Article Text

Download PDFPDF

Improving quality in hospital end-of-life care: honest communication, compassion and empathy
  1. Deb Rawlings1,
  2. Kim Devery2,
  3. Naomi Poole3
  1. 1Palliative and Supportive Services, Flinders University, Adelaide, South Australia, Australia
  2. 2Palliative and Supportive Services, Flinders University Faculty of Medicine, Nursing and Health Sciences, Adelaide, South Australia, Australia
  3. 3Director, Partnering with Consumers, Australian Commission on Safety and Quality in Health Care, Sydney, New South Wales, Australia
  1. Correspondence to Deb Rawlings; deborah.rawlings{at}


Background With over half of expected deaths occurring in acute hospitals, and a workforce not trained to care for them, good quality end-of-life care in these settings is hard to achieve. The National Consensus Statement on Essential Elements for Safe and High-Quality End-of-Life Care has been translated into e-learning modules by the End of Life Essentials project, and this study aims to demonstrate how clinicians interpret the Consensus Statement in their day-to-day practice by answering the question at the end of each module: ‘Tomorrow, the one thing I can change to more appropriately provide end-of-life care is…’

Methods The modules were developed by a palliative care educator with the support of a peer review group and were piloted with 35 health professionals. Pre-post module evaluation data were collected and during a 10-month period from 2016 to 2017 a total of 5181 individuals registered for the project accessing one or more of the six modules. The data from 3201 free-text responses to the post hoc practice change question have been analysed, and themes generated.

Findings Five themes are derived from the data: communication, emotional insight, professional mindset, person-centred care and professional practice.

Conclusion Learners who have completed End of Life Essentials have shared the ways they state they can change their practice tomorrow which may well be appreciated as a clinical response to the work by the Australian Commission on Safety and Quality in Health Care in leading and coordinating national improvements in quality and safety in healthcare in Australia. While intent cannot guarantee practice change, theory on intention—behaviour relations indicate that intentions have a strong association with behaviour. This indicates that the modules have the ability to influence end-of-life care in acute hospitals.

  • end-of-life
  • practice change
  • intent-to-change practice

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


An estimated 50% of deaths in Australia are expected1 with approximately 54% occurring in hospital,2 with the likelihood of hospitals keeping up in terms of good quality end-of-life care in question.3 4 Similar figures are reported elsewhere,5–8 indicating that this is not just restricted to Australia.9 The numbers of deaths are projected to increase10 in line with a growing, ageing population. In conjunction with a system-level default to continue lifesaving treatments, living well towards the end of life in acute hospitals is a challenge.11 Healthcare professionals (HCP) who work in acute hospital settings are often highly trained and skilled in rapidly diagnosing and responding to patients with deteriorating health but usually with a focus on cure.12 There is often limited capacity of these HCPs to recognise dying, with poor communication common, and invasive clinical investigations the norm.13 Arguably though, treatment of symptoms and response to those with a life-limiting illness requires a different set of skills.

In 2015, the National Consensus Statement: Essential Elements for Safe and High-Quality End-of-Life Care (the Consensus Statement) was released by the Australian Commission on Safety and Quality in Health Care (ACSQHC) who leads and coordinates national improvements in this arena. The Consensus Statement contains recommendations for clinical practice in end-of-life care in hospitals, endorsed by the National Health Ministers as the national approach to the delivery of quality end-of-life care.1 While it is a fundamental step towards improved care, providing a blueprint for HCP and the healthcare sector, it does not provide practical steps or translational directions.8 Such direction should include how to develop skills and gain confidence in recognising when patients are approaching the end of life, as well as communicating with and managing the care of patients at the end of life, and their families.

