Article Text

Download PDFPDF

Knowledge of and attitudes towards palliative care and hospice services among patients with advanced chronic kidney disease
  1. Sara N Davison1,
  2. Gian S Jhangri2 and
  3. Jonathan Koffman3
  1. 1Division of Nephrology & Immunology, University of Alberta, Edmonton, Alberta, Canada
  2. 2School of Public Health, University of Alberta, Edmonton, Alberta, Canada
  3. 3Cicely Saunders Institute, Department of Palliative Care, Policy and Rehabilitation, King's College London School of Medicine, London, UK
  1. Correspondence to Dr Sara N Davison, Division of Nephrology & Immunology, University of Alberta, 11–107 Clinical Sciences Building, Edmonton, Alberta, Canada T6G 2G3; sara.davison{at}


Objectives Palliative care is greatly underutilised for patients dying from advanced chronic kidney disease (CKD). This study explored CKD patients’ knowledge of, and attitudes to, palliative and hospice care.

Design and intervention This was a cross-sectional interview-based survey of stage 5 CKD patients.

Setting Both dialysis patients and those managed without dialysis were assessed in a university-based renal programme in Alberta, Canada.

Outcome measures The primary outcome was validated patient knowledge of, and attitudes, towards palliative and hospice care. Secondary analyses of associations between knowledge and potential explanatory variables such as patient demographics, living circumstances, self-assigned ethnicity, socioeconomic position and sources of knowledge were explored.

Results Only 22.2% and 17.9% of 436 patients surveyed possessed a correct understanding of palliative and hospice care, respectively, and perceptions were mostly negative. Sources of knowledge of these services rarely involved healthcare professionals (3.1%–7.7%). Multivariable analysis identified that white participants were more than 12 times more likely to have accurate knowledge than non-Caucasian patients, not accounted for by socioeconomic position. Accurate knowledge was also more likely among patients who had personal experience with palliative care, or obtained their knowledge firsthand from family and friends. After palliative and hospice care were described, the vast majority of patients felt these services were valuable and should be offered to patients with advanced CKD (87.8% and 89.7%, respectively).

Conclusions Based on study findings, CKD patient and family education about the benefits of palliative and hospice services is likely important in optimal use of these services.

Statistics from


Worldwide, the prevalence of non-cancer chronic diseases such as chronic kidney disease (CKD) continues to increase, driven by the tremendous burden of diabetes, hypertension and vascular disease. Despite great need, specialist palliative care is underutilised among patients dying from chronic non-malignant illnesses.1–4 Only 16%–30% of Canadians who die as a result of progressive or advanced diseases access palliative care programmes.5 Of those who do, the majority (85%) are cancer patients.2 ,3 Yet, 62% of all deaths result from chronic, non-malignant disease.3 Of these, only 4%–7% are referred for palliative care services. The situation in both the USA6 and the UK is comparable.7 ,8

Several possible explanations for the underutilisation of palliative and hospice care have been proposed and include a lack of knowledge and understanding of palliative care and related services among patients and their families, which may be socially and culturally patterned.9–11 In South Australia, a survey of 4000 households identified that 58.8% of those surveyed were able to offer a description of palliative care. However, understanding palliative care varied markedly between population subgroups. It was highest among female and younger respondents; Australian or UK/Irish born; those with degrees; those with higher incomes; and lowest among aboriginal populations.12 Although not explored in detail, a survey commissioned by the Center to Advance Palliative Care determined that the US public's self-reported understanding of palliative care is very poor.13 A more recent cross-sectional survey among adult cancer patients in the UK identified that only a fifth of respondents could provide an accurate definition of palliative care and that this knowledge was owned disproportionately by those who resided in materially affluent areas and who self-identified as being Caucasian compared with those who lived in materially deprived areas, and identified themselves as from minority ethnic groups.9

Little is known about the knowledge and awareness of palliative care and related services among those living with chronic non-cancer illnesses such as CKD. Despite evidence that demonstrates the value of palliative care among these patients, uptake is poor4 ,6 ,14 ,15 resulting in poor end-of-life care.16 The purpose of this study was to examine CKD patients’ knowledge of, and attitudes to, palliative and hospice care, and to examine associations between knowledge and potential explanatory variables such as patient demographics, including ethnicity, socioeconomic status and various sources of information.


