The effective ingredients of social support at home for people with dementia: a literature review

Purpose 
 
 
 
 
Recent years have seen an increasing reliance on social support at home for people with dementia and the advocacy of a person centred and integrated approach in its provision. However, little is known about the effective ingredients of this support and how they differ from more generic or health-based services. The purpose of this paper is to explore the existing evidence base. 
 
 
 
 
Design/methodology/approach 
 
 
 
 
A review of relevant literature was carried out, combining a systematic search and selection of articles with a narrative analysis. 
 
 
 
 
Findings 
 
 
 
 
The review identified 14 relevant studies of varying research designs which yielded conflicting findings with regard to the optimal timing of interventions and their overall impact. This highlights the problems of review and generalisability when attempting to compare findings of research in this area. This was exacerbated by the blurred divide between health and social care and ambiguities in the meaning of the latter. 
 
 
 
 
Research limitations/implications 
 
 
 
 
This methodological heterogeneity demonstrates the need for consistency in research approaches if comparisons are to be made. Further questions include the precise components that distinguish social care from health care, the optimal timing for the introduction of this care and whether adherence to good practice in this area can be linked to cost effectiveness. 
 
 
 
 
Originality/value 
 
 
 
 
The review identifies relevant issues in need of further investigation and tentative themes emerging from the literature which suggest the utility of an adequately resourced, integrated and responsive approach to intervention.


Introduction
Due to demographic trends towards an ageing population, dementia care is one of the fastest growing areas of need in the UK with predictions that there will be more than 1.7 million people in the UK with dementia by 2051 (Alzheimer's Society, 2013). These trends have been coupled with a longstanding policy commitment to the community care of vulnerable groups, the development of a system of welfare pluralism and a corresponding purchaser-provider split (Alakeson, 2007), with service users being reframed as active consumers rather than as the passive recipients of welfare (Department of Health, 1990). Arising from these developments has been a progressive increase in the amount of intensive home care provided to older people and the targeting of these services on high-risk and high-dependency groups (UK Home Care Association, 2016). Accompanying these service transformations, have been demands for improvements in the delivery of these services (Department of Health, 2009;2013, 2015, with the widespread advocacy of person-centred and personalised care (Glasby and Littlechild, 2009) and the adoption of integrated approaches in the provision of this care. In recognition of the service fragmentation arising as a result of welfare pluralism, this policy commitment to integrated approaches was reiterated in the Prime Minister's Dementia Challenge (Department of Health, 2015). This encourages more collaborative working and coordination between different care sectors thus promoting continuity and responsiveness in the care provided. Other components of this joined up approach include better trained staff, the creation of dementia friendly communities, the guarantee of a personalised plan of care and the involvement of people with dementia and their carers in the management and control of this care (Woolrych and Sixsmith, 2013). A similar focus on integration has been apparent in the home care guidelines recently published by the National Institute for Health and Care Excellence (2016). These guidelines include recommendations on the length of home care visits and the need for a planned, integrated and person centred approach in care delivery. It also suggests the need for the better provision of information about support options to clients, measures aimed at ensuring the safety of these clients and the more effective recruitment, training and support of home care workers.
In spite of these policies, there are still huge variations in the quality of home support provision and that older people still do not receive integrated, personalised, effective and responsive care. For example, focusing on the front line workforce issues, the All Party Parliamentary Group on Dementia (2014) observed that while the majority of people with dementia live at home, home care workers still lack relevant skills and knowledge in dementia care and tend to be task-orientated in their approach to their role. The report goes on to identify a number of reasons for these deficiencies such as high levels of staff turnover, lack of regulation over required levels of dementia training as well as inadequacies in leadership and resourcing. Moreover, with regard to the broader service context, joint working is often inconsistent or short term and J o u r n a l o f I n t e g r a t e d C a r e 2 multiagency coordination and cooperation is poor with older people still falling through the gaps in services (Care Quality Commission, 2016) and in subsequent need of crisis intervention rather than lower level and preventative support (UK Home Care Association, 2016). As recently observed, these gaps are increasing due to the chronic underfunding of social care services (Humphries et al., 2016) and of home care provision (UK Home Care Association, 2016). These issues are further exacerbated by an apparent lack of clarity amongst service providers on the meaning of person centred and integrated care and how it should be implemented, suggesting the need to establish a shared understanding of this issue if services are to be improved (Care Quality Commission, 2016). Similar issues of conceptual clarity are apparent in the meaning of home care which is poorly understood and little-researched (Godfrey et al., 2000). Thus, at the outset of the review presented here, the following quotation held true: "The necessary conditions for delivering improved home care services for older people with dementia are not fully understood, particularly in comparison with standard service provision" (Rothera et al., 2008: 71). In the light of the above discussion, this review set out to find empirical evidence concerning aspects of home care that might be relevant to setting standards or monitoring criteria for home care for people with dementia. It was undertaken to inform a research programme investigating the ingredients of good home care for people with dementia. Further details of this study can be found at the following website link: http://www.sscr.nihr.ac.uk/PDF/Findings/RF7.pdf