The first five elements of the Consensus Statement (box 1) provided the basis for a major national education project (funded by the Commonwealth Department of Health) End of Life Essentials (EOLE). The project, conceptualised and developed at Flinders University South Australia, in effect, operationalises the essential elements identified in the Consensus Statement and builds on the work already identified. EOLE provides free evidence-based, peer-reviewed online learning modules to help build the capacity of HCPs in Australian hospitals.14 The modules are designed for doctors, nurses and allied health professionals (AHP) working in acute care settings and cover: dying (a normal part of life), patient-centred communication and shared decision-making; recognising end of life; goals of care; teamwork; and responding to concerns. The modules are case based with evidence-based resources and web pages available to support learning and were released in two stages: modules 1–3 in June 2016 and modules 4–6 in October 2016.

Box 1

The Consensus Statement—processes of care elements

  • Patient-centred care.

  • Teamwork.

  • Goals of care.

  • Using triggers.

  • Responding to concern.

To inform future module development it was important to know what learners do with their new knowledge and what they plan for changes in practice. At the end of each module, a question is posed: ‘Tomorrow, the one thing I can change to more appropriately provide end-of-life care is…’ This statement changes slightly to make it relevant to each module, allowing the learner to consider what they have learnt and how practice can change immediately. It also allows the researchers to understand the impact of learning and the scope for practice change. The responses from this activity are the focus of this paper, highlighting how the Consensus Statement, as a guide to improve quality and safety and the education derived from it, can influence change in attitudes, behaviour and practice, which in turn impacts end-of-life care and the patient experience.


Developing the education modules

From inception, quality, consultation and industry partnership were paramount. A National Advisory Group (NAG) included the project coleads, ACSQHC delegates and key stakeholders representing health professionals. The modules were developed by a palliative care educator with the support of a peer review group (invited by the NAG) and were multidisciplinary hospital clinicians from across Australia. Following this, 35 doctors, nurses and AHPs piloted the modules. In order to elicit the changes to their own practice that acute hospital HCPs recognise from education derived from the ACSQHC Consensus Guidelines, a question was incorporated into routine evaluation, where learners voluntarily answered, in free text, the following question: Tomorrow, the one thing I can change to more appropriately provide end-of-life care is…

Patient and public involvement

Patients or families were not included in this process, having been involved in development of the Consensus Statement from which the education is derived.

Data collection and analysis

Data have been included from any health professional (self-selected) who completed one or more of the six modules from 1 November 2016 (following the final module release) to 28 September 2017 (the date of the data extraction). A total of 5181 individuals registered for the project during the time period. Data were extracted and deidentified by an administration officer. Data were analysed and a coding scheme was developed independently by a research assistant with no prior knowledge of module content. This was reviewed by the broader team who consolidated the coding scheme and examined emerging patterns. Internal homogeneity and external heterogeneity were established by verifying the meaningfulness and accuracy of the placement of data in the various codes.15 Analytical categories were derived inductively from the data, and the data were read and read again to recognise and code themes and subthemes.16 NVivo V.11 was used to assist with the sorting, organisation, reorganisation and storage of data, and thematic analysis informed our data analysis.17 Themes were systematically and coherently compared to identify as many nuances as possible.18


The themes from 3200 responses were organised in five broad categories concerning intent-to-change practice: communication, emotional insight, professional mindset, person-centred care and professional practice.


This theme included a variety of ways respondents intended to change their clinical communication. Many detailed that they would aim to be more open to questions from patients, families or even their teams—effectively wanting to change their willingness to broach, and respond to, end-of-life issues. A subtheme curbing own avoidance to end-of-life issues in conversations with patients was stated as key to providing appropriate care, even being able to say the word ‘dying’ in conversations was reported as more appropriate. The subtheme of end-of-life discussions holds statements of respondents who aimed at being open, honest and providing full information and advocating on behalf of patients and families as steps that could normalise dying in hospitals. Listening effectively was another subtheme where respondents acknowledged that listening is a powerful part of communication and end-of-life care—hearing and listening to what the patient wants and wishes is key. However, it was identified that such willingness and open communication also posed potential associated risks. While stressing the importance and positive impact (for patients’ care and families’ bereavement) of being able to communicate end-of-life issues, respondents also described the perceived side effects of the emotional nature and labour of such communications. Discussing dying and death in hospitals is essential but can trigger emotional responses in staff and conflict in clinical teams or even with families. Respondents stated clearly that these risks can be mitigated by being prepared, skilled and equipped for conflict and high emotion. Non-verbal communication, another subtheme, captures techniques that could improve practice. Sitting, appropriate touch, a smile or even an attentive look or body posture were mentioned as areas where they could change practice and strengthen end-of-life care.