Setting and participants

This cross-sectional interview-based survey involved prevalent patients with stage 5 CKD (estimated glomerular filtration rate <15 mL/min/1.73 m2) in the Northern Alberta Renal Program, a university-based renal programme that serves approximately 1500 stage 5 CKD patients in Alberta, Canada. Alberta has a fully integrated community-based model of palliative care that provides publicly funded access to palliative care services in the home and in all region healthcare institutions. Participants included patients on dialysis and those being managed conservatively (without dialysis); were older than 18 years of age; and able to complete the interview in English. Ethics approval was obtained from the University of Alberta Research Ethics Board.

Recruitment and data collection

Eligible patients were consecutively recruited as they presented to nephrology clinics between July 2009 and September 2011. After providing written informed consent, demographic variables were collected and patients completed the interview-based survey in English with qualified research staff. This was often done while on dialysis or attending a clinic. We estimated socioeconomic position using the median household income from each participant's postal code, based on 2006 Canadian census summary data.

The survey asked participants how familiar they are with the terms ‘palliative care’ and ‘hospice care’ on a 5-point Likert scale: 1=‘very familiar’, 2=‘somewhat familiar’, 3=‘unsure’, 4=‘somewhat unfamiliar’ and 5=‘very unfamiliar’. If participants reported being somewhat or very familiar with the term, their knowledge was validated by asking them to describe what the term meant to them. They were also asked the source of their knowledge. Participants were then given written and verbal descriptions of palliative and hospice care, based on WHO definitions.17 Using these descriptions, they were asked to share their views on the importance of offering these services to patients with advanced CKD.

Data analysis

Demographics were described as number and percentage or as a mean and SD. Descriptive statistics were used to explore the percentage of patients who believed they understood the terms ‘palliative care’ and ‘hospice care’. Differences in proportions were examined using a χ2 test.

A coding strategy was developed to analyse free-text comments from the open-ended questions.18 All responses were indexed independently to identify categories. Respondents’ knowledge was deemed accurate if they were able to verbalise key concepts; for palliative care, this meant care that was focused on quality of life and symptoms. For hospice, this meant similar care but for those who were very close to the end-of-life or dying. Merely stating ‘end-of-life’ was insufficient if the notion of ‘care’ was absent in their description. For example, stating, ‘it is just a place to go to die when there is ‘nothing else to be done’’ was not considered an accurate description of hospice care. Categories of inaccurate descriptions were discussed and compared and then reduced in number by grouping them together to form themes.19 Verbatim quotes from the comments were selected to illustrate the range and diversity of views.

Logistic regression univariate and multivariable analysis were used to examine associations between actual knowledge of palliative and hospice care and potential explanatory variables such as patient demographics and source of information. Multivariable models were adjusted by age. Analyses were conducted using SPSS 20 and a two-tailed p<0.05 was considered statistically significant.


A total of 436 of 576 approached eligible patients (75.7%) consented to participate in the survey and completed the interview. Patients’ demographics and familiarity with the terms palliative and hospice care and their sources of information are summarised in table 1. While 63.4% of people felt they were familiar with the term palliative care, only 22.2% were able to correctly describe it. Participants were less knowledgeable of hospice care and only 17.9% were able to correctly describe it. Participants’ sources of knowledge of palliative and hospice care rarely involved healthcare professionals (figure 1). The most common sources of knowledge were from family/friends and personal experience for both palliative and hospice care.

Table 1

Patients characteristics and knowledge of palliative and hospice care (n=436)

Figure 1

Source of palliative and hospice care knowledge.

The themes that emerged from participants’ descriptions of palliative and hospice care and examples of incorrect definitions are presented in table 2. Figures 2 and 3 present the frequency of these themes among descriptions of palliative and hospice care, respectively. The most commonly held view was that palliative care and terminal care were synonymous. However, participants’ conceptions of palliative care were also linked to location of care; they specifically associated this type of care with acute hospital settings during the terminal phase of someone's illness. Other common misperceptions included beliefs that palliative and hospice care were akin to nursing home placement, caring for the elderly, home care for basic activities of daily living, and aggressive curative or rehabilitative treatments (including critical care).

Table 2

Examples of incorrect descriptions of palliative care and hospice care

Figure 2

Themes for incorrect descriptions of palliative care.

