example, a rejected study focused on the difficulties faced by renal nurses in end-of-life care); (5) did not address carer-specific issues (for example, studies that looked exclusively at patient outcomes for palliative care). Conference abstracts and letters to the editors were also excluded. We extracted the following information: study design, study duration, study setting, study population, data collection methods, study aim, and barriers and facilitators to PC practice.

Quality and Risk of Bias Assessment

The quality of eligible articles was evaluated by 2 independent reviewers using Joanna Briggs Institute’s (JBI) critical appraisal tools. ¹⁷ Checklist scores were converted to percentages to compare quality across study design, and a percentage less than 30% was defined as unacceptable, although no studies fell below that threshold.

Data extraction from articles identified the process by which PC staff worked with or integrated into usual medical care in the inpatient setting and identified potential barriers and facilitators to PC referral and use. From the extracted data and in consultation with palliative care specialists, barriers and facilitators were categorized into 7 domains: (1) Patient concerns, (2) Family concerns, (3) Clinician knowledge, education, and experience in palliative care use, (4) Recognition and acceptance of prognosis when a patient was dying or needing end of life treatment, (5) Reconciliation of individual and professional values around PC, (6) Clinician access to resources for PC in the hospital, and (7) Communication between the PC team and hospital ward clinicians. These domains were developed to fit the specific issues of the use of PC services in hospitals but were adapted to this context from The Theoretical Domains Framework, which applies psychological theory to healthcare implementation. ¹⁷ We charted data in a table to visually convey which studies informed each of the theoretical domains. ¹⁸

Results

Our search returned 1029 studies, 34 of which met inclusion criteria. A PRISMA diagram in Figure 1 summarizes this process.

Figure 1.

PRISMA diagram of the study selection process.

Study Characteristics

Most studies collected data in Australia’s 3 most populous states, with eleven from Victoria,^6,13,19-26 8 from New South Wales,^27-34 and 6 from Queensland.^35-40 Two studies were completed in Western Australia,^41,42 2 in South Australia,^43,44 and 1 in the Australian Capital Territory. ⁴⁵ Two studies used data from national surveys of clinical professionals but did not report how these professionals were distributed across the country.^46,47 The locations of the hospitals involved in the 2 remaining studies were not reported.^10,12

Regarding sample size, the studies ranged from large samples that included medical record audits, registry data analysis, and population surveys, to smaller sample semi-structured interviews of both patients and clinicians and/or focus groups on palliative care use. Specifically, twelve studies reported using retrospective audits of hospital administrative data or medical records, with the largest study using records from over 150,000 patients and 4 other studies reporting samples larger than 20,000. An additional 4 studies linked hospital and population health registries in a cohort design, with the majority using samples that ranged from 100 to 5000 patients. Another 9 studies used qualitative data taken from semi-structured interviews with between 9 to 54 respondents. Five studies used surveys of clinical professionals, with moderately sized samples between 200 to 500 respondents. Finally, 3 studies used focus groups, and 1 included open-ended interviews. Table 1 summarizes study details of the full set of 34 studies reviewed here.

Table 1.

Characteristics of Reviewed Articles.