Virtues and emotional insight

Virtues and emotional insight were a theme used to group and capture the respondents’ considerable descriptions of thoughts, feelings, empathic approaches and moral awareness. This included intent to alter or heighten self-awareness and reflect on thoughts and feelings while thoughtfully responding to families, patients, friends and staff on end-of-life issues. Awareness of own emotions was another subtheme capturing capabilities of being reflective and thoughtful about what could be said, and the impact of words included in conversations. Also, being extra aware of others’ emotional state, another subtheme, was an additional implementation skill. Respondents mentioned being brave—as listening to and being aware of another person’s sorrow, anxiety and ‘being there’ in the face of their suffering requires a strength in or braveness of professional character.

Compassion is a subtheme that included responses regarding the ability to be open to the suffering of others in the face of one’s own fear while providing care to alleviate suffering. Also mentioned was ‘not avoiding’ discussion on dying, showing compassion by answering questions honestly and objectively. In order to relieve suffering, the respondent told us they could be brave and compassionate in their ‘feelings, skills, thoughts and words.

Another subtheme was being aware of the concerns, fears, emotions and feelings of thepatient. The business of providing care and service in a hectic acute care hospital can mean that conversations and identification of needs can be kept at a ‘surface’ level. However, the respondents stated that a commitment to a deeper level of understanding of the dying patient’s needs, fears and feelings could be translated to their practice to improve the quality and safety of their practice. Not being afraid, as practitioners or clinicians, to use the words ‘dying’ or ‘death’ in conversations, acknowledging patient’s concerns and being in tune with the patient with a heightened sense of mortality were all translatable capacities that were stated by respondents. Increasing one’s empathy, another subtheme, was specifically mentioned by respondents as an area of practice improvement. Being able to embrace the patient’s experience or walking in other person’s footsteps was important. Such a heightened awareness can raise challenges even for the most experienced clinicians and this was captured in the subtheme controlling my own emotions. Translatable skills and the capacity to ‘step back’ reflect and attending to self-care and one’s own emotional state were offered by respondents as strengthening practice change points.

Dignity, valuing and respecting the patient for his or her own sake, was another subtheme where respondents specified they could change their practice. Being able to honour the personhood of the individual patient and their specific needs was a dignified approach to end-of-life care that could strengthen in professional practice. While no respondent admitted to being dishonest, honesty was mentioned as an area of practice change. Trying not to avoid using the word ‘dying’ was yet again mentioned by respondents. Also, respondents stated that answering questions and responding directly was a strengthening practice in quality end-of-life care—questions that arise in clinical end-of-life practice are commonly about the future (what will happen to me?), rather than skirting and dodging the question acknowledging it with a respectful and honest response is required.

Professional mindset

This theme included a variety of ways respondents intended to change their professional mindset. A subtheme was confidence in professional practice which held descriptions of intentions not to be scared or frightened, to be brave and truthful. The respondents stated that their approach to care could be altered to enhance these virtues which underlay quality end-of-life care. Such virtues that underpin quality care are held together by professional self-care and respondents described the importance of self-care, another subtheme. Proactive investment in self and energising practices were descriptors provided by respondents on how they could invest in their own capacity. Another subtheme was open-mindedness; respondents specified that they could alter practice by not being judgemental of various cultural and religious beliefs and respecting the viewpoints of others are ways that could improve practice. Upskilling was another subtheme developed to capture respondent’s descriptions of working towards excellence and developing practice over time. To learn, to develop and to educate others were all included in this subtheme. Also important is to recognise the qualities and capacity already held in clinicians—the respondents wrote it is important to pat yourself on the back while also aiming high for clinical practice improvement.

