The support needs of terminally ill people living alone at home: a narrative review

Context: The number of terminally ill people who live alone at home and without a caregiver is growing and exerting pressure on the stretched resources of home-based palliative care services. Objectives: We aimed to highlight the unmet support needs of terminally ill people who live alone at home and have no primary caregiver and identify specific models of care that have been used to address these gaps. Methods: We conducted a narrative review of empirical research published in peer-reviewed journals in English using a systematic approach, searching databases 2002–2013. This review identified 547 abstracts as being potentially relevant. Of these, 95 were retrieved and assessed, with 37 studies finally reviewed. Results: Majority of the studies highlighted the reduced likelihood of this group to be cared for and die at home and the experiences of more psychosocial distress and more hospital admissions than people with a primary caregiver. Few studies reported on the development of models of care but showed that the challenges faced by this group may be mitigated by interventions tailored to meet their specific needs. Conclusion: This is the first review to highlight the growing challenges facing community palliative care services in supporting the increasing number of people living alone who require care. There is a need for more studies to examine the effectiveness of informal support networks and suitable models of care and to provide directions that will inform service planning for this growing and challenging group.


Introduction and background
The fundamental aim of palliative care is to achieve the best quality of life possible and support for terminally ill people and their families (World Health Organization, 2007). (National Hospice and Palliative Care Organization, 2008;National Institute for Health and Clinical Excellence, 2004;Palliative Care Australia, 2005). Family or informal caregivers provide unpaid help and support to a relative, friend, or neighbor receiving palliative care who could not manage otherwise because of frailty, illness, or disability. These caregivers commonly play a central role in homebased palliative care undertaking tasks such as symptom assessment and management, personal care, and administering medications Aoun, Kristjanson, Hudson, Currow, & Rosenberg, 2005;Funk et al., 2010;Hauser & Kramer, 2004;Stajduhar et al., 2010). In so doing, they provide much of the support needed by the patient as from less than a year to 60 years. Nearly 80% of the participants reported that they were living alone by choice rather than by circumstance.

Challenges in provision and access to care
The proportion of patients supported at home until death, if this is their preference, remains an important performance indicator for many community palliative care programs (Brogaard, Neergaard, Sokolowski, Olesen, & Jensen, 2013). A large survey of 9344 people residing in several European countries reported that 68.2% of participants indicated a preference to die at home if they were faced with terminal cancer (Gomes et al., 2012). A similar study in Canada also showed that 70.8% of 1203 respondents preferred to be at home at end of life (Wilson, Cohen, Deliens, Hewitt, & Houttekier, 2013). Focus groups and interviews with adults (65 years of age and older) in the UK showed that, despite home being the preferred place of care, the presence of informal caregivers was cited as a mediator to achieving this outcome (Gott, Seymour, Bellamy, Clark, & Ahmedzai, 2004).
Dying at home is also the stated preference of many people receiving palliative care, at least in the earlier stages of their final illness. A recent systematic review of the literature on preferences for places of death showed that most adults at end of life indicate a preference to die at home, and most do not change their preference during disease progression (Gomes, Calanzani, Gysels, Hall, & Higginson, 2013). A study of Danish cancer patients demonstrated that 84% preferred to be cared for at home and 71% preferred to die at home (Brogaard et al., 2013).
However, terminally ill people are far more likely to die in a hospital or a similar facility rather than at home. A study of all cancer deaths in England from 1993 to2010 showed that hospital was the most common place of death (48%), followed by home (24.5%) and hospice (16.4%) (Gao, Ho, Verne, Glickman, & Higginson, 2013). Similarly, a study in Spain showed that only 17% of cancer patients died at home; the majority (74%) died in hospital and the remainder died in either a hospice (6%) or a nursing home (3%) (Alonso-Babarro et al., 2013). However, home-based palliative care programs may assist people to stay at home for longer and to die at home, if this is their wish. For example, people receiving home-based hospice care in Taiwan were more likely to die at home (60.8%) than in a hospital (39.2%) (Lee, Hu, Loh, & Hwang, 2013).
Despite these preferences, facilitating place of death wishes is a challenge for services. A postal survey of Australian health professionals working in home-based palliative care (Aoun et al., 2007) demonstrated that more than 60% reported spending at least 15-30 minutes of additional time per visit to support clients without a caregiver with symptom control, medications, mobility, transport, and social support. Furthermore, nearly half (43%) reported spending one hour or more of additional time per visit on activities of daily living, while 25% reported taking one hour or more of additional time per visit for symptom control, housekeeping, and emotional support. The nurses reported concerns such as limited staffing, lack of a social worker and housekeeping support, and client safety. Postal surveys and interviews with Australian service providers regarding their perspectives on resources needed to look after home alone dying clients revealed that, while service providers expressed a respect for the clients' autonomy and independence, they felt pressured to ensure that safe and attentive care was possible. They identified the inability of home alone clients to anticipate their needs should their condition deteriorate, and they struggled to make them understand the limitations of their situation (Aoun, Wall, Kristjanson, & Shahid, 2013).
Given the aging population, the growing reliance on informal carers, and the preference to be cared for and die at home, it is likely that home-based palliative care services are facing increasing challenges in servicing the needs of such clients. However, there is limited evidence base upon which service providers may draw in order to inform service planning for a growing population.

