Background
Methods
Design and eligibility criteria
Searches
Study identification and data extraction
Analysis
Critical appraisal of the methodological quality
Reference | Aim | Country | Setting | Method | Participants | N | Informa-tion on 1,2, 3,b | Critical appraisal scorec |
---|---|---|---|---|---|---|---|---|
[31] | To know more about the content of advance directives completed by homeless people who participated in a guided intervention arm | USA | Homeless drop-in shelter | Qualitative analysis of participants’ responses to individual items in an advance directivea | Homeless people with a terminal illness | Homeless people (n = 17) | 1, 2, 3 | 24/20 Moderate |
[32] | To identify the observed changes in general condition or behaviour of homeless people with advanced liver disease who may be in deteriorating health and approaching the end of life in order to better recognize an increased likelihood of death and to explore staff’s experiences of death of residents | UK | Homeless shelter | Case note review, focus groups (qualitative) | Case notes about homeless people with advanced liver disease, staff members of a supporting home for homeless people | Case notes (n = 27) staff members (n = 13) | 1, 2, 3 | 22/21 Moderate |
[37] | To explore the staff members’ experiences of and reasoning about the palliative care they provided | Sweden | Support home for homeless | Paired and individual conversations (qualitative) | Staff members of a support and housing home for homeless | Staff members (n = 12) | 1, 2, 3 | 34/31 Good |
[49] | To describe challenges of caring for homeless veterans at end of life as perceived by Veterans Affairs Medical Centre (VAMC) homeless and EOL care staff | USA | Veterans Affairs Medical Centres (VAMC) with programmes for homeless veterans with a short life expectancy | E-mail survey (quantitative) | Care staff of homeless and EOL programmes | 50 VAMCs | 2 | 28/23 Moderate |
[43] | To assess the extent to which homeless persons may underuse healthcare services even when they are at a high risk of death and to examine potential opportunities for intervention in this population | USA | Boston Health Care for the Homeless Program | Review of medical records (quantitative) | Deceased homeless adults | Medical records (n = 558) | 2, 3 | 27/25 Moderate |
[23] | To explore the views, concerns, and needs regarding advance care planning among older homeless adults | USA | Transitional housing facility | Semi-structured face-to-face interviews (qualitative) | Homeless adults aged 60 and older | Homeless (n = 21) | 1, 2, 3 | 30/29 Good |
[24] | To explore older homeless adults’ perspectives toward good and bad deaths and their concerns regarding their EOL care needs | USA | Transitional housing facility | Semi-structured face-to-face interviews (qualitative) | Homeless adults aged 60 and older | Homeless (n = 21) | 1, 2, 3 | 30/33 Good |
[36] | To explore how access to Toronto’s palliative services can be improved to better serve the city’s homeless | Canada | Providers of care for the homeless | Semi-structured interviews (qualitative) | Homeless care providers with extensive experience and experience dealing with death and dying | Registered nurses (n = 3) outreach workers (n = 4) | 1, 2, 3 | 19/18 Poor |
[41] | To determine the rate of advance directive completion using a one-on-one counsellor-guided intervention | Canada | Shelter for homeless men | Counsellor-guided intervention (quantitative)a | Chronically homeless individuals in a managed alcohol harm reduction programa | Shelter residents (n = 205) | 1, 2 | 31/33 Good |
[47] | To identify best practice for managing the palliative care needs of clients experiencing homelessness in a community setting and to guide the development of policies for a community-based palliative care service working with these clients | Australia | Community-based palliative care service | Semi-structured individual interviews (qualitative) | Workers from hospital and community organizations | Staff members (n = 6) | 2, 3 | 27/24 Moderate |
[30] | To explore and describe aspects of social networks that have a potential for caregiving during the terminal phase of a disease | USA | Patients of two medical centres, living in single room buildings | Semi-structured individual interviews (qualitative) | Homeless who had been diagnosed with unresectable lung cancer | Homeless (n = 8) | 2 | 21/19 Moderate |
[26] | To identify challenges health and social service providers face in facilitating and delivering end-of-life care services to homeless illicit drug users | Canada | Health and social care services | Semi-structured individual interviews (qualitative) | Health and social services professionals involved in end-of-life care services delivery to homeless persons | Healthcare professionals and managers (n = 50) | 2, 3 | 29/31 Good |
[27] | To identify barriers to the end-of-life care system for homeless populations and generate recommendations to improve their access to end-of-life care | Canada | Health and social services | Semi-structured individual interviews (qualitative) | Health and social services professionals involved in end-of-life care services delivery to homeless persons | Healthcare professionals and managers (n = 54) | 2, 3 | 32/35 Good |