Person-centred care

This theme included a variety of ways respondents intended to strengthen their capacity to deliver person-centred care. Understanding the goals, needs and expectations of thepatient was a subtheme where intentions were collected about allowing patients to be heard, to be in tune to the patient’s needs and their wishes for the future. Goals of care are for and designed by the patient, not the healthcare teams. Being able to respect thepatient, their family and wishes is a subtheme that included a variety of ways respondents intended to improve their practice by empowering and respecting patients’ choices. Even if patients’ wishes are different from an individual clinician’s quality end-of-life care it embraces respect and upholds patient autonomy. Being able to separate thepatientfrom the illness was another subtheme based on respondents’ replies regarding the ways to acknowledge the person, not the illness or disease, an enabler of person-centred care. Supporting thepatientand the family members is a subtheme that collected the respondent’s descriptions of the various ways support can be offered at the end of life. This included providing symptom control, referral for spiritual or pastoral care, a listening ear and being a mediator when miscommunication happens. Taking time to be with thepatient was another subtheme based on the respondents’ replies, where allowing and giving time was mentioned as an area, they could change to deliver quality end-of-life care. Taking time to listen, to leave ‘to-do lists’ and just ‘be’ were mentioned as strengthening along with not offering quick reassurances to end-of-life issues. Allowing time just for silence was also offered as an area to improve appropriate quality of practice.

Professional practice

Professional practice was a theme developed to capture the specialised qualities of practice needed to deliver end-of-life care. Advocacy of thepatient was a subtheme where respondents wrote the importance of actively supporting the patient in team meetings or other fora where patient’s preferences and wishes are being overtaken by bureaucratic or organisational and professional preferences. Advocacy of the patient also included ensuring other HCPs can see the patient’s point of view. Acting quickly is a subtheme that includes the realisation that end-of-life care can be urgent care, where occurrences of suffering or uncontrolled symptoms means that escalation to other teams in the hospital can alleviate ongoing distress and offer appropriate treatments. Being able to question colleagues or hospital standard operating systems is another area where practice can be altered to improve care, and these subthemes were collected in asking questions. Understanding illness trajectories and not missing opportunities to broach and discuss the patients’ future care plans is a subtheme called identify end-of-life triggers. Telling the truth with kindness, as one respondent stated, can happen at opportune times that allows quality patient-centred care at the appropriate time. Respondents also wrote about not missing an opportunity to offer to discuss end-of-life issues with patients and triggers are key to this end.

Includingpatientsand families in thehealthcareteam is another subtheme where intentions to alter practice for the better were collected. Feeling like you are not alone and recognising and using a team where decisions and goals can be made respectfully have the potential from respondent’s perspective to alter practice for the better. Effective leadership was a subtheme used to collect the practice change intentions of respondents. Clear goals, clear roles, trust, respect and a cultural readiness to allow patients to steer care were all mentioned as practice change points that could be instigated.

Qualitative comments by theme are detailed in table 1.

Table 1

Intent-to-change end-of-life care practice as identified by 3200 healthcare professionals, by theme and subtheme


Overall, our findings suggest that most respondents thought they could change their practice for the better, nominating what they could do to improve appropriate care. There were no responses or suggestions that practice changes could not be implemented or that nothing could be changed. Respondents were generous with their comments, and optimistic that change could be achieved. One thing I can change was interpreted at an individual level as there was no reference to other HCP or the organisation itself had roles in these changes. Findings regarding the professional virtues of compassion, dignity, empathy and honesty underscore how important HCPs see these qualities in delivering end-of-life care.

Definitions of what constitutes quality end-of-life care vary and are subjective.3 19 A study of 49 acute care nurses in America20 detailed eight nursing actions to promote dying well: communication with disciplinary team/nursing staff, provision of optimal physical care, demonstration of caring and compassion, supporting dignity in death for patient/family, education of patient/family to support dying well, emotional support for patient/family, advocacy for dying well and fostering a peaceful environment. Conversely, a study conducted in Malaysia21 with 553 acute care nurses highlighted the main barrier to quality end-of-life care was the emotional work required in ‘dealing with a distraught family,’ while focus groups/interviews in acute care hospitals in the UK3 found a main deterrent to good quality care was in delays in diagnosing dying, all of which are reflected in themes or subthemes in our study.