Methods
Systematic reviews are gaining increasing popularity as a reproducible, comprehensive and unbiased means of evaluating and summarising research through the adoption of predetermined stages and criteria and through the metaanalysis of quantitative data. However, there are many challenges in systematically reviewing literature in this area (Pawson et al. 2003) due, for example, to the fact that, unlike health care literature, relevant sources for literature on social care interventions are not clear cut and relevant studies are hugely diverse in their design and contexts. Therefore, for pragmatic reasons, the mode of working adopted in this review combined a systematic literature search and selection process with other methods of research synthesis (Petticrew and Roberts, 2008). These included elements of a scoping review in order to identify the range of home care ingredients and a qualitative, narrative review to examine the nature of these elements. Searched databases included MEDLINE (PubMed), EMBASE (www.embase.com) and PsycINFO and a full report on the initial literature search can be found here: https://nottingham.ac.uk/research/groups/dementia/projects/fidelityindex/index.asp x. In order to identify all potentially relevant studies, key words for 'home care' and 'dementia' were the relevant terms used in each database which were each searched from January 1991 to July 2011. These dates also incorporated the twenty years timespan since the launch of the NHS and Community Care Act (Department  Health, 1990) which led to the sustained transformation of domiciliary support services. Potentially relevant articles were reviewed against predetermined inclusion and exclusion criteria which included all types of dementia and study type and excluded issues of unpaid care, studies with a solely health care or economic focus and those not written in the English language. The final screening process involved checking the remaining references against this criteria. Where there was a difference of opinion regarding the inclusion of literature, these were referred to the team for resolution.

Results
The initial search process yielded 7867 references and after removing duplicates, 5135 articles were selected for further screening which was performed independently by two reviewers. Following a screening of records, a further 3561 were excluded leaving a total of 1574. From these 1574 another 1310 were excluded after abstract screening. During the final screening process of the remaining 264 potentially relevant references, 252 references were excluded as they were duplicates (5), there was no full text availability (76) or the text available was unclear (41). Many references were excluded because they had an inadequate focus on social support at home for people with dementia (130). A further two studies (Brooker et al., 2011;Riordan and Bennett, 1998) were identified and added by the team. This was because they were considered to be relevant to the review. The large number of exclusions and the fact that two key studies were not identified by the initial search highlights the challenges of conducting a systematic review of literature into social care issues (Pawson et al., 2003). Of the 14 included studies, four were randomised controlled trials, 3 were cohort studies, two were quasi experimental studies, 3 were qualitative studies and two were case reports. In addition, 8 were conducted in the UK and 6 outside the UK.

