Review of the literature

Parts of this chapter have been reproduced from Butler et al.15 © Article author(s) (or their employer(s) unless otherwise stated in the text of the article) 2018. All rights reserved. No commercial use is permitted unless otherwise expressly granted. This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The text below includes minor additions and formatting changes to the original text.

Introduction

This chapter reports the review of the literature to understand existing models of HAH services in the UK and their evaluation.

Background

Providing patients with choice about where they receive their care at the end of life remains central to UK policy3 and HAH services have been introduced to support patient choice if their wish is to die at home. The number of patients wishing to die at home has been increasing.5,6 Home palliative care increases the chance of dying at home and reduces patient symptom burden and grief for family carers post death.19

Stosz10 conducted a literature review in 2008 to establish the evidence base for HAH services. The characteristics identified and the terms describing the services operating included the following: a palliative care service provided in the home environment, OOH, hospital at home, community specialist palliative care, crisis intervention and rapid-response teams. The recommendations from that review were that a successful intervention should include the following:

• a service operating in addition to existing community services that is available throughout the course of the end stages of illness (particularly in the last few weeks of life when crises may occur)
• rapid access to specialist input at all hours
• providing access to medication and equipment
• viewing the informal carer as integral to the care team and recognising carer burden.

Based on the findings and recommendations of the Stosz10 literature review, Pilgrims Hospices in East Kent developed a new HAH ‘rapid-response’ service caring for adults in the last 72 hours of life.20 Evaluation of this service was conducted through a pragmatic quasi-controlled trial.11 Although the new service was cost neutral and enabled more people to die at home, it did not improve hospice patients’ chances of achieving PPOD (primary outcome).

In 2012, the NAHH and Help the Hospices (now Hospice UK; London, UK) collaborated to conduct a multiservice survey (across 76 services in England), which started to describe the landscape of HAH services across the country (Heather Richardson and Andrew Thomson, Hospice UK, 2014, personal communication). The conclusions from this work were that HAH services were not homogeneous and that there were at least two models of care, despite the shared name of HAH. A clear distinction was found between one set of services, delivering high numbers of completed episodes of care (> 50 episodes of care per service per month), and the other set, providing significantly fewer completed episodes of care (< 50 per month). In addition, there were notable differences relating to reasons for referral, episode duration, who was involved in care, and knowledge regarding preferences and PPOD. The recommendations from this survey acknowledged that there was a need to further increase understanding of HAH. There was lack of clarity about what was the best model of care, for example there was uncertainty about skill mix and to what degree teams should incorporate senior staff alongside more junior nurses and social carers (Heather Richardson and Andrew Thomson, personal communication).

The NAHH also published recommendations in the form of national standards for HAH services, which they developed through workshops with HAH service professionals in May 2011, November 2011 and May 2012. These resulted in six agreed core standards, with examples of structural, process and outcome criteria underpinning them:21

1. The HAH service has a workforce management, education and development strategy that ensures the competence and confidence in practice of its employees to deliver and support high-quality clinical services.
2. The HAH service is integrated into the local EOLC service provision and involved in providing co-ordinated care for patients and families.
3. The HAH service clearly defines and communicates referral criteria and pathways to all referrers, key stakeholders and other partners.
4. The HAH service ensures that patients, and their families and carers, receive the service information required to enable them to make informed choices in relation to their preferred place of care and support, including at the end of life.
5. The HAH team’s care and support service, in partnership with other agencies, meets the assessed needs of patients, carers and families.
6. The HAH service has systems and processes to ensure pre-and post-bereavement support for patients (when appropriate), carers and families.15

The findings from these projects indicated that there was value in HAH as a concept, but led to the broader question of what would be the most successful and cost-effective model of HAH that could improve the outcomes for an even higher proportion of patients whose preference was to die at home, in their area. This prompted a further review of the existing literature, to understand what different HAH models existed in the UK and their value, that is whether or not any comparative data or assessment of optimum HAH service model delivery existed. This review of the literature, initially conducted in 2014, and updated in 2017, 2019 and 2020, is described in the following sections.