The learners who have completed EOLE have shared the practical ways they state they can change their practice tomorrow (ie, intent to change), which could then be appreciated as a clinical response to the ACSQHC work in this space. EOLE has provided a way for HCPs in hospitals to learn more about the essential elements of end-of-life care (box 1), handed down by the Consensus Guidelines and expressed and interpreted through a suite of free online learning modules. Themes were subsequently derived from the evaluation data, demonstrating how clinicians have reinterpreted these essential elements and intend to use them in practice (table 1).


Communication is reported throughout and has been cited in the literature as a main factor that can influence and help improve end-of-life care in hospitals3 so its importance cannot be underestimated. Aspects of communication were highlighted by respondents; however, having open and honest conversations is something that many HCPs either struggle with22 or are unaware that this is wanted or required. End-of-life communication was raised as a barrier in a study with 533 Malaysian hospital nurses including the difficulties inherent in discussions with patients and families,21 evidenced in our study via a subtheme of ‘curb my own avoidance’.

Emotional insight

Emotional insight, which is missing from the Consensus Statement elements, was an important emerging theme, including discussions of prerequisites that enable safe and quality care to flow. The professional and, arguably, human virtues that respondents say are important when working in large bureaucracies include compassion, empathy and honesty, which are inherent in end-of-life care.23 HCPs in generalist areas are often unfamiliar with the importance of acknowledging the often strong emotions of others (staff, patients, families), and can more easily struggle with their own emotions, becoming susceptible to burnout.24 For example, it was insightful that some respondents were concerned that their own emotions would impact or over-ride that of the patients in emotional situations.25

Professional mindset

Expected deaths will continue to occur in hospitals and it is important that HCPs providing care strengthen their knowledge, confidence, open-mindedness and skills to provide best practice end-of-life care.3 The term ‘silent witnesses to care’, although in relation to student nurses in Canada facing difficult end-of-life situations where they feel unprepared and powerless to effectively engage with patients and families,26 arguably applies to many HCPs.

Person-centred care

Person-centred care sees HCPs acknowledging the patient as a partner in care planning and decision-making.27 In our study, respondents recognised the need to respect the wishes of patients and families, to support them and to take time to be with them. Sharma and colleagues28 indicate the need to get to know the person which has come across in our study as the need to separate the person from the illness. With an emphasis on patient-centred approaches to care, advocacy roles can be more easily facilitated as patients’ wishes and views are known,27 and this came across in respondents’ intentions.

Professional practice

Interestingly, teamwork and the need to include patients/family as part of this was highlighted, tying back again into person-centred care, which in turn is seen as integral to collaborative end-of-life care.29 Acute care clinicians can recognise dying, although this does tend to be at a late stage,11 so unsurprisingly this is conveyed in the results as the need for ‘triggers’, ‘trajectories’ and ‘tools and resources’ to help with this.

Practice change

It is important to note that intent-to-change practice, a form of human behaviour,30 cannot be measured in this study; however, theory on intention-behaviour relations indicates that intentions have a strong association with behaviour, and that the greater control we have over our behaviour the more likely it is that change will follow.31 The EOLE education is targeted at individual HCPs, recognising that this is the level at which they have most control, and in recognising that systems change is required to improve dying in hospitals.32 While HCPs self-select into the education and are empowered to improve their own practice, influencing change or implementing new initiatives even at a local level can be challenging, dependent on the culture of the team and work setting.33


People are living longer than ever before, often at the cost of living with debilitating multiple comorbidities requiring complex care or interventions (likely to require hospitalisation), which may well result in end-of-life care.12 34 The rhetoric around dying is that it is everyone’s business,35 but care of the dying in hospitals is undertaken by HCPs who are unlikely to have palliative care experience; be working in organisations that integrate a palliative approach to care,36 or necessarily refer to (or have access to) specialist palliative care.37

This study demonstrates the complex emotional nature of the end-of-life care work38 and its impact on the workforce at an individual, team and organisational level.39 Educators, managers, policy writers and government should appreciate the complexity of this type of care and, wherever possible, acknowledge, promote and support the workforce.40


Education is vital, however alone it does not bring about change.41 For EOLE, this is achieved in part by offering the education in conjunction with an implementation toolkit that provides learners with practical ways in which they can change their practice in an ongoing manner.14 Whether this change is sustainable in practice is still yet to be understood and this is an area open to further investigation.