Study Lead Author, Year	Design	Study Duration	Setting	Study Participants (Numb. In Parentheses)	Method of Collection	Aim of the Study
Assareh, 2020	Retrospective cohort	2010-2015	73 NSW hospitals	All adults who died as inpatients (91,840)	Linked hospital admission and mortality data	To examine trends impacting the uptake of palliative care and any inter-hospital variations.
Batten, 2018	Medical record review	2010-2015	One NSW hospital	Adult med oncology inpatients (295)	Care plans and goals of care	To examine patterns of end-of-life care and quality of care for medical oncology patients.
Broom, 2012	Physician interviews	2010	A large urban palliative care unit	Physicians (20)	45-60 mins semi-structured interviews	To examine the logics explaining timing of referrals to palliative care for a range of medical specialists.
Broom, 2015	Nurse interviews	NA	Brisbane QLD hospitals (2)	Nurses (20)	30-60 min semi-structured interviews	To understand the role of nurses in palliative care referrals.
Button, 2014	Chart review and survey	2012-2017	Brisbane hospitals (2); transplant centres (16)	Patients dying in haematology units (613); APNs from centres (14)	Patient episode data; nurse surveys	To identify characteristics of patients dying after stem cell transplantation and the role of EOLC for these patients.
Cahill, 2021	Qualitative interviews	2017 -2018	A specialist palliative care inpatient unit in Australia	Palliative care inpatients and family members (22)	Semi-structured interviews	To explore inpatients receiving palliative care and their families’ experiences of participation in a patient-centered family meeting, where the patient sets the meeting agenda.
Chambers, 2018	Prospective cohort	2015-2016	One QLD hospital	Patients in kidney health service (19)	Medical records	To track and quantify health service utilisation of people with ESKD stages 3-5 during the last 12 months of life
Collins, 2018	Population cohort	2000-2010	All VIC hospitals	Patients with lung, breast, and prostate cancer (46,700)	Discharge data and death certificates	To develop more standardised referral practices to EOLC by examining current transition points for those with cancer
Collins, 2022	Mixed method randomised trial	Not reported.	Four metropolitan tertiary cancer services in Melbourne	Patients with advanced breast, prostate and brain cancers (65)	Medical records and semi structured qualitative interviews.	To test the feasibility of using defined triggers for the integration of standardised, early palliative (STEP) care across 3 advanced cancers.
Duggan, 2019	Retrospective cohort	2006-2012	SW sydney cancer registry	Patients with end stage cancer (923)	Medical records	To describe patterns of palliative care use and to identify factors predicting palliative care use and survival.
Eastman, 2013	Retrospective cohort	2008	One VIC hospital	Patients in stroke care unit (544)	Medical records	To investigate factors associated with referral of patients from stroke unit to a palliative care service.
Fox, 2016	Qualitative interviews	NA	One QLD public hospital	Patients/care-givers in support group (16); clinicians (13)	Semi-structured interviews	To examine perspectives on the transition to palliative care in the context of end-stage cancer.
Grbich, 2006	Retrospective cohort	3 months	One SA teaching hospital	Patients dying during hospital stay (20) and nurses (40)	Case notes	To identify deficits in EOL care provision and develop a palliative care pathway or model for care.
Le and Watt, 2010	Retrospective audit with interviews	3 months	One VIC hospital	Patients dying in hospital (171)	Chart audit and interviews	To assess EOLC and understand senior clinician decision-making around referral.
Maubach, 2019	Retrospective observational	2010-2014	One ACT hospital	Patients dying in hosp (529), exclud ED and ICU	Medical records and death certificates	To document patterns in end-of-life care for patients dying across medical units.
McGrath, 2009	Qualitative interviews	N/A	Regional cancer care centre (QLD)	Clinicians, incl MD, social worker, and nurses (4) and patients (5)	Semi-structured interviews	To explore differences between pall care providers in Kerala, India and Queensland, Australia.