[25] | To explore the role of harm reduction services in end-of-life care services delivery to homeless and marginally housed persons with problematic use of alcohol and/or illicit drugs | Canada | Health and social services | Semi-structured individual interviews (qualitative) | Health and social services professionals involved in end-of-life care services delivery to homeless persons | Healthcare professionals and managers (n = 54) | 2, 3 | 32/33 Good |
[45] | To determine the benefits and barriers of in-shelter palliative care and possible enablers to future implementation in Toronto | Canada | Three shelters | Semi-structured individual interviews (qualitative) | Shelter-based social service providers | Case workers, social support workers, shelter managers (n = 5) | 2, 3 | 23/19 Moderate |
[40] | To examine the treatment preferences of homeless (in comparison with preferences of physicians likely to be providing care for homeless persons and patients with oxygen-dependent COPD) | USA | Homeless shelters, hospitals | Cross-sectional survey (quantitative) | Visitors of homeless shelters, physicians providing care to homeless persons, patients with COPD | Homeless (n = 229), physicians (n = 236), COPD-patients (n = 111) | 1, 3 | 31/32 Good |
[44] | To improve the understanding of elderly homeless persons and to describe the living circumstances of the group, especially housing | USA | Multidisciplinary Street Team of Boston | Analysis of an interventiona (quantitative) | Elderly homeless individuals (> 50) | Homeless (n = 30) | 2, 3 | 15/13 Poor |
[48] | To explore if effective shelter-based palliative care could be provided to terminally ill homeless individuals at substantial cost savings | Canada | Shelter-based palliative care hospice | Analysis of records of a cohort and a five-member panel (quantitative) | Terminally ill homeless | Records of homeless (n = 28) | 2, 3 | 32/30 Good |
[38] | To explore the importance of end-of-life care for homeless people and the type of concerns | USA | Sites for homeless in Minnesota | Focus groups (qualitative) | Homeless individuals | Homeless (n = 57) | 1, 2, 3 | 18 Poor |
[33] | To understand the viewpoints of people who are homeless regarding end-of-life issues, to elucidate the barriers to good end-of-life care, and to offer insight into the most basic needs and wishes. | USA | Homeless shelter, two service organizations for homeless | Focus groups (qualitative) | Homeless individuals and social workers | Homeless (n = 11) service providers (n = 9) | 1, 2, 3 | 23/24 Moderate |
[29] | To explore the experiences and attitudes toward death and dying among homeless persons. | USA | Social service agencies which serve homeless | Focus groups (qualitative) | Homeless individuals | Homeless (n = 53) | 1, 2, 3 | 30/35 Good |
[28] | To examine how homelessness influences concerns and desires about care at the time of death. | USA | Social service agencies which serve homeless people | Focus groups (qualitative) | Homeless individuals | Homeless (n = 53) | 1, 2, 3 | 29/32 Good |
[46] | To improve the EOL decision-making process for homeless persons by facilitating ACP | USA | Drop-in centre | RCT comparing two types of interventionsa (quantitative) | Homeless individuals | Homeless (n = 59) | 2, 3 | 32/35 Good |
[42] | To determine whether homeless persons will complete a counselling session on advance care planning and fill out a legal advance directive designed to assess care preferences and preserve the dignity of marginalized persons | USA | Sites serving homeless persons | RCT comparing two type of interventionsa (quantitative) | Homeless individuals | Homeless (n = 262) | 1, 2, 3 | 31/35 Good |
[39] | To increase healthcare providers’ understanding and insight into how to better provide EOL care for homeless people. | USA | Free urban healthcare clinic for homeless individuals | Focus groups (qualitative) | Homeless individuals | Homeless (n = 20) | 1, 2, 3 | 31/28 Good |
[34] | To identify and examine the needs of older people who are homeless or who have previously experienced homelessness as they age and are faced with the issues of serious ill health, dying and death. | Ireland | Community where care, accommodation and support are being provided for people experiencing homelessness and those at risk | Interviews (qualitative) | Homeless individuals | Homeless (n = 16) | 1, 2, 3 | 22 Moderate |
[35] | To explore the views of hostel staff regarding palliative and end-of-life care for the homeless population | UK | Intermediate or long stay hostels | Semi-structured individual interviews (qualitative) | Hostel workers | Hostel workers (n = 7) | 1, 2, 3 | 28/33 Good |
Results
Review selection
General characteristics of studies
Setting and participants
Concerns, care needs and future preferences for care and treatment of seriously ill homeless people
Concerns | Care needs | Preferences for future care and treatment |
---|---|---|
