What happens when care homes close? A review of the literature

In a fraught fiscal and post-pandemic context, English adult social care is likely to see further care home closures. To make sure that closures minimise distress and potentially negative outcomes for residents, this article reports findings from an analytical narrative review of the existing research evidence about the process and outcome of such closures in the UK. Despite the importance of these issues, there is little underpinning research to draw on when establishing good practice, with particular gaps in terms of understanding outcomes for older people, the experience of care staff, economic evidence and the perspectives of social care leaders.


Introduction
Care homes are a central component of the English adult social care system, with recent data showing that more than 390,000 older people reside in such service settings (ONS, 2021). However, many care homes cease to operate every year for a variety of financial, regulatory, practical and political reasons (Leyland et al, 2016). This raises profound questions about what happens to older people and their families (and indeed care staff) when homes close, and how best to manage such difficult situations to minimise potentially negative outcomes. Closure may be the consequence of an emergency (for example, a fire or flood), due to the costs of service delivery or building maintenance being prohibitive, or the result of intervention by regulators or local commissioners following the discovery of poor care. Most care homes in England are also privately owned (for further discussion, see later). As such, closures can result from business decisions (for example, a large chain of care homes deciding to close an individual home that is deemed to be performing poorly, either financially or in regulatory terms, or an individual care home owner deciding to retire, sell the property, convert the buildings for other purposes and so on). While it is notoriously difficult to get accurate figures on the number of closures each year, the number is high, with potentially thousands of beds closing each year (LaingBuisson, 2019).
Over time, the care home sector has become dominated by private providers (Robinson et al, 2013), following a process of marketisation with its origins in the 1980s (Hudson, 2021). Indeed, adult social care provision in England is now a market commodity, 1 and the logic of markets is such that the risk of failure must be possible to incentivise cost-effectiveness. Over time, there has also been an associated trend towards a smaller number of large providers, with as few as five offering around 20 per cent of the total provision (see Hudson, 2021: 27). In this regard, there are profound consequences should even one provider fail -as was the case when Southern Cross became insolvent in 2012, with the risk that some 30,000 beds might be lost all at once and in most areas of the country (Glasby et al, 2018).
Following almost a decade of austerity that left the sector in a precarious position, adult social care was then severely impacted by the COVID-19 pandemic. In the first year of the pandemic, more than 41,000 deaths of care home residents from COVID-19 were recorded by the spring of 2021 (the time most care home residents had received two COVID-19 vaccinations) (Health and Social Care and Science and Technology Committees, 2021). This partly reflected the association between old age and the heightened risk of serious illness or death from COVID-19. However, government policy of discharging hospital patients -seeking to protect National Health Service (NHS) acute-care capacity in the early stages of the pandemic -also had the unintended consequence of spreading COVID-19 in care homes (see Calnan and Douglass, 2022: 79-82) and caused associated alarm about group-based service models in adult social care. Later in the pandemic, following much more cautious government policy, care homes were unable to accept new arrivals, with beds remaining empty (Booth, 2022b).
In terms of the prevailing model of commissioning and provision in adult social care overall, Hudson (2021: 117) describes the pandemic as 'a colossal stress test of everything previously taken for granted about the economic, political and social order'. It is true that additional financial support was made to care providers during the pandemic (Department of Health and Social Care, 2021a), which might have extended the viability of some care homes in the short term. However, the strains and consequences of the COVID-19 pandemic have caused the closure of over 130 care homes and, as such, the estimated loss of 1,600 beds since the summer of 2021 (Booth, 2022a).
Additionally, a growing cost-of-living crisis also has potentially disastrous consequences for the sector, with concerns about spiralling operational costs (including for energy and food) and staff retention (ADASS, 2022;Melley, 2022). The government in England has pledged to ensure that there is a 'fair cost of care' (Department of Health and Social Care, 2021b;Curry, 2022). However, providers and local councils are at an impasse as to how to achieve this in practice in the current context and in a sector that has been historically underfunded. Importantly, although we recognise that marketisation in adult social care, the considerable impacts of the pandemic on the sector and rising living costs are relevant to other contexts internationally, looking across the dimensions discussed, the UK (and particularly the English) care home sector is very specifically positioned. In such a political, fiscal and post-pandemic context, future care home closures are to be expected.