McNamara, 2018	Qualitative interviews	2015-2016	Population-based, WA	Clinicians caring for those with schizophrenia (16)	Semi-structured interviews	To examine experiences for people with schizophrenia who have advanced life limiting illness.
Michael, 2019	Retrospective cohort	2012-2016	Hospitals (VIC)	Patients with end stage cancer (278)	Administrative data	To identify the association between timing of referral to pall care and extent of cancer care received.
Michael, 2016	Qualitative focus groups	2014	A combined inpatient and community palliative service (VIC)	Clinicians in focus groups (32)	Interviews, focus groups, questionnaires	To explore staff experiences of transition to specialist service with early referral of patients.
Möller, 2020	Retrospective cohort study	2010-2015	118 NSW hospitals	Inpatients requiring pall care (74,057)	Admission records and death records	To identify if patient’s country of birth affected use of pall care services.
Monterosso, 2016	Cross-sectional survey	2013-2014	One teaching hospital	Clinician and pastoral care staff (302)	Close-ended survey	To identify clinicians’ perspectives, education, and support needs related to palliative care provision.
Moon, 2021	Qualitative interviews	Not reported	A general medicine unit in a large university hospital in Melbourne, Australia	Clinicians and bereaved family members of deceased patients with dementia (32)	Semi-structured interviews were conducted at the hospital, families’ homes, or over the phone.	To assess clinicians and families’ perceptions of dementia as a terminal condition in relation to end-of-life admissions.
Morgan, 2019	Cross-sectional survey	2015-2016	Population-based	Allied health workers (217)	Close-ended survey	To elicit clinicians’ views on palliative care relevance to their clinical practice.
Noble, 2018	Qualitative interviews	2015	One QLD hospital	Clinicians (28)	Semi-structured interviews (18) and 2 group interviews (10)	To investigate EOL care excellence and barriers and enablers to implementing guidelines for dying patients.
Panozzo, 2023	Qualitative interviews	2017 - 2018	A metropolitan tertiary teaching hospital in Victoria	Nurses, physicians and allied health professionals providing care for indigenous patients at the participating hospital site (54)	Face-to-face, semi-structured focus groups and interviews	To explore the experiences and perspectives of indigenous and non-indigenous health professionals regarding the provision of palliative care for indigenous people with advanced disease.
Philip, 2006	Qualitative focus group	N/A	Two VIC hospitals	Palliative care clinicians (45)	Focus group	To identify characteristics of ideal PC care and dynamics of practice
Philip, 2019	Retrospective cohort	2000-2010	One hospital	Patients dying with non-hodgkin lymphoma (4380)	Medical record and audit data	To document EOLC for patients with non-hodgkin lymphoma.
Rosenwax, 2016	Retrospective cohort	2009-2010	Population-based, WA	Patients with chronic disease (23,852)	Administrative data	To examine palliative care provision across 10 years for those with non-cancer conditions.
Smallwood, 2018	Retrospective cohort	2004-2015	One urban and 1 rural hospital (VIC)	Patients dying with COPD (343)	Medical record audits	To determine if symptom palliation or a palliative approach were delivered during the final hospital admission.
Stubbs, 2019	Retrospective cohort	2010-2015	NSW hospitals	Patients aged 50 year or older who died (150,770)	Administrative data	To describe hospital use in the final year of life, including timing and use of palliative care.
To, 2011	Cross-sectional survey	NA	Three SA hospitals	Patients (496)	Close-ended surveys	To identify those referred for palliative care.
Trankle, 2020	Cross-sectional survey	2018	Australian hospitals	Clinicians (408)	Semi-structured surveys	To identify barriers and enablers for end-of-life discussions.
Virdun, 2020	Qualitative study	2018-2019	5 NSW hospitals	Adults with palliative care needs (21)	Semi-structured interviews	To use patient experience to inform the relevant domains of optimal inpatient palliative care.
Virik, 2002	Prospective cohort	1996	One NSW hospital	Patients (50)	Medical records	To quantify palliative care use and assess use of a simple assessment for impact of pall care.