Physical domain • Concerns about serious illnesses and physical distress related to specific illnesses, e.g. heart disease, open heart surgery, multiple broken bones [28, 29, 31] • Concerns about losing control over basic physical functions [24] • Concerns about being off medication [31] Psychological domain • Fear of death and dying, partly due to bad and lonely deaths of other homeless people [24, 28, 29, 32‐34, 37] • Concerns about psychiatric disorders, in particular schizophrenia, mental illness, depression, affective disorder, anxiety, hearing voices, PTSD, bipolarity, uncontrolled anger [24, 31] Social domain • Worries about relationships with friends and family, e.g. family not being notified, leaving a wife and children behind, lack of resources to cover burial costs, being alone, family may not show up [31, 33] • Fear that family may not know wishes, peers might help to a certain extent, but no assumptions of this help [33] • Concerns about being homeless [31] Spiritual domain • Fear of the unknown [31] • Fear that the death rituals for their culture may not take place [33] Care domain • Many patients had bad experiences from previous healthcare and social service encounters, homeless persons believe that care will be poor at the end of life [23, 29, 36, 37] • Concerns about lack of insurance and receiving sub-optimal treatment due to discrimination by HCP’s/insurance companies [31, 39] • Concerns about what will happen to the body after death, fear that their body will not be respected or taken care of [28, 33] • Homeless people who completed an advance direction worry more about the care they would receive if seriously ill or dying [36] • Fear of what will happen if no-one can speak for them [33] • Fear of being transferred to a nursing home [34] | Attitudes/behaviour of healthcare professionals • Homeless patients want to be treated with respect and dignity, e.g. treat patients like others, no judging/labelling, accept patients for who they are [28, 31, 38] • Wish for companionship at the end of life, seeking relationship-centred, compassionate care [28, 39] • Acknowledging emotions; many homeless people have experienced tremendous losses in life. Intensifying of emotions could interfere with participants’ future decision-making process [39] • Providers who tell the truth [31] • Providers who respect privacy [31] • Providers should recognize cultural differences, this will serve as the basis for increasing sensitivity and trust [23] • Death and dying are perceived to be temporary matters, and many thought dwelling on the end of life situation was undesirable [23] • Patients prefer to use a GP who specializes in the care of the homeless [32] Involvement of family • Some of the homeless persons want family nearby, others (often a majority) do not want to burden their families [28, 38] • Requests for some form of social contact with family and friends and resolving remaining issues and disagreements before dying even if they were estranged [24, 32] • Participants who are not in contact with their family desire to be placed in a familiar environment where they could be surrounded by a social support network [24] Treatment/care options • Desire for advance care planning/documentation; this relates to several concerns (anonymity, estrangement, maintaining control, discussion with significant others), with trust as an important condition [28, 29, 33, 39] • Requests for detoxification [32] • Patients predominantly interact with GPS for prescriptions [32] • End-of-life care focus on pain control [28] • Asking how they would like to be remembered, including post-death wishes [31] After death • Explicit and detailed desires that homeless people’s bodies be laid to rest in a personally and culturally acceptable manner (due to the misconceptions and fears about body disposal) [28] | Treatment preferences • Resuscitation: - Almost all homeless persons expressed a preferences to receive cardiopulmonary resuscitation (CPR) in the event of cardiorespiratory arrest if there was a chance of returning to their current state of healtha [41] - Homeless people want resuscitation more than physicians and patients with COPD [40] - Homeless men are more likely to want resuscitation than homeless women [40] - Non-white homeless people are more likely to want resuscitation or life-sustaining treatment than white homeless people [40, 42] • Life sustaining treatment: - Nearly half of the homeless participants (8/17) indicated that they would want all measures taken, a smaller proportion (7/8) would prefer limited treatment [31] - Between 20% and 37% want life-sustaining treatment depending on condition (lowest in case of dependence, highest for unconsciousness [42] - 31% desired no life-sustaining treatment if dying [42] - In the scenario of a permanent coma or severe dementia, homeless people are more likely to want CPR or mechanical ventilation than physicians [40] Wishes for the dying process • A natural death (dying in sleep, no artificial medical interventions to prolong life, avoiding heroic measures such as prolonged life support without hope of functional recovery) [24, 33, 38] • Homeless people want to have their wishes represented when they become incompetent and/or dying [23, 39] • Dying peacefully, taking care of inner conflicts, being able to express love, apologizing to family and others [24] • Death without suffering [24] Proxy decision-makers • A significant proportion of homeless people named a proxy decision-makera [41] • Nearly all chosen surrogate decision-makers were not related; most often they were service providers, friends or (occasionally) romantic partners [28] • 29% to 34% of homeless participants showed a (written) preference for surrogate decision-making [42] • 87% of homeless participants named a family member as a surrogate decision-maker in their completed advance directives [42] |