Previous reviews
Ten years ago, Holder and Jolley (2012) conducted an important and comprehensive international review of the literature on 'involuntary relocation' (though not solely within a social care context) (for an earlier example, see Smith and Crome, 2000). The review aimed to identify international empirical evidence about the consequences and impact of relocation, as well as good-practice recommendations to shape future approaches -albeit with a focus particularly on forced relocation between nursing homes, though also involving residents transferring inter-service (including hospitals and community-based settings). This is crucial work, as the received wisdom is that some forms of relocation can have extremely negative consequences for older people, possibly even increased mortality, if the move is handled less than optimally.
However, in practice, very few of the studies included in previous reviews focus on closures of care homes for older people. Instead, many of the insights generated focus on older people moving into existing services, or on relocations from a variety of different health and social service settings. Thus, despite reviewing over 3,500 initial results, Holder and Jolley found no studies about the outcomes of care home closures in the UK and only one previous UK research team exploring closure processes, while Smith and Crome identified only one UK study that specifically explores care home closures. This was a four-page article from the early 1990s, where 59 residents moved from one large care home and from one psychiatric service initially opened to rehouse long-stay hospital patients -perhaps more of an article on deinstitutionalisation than on the closure of current care homes. Therefore, when Glasby et al (2018) published the results of an evaluation of a single-site closure programme in 2018, they concluded that this was one of only a handful of studies conducted in the UK and probably internationally, with the issue of care home closures still a largely evidence-free zone. In particular, the literature at that time seemed extremely limited in terms of insights around the outcomes of care home closures. Previous studies were also predominantly clinical in nature/professionally led, making virtually no attempt to involve or consider the different experiences of older people, families and care staff. However, these contentions require empirical analysis through a structured review.
Ten years on from Holder and Jolley, this article revisits and reviews literature relating to care home closure, seeking to contribute to improved outcomes for older people and to minimise the risk of harm by ensuring that future closures are planned in an evidence-informed manner. This is the backdrop to a new national study on care home closures funded by the National Institute for Health and Care Research (known as 'Achieving Closure') but is also particularly timely considering the increased rate of closures that is anticipated following the pandemic and in response to the current cost-of-living crisis. In doing so, we empirically demonstrate what is known about the process of care home closures, the perspectives and experiences of the different stakeholders involved, potential outcomes, and any recommendations as to what might constitute good practice in such a complex and contentious area of policy and practice. Given that service models, policy priorities and terminology all vary so much across different systems, we focus here on evidence from England (which, as set out earlier, has a very particular type of care home market, with some in-built tensions and complexities). However, we recognise that there are likely to be insights emerging from this context that speak to, or can influence, care home closure processes and outcomes internationally.

Methods
This review set out to answer three key questions: • To answer these questions, we undertook a narrative analytical review, replicating a number of searches previously published (Ince et al, 2022) and adopting an approach used in previous research for the Department of Health/National Institute for Health and Care Research into the discharge of older people from hospital, the appropriateness of older people's admissions to hospital and the discharge of people with learning disabilities from long-stay hospitals (Glasby et al, 2006;Thwaites et al, 2017;Ince et al, 2022). The search was undertaken by a specialist health and social care library and literature searching team at the authors' institution, searching the following databases: • Applied Social Sciences Index and Abstracts (ASSIA) • Health Management Information Consortium • Medline • Scopus • Social Policy and Practice (including CareData, Social Care Online and AgeInfo) • Social Services Abstracts • Social Sciences Citation Index Search terms varied according to the details of the specific terms used in specific databases but sought to combine different terms for 'older people'/'elderly', 'care homes'/'long-term care' and 'closure'/'decommissioning'/'disinvestment'/'relocation' /'transfer' (including variations in each of these terms) (for a worked example, see Table 1). The filters used ensured that the outputs were published in English between 1990 and July 2022. Outputs were then included if they met the following criteria: • reporting original empirical research around the process and/or outcome of care home closures in the UK, or providing a structured review of such research; • focusing on older people (aged 65 and over) with or without additional needs; • having been published since 1990 (this year saw the passage of the NHS and Community Care Act, which essentially created the current English social care system and led to more market-based approaches in which the potential for closure was more prominent than previously); and • investigating outcomes associated with care home closures and residential home closures involving stakeholder outcomes, such as carers, councils, providers, social workers and any other groups involved in care home closures.
Outputs were excluded if they: • focused on other kinds of closures (for example, nursing homes, which are differentially regulated in England [for a discussion of different care settings, see Care Quality Commission, 2022]) or other groups of people (for example, people of working age with learning disabilities or mental health problems); and • did not report original empirical data.
The reference lists of all included articles were also searched, and any additional studies that met our criteria were included in the review. After carrying out these searches, all titles and abstracts were reviewed independently (by Shazia Zafar and Tom Douglass), with any differences discussed in a team meeting and resolved by consensus. Where a decision could not be made based on a title/abstract, the full output was obtained and screened. Once a decision had been taken to include an Total output in the review, data were extracted using a pro forma, which recorded such information as: the nature of the care home closures; the study design and methods, as well as results or evidence relating to the closure process; the outcomes of closure; the experiences and perspectives of a range of stakeholders, including older people, families, care staff and social workers; and policy and practice implications. Content was independently extracted from five papers (by Shazia Zafar and Tom Douglass) to help ensure consistency and rigour. These pro formas were then sent to a third reviewer (Jon Glasby) to further scrutinise for consistency. After this, the remaining outputs were reviewed by both Shazia Zafar and Tom Douglass, with Jon Glasby reviewing the final documents before analysis.
The initial database searches, abstract and full-text screening were undertaken during May and June 2021, with pro forma extraction finalised by February 2022. The searches were then refreshed in July 2022 to identify any new articles written since the initial searches.