Figure 3

Themes for incorrect descriptions of hospice care.

Caucasian patients compared with non-Caucasian patients were significantly more likely to be knowledgeable about palliative (26.4% vs 2.4%, OR=12.6) and hospice care (20.4% vs 6.9%, OR=3.43) (table 3). Patients who live in a higher-income neighbourhood were more knowledgeable about palliative care, but not hospice care. Patients whose knowledge resulted from (a) personal experience, (b) family and friends or (c) paper resources (ie, newspapers, books, brochures) were significantly more likely to have accurate knowledge about both palliative and hospice care. Results of a sensitivity analysis using self-reported familiarity instead of actual knowledge were similar (results not shown).

Table 3

OR and 95% CI of actual knowledge of palliative and hospice care

In the multivariable analysis, patients who self-identified as white/Caucasian were 12.4 times more likely to understand the term ‘palliative care’ compared with non-Caucasian patients, and those residing in assisted living were 3.5 times more likely to understand this term than those living at home with others. Similarly, accurate knowledge of palliative care was at least 4.0 times more likely among patients who had personal experience with palliative care, or obtained their knowledge firsthand from family and friends or from newspaper resources. In contrast, patients who lived at home alone were 2.1 times more likely to be knowledgeable about hospice care. Similarly, accurate knowledge of hospice care was significantly higher among those who possessed personal experience with these services (OR=8.77, p<0.001), or obtained their knowledge from family and friends (OR=14.2, p=0.005), newspapers (OR=4.66, p<0.001), or from television/media (OR=21.8, p=0.033). Gender and ethnicity had only borderline significance with knowledge of hospice care (results not shown). Female patients and those who self-identified as ‘white’ had greater knowledge of hospice care than male (OR=1.68) or ‘non-Caucasian’ (OR=2.73) patients. Knowledge of palliative care and hospice care was not different based on neighbourhood-income after controlling for other covariates.

After palliative and hospice care were described, more than 87% of participants felt that palliative and hospice services were valuable and should be offered to patients with advanced CKD. There were no differences associated with age, gender, marital status, living circumstances, ethnicity or neighbourhood-income with the exception that white participants were more likely to report hospice services as being valuable.


In this paper, we explore knowledge and understanding of palliative care and related services among patients living in Alberta, Canada with advanced CKD. Our findings must be viewed within the following limitations. Participants had to complete the interview in English, introducing a selection bias. White patients were slightly over-represented, as they comprised 83.4% of the study population compared with approximately 75% of the total Alberta dialysis population. Further, there is an assumption that only poor people live in areas of social and material deprivation and only those with greater resources live in corresponding wealthier areas. This can lead to the ‘ecological fallacy’—the mistake of assuming that there are no individual class differences within specified geographical locations.20 A closer examination of the social capital available to CKD patients may provide important complementary data to understand the complex relationship between living circumstances and awareness of local services.21

Even after accounting for these study limitations, our findings are clinically important and add to a small but growing literature on the underutilisation of palliative and hospice care for patients dying from chronic illnesses.1–3 Despite high physical and psychosocial symptom burden and evidence that demonstrate the value of palliative care among patients with advanced CKD, dialysis patients comprised only 2.9% of patients admitted to hospice with non-cancer diagnoses in a 2004 Report to the Congress from the USA.14 Although the number of hospice users has increased between 2000 and 2009, partly due to enrolment of patients with non-cancer diagnoses, CKD patients remain under-represented. Only 13.5% of dialysis patients in the USA who died between 1 January 2001 and 31 December 2002 used hospice/palliative care services.4 These patients were rarely admitted to hospice unless they had opted to discontinue dialysis. Based on 2010 Centres for Medicare and Medicaid Services death notification data, it is estimated that 22.9% of dialysis patient deaths in the USA were preceded by hospice care.22 However, in a recent study, only 5% of dialysis patients admitted to a tertiary referral hospital in the USA were referred to the hospital's inpatient palliative care service, and 54% of these patients withdrew from dialysis while in hospital.15 Specialist palliative care was only involved in 38% of advanced CKD patients who died in a tertiary referral centre in Ireland. These patients were referred late, a median of 12 days before death.23 In a US study, even when offered, 24% of dialysis patients for whom dialysis was discontinued refused a hospice referral suggesting they did not understand the benefits that hospice could provide. There was no difference by race.24