It is also envisaged that timely reminders can be built into the learning management platform to prompt further module completion and engagement.42 The feasibility of local change champions is also under investigation, to support learners in practice and improve quality care. 43 There is also the need to directly ask learners who have completed the education about challenges to practice change, and an evaluation is currently underway to address this via an online survey and semistructured interviews.

Hospitals are complex bureaucracies which will need complex multifaceted drivers for change. Our research shows that, with education, individual clinicians can devise appropriate and honourable change elements for their practice. In addition, the ACSQHC has integrated these Elements of Processes of Care (used as the foundation of EOLE) into the National Safety and Quality Health Service Standards, Second Edition, thus elevating comprehensive care into the accreditation processes. Hospitals are now required to provide evidence in a quality assurance mechanism on end-of-life care.


Of the 5181 learner registrations, we received 3200 responses to the particular question. Answering the question was voluntary. However, the same kind of question was posed at the end of all six modules and respondents could have opted out, with questions viewed as repetitive if modules were taken sequentially. There are also limitations in that this is a self-report measure and learners may be motivated to change practice at the end of the education but less motivated when faced with any barriers to actual practice change. Practice change is governed by many drivers and barriers, including, but not exclusive to, workplace culture, policies and procedures, staffing numbers and mix of disciplines, and funding arrangements. We have reported on one element of a complex system—how individual staff identified how they can change their own practice.

Learners are also responding to a single question raised at the end of a learning module, where they could easily quote what they had just learnt, without critically analysing what it actually means for patient care. Some may well have done this, but others have looked closely to how they can change their clinical practice.

While the concepts highlighted in the education modules predetermine to some extent the results, the themes that emerged in coding were initiated by a research assistant with no prior knowledge of the module content.


The EOLE education modules have been well received and self-reported practice change intentions indicate how clinical care can be improved. This study also serves to demonstrate how the Consensus Statement, as a guide to improve quality and safety and the education derived from it, can influence an intent by HCPs to change attitudes, behaviour and practice, which in turn impacts end-of-life care and the patient experience. We report tangible actions and moral dimensions that provide a foundation to quality end-of-life care. These practice change actions and dimensions can provide examples for other clinicians who are keen to improve practice. The findings are also valuable to organisations and policymakers who can support positive change in workplace culture and are wanting to change healthcare outcomes and improve the quality of dying in hospitals. The human virtues of honesty, compassion, dignity and empathy are central to quality end-of-life care in our hospital systems and these virtues can be implemented into practice.


Georgia Middleton is acknowledged for her work in data analysis.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 40.
  42. 42.
  43. 43.


  • Contributors KD and NP were involved in the original project conception and grant. KD developed the education modules. DR and KD conceptually developed the paper, which was drafted by DR. All authors commented on, reviewed and approved the final draft.

  • Funding EOLE is funded by the Australian Government Department of Health.

  • Disclaimer The funding body and ACSQHC have no role in the development or writing of the article.

  • Competing interests KD and DR have received salary funding from the End of Life Essentials (EOLE) project. NP is employed by the Australian Commission on Safety and Quality in Health Care (ACSQHC).

  • Patient consent for publication Not required.

  • Ethics approval Ethics approval for the project was granted by the Social and Behavioural Research Ethics Committee at Flinders University, Adelaide, South Australia (Project 7012).

  • Provenance and peer review Not commissioned; internally peer reviewed.

  • Data sharing statement Data are available upon reasonable request.