Background
Tens of thousands of people in North America experience homelessness every year [
1,
2]—that is, live in conditions unfit for human habitation or temporary or emergency accommodations without housing alternatives [
3]—and many thousands more are at risk of homelessness at any given time [
1,
2]. Homeless populations in North America, while highly heterogeneous in terms of age, gender, sexuality, and ethnicity, suffer disproportionately from poor health in comparison to housed populations [
4]. Chronic health conditions that are common among the homeless include chronic lung diseases [
5], circulatory diseases [
6], and diabetes [
7]. Homeless persons also experience higher incidences of substance use [
8,
9], severe mental illness [
10,
11], and infectious diseases such as HIV/AIDS [
12,
13] and Hepatitis C [
14]. Daily challenges associated with homelessness (e.g. food insufficiency, exposure, etc.) [
4,
15,
16] and barriers to accessing health care services (e.g. discrimination, lack of insurance, etc.) [
4,
17,
18] make it difficult to manage medical needs, leading to further deteriorations in overall health. Homeless populations subsequently have among the highest all-cause mortality rates of any population in North America [
19‐
23].
Homeless persons have a high level of need for end-of-life care services [
24,
25] and these needs may be increasing due to the steady growth in the number of homeless older adults [
26,
27]. It is estimated that more than 58,000 seniors (i.e. 62 years or older) will experience homelessness annually in the US by 2020 [
26] and, while estimates are not available for Canada, researchers in various cities have observed upward trends [
27]. High levels of morbidity among homeless older adults [
28], in combination with the natural progression of health challenges common among this population (e.g., HIV/AIDS, HCV, etc.), suggest that the end-of-life care system will likely see an increased demand for its services among the homeless in the immediate future.
While the demand for end-of-life care services may be growing among the homeless in North America, this population faces many barriers to accessing end-of-life care services [
24,
25,
29,
30]. In North America, the end-of-life care system is largely premised on a series of assumptions that do not reflect the experiences and circumstances of homeless populations. Specifically, the end-of-life care system generally assumes that prospective clients are housed, supported by family and friends, and able to pay for supplementary care. In Canada, where our research was conducted, hospice and palliative care services are underdeveloped [
31] and are structured in ways that limit access for homeless populations. For example, existing service structures emphasize family caregivers and dying-in-place (i.e., the home) [
31,
32]. Accordingly, in many regions, end-of-life care services are oriented toward providing home care support and potentially limit access for homeless or precariously housed persons. Hospice and hospital-based end-of-life care services are also available to provide an additional source of care in many communities, especially in urban centres [
31]. However, homeless populations are often unable to access hospice or hospital-based end-of-life care due to rules and regulations (e.g. anti-drug policies, codes of behaviour, etc.) that exclude substance-using populations [
29,
30]. For example, research suggests that, among those who continue to use alcohol or illicit drugs at end-of-life, abstinence-only policies and discrimination within hospice and palliative care settings serve as a significant barrier to care [
29,
30]. Furthermore, homeless persons who use alcohol or drugs may also refuse referrals to hospice and palliative care services due to a range of factors, including real or perceived discrimination in these settings or the preference to die in a familiar environment (e.g., emergency shelter, hostel, etc.) [
30].
Due to these barriers, homeless persons often die with poor health care support [
33] and without accessing the end-of-life care system [
34]. Song et al. [
35] have reported that these barriers may serve as a source of anxiety among homeless persons—namely, that they might have poor access to necessary care (e.g., pain and symptom management) due to financial barriers. Several ways to improve end-of-life care services delivery to homeless populations have been previously identified, and have included individual-level and environmental recommendations. For example, researchers have recommended that advance care planning be undertaken with homeless persons and noted that this population is willing to document its end-of-life care preferences [
35‐
39]. Researchers have also documented the benefits of emergency shelter-based end-of-life care services delivery, including cost savings [
24] and cultural competence [
24,
30,
40]. And yet, research has not been conducted that has explored ways to improve the end-of-life care system as a whole for homeless populations. Research providing systems-level recommendations is urgently needed to identify structural changes that have the potential to increase access and equity in end-of-life care services for homeless populations.
This article presents recommendations for improving the end-of-life care system for homeless persons based on research conducted in six Canadian cities as part of a national study of homelessness and end-of-life care. The main objectives of this study were to identify barriers to end-of-life care services delivery to homeless persons and identify recommendations to improve the end-of-life care system for this population. The findings presented here take into account the perspectives of health and social services professionals providing care to homeless persons at end-of-life. While this study was carried out in a country with universal health insurance, our findings provide insights that may strengthen end-of-life care services delivery to homeless persons elsewhere given the barriers they face to accessing care even when healthcare coverage is available [
41].