Table 1: Quantitative studies included in the review
Within this category, a randomised controlled trial by Brooker et al. (2011) and two quasi-experimental studies (Riordan & Bennett, 1998;Challis et al., 2002), possibly, provided the strongest levels of evidence from the UK for this review. It is notable that all three of these studies involve augmentation significantly beyond what is generally provided by English social care support at home and the extent to which their results may be applicable to less intensive home care remains open to question. However, they do provide indications of what might, or might not, better enable such care to work at its best when unaugmented. For example, effective leadership and relevant staff training (Brooker et al., 2011), a systematic approach and individualised packages of care (Challis et al., 2002) and support for the principal carer (Riordan and Bennett, 1998), are all key enablers identified in the above studies.
In contrast to the preceding UK-based studies, which tended to focus on the evaluation of specialist models of care and their impact on the wellbeing of particpants, relevant included studies from outside the UK largely focused on issues of care transitions and case management. In addressing these issues, the non-UK evidence also considered the impact of home care interventions on carer support, commissioning and flexible and responsive services. Research designs included three randomised controlled trials (RCTs) which were conducted in Finland and the US (Eloniemi-Sulkava et al., 2001;Miller et al., 1999) and three cohort studies all carried out in the US (Gaugler et al., 2005;Pot et al., 2005;Temple et al., 2010). Of these, Eloniemi-Sulkava et al. (2001; actually apply 'health' interventions to augment social care and, thus, seem not to illuminate what could make current straightforward English social care support work better. However, they do indicate the possibilities that arise from integrated approaches involving cooperation and coordination between health and social care services and targeted within an individualised plan of care. In the United States, Gaugler et al. (2005) followed a cohort of caregivers over three years, with the aim of determining whether being in receipt of community-based services early in the dementia caregiving career delayed time to nursing home placement. Miller et al. (1999) conducted a randomised controlled trial in order to determine whether an intervention aimed at improving caregiver outcomes through case management and subsidised community services affected the nursing home entry rate of clients with dementia. Temple, Andel and Dobbs (2010) aimed to examine risk of nursing home placement among older adults with dementia according to whether they received community-based home care input or assisted living support. While most studies in this section have focused on aspects of case management and on the transition from home to institutional care, in their cohort study Pot et al. (2005) focused on transitions in paid home care and its impact on the wellbeing of the principal care givers of relatives with dementia. They also recognised that, due to such things as long gaps between first and second followups and the fact that the cohort of carers were self-selecting, their study had a number of limitations. Three UK based qualitative studies evaluated specialist home care services which were community based (Gladman et al., 2007) multi-agency (Rothera et al., 2008) or provided on a respite basis for the carers of younger people with dementia (Parahoo et al., 2002). They reported that carers valued home care services that specialised in working with people with dementia. Five desirable characteristics of such services emerged from these qualitative evaluations which taken together offer tentative criteria for evaluating home care provision.

Qualitative studies and case reports
These include integrated and multidisciplinary input, care worker autonomy, continuity of personnel, continuous client reassessment and a respect for 'personhood'. Two case reports on a domiciliary respite service (Ryan et al., 2008) and a specialist home care service (Russell et al., 2002) reported similarly positive impressions on the part of staff, family carers and people with dementia in the UK.