In this study, CKD patients possessed an inadequate understanding of what palliative and hospice care entails. For some, there may be a tension between receiving a life-sustaining treatment such as dialysis and contemplating palliative care. Healthcare providers who avoid discussing issues relating to suffering, symptoms, death and dying until late in the illness may perpetuate this. However, we have evidence that the vast majority of dialysis patients welcome the opportunity to engage in end-of-life discussions with their physician.25 It is possible that a similar phenomenon is present to explain why only 8.7% of patients receiving curative cancer treatment in a UK survey were able to provide an accurate description of palliative care.9

The most common belief among our participants, that palliative care and terminal care are synonymous, has also been described among healthcare providers.26 This may be propagated by gate-keeping of palliative care programmes that will not register patients who are in the early stages of illness. This may not be appropriate for patients with advanced CKD who frequently have tremendous palliative care needs for years before death.

Palliative care involves services aimed at improving the quality of life for patients facing a life-threatening illness, at any age and at any stage of illness. It can be provided together with other therapies that are intended to prolong life, such as dialysis. Hospice care is under the larger umbrella of palliative care, but is typically made available only to patients who are believed to be within months of death27–29 but shares the same philosophy as palliative care—to enhance quality of life and prevent suffering. When palliative care is understood more broadly to encompass patients from the time of diagnosis, depending upon patient-centred needs rather than expected survival, unpredictable illness trajectories should no longer be a barrier to referral and use of appropriate services.

Even among participants who understood that palliative and hospice care were somehow related to the end-of-life, there was a lack of appreciation of the types of services provided and of the benefit to patients and family. Perceptions were often negative and the concept of ‘care’ was missing from most descriptions. There was also a lack of appreciation that palliative and hospice care can be provided anywhere—at home, in a residential hospice, in hospital or in a long-term care facility. The notion of palliative and hospice care as services primarily for cancer patients was not a prevalent belief among study participants.

Our findings of poorer knowledge among CKD patients from non-Caucasian backgrounds have been well documented in other areas of service provision. Lower levels of utilisation of health promotion activities among minority ethnic communities and less affluent sections of society are mirrored by poor knowledge of available services.30 In palliative care, two recent qualitative studies have alluded to inadequate knowledge about end-of-life care among minority ethnic communities living in the USA and UK.31 ,32 Although various reasons have been postulated as to why the use of palliative care is lower among certain population groups,1 it is commonly regarded that the influence of socioeconomic position is central to understanding the processes of observed ethnic inequalities in relation to healthcare.33 However, for these CKD participants, socioeconomic position did not explain the substantially poorer knowledge of palliative and hospice care service among non-Caucasian patients. Other variables beyond the scope of this study such as health literacy may have a role to play and warrant further investigation.

These findings progress palliative care studies that only controlled for ethnicity.34 ,35 The data we present identify that ethnicity alone, while a significant factor, did not account for all disparity in palliative and hospice care knowledge. Prior experience with palliative care services, either through professional or personal lives, was the predominant source of accurate knowledge. Only a small proportion of patients were informed about relevant palliative care services from healthcare professionals. Poor knowledge is possibly compounded by healthcare provider confusion about palliative and hospice care, and their relevance to their CKD patients.

These CKD patients poorly understood palliative and hospice care, yet had very positive attitudes towards these services once they were provided with a description. Their misconceptions about the philosophy and goals of care may partially explain the underutilisation of palliative care. Opportunities for healthcare professionals to confidently and competently engage with patients on appropriate palliative care services to address physical and psychosocial needs are urgently required. Improved integration of palliative care and nephrology will likely require better patient, family and healthcare professional education resulting in an appreciation of the relevance and benefits of palliative and hospice care.


Supplementary materials

Related Data


  • Contributors SND is responsible for the overall content of this manuscript as guarantor. She planned, conducted and reported the study. GSJ did all the statistical analyses and contributed to the writing of the manuscript. JK contributed to the initial development of the study and was involved in the writing of the manuscript.

  • Funding Canadian Institutes of Health Research. Grant number 89801.

  • Competing interest None.

  • Ethics approval University of Alberta Research Ethics board.

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

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.