The care provided: barriers and facilitators
Relating to the homeless people | Relating to the interaction between homeless people and healthcare professionals | Relating to the healthcare professionals |
---|---|---|
In relation to receiving healthcare • End-of-life care is not a priority; to obtain the basic necessities of survival and living on a day-to-day basis takes precedence over efforts to obtain health and/or end-of-life care [23, 26, 34, 38, 43] • Drug and/or alcohol dependence and non-disclosure of illicit drug use may lead to decreased opportunities for persons to remain in their usual abode or to receive and/or adhere to treatment at traditional end-of-life services due to anti-drug policies [26, 27, 34, 47] • Planning care activities and attending for hospital appointments is often difficult: patients frequently do not adhere to expected routines, arrangements for health service activities, GP and hospital appointments and often have to be reminded about their condition, homeless people are reluctant due to a long waiting time and/or they self-discharge [32, 34, 37] • Very late stage of seeking help and thus medical problems that are difficult to handle and multiple admissions before death [37, 43, 47] • Pain and symptom management of homeless persons who use illicit drugs (high levels of opioid tolerance) and specialists who are unable or unwilling due to fears that they would be liable for adverse reactions [26] • Lack of health insurance [43] In relation to social relationship • No support from family members or relatives and small networks and many without trusted peers [30, 33, 37] • A lot of homeless people who have psychiatric illness and are paranoia, refuse multiple offers of housing [44] • Travel and access to transport when living in a rural area [34] • Relationships between healthier and sicker patients are complex and sometimes manipulative to gain access to further alcohol [32] • Death and dying does affect other homeless patients [32] In relation to (health) behaviour • Problems relating to alcohol and/or drug addiction, such as denial of addiction, bingeing, ignoring of risks of overdose [32] • Aggressive or changing behavior [32] • Unwillingness to pay attention to their personal hygiene [32] | • Feelings of being ignored, discriminated and disrespected by healthcare providers and a lack of trust and suspicion (e.g. shown disrespect, withholding of pain medication, inappropriately short hospital stays, not respecting wishes) that initially has to be overcome before any treatment could be started [26, 33, 35‐38, 45] • End of life is an uncomfortable topic; some homeless persons do not want to know about their own diagnoses, do not want to talk about their health concerns or are incapable of talking comfortably about death and dying [23, 35, 36] • Barriers to achieving the level of communication and connections homeless people desired, e.g. too little time to chat with staff and volunteers because they were busy [34, 39] • Patients engage with internal services such as key and substance misuse workers but rarely with mental health or social workers [32] • Homeless people express many misperceptions and uncertainties about surrogate decision-making [28] • Homeless persons often describe their problems in a jumbled manner, understanding the most prioritized needs is thus not always easy [37] | Knowledge and skills • It is difficult for staff to determine when a patient is nearing the dying phase and to establish palliative care needs; staff members’ notions of palliative care vary and opportunities to prevent deaths are being missed [32, 33, 35, 37, 43] • Hostel staff are often not able to plan for end-of-life care with patients [32] • Medical intake personnel (in hospital) do not know how to deal with a homeless person 47] • Little opportunity for funding or training shelter staff in palliative care [45] • Working with limited medical information [35] • Staff of healthcare services not being knowledgeable about the unique issues facing the homeless [36] • Often difficult to interpret reaction of patients suffering from mental illness and/or illicit drug use [37] • Trying to solve all of a patient’s problems at once is seldom successful [37] Organization • Access to palliative care, primary care and/or preventive services is minimal (due to competing priorities, attitude of healthcare professionals, anti-drug policies, not conforming to procedures, healthcare system’s nonadherence to harm reduction strategies, a lack of caregiver support and/or financial resources) and a significant proportion of homeless persons may be underusing healthcare [25, 27, 28, 32, 36, 38, 43, 47] • Lack of appropriate housing, beds, respite or hospice facilities and programmes and care sites for homeless people at the end of life and limited resources for providing end-of-life care [25, 28, 44, 45, 49] • Poor coordination and/or communication between secondary care and hostel staff or homeless programmes and end-of-life programmes [32, 35, 49] • Setting treatment goals according to routine guidelines were often regarded as unrealistic in this context [37] • In-shelter palliative care means more work for staff and a greater burden for a workforce already thinly stretched [45] • Cost of medications that was not covered by the benefits and had to be paid for in cash [30] |