Discussion
This review has provided evidence of a number of effective ingredients of paid social support at home for people with dementia. At an individual level, expert opinion and policy recommendations (NICE, 2016) supported by the findings presented in this review suggests the importance of care worker autonomy, continuity of the relationship between individuals and their care workers and frequent reassessment. Diagnosis should also be prompt in order that appropriate services can be introduced at an optimal stage in the dementia trajectory and the intensity of the service needs to be 'person-centred' and responsive to users needs and aspirations (Care Quality Commission, 2016). In order for this responsiveness to be achieved, commissioning practices need to be similarly flexible and adequately resourced (Humphries et al., 2016). For while policy recommendations have consistently supported the need for early diagnosis (All Party Parliamentary Group, 2014), the lack of availability of appropriate post diagnosis support is likely to undermine the goal of prompt intervention that such measures aim to promote. The review has also pointed to the likely benefits that arise from an integrated and multidisciplinary approach involving health and social care providers in the delivery and co-ordination of services (Brooker et al., 2011;Department of Health, 2015), thus reflecting the efficacy of an integrated approach to home care recently recommended by NICE (2016). If some or all of the above criteria are fulfilled in services provided to people with dementia living at home, evidence from this review suggests that their admission to residential or nursing home care might be delayed (Riordan & Bennett, 1998;Challis et al., 2002;Eloniemi-Sulkava et al., 2001Gaugler et al., 2005;Temple, Andel and Dobbs, 2010).
Nevertheless, further questions still remain, including the optimal amount of home care input for specific needs, as well as the optimal stage in the dementia trajectory to introduce this input, with included studies yielding opposing evidence on this latter point. Thus, Gaugler et al. (2005) suggest that, if interventions are delivered early in the trajectory, they are more likely to defer institutionalisation. Conversely, Eloniemi-Sulkava et al. (2001) recommend that services should be targeted at those in later stages of the dementia trajectory if this deferral is to be optimally achieved. The conflicting findings of included studies may well be due to the differing characteristics of the interventions adopted and corresponding ambiguities in the interpretation of central concepts such as home care and integrated care and the policy and social contexts in which the interventions were located For example, as Woolrych, and Sixsmith (2013) observe, not only are there many approaches to flexible and integrated care but the context in which this care takes place is likely to influence the impact that it has. Furthermore, many of the included studies used 'ideal type' interventions, including intensive service provision involving a range of professionals (all the UK studies), while the two Finnish studies applied health interventions to augment standard social care for people with dementia. In spite of their benefits, these types of intervention may be more costly to provide (Brooker et al., 2011) and financially unsustainable in the long term (Challis et al., 2002;Gladman et al., 2007). Such issues of cost are also relevant to the type of research design adopted and the different criteria used in determining the success of the intervention being evaluated. Thus, UK studies often included process measures of home care and its acceptability to participants. However, this may neglect its impact on such things as acute hospital admissions as well as the cost of the service being provided (Gladman et al., 2007). In contrast, non-UK studies mainly reported the impact of home care intervention. However, this focus not only overlooks the acceptability of the service to clients and their carers, it may also overlook the impact of intervening variables which are likely to influence care transition, such as the availability of wider supportive services.
These problems of generalisability when attempting to compare the findings of research in this area demonstrate the need for consistency so that relevant comparisons can be made both with regard to measuring the quality of interventions as well as in evaluating their impact. Similar recommendations for consistency have been made by the Care Quality Commission (2016) both with regard to defining the meaning of integrated care and in the methods used to identify those in need of this care. In doing so, they also highlight the importance of care recipients identifying their own care needs and outcomes thus indicating a move away from the professionally-led needs assessments (Alakeson, 2007;Glasby and Littlechild, 2009;NICE, 2016).
The advocacy of a user-led approach rather than one which is professionally defined is compatible with the recommendation by Pawson et al. (2003) for the adoption of a wide and inclusive range of knowledge when researching social care issues. Such an inclusive approach has also been adopted in this review, in order to maximise coverage of the limited range of relevant literature while also highlighting the significant problems of gathering evidence in this area. These challenges arise from the increasingly blurred distinction between health and social care provision (Glasby and Littlechild, 2009) the different criteria used to judge the quality and relevance of knowledge domains in these two areas (Pawson et al, 2003) and ambiguities in the meaning of central concepts such as home care. Related to these issues is the methodological heterogeneity of the included studies, due for example to their varying choice of research design and subject groups, meaning that a meta-analysis of the data was not seen as appropriate or feasible. Moreover, as with other literature reviews, the material incorporated was inevitably constrained by the dates adopted in the initial literature search, although it is likely that the themes emerging will be reflected in more recent studies not included in this review. Consequently, while this review does not yield definitive or clearly generalisable answers, it does indicate areas in need of further investigation which are outlined below.
No criteria were included in this review regarding the stages of dementia covered. Therefore, defining the critical period within which home care can help is one challenge for future enquiry. Similarly, the optimal amount of home care needed in order to meet specific needs merits exploration. Due to current ambiguities in their respective meanings, another question is the definition of the nature and extent of care inputs which might be deemed 'social' and 'integrated' and the potential barriers and facilitators to their implementation within diverse policy and social contexts. A further question is whether demonstrable adherence to best practice in home care can be linked to cost-effectiveness, with most of the included studies focusing on the impact of short term 'ideal type' interventions and not on their long term effectiveness or feasibility. With regard to future research, secondary analysis of existing datasets could be a relatively inexpensive way to increase our understanding of the impact of home care. The data kept by local authority commissioners and providers of home care concerning their interventions with clients over time provide a valuable resource for resolving many questions. These include whether home care is more costeffective in delaying institutional admissions if introduced early, rather than late, and what is the ideal combination between home care and informal care provision. Within the increasingly diverse home care market (Alakeson, 2007), there may also be opportunities to conduct large, well-controlled studies of the active ingredients and benefits of different types of home care. These should take account of the social context in which the services operate, the characteristics of the people for whom they care, the impact of those services on these individuals and their subsequent role in deferring transition to long term care.