Search strategy for hospice at home models, comparators and outcomes

The search sought to identify any type of literature or study that aimed to describe or evaluate a HAH model in the health and social care setting of the UK that was providing care to adults with a life-limiting illness who wished to die at home. The service could be described as a HAH service by name or could potentially be a community service under a different name. Therefore, the search strategy included concepts that could identify these services in the literature. The search concepts were chosen based on the previous literature review.15

The criteria for selection of articles were as follows:

• HAH service
• community service under a different name with clear HAH characteristics:
  • rapid response
  • crisis management
  • 24-hour coverage
  • staff in service were palliative care specialists who were hospice trained.
• UK based.

A set of searches was carried out on several databases of academic publications, grey literature and current research (Figure 1; also see Appendix 1, including Tables 19–21, for further details).

FIGURE 1 The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 flow diagram. (Figure not reproduced here; see the original article.)

Scoping the literature on service models and evaluations

Sixty-three papers/grey literature sources were identified from the searches. The articles were analysed by recording key information from each article relating to (1) study design and (2) service description (Table 1). This provided information to scope what types of services existed and what work had been undertaken to evaluate them. Any identified barriers to and facilitators of accessing the service, or the service achieving its aims, were also included.

Many articles included were evaluations or descriptions of one service model in one locality. However, 13 articles (involving 11 studies) looked at several models of care. Two of these were multiservice surveys to understand and scope HAH models: the NAHH/Help the Hospices survey mentioned previously (Heather Richardson and Andrew Thomson, personal communication) and the survey conducted as phase 1 of the OPtimum hospice at home services for End of Life care (OPEL) study.22 Eight were literature reviews or syntheses.5,10,18,23–27 Hashem et al.18 applied a realist logic of analysis to their review of HAH. Taylor et al.23 conducted a literature review of the international evidence for models of care supporting effectiveness in reducing inappropriate/non-beneficial hospital bed-days for people nearing the end of life; they concluded that such evidence was generally limited or absent. HAH was one type of care described in the review.23

The criteria for ‘home palliative care’ services were much broader than the definition used for HAH in some of the reviews, and looked at literature beyond the UK.5,24,25 Sarmento et al.25 undertook a review of qualitative evidence to understand patients’ and family caregivers’ experiences, and the key components of care that shaped the experiences of service users. Shepperd et al.5 undertook systematic reviews of the trial and controlled study literature on home-based EOLC. Bainbridge et al.24 identified components of home-based programmes: a total of 30 unique components were identified by a content analysis of the literature. Efficacious programmes included multiple components; the most common were linkage with acute care, multidisciplinary nature, end-of-life expertise and training, holistic care, pain and symptom management, and professional psychosocial support. Luckett et al.26 looked to understand elements of effective palliative care models in a range of settings, not just home care. They identified essential attributes of effective palliative care models to be communication and co-ordination between providers, rapid response to individuals’ changing needs and preferences over time, skill enhancement and specialist expertise. Another study looking at care models provided a set of criteria to define and compare models of UK specialist palliative care, which distinguished home-based care from other forms of care as one criterion, with several other cross-cutting criteria.28 Another looked at a number of models of care, but only within a primary care setting.29,30