Objectives
The objectives of the review were to (1) highlight the unmet support needs of terminally ill people who live alone at home and have no primary caregiver and (2) identify specific models of care that have been used to address these gaps.

Methodology
Using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement as a guide, a literature search was undertaken of the following databases: Ovid MEDLINE and EMBASE, ProQuest Medline, NLM MEDLINE, PyscINFO, CINAHL, and CareSearch. We considered national and international literature published, or available online in 2013, in refereed journals in the last 11 years (January 2002-October 2013. We included only articles published in English. Searches were undertaken using the following key words: palliative care, hospice care, end of life, supportive care, life-limiting illness and terminal illness, models of care, and interventions, in conjunction with associate keywords: home alone, living alone, no carer/caregiver, home care, place of care, and place of death. Figure 1 shows the search process and outcomes. We conducted a comprehensive narrative review of empirical research on the support needs of terminally ill people living alone at home, with a systematic approach that utilized several clear inclusion/exclusion criteria. Exclusions were not made based upon assessments of the rigor, validity, and reliability of studies and their findings, as would be typical in a systematic review. All three authors participated in the coding of the identified articles. We took a directed content analysis approach (Hsieh & Shannon, 2005) in order to group the articles into themes relating to either unmet support needs or models of care. We discussed all differences and achieved consensus, indicating the trustworthiness of the coding scheme (Mays & Pope, 2000). Our goal was to provide a broad, comprehensive overview of published research in this field, as opposed to a systematic review that focuses on generating a definitive answer to a narrowly defined research question.
All search results were downloaded into a reference manager (EndNote X5, Thomson Reuters). This initial search elicited 547 records for further review. After deletion of duplicate records, we removed articles that were not available in English, involved pediatrics, or were not obvious original research. Preliminary screening to exclude non-relevant articles and further non-original research articles led to the exclusion of 452 papers, resulting in the retention of 95 full articles for eligibility assessment for inclusion in the review. We excluded 62 articles due to the absence of variables concerning home alone and palliative care. A further hand search of the relevant references of these articles was undertaken to capture articles that might have been missed on the initial search. Initially, 49 studies met the inclusion criteria for our review. However, 12 studies were later excluded because, although they described samples wherein a considerable proportion of their study population was living alone, the authors provided no comparisons or analyses relevant to places of care and death or physical and psychosocial health and well-being. Therefore, 37 studies are included in the results section.

Results
This section describes the unmet support needs of terminally ill people who live alone at home in terms of their places of care and death, and also their physical and psychosocial well-being and identifies specific models of care that have been used to address these gaps.