Conclusion
Due to demographic trends and policy developments, recent years have seen the rapid growth of social support at home for people with dementia. However, community based home care services are not always designed to meet the specific needs of people with dementia and their carers. It has, therefore, been the purpose of this review to explore the ingredients of social support at home for this client group, in order to inform research and practice. Clearly, this exploration has faced many challenges due to the relative paucity of literature on this topic and the heterogeneity of relevant studies both in terms of their design and focus. This has led to conflicting findings and a lack of generalisability of these findings. These challenges have been exacerbated by issues of definition, with the blurred divide between health and social J o u r n a l o f I n t e g r a t e d C a r e 8 care, and the different criteria for judging knowledge in these two areas forming a barrier to the performance of a systematic review on this issue. Nevertheless, while the evidence from this literature review does not indicate complete answers or definitive solutions, its findings do suggest the effectiveness of a flexible, responsive and person-centred approach towards home care for people with dementia. It also highlights the role of inflexible commissioning practices and inadequately resourced or poorly integrated service provision in impeding the development and subsequent impact of these types of interventions.

Finland
To determine whether community residence can be prolonged by a 2-year multicomponent intervention programme.
Randomised controlled trial with 125 couples one of whom had dementia who were allocated to the control or intervention group.
At 1.6 years, a larger proportion in the control group than in the intervention group was in long-term institutional care. The intervention led to a reduction in use of community services but when the intervention costs were included, there was no difference between the two groups.

USA
To examine the association between family caregivers' decisions to initiate or stop home care provision and their stress and psychological wellbeing.
Naturalistic, observational study in which a cohort of 264 caregivers completed up to 3 interviews during 1 year Taking up home care was associated with increases in carer worry and strain and ending it was strongly associated with a decrease in depressive symptoms. Sustained use of paid home care was related to reduced overload. Riordan and Bennett (1998), UK To evaluate an augmented home care service for people with dementia.
Quasi-experimental study with 38 clientcarer matched pairs. 19 pairs received the intervention.
Clients in the intervention group were able to continue living at home for longer than the control group.

USA
To examine risk of care home placement among older adults receiving publicly funded home and communitybased services or assisted living and to explore whether these settings of care modify the relationship between dementia and risk of care home placement.
Cohort study over 5 years of Medicare and Medicaid beneficiaries (not exclusively with dementia) who received home and community based services or resided in assisted living.
The setting of care modified the relationship between dementia and care home placement: dementia was associated with a 50% increased risk of care home placement from home-based settings but not associated with placement from assisted living settings. To report on a domiciliary respite service for people with dementia using the senses framework and relationship centred care Case report drawing on the views of clients, carers and staff The service was successful in meeting the needs of people with dementia and their carers. It provided high levels of job satisfaction for staff. These mutual benefits were described as providing a sense of continuity, security and belonging.