A smaller number of articles (seven) identified in the review looked at variations of the same service model: the Midhurst Macmillan Specialist Palliative Care at Home Service31–33 and the Marie Curie Delivering Choice Programme;34–37 realist evaluation principles were used in one evaluation of each service model.32,37 The Marie Curie programme was implemented across two counties, and included co-ordination centres, a telephone advice line, ‘discharge in reach’ nurses, a specialist community personal care team and nurse educators. The importance of having ‘highly skilled’ palliative professionals with ‘dedicated and sufficient time’ to support informal carers in navigating the system was noted. The whole-system approach of the Delivering Choice Programme underpinned its success, which relied on the collective effort of senior and front-line professionals across hospices, the NHS and social care services. In contrast, Johnston et al.32 found variation in the implementation of the Macmillan service across its six sites. Overall, they concluded that users of the service were more likely to die at home, and identified the importance of rapid response, early referral, good leadership, flexible working and the added value of health-care assistants (HCAs) and volunteer roles within the service, in particular for psychosocial support. These studies assessing specialist palliative care models highlighted the variation in the components of HAH or home-based palliative care services.

Description of services

The majority of HAH services offered service provision that had long periods of involvement (i.e. not just for the last days or weeks of life) and did not provide a crisis management element or a 24-hours-per-day, 7-days-per-week (24/7), rapid-response service.

Common service characteristics identified in the literature were as follows:

• enabling patients to be cared for and to die in their place of choice, namely home
• specialist staff [whether HCA or registered nurse (RN) core staff] with high levels of palliative care expertise
• ability to provide more staff time with a patient
• some HAH named services offered ‘sitting’ services or assisted discharge from hospital.

A diverse range of multifaceted services was described in the literature based on locally perceived need, for example population/geography, which tended to complement other existing services. However, some inequality of access was observed, for example:

• Association between greater deprivation/lower socioeconomic status and lower rates of access to HAH.38,39
• Inequality in referral practices in primary care – difficulties in prognosis and identifying terminal phase of non-malignant diseases. The majority of patients seen by HAH were cancer patients.39,40

Some publications described the process of service development20,41,42 through learning from the evidence base, listening to their local stakeholders and service users or by replicating service models that seemed to work elsewhere. An example of sharing of lessons learned for service development was published by a service in north-west England, which provided its own 10 steps to develop an effective HAH service: preparation, being clear on what it can offer, clinical leadership, staff have community or palliative care experience, comprehensive induction with the hospice, support for staff, good lines of communication with primary care teams, reassurance to other health professionals (e.g. about not ‘taking over’), clear referral criteria agreed by all stakeholders and publicity among the public.43 The provision of services was often still evolving, and services were being evaluated in the light of the need to secure further funding to continue.44,45

Evaluations of services

The majority of evaluations of single HAH services were descriptive, capturing views of service users and/or the service staff.13,14,43,45–47 They did not have a control group and had small sample sizes. Tyrer and Exley40 focused on the demographics of the service users, referrals and service use.

Some evaluations captured the views of bereaved carers.12,20,48–50 Grande et al.51 looked at the impact of HAH on carer bereavement. Other studies included views of referrers to the service, such as community nurses/district nurses (DNs) and general practitioners (GPs).52–56 These descriptive studies tended to use surveys or qualitative methods such as focus groups and interviews.

Buck et al.39,44 evaluated services by reviewing case notes, and Koffman et al.47 measured clinical and psychological changes at the time of referral to HAH and after receiving the service, for patients with advanced human immunodeficiency virus/acquired immunodeficiency syndrome using the support team assessment schedule.47 Others reviewed their current service provision by how well they had met key performance indicators or other objectives of their services, or patient outcomes.42,57,58

Some evaluations assessed the extension of already established HAH services. These included the introduction of additional OOH support,59,60 a respite service61 and combining existing services to enable cross-working between multidisciplinary teams.62 Strategic changes played a key part in the success of one service.60

The difficulty of trial designs in palliative care was acknowledged in the literature, and only two of the evaluations reported were trial designs. One of these was in east Kent, where a pragmatic, quasi-experiential, controlled trial to evaluate a rapid-response HAH service was undertaken.11,20 The second trial was of a ‘hospital at home’ service in the Cambridgeshire area in the late 1990s.38,63–65 The latter service was more focused on the provision of respite care, rather than rapid response, and was not a ‘specialist’ service as is characteristic of the more recent HAH services. There were two retrospective cohort studies in which patients were not randomised: one observed outcomes of patients who accessed the service43 and the other compared the outcomes of patients who accessed the service with those of patients who did not.36 One evaluation did attempt a before-and-after study, but had to abandon it because of small numbers.41