The unmet needs of terminally ill people living alone
Thirty-five (95%) of the included studies described the unmet needs of terminally people who live alone. Although different methodologies were used in these studies, ranging from prospective designs to retrospective review of health data, the proportion of people living alone in their study population was considerable. Our review of these studies revealed two main disadvantages concerning the unmet needs of terminally people who live alone: places of care and death (Table 1) and physical and psychosocial well-being ( Table 2).

Places of care and death
Several studies (n = 28, 76%) provided data on places of care and death. Irrespective of methodology and country of origin, these studies demonstrated that people were less likely to die at   Longitudinal study with four data collection points (1984, 1986, 1888, and 1990) (n = 7527) Not stated Older adults who lived alone reported receiving significantly less assistance with activities of daily living than those who lived with others or were married (9) Kharicha et al. (2007) UK Adults (65 years and older) living without disabilities in the community Retrospective analysis of baseline data from an RCT of health risks in older people (n = 2641)

33%
People living alone were significantly more likely to report a range of psychosocial difficulties in relation to activities of daily living, mood, memory, physical activity, vision, diet, alcohol use, multiple falls, and social isolation a Rounded to nearest whole number.

Health Psychology & Behavioural Medicine
home when they lived alone (Table 1). In several studies, people living alone were described briefly as a subgroup of the total study sample of people receiving palliative care having a decreased likelihood of a home death (Ahlner-Elmqvist et al., 2004;Brink & Frise Smith, 2008;Cohen et al., 2006;Grundy et al., 2004;Gyllenhammar et al., 2003;Jordhøy et al., 2003;Tang & McCorkle, 2003;Tiernan et al., 2002). These studies demonstrating disadvantage in place of care and death are complemented by subjective data. For instance, interviews with terminally ill patients who live alone in the USA (Tang, 2003;Tang & McCorkle, 2003) and Australia  revealed that, while most preferred to die at home, some reported feeling unable to do so due to limited supports. These perceptions were corroborated in a longitudinal study of end-of-life preferences of Australian home alone clients who demonstrated a preference to die at home (49%) over dying in a hospice (23%) or a hospital (12%); 16% indicated no preference . However, only 14% died at home, while 56% died in a hospice and 22% in a hospital. Overall, congruence between preferred and actual place of death decreased from 53% to 41% during the course of the study, possibly due to clients growing more confident of achieving a home death after using home-based care for a period of time.
Additionally, the design of some studies rendered this group overlooked, for instance, by sampling caregivers to determine variables associated with home death (Brazil, Bedard, & Willison, 2002;Carlsson & Rollison, 2003;Masucci et al., 2010) or by not reporting home alone status and/or inferring it from the absence of a caregiver or being single (Aabom et al., 2005;Alonso-Babarro et al., 2013;Brogaard et al., 2013;Cohen et al., 2010;Gao et al., 2013;Goodridge et al., 2010;Houttekier et al., 2009;Jakobsson et al., 2006;Jayaraman & Joseph, 2013;Lee et al., 2013;Neergaard et al., 2009;Pinzón et al., 2011;Tang, 2003;Taylor et al., 2012). Only one study (3%) focused on the place of death outcomes of this subgroup and reported that, compared to patients with a caregiver, those without a caregiver were less likely to die at home (35% compared to 57%), twice as likely to die in a hospice, and 2.5 times as likely to die in a hospital (Aoun et al., 2007).