Relating to the homeless people | Relating to the interaction between homeless people and healthcare professionals | Relating to the healthcare professionals |
---|---|---|
• Primacy of religious beliefs and spiritual experience or connection; religious beliefs are a core component of homeless people’s end-of-life beliefs and experiences; it provides comfort and solace through spirituality/religion [24, 28, 39] • Allow for patients to have “unscheduled” space to share their life stories and to acknowledge those stories [37] • Freedom is essential to homeless people [33] • Other homeless patients could become involved in the care of fellow residents who are unwilling to work with services [32] • Among homeless people who filled out an AD, there were increasing in plans to write down end-of-life wishes, plans to talk about these wishes with someone and less worrying about death [46] | Attitude towards homeless people • Building and establishing trusting and/or family-like relationships and contact by interacting with patients in everyday situations and staff taking a supportive and/or advocating role in encounters with other health providers [25, 32, 35, 37] • Upholding homeless residents’ dignity and maintaining pride by showing human kindness, respect, love, comfort and to name accomplishments and elements of character [29, 31, 35, 37] • Staff must never judge a homeless person as impossible, or in terms of failure, and always patiently give them a new chance [37] • Persistence to engage the patient and to keep them engaged, with a constant effort required for effective follow-up [47] Treatment of homeless people • A pragmatic approach by staff, facilitating flexible care solutions, such as the choice where to die and accepting that planned activities may not happen or need to be cancelled [25, 37, 45] • Compassionate healthcare providers who are present (e.g. not leaving the individual alone during or after death [25, 28, 37] • Staff can respect individual’s habits and needs (also if rather unconventional, friends and preferred surroundings (e.g. stay in the hostel) when they are at the end of life [32, 36] • Staff only contacting family members at the end of life if the patients so request [37] • Formulating simple messages towards patients about death and dying [37] Activities/therapies • Advance directive completion rate is higher when counsellor guided that compared to no counsellor guidance [41, 42, 46] • Harm reduction services (e.g. clean needle exchange, medically prescribed alcohol) are a critical point of entry to and source of end-of-life care and support for homeless people who use alcohol and/or illicit drugs and are unable to access services [25, 48] • Physical contact can enable feelings of safety and appreciation in patients (not all patients) [37] • Memorial services held by staff to give staff members and other patients or visitors a moment to remember and say farewell [37] | Knowledge and skills • Optimizing management of pain, symptoms and functional decline, e.g. by palliative care consultations [48, 49] • End–of-life care and addiction training [26] • To preserve integrity in being close to patients [37] • Treatment for symptoms and distress is often provided simultaneously with the use of illicit drugs and/or alcohol, this necessitates special skills for identification of signs and symptoms and treatment regimens [37] Organization In-shelter hospice care; without it, a large part of homeless patients might not have sought care or received services and died homeless with no pain and symptom management [48] • Costs of in-shelter hospice care are substantially less than the estimated costs of traditional care for the same patients [48] |
Recommendations for practice
Training, education and knowledge | Delivering care | Overall organization |
---|---|---|
• Training regarding providing palliative care for (older) homeless people and their specific needs • Education about addressing preferences, advance directives, after death wishes and surrogate decision-makers | • Patient-centred approach • Trustful and respectful relationships • Reliability, experience, sensitivity and commitment of healthcare professionals • Attention to various areas of concern of homeless people • Flexible programmes and availability • Support after death | • Availability of accommodation • People involved and coordination • Hospital discharge policies • Policies and guidelines • Partnering and exchange of knowledge between organizations |