Facilitators and Barriers

In reviewing the barriers and facilitators to PC service use, we developed 7 domains to define types of barriers and facilitators: (1) Patient concerns, (2) Family concerns, (3) Clinician knowledge, education, and experience on palliative care use, (4) Recognition and acceptance of prognosis when a patient is dying and end of life treatment is needed, (5) Reconciliation of individual and professional values around PC, (6) Clinician access to resources in the hospital, and (7) Communication between PC and clinicians.

Domain 1: Patient Concerns

Patient reluctance to engage with palliative care was a commonly identified theme, and studies highlighted several reasons for this.^25,38,39 McGrath et al ³⁸ noted that feelings of pride and independence and the stigma associated with dying, in addition to preferences to stay home rather than enter hospice or inpatient PC, made patients reluctant to seek palliative care support. Michael et al ²⁵ and Fox et al ⁴⁰ also reported that patients had strong negative connotations regarding palliative care. Fox et al ⁴⁰ reported that clinicians heard negative views of palliative care from patients, which subsequently led clinicians to be reluctant in making referrals.

Negative views of palliative care sometimes arose from misperceptions, which were discussed in articles by Fox et al, ⁴⁰ Monterosso et al, ¹² and Stubbs et al ³¹ Fox et al ⁴⁰ found that patients often believed that PC focused only on physical symptom control and ignored or neglected psychosocial and spiritual dimensions of care. They also found that patients believed PC were appropriate solely for terminal or near terminal care and that a referral to PC meant that no further formal medical treatment was available for their disease. In other words, patients associated PC with the end stages in life journeys, ⁴⁰ which was further supported in the research by Stubbs et al. ³¹

Patient or family resistance to palliative care services sometimes led clinical staff to delay PC referral. For example, Broom et al ³⁵ found that medical oncologists felt that family or patient resistance prevented referral even when the oncology team was prepared to refer. Age of the patient had some impact on decisions to refer to PC, although there was conflicting evidence on the direction of effect. Duggan et al ²⁹ reported a 20% increase in the likelihood of receiving early PC referral among those aged 85 and over as compared to people aged between 75 and 84 (RR1.2, 1.07-1.35, P = .0017). In contrast, Rosenwax et al ⁴² found a decreased likelihood of accessing PC for older patients. Two studies also found that men were less likely to use PC.^8,42

Domain 2: Family Concerns

Family resistance to PC, particularly if they associated PC with a terminal prognosis, was noted in multiple studies, including Broom et al ³⁵ and McNamara et al ⁴¹ Phillip and Komesaroff ²³ also found that healthcare staff were reluctant to discuss palliative care if families discouraged discussion of prognosis or diagnosis. In some instances, family resistance arose from cultural beliefs that dying persons should not be told they were dying; McGrath et al ³⁸ found such beliefs present among some ethnic groups from India, for example. McGrath et al also found these cultural values were occasionally present in the Australian context as well, although norms for clinical care in Australia support providing patients with transparent prognostic information. Finally, Grbich and colleagues ⁴⁴ discussed how unique family dynamics could lead to reluctance to use PC services. For example, in 1 case, a daughter expressed concern that PC could not accommodate her wish to take her mother home. More generally across families, Grbich et al ⁴⁴ noted several caregiver factors contributing to concern about PC, including distress or concerns regarding prognosis, the impact of medications on drowsiness, and dissatisfaction with the quality of care received by their loved ones. Family concerns could also be disease specific – Moon et al ⁷ found that families of patients with dementia were often unaware of its status as a life-limiting illness, which could lead to the continuation of futile life-prolonging treatments if this misunderstanding was not appropriately rectified by clinicians.

On the other hand, family members were also sometimes facilitators of PC use. McNamara et al ⁴¹ noted that family members played important roles as advocates of PC services. Similarly, Virdun and colleagues ³³ reported that patients stressed the importance of family member engagement in PC. It was noted that when patients were cognitively impacted by their medical treatment, family members were indispensable for assisting the patient to engage in treatment discussions. Further, patients mentioned several ways in which hospitals can facilitate family caregivers taking on advocacy roles, including flexible visiting hours and regular ward round times, so that family members could balance visits alongside external commitments. Cahill et al ³⁴ described formalized family meetings with an interdisciplinary team, which provided a forum for family concerns to be shared and addressed. These meetings sometimes facilitated greater acceptance of PC involvement.

Domain 3: Clinician Knowledge, Education, and Experience

The knowledge, education, and experience of the clinicians treating the patient was also influential. Broom et al, ¹⁰ McGrath et al, ³⁸ and Monterosso et al ¹² interviewed clinicians who emphasized that a lack of knowledge, experience, or inclination from clinical staff can delay or prevent PC referral. Stubbs et al ³¹ likewise found that shortcomings in the clinical judgement of doctors and nurses led to late PC referral. Michael et al ²⁵ noted that poor clinician knowledge can result from insufficient training on what patients need with regards to palliative care, including uncertainty about how the principles of comfort care integrate with standard care, which is more often characterized as curative. On a related note, Panozzo et al ²⁶ found that a lack of knowledge of indigenous cultural values sometimes resulted in sub-optimal relationships between PCS and indigenous Australians, which could complicate their access to care.