Physical and psychosocial well-being
Nine studies (24%) highlighted the considerable health and psychosocial disadvantages experienced by people living alone at end of life (Table 2; NB: Two of these studies also provided data on places of care and death and therefore appear in both tables). People living alone at the end of life reported more distress, poorer adjustment to diagnosis, and reduced quality of life, and received less help with activities of daily living than those living with others (Chibnall et al., 2002;Johnson et al., 2004). Interviews with adult palliative care clients living at home without a caregiver in the UK (Hanratty et al., 2013) and Australia  showed that patients faced challenges in meeting their care needs and perceived they were disadvantaged in terms of receipt of practical and emotional supports.
People living alone at end of life were also more likely to experience a range of problems with falls, diet, smoking, social isolation, and chronic health conditions Kharicha et al., 2007). Notwithstanding these additional struggles, they received half as many home visits, despite being enrolled in the service for an average of 20 days longer, and were more likely to be admitted to hospital. They required more equipment, support with hygiene, home help, and liaison with other health professionals (Aoun et al., 2007;Currow et al., 2008;Goodridge et al., 2010). Thus, these studies showed that the presence of a caregiver in the home was related to better physical and psychosocial well-being compared to those without a caregiver and more timely and improved access to required treatments at end of life.

Models of care for terminally ill people living alone
Two studies (5%) explored models of care for terminally ill people who live alone (Table 3). These models of care were based upon formative data provided by three studiesone focusing on the perspectives of clients  and two on service providers (Aoun et al., 2007;Aoun, Wall et al., 2013). The first study was a pilot intervention using a randomized trial design to test two models of care (personal alarms and additional care-aide support) compared to standard care . The findings indicated that those who received care-aide support tended to have improved scores (lower scores) in appetite problems and fatigue compared to the other two groups. The second study sourced qualitative feedback from these participants on the benefits and barriers of these interventions (Aoun et al., 2012). The care-aide model of care resulted in benefits such as easing the burden of everyday living, supporting well-being, preserving a sense of dignity, and reducing loneliness and isolation, while the personal alarm model of care imparted a sense of security, provided peace of mind, and helped clients manage feelings of isolation. Importantly, participants in both groups felt they could remain at home longer. By providing a safer, more secure environment through the use of either model of care, clients were able to continue their activities of daily living, and could build a sense of 'normality' into their lives with a degree of independence through support and dignity. Together, these two studies showed that the challenges faced by terminally ill people who live alone may be mitigated by interventions tailored to meet their specific needs.

Discussion
This review has described the unmet needs of terminally ill people living alone, the disadvantages they face in terms of their physical and psychosocial health and well-being, and their reduced likelihood to be cared for and die at home. It also reported on potential management options to enable this group to remain at home for as long as possible. Additionally, this review has highlighted three factors that would intensify the challenges of providing community and home-based palliative care services: (1) an increase in the number of people living alone who require care; (2) a decrease in the provision and availability of family caregivers; and (3) people's preference to be supported to die in their own homes. For people living with a terminal illness, the presence of a caregiver and family support proves to be the strongest independent factor associated with home death (Gomes & Higginson, 2006). Living alone raises both the probability and the necessity for living and being cared for in institutions (Rolls et al., 2011), which may be less preferable from a patient perspective and more expensive as home care tends to be more cost-effective than hospital care (Higginson, Jarman, Astin, & Dolan, 1999). Furthermore, hospitals are not recognized as ideal places for terminal care due to a number of problems including communication difficulties between hospital staff and patient/family, being overlooked by staff on wards, and lack of privacy (Jakobsson et al., 2006;Tiernan et al., 2002). However, nearly half of the home alone clients in one study did not choose home as their preferred place of death , prompting the authors to conclude that 'the ability to die in the place of choice needs to be looked at as a possible indicator of meeting patient needs or as a quality measure in end of life care' (p. 5).