Only four evaluations included a health economic component to try to assess the cost-effectiveness of their service.53,66,67 The Spiro et al.67 pilot study suggested a model that could offer an economic propostion, but concluded that assessing the cost of EOLC was complex.67 Gage et al.66 and Addicott and Dewar68 found the services they were evaluating to be cost neutral, but offered an increased likelihood of achieving death at home. Grady and Travers53 were not able to draw firm conclusions because of the limited number of data.

Place of death was a common outcome measure for evaluations to assess what proportion died at home and prevention of admission to hospital/hospice. Otherwise, outcomes were weaker and looked at ‘impact’ or ‘strengths and weaknesses’ in line with the descriptive nature of the majority of studies.

Themes identified in the literature

The following themes were identified as features of HAH services that work well, but there were also challenges:

• Staff offered specialist knowledge and something over and above other service provision at home. A particular feature was that HAH services were able to spend time with the patient that other services visiting the home could not provide. Good communication was also key.
• A minority were rapid-response/24-hour services; only seven offered rapid response.40,42,43,45,48,53,69,70 Services providing rapid response reported effectiveness in enabling patients to remain at home.
• Eight services in the literature offered 24/7 OOH provision. Some provided full service, whereas others offered a reduced service OOH, for example a telephone advice line or voluntary staff. OOH provision was seen as desirable for many services that were not offering it. However, difficulties continued to be identified, even for those that were offering 24/7 services, for example access to medication, fewer staff and less medical support OOH.
• Instead of 24/7 rapid response, services tended to offer ‘sitting’ respite, appointment-based services or assisted discharge from inpatient units to allow a patient to be at home.
• The role of the carer was key: HAH services helped to provide physical, emotional and social support to relieve carer burden, and also provided bereavement support in some cases.
• Working with primary care teams (GPs and DNs). Patients remained under the care of the primary care teams, so HAH services complemented this. Communication was key to reassure them that the HAH service was not ‘taking over’.
• Issues of timeliness in receiving equipment into, and removing it from, the home.

Conclusions of the review of the literature

Hospice at home is an umbrella term with no clear service specification. Many hospices have adapted and used elements of what could be described as HAH, resulting in many different models of HAH being implemented in practice.

This literature review set out to identify the literature that described HAH service models in the health and social care setting of the UK, or community models under a different name that had HAH ‘characteristics’ in terms of rapid response, crisis management, 24-hour coverage and palliative care specialist staff who were hospice trained. A limited number of studies of services, described as HAH, met all these characteristics. Most often, HAH services shared the objective of enabling a patient to die at home if that was their place of choice, but they were less likely to provide the service on a 24/7 basis that offered rapid-response crisis management. Other elements that these services did provide were appointment-based services such as ‘sitting’ respite-type services or assisted discharge from inpatient units to allow a patient to be at home. A theme present through all the service models was staff with high levels of palliative care experience. The additional time they were able to spend with patients, which other services visiting the home could not provide, was a highly regarded element of HAH service provision, whether the core staff were RNs or HCAs.

The literature supported the proposition that HAH services at the end of life are valuable and complement existing service provision, but much of the literature was limited and the evidence was relatively weak.

Summary

The literature endorsed the value of HAH services in supporting patients to remain at home to receive their care at the end of life. However, the review of the literature posed important outstanding questions and highlighted continuing gaps in evidence about the most successful and cost-effective service configuration and activity. These questions cover the following topics: staffing profile, working patterns, communication and co-ordination with other local services, and support for carers. The review informed the funding application to Health and Social Care Delivery Research for the development of the OPEL study to identify optimum HAH services at the end of life.15