A facilitator for PC service integration was improvements in knowledge and experience around end-of-life discussions. Trankle et al ⁴⁷ found that improvements in training and experience around end-of-life discussions could facilitate PC referral and could be gained through either job or life experience. They suggest such experience could be facilitated through peer mentoring, staff workshops, and role play.

Domain 4: Recognition of Dying

Multiple studies also identified clinicians’ failure to accurately recognize dying as a barrier to PC referral and as a cause of unnecessary interventions in the last days of life. Stubbs et al ³¹ compared the proportion of hospital decedents who received PC with the proportion who could have benefited from PC referral, and found that many dying patients were not appropriately identified as such. Stubbs et al hypothesized that an associated barrier was clinicians’ tendency to rely on their subjective clinical judgement of prognosis over objective diagnostic tools to identify eligible patients. Grbich et al ⁴⁴ further found that recognition and documentation of the terminal phase occurred in only approximately 1 out of every 2 studied deaths. Maubach et al ⁴⁵ found that rates of recognition of dying varied across specialties, with oncologists most likely to recognize dying and provide and communicate accurate forecasts.

Good documentation of advance care plans and comfort care plans was a facilitator for PC referral. Maubach et al ⁴⁵ reported that documentation of comfort care plans was significantly higher in oncology wards compared to respiratory and cardiology wards. This study identified that the presence of goals of care documentation decreased use of active medical management, such as “dialysis, intubation and [Medical Emergency Team] calls” in the final 48 h of life. ⁴⁵

Domain 5: Reconciliation of Individual and Professional Values Around PC

Clinicians’ professional values were frequently described as barriers to PC referral.^35,40,47 Fox et al ⁴⁰ found that oncologists delayed referral to PC if they saw themselves as effectively managing all patient needs. Broom et al ³⁵ found that healthcare professionals would sometimes take on personal crusades against diseases or perceive an urgent need to treat in the case of some cancers, which then created a barrier to timely PC referral. Similarly, Trankle et al ⁴⁷ found that clinicians perceived avoiding death as paramount, while Le and Watt ²¹ found that clinicians often became emotional about a patient’s dying. Consequences of this barrier could be significant.^39,40 Noble et al ³⁹ found that clinicians often focused too much on futile attempts at prolonging life, which led to delaying end-of-life discussions and/or PC involvement until too late in a patient’s illness.

Individual values or perceptions about PC also created barriers.^21,31,35,40 Broom et al ³⁵ reported that clinicians would delay PC referral to avoid the task of explaining PC to the patient. Stubbs et al ³¹ echoed this point, indicating that clinicians had difficulty initiating conversations about death, while a clinician interviewed by Fox et al ⁴⁰ described not wanting to scare patients with a PC referral. And, finally, Philips and Komesaroff ²³ described 1 GP who preferred to consider his patient as a friend, which ultimately resulted in failing to provide the patient with optimal medical care.

Domain 6: Access to Services and Resources

Access to PC services and resources was a key facilitator for providing PC referrals. For example, in studies by both Broom et al ¹⁰ and Assareh et al, ⁹ results indicated that the lack of onsite palliative care services at the hospital was a major barrier. When PC is present, Noble et al ³⁹ found that hospital staff could still experience difficulty navigating the bureaucracy around PC referrals. In both qualitative interviews and focus groups, this study found that staff were confused about when to appropriately make palliative care referrals.

Several studies found that geographic location could be a barrier to receiving PC^9,38,42 Assareh et al ⁹ found that patients from non-major urban areas received less PC and experienced delays in PC engagement compared to their urban counterparts. Similarly, Rosenwax et al.’s retrospective cohort study ⁴² found a statistically significant decrease in referrals to specialist palliative care for patients living in more remote areas and for patients living in aged care facilities. In McGrath et al.’s qualitative study, ³⁸ patients indicated that financial costs of travel were a barrier to receiving PC.