Implications for palliative care services
Community-based care is a vital component of palliative care. However, unless additional funds and resources are provided for community-based palliative care services, the demographic projections indicate that these services will face increasing challenges in servicing the needs of clients living alone and with no caregiver. Palliative care support needs for individuals living alone without a caregiver remain somewhat obscured because, until recently, those living home alone have been an invisible yet disadvantaged subgroup. While places of care and death are important outcome variables for examination, status of living and caregiver arrangements need to be more than of peripheral interest. Although studies have been conducted to assess the characteristics of clients who are receiving home-based palliative care, limited research has been conducted to specifically assess the needs of those who live alone and who do not have access to a primary caregiver, despite this group making up a significant percentage of the overall population under study. Studies highlight that the changing living arrangements of older peoplesuch as the increase in living alonehas important implications for planning and provision of care and treatment for cancer sufferers (Grundy et al., 2004;Jakobsson et al., 2006); however, few suggested models of care addressed these implications. Given the aging populations, maintaining and extending the proportion of home deaths will likely result in more work for family doctors, district nurses, social services, and palliative home care teams (Higginson et al., 1999). Furthermore, it has been argued that general improvements in home care support may only help those who are already at an advantage (Grande, Addington-Hall, & Todd, 1998). Real progress may only be achieved by identifying the factors behind the disadvantage of certain groups (those living alone being one of them) and targeting interventions to address the disadvantages (Gao et al., 2013).
The ability of services to meet the needs of this group is further complicated by a lack of definitional clarity. In most of the reviewed studies, living alone status was inferred either by absence of spouse, or being single, or by being home alone but having caregivers. Nevertheless, in studies with a clearer definition, the wide variation in the degrees of 'home aloneness' created an impediment to evaluating the effectiveness of interventions using a randomized controlled trial (RCT) approach  and this variation confounded the findings in terms of how much informal support each patient was getting over and above the implemented models of care that were randomly allocated. Therefore, the authors concluded: 'there is a need to develop a scaling system of patient need, depending on the extent and frequency of informal support being provided by family and friends, before further trials are undertaken' (p. 188). Rolls et al. (2011) highlighted the vast range of informal support and non-blood family upon whom older people living alone drawfriends, neighbors, volunteers, church members, and former unmarried partners and ex-in-laws to name a few. Given the complexity of the definitional criteria of this group, the RCT approach used in Aoun, O'Connor et al. (2013) was not considered appropriate for the 'home alone' palliative care population who would have been better supported by providing each participant with a personalized model of care that would have met their particular needs. This would have avoided offering a model of care that was inappropriate for the stage of their illness or unwanted by the patient in some cases .
There is a small but growing body of literature dedicated to examining the service needs of palliative care patients who live alone without a caregiver focused on specific support services, end-of-life preferences in terms place of care and death, and interventions to address the disadvantages of being home alone at end of life. Two studies (Aoun et al., 2012 tested two models of care to enable this group to remain at home for as long as possible and provided some directions to inform service planning for this growing and challenging population group. There is a clear need for more studies to examine the effectiveness of informal support networks, interventions, and models of care. Adequate and timely services based on evidence will then lead to more care being delivered at home, a reduction in hospitalizations, a better quality of life, and a capacity to die at home if this is the patient's wish (Aoun et al., 2007).

Limitations and future directions
Our purpose in this review was to provide a broader, comprehensive overview of the field of support needs at the end of life for those living alone at home, as opposed to a systematic review designed to weigh the evidence relating to a specific question. The comprehensive approach used to reviewing this literature optimized the rigor of the search processes and robustness of the conclusions. Most research in this field is difficult to grade using traditional levels of evidence for systematic reviews  as there are challenges that constrain the random selection of samples, and sample sizes tend to be relatively small (Aoun & Nekolaichuk, 2014). Furthermore, while international work was sought in the search for articles on this topic, a considerable proportion of the research generally, and the testing of the two models of care specifically, was conducted in Australia. Although the findings seem particularly pertinent to one location, they nevertheless are likely to have international application, and therefore there is a need for more international research on this under-examined topic.

Conclusion
A significant proportion of older people in the developed countries will spend the last year of their life in poor health and with a considerable burden of palliative care needs due to social isolation, co-morbid conditions, and frailty. This has raised and continues to raise challenges for the provision and delivery of health and community services at the end of life. The growing population of people living alone and the reduced availability of informal caregivers together mediate the capacity to remain at home at end of life. This is the first literature review on the support needs of terminally ill people living alone at home and it has provided a comprehensive background for understanding current knowledge and highlighting the significance of potential research in the field.

Funding
The second author is supported by the Australian Research Council (DE120101640).