Methods
Study design and participants
We conducted qualitative interviews with health and social services professionals in six Canadian cities between February 2007 to August 2008 in which we explored the social and structural factors that impact end-of-life care services delivery to homeless persons. In particular, we aimed to identify barriers to providing end-of-life care services to this population, as well as identify recommendations improve equity in the end-of-life care system. We invited health and social services professionals involved in end-of-life care services delivery to homeless persons in Halifax, Ottawa, Hamilton, Toronto, Thunder Bay, and Winnipeg to participate in this study. An advisory committee comprised of regional experts (i.e., senior health and social services administrators for homeless service organizations), as well as existing relationships with homeless service organizations in Ottawa and Toronto, helped us to identify key informants in those cities. We identified key informants in the remaining cities by conducting a scoping review of health services for homeless persons in those cities.
Letters or emails were sent to seventy-three potential participants to provide them with information about the study and invite them to participate. Fifty-four individuals (74%) agreed to participate, representing a range of professional roles, including physicians (6), nurse practitioners (2), nurses (16), social workers (5), emergency shelter or supportive housing executive directors and/or senior managers (9), harm reduction specialists (5), outreach workers (7), and personal support workers (4). Fifteen participants worked in low-threshold hospices (i.e. allowing onsite alcohol use, providing clean syringes, and permitting off-site illicit drug use) and the remainder of participants worked primarily in other community settings where they provided care to homeless persons unable to access the end-of-life care system. A minority of participants (6) worked in both hospital and community settings but reported that they seldom provided end-of-life care to this population in hospital.
Data collection
Semi-structured qualitative interviews were conducted with participants at their workplace or alternate location of their choosing. The majority of interviews were conducted by the lead author (RM), a qualitative health researcher, while the remaining interviews were conducted by the study principal investigator (MGY), a clinical psychologist. The interview guide focused on the end-of-life care needs of homeless persons and barriers and facilitators to providing end-of-life care to this population. Participants identified few facilitators and instead tended to make recommendations to improve care. Participants were asked throughout the interview to identify strategies to improve the end-of-life care system for homeless persons. An abridged version of this interview guide is provided in Table
1. Interviews were audio recorded and ranges in length from 45 to 120 minutes, although the majority were approximately 60 minutes in length. Interviews were transcribed verbatim by research assistants and we conducted a data fidelity check by reviewing transcripts while simultaneously listening to the audio files.
Table 1
Examples of interview questions
Questions
|
Probes
|
What do you feel prevents homeless persons from receiving referrals to palliative care services? |
Why do you feel they are not referred to:
|
▪ Hospice care?
|
▪ Hospital-based care?
|
What do you feel prevents homeless persons from receiving referrals to palliative care services? |
How do you feel their access is limited by:
|
▪ Real or perceived discrimination?
|
Alcohol or illicit drug use?
|
What do you feel are the challenges to palliative care services delivery to homeless persons? |
How is palliative care services delivery to homeless persons adversely affected by:
|
▪ Alcohol or illicit drug use?
|
▪ Mental illness?
|
Topic Area: Facilitators to accessing the end-of-life care system
|
Questions
|
Probes
|
What do you feel facilitates access to palliative care services for homeless persons? |
What makes it easier for them to access palliative care services?
|
What do you do to help homeless people access palliative care services? |
Can you describe when you helped someone access palliative care services?
|
How did you help them?
|
Probes: Ask throughout as appropriate
|
Questions
|
What changes do you feel would minimize the impact of [name of barrier]? |
What types of programs do you feel would minimize the impact of [name of barrier]? |
Data analysis
The goal of analysis for this article was to identify perceived barriers to the end-of-life care system for homeless persons, as well as recommendations to improve services for this population. We imported the transcripts into NVivo qualitative data analysis software (version 8) to facilitate coding. A preliminary set of three categories (e.g. access to end-of-life care, community partnerships, and education and training) was extracted from lead author’s field notes and used to provide an initial framework for the analysis. Two of us (RM & LBD) independently coded the data by drawing on constant comparison methods, wherein preliminary categories were revised and emerging categories were identified and expanded through constant comparison to the data [
42,
43]. We regularly met to discuss emerging categories, with any revisions to the coding framework made by consensus. All authors discussed emerging themes to aid in framing the findings in relation to existing literature. Once the final categories were established, one of us (RM) re-coded sections of the data to ensure the credibility of these categories.
Ethics
This study was approved by the research ethics committees at the University of British Columbia and Saint Paul University. Informed consent was obtained prior to interviews and participants retained a duplicate copy of the informed consent protocol.
Discussion
Previous studies have identified interventions to improve the end-of-life care services for this population, highlighting the benefits of completing advanced planning directives [
35‐
39] and emergency shelter-based end-of-life care [
24,
29], but have not focused on larger changes needed to the end-of-life care system. Our findings add to the literature by identifying systematic barriers to the end-of-life care system for homeless persons and articulating recommendations that have the potential to improve access and equity in this system for this population. Our study draws attention to the fact that poor access to the end-of-life care system is in part due to the fact that this system is in underdeveloped in Canada. Presently, the Canadian end-of-life care system is struggling to meet the demand its services due to the combination of an aging population and low levels of government investment in these services [
44‐
46]. It is estimated that as few as 16% to 30% of people in Canada who die receive end-of-life care services [
44,
46]. It is apparent that a critical step of improving access to end-of-life care services for homeless populations is improving overall access to end-of-life care services through increased investment in this area.