Likewise, hospital resource availability could also be a significant barrier to receiving PC. Grbich et al ⁴⁴ quoted a nurse who felt that the hospital staff could not provide sufficient staff to sit with dying patients as would have been needed to help provide dignified care for those at the end of life. Further, Fox et al ⁴⁰ identified time constraints as key to providing clinical care, while in qualitative interviews with patients, Virdun et al ³³ identified the lack of private rooms and available nursing staff trained in palliative care to sit with patients long enough to sufficiently assess needs as important barriers. In Noble et al.’s study, ³⁹ interviewed subjects reinforced the need for more available palliative care clinicians.

Hospitals where palliative care specialists worked alongside other healthcare professionals, sometimes in multidisciplinary teams, were found to have improved facilitation of PC referrals in several studies. For example, Duggan et al ²⁹ found that access to integrated centers with both oncology and PC services facilitated PC use. Michael et al ²⁵ reported that access to a PC consultative service led to earlier access to palliative care as needed for symptom management. Finally, Noble et al ³⁹ found that involvement of palliative-care trained nursing staff in multidisciplinary team meetings facilitated PC referral and access.

Domain 7: Communication Between PC and Clinicians

Dysfunctional relationships between PC and other healthcare professionals were commonly cited as a barrier to timely PC integration.^10,23,35 Both Broom et al ¹⁰ and Philip and Komesaroff ²³ identified the lack of working relationships between hospital clinicians and palliative care specialists as a barrier to effective PC to the extent that the clinical judgement of specialists was undermined, and patients perceived the lack of collegial relationships. In some cases, hospital clinicians felt that their care was criticized and were less likely to refer to PC, although Philip and Komesaroff noted that such issues were generally minor.

Multidisciplinary discussion between PC staff and other hospital clinicians was widely found to be a facilitator for PC integration.^29,33,41,47 Duggan et al ²⁹ found that multidisciplinary team discussion meetings were significantly associated with early PC usage. Multidisciplinary team meetings were found to improve visibility of the PC team and facilitate working relationships and referral pathways with hospital staff. Virdun et al ³³ reported that patient perceptions of greater communication between PC and other hospital clinicians facilitated better integration of care. McNamara et al ⁴¹ found that involvement of general (family) practitioners, PC staff, and mental health workers in case conferencing and the sharing of lead roles across these parties facilitated improved referral and use of PC for patients with schizophrenia. Le and Watt ²¹ found that allowing nursing staff and allied health to communicate directly with the PC staff facilitated care. In addition, good working relationships with PC also facilitated improvement of PC competencies among non-palliative care specialists. For example, Le and Watt ²¹ noted that PC staff helped junior medical staff communicate with families and patients. Finally, Collins et al ¹³ found that formalized trigger points which mandated a referral to PCS could be an effective method of overcoming psychological communication barriers that might otherwise prevent these referrals being made.

Discussion

While early integration of PC with disease-targeted management has been linked to improved patient outcomes, PC remains underutilized. This is not surprising given the number of barriers identified in studies of Australian hospitals, which indicate significant issues in providing timely and needs-based PC in this context. Given the complexity of potential issues with implementation, additional training and education for clinicians alone will not suffice to facilitate implementation. Instead, hospitals need to be attentive to how to address patient and family concerns, how to provide greater resources for ongoing PC service use, and finally, how to influence clinician perceptions regarding the use of PC. Only by identifying and understanding the full set of barriers and facilitators, as we have attempted in this study, can we begin to move towards solution generation.

We identified perceptions among Australian patients (and to a lesser extent clinicians) that palliative care was either appropriate only for terminal care or was incompatible with standard or curative medical treatment. In the international literature, a distinction is made between palliative care, which manages the needs of patients with a serious or life limiting illness at any stage of their illness, and hospice care, which is explicitly reserved for patients nearing the end of life. ⁴⁸ Both terms represent an approach to care and can be provided at home as well as in a facility. Conversely, the term hospice in an Australian context usually refers to a dedicated inpatient facility that delivers palliative care. ⁴⁹ To members of the public, the terms are likely synonymous, which may contribute to the confusion and barrier to engagement that we identified in our review.

Each study addressed some significant issues, but so far no study provides a comprehensive approach to comparing the importance of all the possible barriers and facilitators within a specific hospital context. Two studies^35,47 discussed barriers from 6 of the 7 domains but most studies focused on fewer than 5 domains. Table 2 provides a summary of which domains were identified in each study. The facilitators and barriers relevant to PC use in hospitals ranged from patient and family concerns to physician responses to changes in PC delivery, and organizational resource availability. Further research is needed to identify which factors are most important for earlier referral to PC and ultimately, for improving patient experience and outcomes.

Table 2.

Identified Domains of Barriers and Facilitators by Paper.

Domains	Assareh 2020	Batten 2018	Broom 2012	Broom 2015	Button 2014	Cahill 2021	Chambers 2018	Collins 2018	Collins 2022	Duggan 2019	Eastman 2013	Fox 2016	Gribich 2006	Le 2010	Maubach 2019	McGrath 2009	McNamara 2018	Michael 2016	Michael 2019	Moller 2021	Monterosso 2016	Moon 2021	Morgan 2019	Noble 2018	Panozzo 2023	Philip 2006	Philip 2019	Rosenwax 2016	Smallwood 2018	Stubbs 2019	To 2011	Trankle 2020	Virdun 2020	Virik 2002
Patient concerns	X		X	X	X					X		X		X		X		X			X							X	X	X
Family concerns			X	X		X							X			X	X			X	X	X				X						X	X
Knowledge, education and experience		X		X	X			X			X	X	X	X	X	X	X	X			X			X	X	X			X	X		X		X
Recognition of dying		X					X					X	X	X	X															X		X
Reconciliation of individual and professional values around PCS			X	X	X							X	X	X							X			X		X				X		X
Access to services and resources	X			X	X					X		X	X			X		X					X	X				X	X		X	X	X
Communication between PCS and clinician			X	X			X		X	X				X			X							X		X	X				X	X	X	X

Many of the studies that examined a range of barriers and facilitators drew heavily on qualitative analyses. For example, Broom et al ³⁵ used semi structured interviews with 20 ward nurses to identify barriers to transitioning patients to palliative care. The qualitative studies we found examined multi-disciplinary factors but were limited by their specialized contexts. These studies also frequently lacked attention to deficits in local resources. In contrast, the included quantitative studies focused frequently on timing of referrals to palliative care and patients’ demographic variables associated with referral timing, but they lacked insight into the complexity of multi-disciplinary factors influencing PC referral. Monterosso et al ¹² developed and used a survey tool that produced quantitative findings, but those study results were not directly relevant to interactions with the PC team. Trankle et al ⁴⁷ provide the most comprehensive survey results on barriers to referrals using a survey of almost 400 clinicians, but their study was limited to clinicians’ communication around end-of-life issues. In summary, while the qualitative studies provided insight into the barriers and facilitators to PC delivery, there is limited understanding of the relative prevalence of these barriers and facilitators in Australian hospitals.

From this scoping review we intend to utilize the identified domains to develop a survey tool to collect data on facilitators and barriers to PC integration in Australian hospitals. This standardized survey could be applied across multiple sites and allow for future consistent identification in the literature of multidisciplinary barriers and enablers to palliative care referral. This would allow for comparisons between hospitals which cannot yet be made based on the studies in this review given they used unique question sets. It is hoped that this survey tool could contribute to filling the literature gaps identified in this review by producing estimates of the prevalence of specific barriers and facilitators to palliative care integration in Australian hospitals. Prevalence estimates will allow prioritization of future research. The survey and our initial findings will be reported in a subsequent paper.

Conclusion

This scoping review identified 7 domains of barriers and enablers to the early integration of palliative care services into patient care. Further efforts should be made to estimate their prevalence, in order to prioritize and design interventions to address barriers, and to leverage enablers to improve patient outcomes.

ORCID iDs

Edward Meehan https://orcid.org/0000-0002-9340-819X

Naomi Katz https://orcid.org/0000-0001-8326-1895

Data Availability Statement

This scoping review relied on published papers which are available to researchers through the relevant journals. We retain copies of the original data extraction spreadsheets, and we can provide these to researchers upon reasonable request.