Our findings also suggest that homeless populations encounter additional barriers to accessing the end-of-life care system that point to the need for changes in the organization and operating policies of this system. Across Canada, policymakers have increasingly championed death-at-home as the optimum outcomes of end-of-life care system and this desired outcome [
46,
47], which may inform how services are organized (e.g., home care or pay-for-service assisted living facilities). As a result, the end-of-life care system in some regions may not be developed in a way that accommodates services delivery to people who are homeless or marginally housed (e.g., living in substandard or transitional housing). A range of strategies, including hospital and non-profit hospice care, are be needed to ensure that those unable to access home care services or assisted living facilities have access to end-of-life care.
Furthermore, our research underscores the need for changes to operating policies within the end-of-life care system in order to better accommodate people who use drugs and/or experience mental illness. Our findings suggest that harm reduction strategies are perceived as a potential strategy to improve access to end-of-life care services for homeless persons because they remove one of the main barriers to hospice and end-of-life care services (i.e. anti-drug policies). Researchers have previously acknowledged that harm reduction strategies improve end-of-life care services delivery to homeless populations [
24,
29]. For example, Podymow et al. [
24] found that integrating harm reduction approaches into a shelter-based hospice (i.e. permitting onsite alcohol use, providing sterile syringes, and permitting off-site illicit drug use) decreased overall healthcare costs by reducing the need for hospital and emergency medical services. More recently, researchers have observed that harm reduction services play a critical role in mediating access to end-of-life care services [
29,
30] and have called for the integration of supervised drug consumption services (e.g., permitting the use of pre-obtained illicit drugs under medical supervision) into end-of-life care services [
29,
30]. These strategies warrant careful consideration and further research is needed to identify the strategies or combination of strategies (e.g. syringe exchange and distribution, methadone maintenance treatment, medically-supervised drug consumption services, etc.) that best mediate access to the end-of-life care system for this population.
Our findings further emphasize the need for improvements in continuity of care and mental health and substance use training. The end-of-life care system may benefit from replicating interventions (e.g. intensive case management, integrated services, etc.) [
48,
49] shown to enhance continuity of care for homeless populations. In particular, patient navigators (i.e. trained peers or healthcare professionals who work with clients to help them overcome barriers to health care services [
50]) might serve as important advocates for homeless persons as they try to navigate the end-of-life care system and help minimize the impact of discrimination and/or exclusionary policies [
51]. Furthermore, formal links between end-of-life care and public health services (e.g. community committees) might enhance collaboration. Finally, our findings echo those previous studies by identifying a need for increased training in mental health and substance abuse among end-of-life care professionals [
24,
52].
Limitations
This study has several limitations that should be taken into consideration. Our findings may have limited generalizability due to limited sample size. Also, several recommendations may have limited generalizability to settings that lack universal healthcare coverage. Participants were recruited largely from community settings and our findings only partly reflect changes necessary to improve mainstream end-of-life care services delivery to the homeless. Further research with mainstream end-of-life care providers is needed to get their perspective on end-of-life care services delivery to this population, and in particular why homeless populations are underserved by this system. Finally, this article is not based on interviews with homeless persons who had experience with the end-of-life care system. We acknowledge that further work is needed to explore this population's perceptions of and experiences with the end-of-life care system. We completed preliminary interviews (n = 5) with homeless persons receiving care at a low-threshold hospice but suspended this part of our study due to concerns about the quality of data (e.g., consistency of accounts, recall of events, etc.). Future research, and especially that aiming to identify ways to improve the end-of-life care system, could benefit from better drawing upon the experiences of homeless end-of-life care recipients.
Conclusions
This article documented health and social services professionals views of the barriers that homeless persons face to accessing the end-of-life care system, as well as recommendations to improve access to this system for this population. While participants identified several barriers (i.e., insufficient availability of services, exclusionary operating policies, and poor continuity of care), they made key recommendations for improving the end-of-life care system for this population. In particular, participants identified the importance of structural changes to the delivery of end-of-life care services, emphasizing the importance of expanding services, integrating harm reduction approaches, and fostering partnerships with the public health system. These observations have the potential to be translated into policy and programmatic responses, notably the expansion of end-of-life care services, implementation of patient advocate programs, and adoption of harm reduction policies. For policymakers and health administrators concerned with increasing equity in the end-of-life care system for homeless populations, these recommendations present a possible way forward.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
MGY and RM designed the study and conducted data collection. All authors contributed to the data analysis. RM wrote the first draft of the manuscript. All authors contributed to the critical revision and approved the final content.