Results
The nature of the evidence Overall, we identified 323 initial outputs for title/abstract review (which reduced to 248 after 75 duplicates were removed [see Table 2]). Ultimately, 12 outputs met our criteria for inclusion in the review, and these form the basis of this subsequent article. Importantly, these 12 studies emerged from eight overarching studies or programmes of research (for summaries, see Table 3). When we refreshed our searches in July 2022, another 67 abstracts were reviewed, but none were identified for inclusion (for an overview of this process, using the PRISMA framework [see Hewison et al, 2020], see Figure 1). As discussed later, this feels like a very small number of outputs/studies for such a crucial (and growing) policy and practice issue, and many of the individual studies are either very dated or provide only brief snapshots of the issues at stake.

The outcomes of care home closures
Of the 12 studies, only five reported empirical outcomes from care home closures, including both quantitative and qualitative outcomes (for a summary, see Table 4).   Gascoigne and Mashhoudy (2011) Quantitative. Comparing the mortality rates (after one year) of a sample of older people from private residential homes who were relocated involuntarily (n = 71) with those of a sample of older people who experienced no relocation in their first year of residential care and/or had just moved to a private residential home (control group: n = 412). Hallewell et al (1994) Quantitative -controlled, prospective design. Before residents moved, age, sex, length of stay and physical and psychiatric diagnoses were recorded. Care home staff also interviewed using version of Crichton Geriatric Behavioural Rating Scale (CGBRS). This was repeated at three months after moving, and then again one year after moving. A total of 62 people lived in the two homes that closed (though three were excluded from the research) and were relocated to 19 different care homes. As such, 30 residents from the first home and 29 from the second home were included in the sample, with a control sample of 30 (n = 90).

Holder and
Jolly (2012) Systematic review of literature published between 2000 and 2012 on involuntary relocation between nursing homes and other settings. Review aimed to identify and evaluate empirical evidence on the consequences and impact of relocation on nursing home residents. It also aimed to identify interventions/recommendations to maximise the safety and well-being of residents.
Leyland et al (2016) A total of 34 semi-structured interviews with a range of stakeholders (11 residents, two relatives, six care home managers, 13 staff and two independent advocates). Study aimed to analyse if a protocol designed by a local council (informed by existing research) was adhered to and the influence it had on the experiences of residents relocated between care homes.

Personal Social Services
Research Unit (PSSRU), University of Kent 2005;2007) A total of 33 closure guidelines (with information collected from 55 of 150 councils in England) were analysed using qualitative and quantitative approaches to identify their purpose and scope, the ways in which roles and responsibilities were defined and allocated, and suggestions and recommendations for good practice. Additionally, 45 residents, relatives and unpaid carers were interviewed (using a semistructured approach) in connection with eight care home closures (with seven participants discussing a further seven homes that the authors call 'non-case study' home closures). Further, 21 care management staff were interviewed using semi-structured interviews. Three senior strategic managers within councils were additionally interviewed. Seven managers also kept a log or diary of their activities and time spent considering face-to-face contact with service users, relatives and staff, telephone calls, and admin. Reed et al (2003) Mixed-methods design and part of a larger project. Quantitative methods analysed incidence of relocations. Questionnaires sent to managers of all care homes in the local authorities (n = 149; 56.4 per cent response rate) concerned with the number of residents, registration categories, resident relocations and management strategies. From this, 10 care homes were approached for interviews with residents who had relocated. Qualitative methods were used to delineate and explore relocating residents' experiences and perspectives of moves. This study focuses on relocations (rather than closures per se) but includes some perspectives from homes that were closing and so has been included in this review. Glasby et al (2011) and Robinson et al (2013) Qualitative. Semi-structured interviews with Association of Directors of Adult Social Services (ADASS) and senior nominees with experience of care home closures (n = 12 from nine local authorities). Focus on what constitutes good practice in closure processes. The 2011 output is a linked national good-practice guide that explores the same underlying data.
(Continued) Mixed methods. Semi-structured interviews with older people, families, care staff and social workers on the closure process in one care home and a linked day centre. Longerterm outcomes were measured via EQ-5D and a questionnaire based on older people's experiences of care services for older people being resettled in the first 13 homes to close as part of a broader council-wide closure programme (n = 314, of which 131 people completed questionnaires at all time points). Data were collected at three time points: 1. baseline first assessment 2. 28-day review 3. 12-month follow-up after moving to new services A sixth provided a structured review of outcomes evidence from a broader series of relocations (Holder and Jolley, 2012).
In 1994, Hallewell et al studied the residents of two care homes for one year after closure (which closed at short notice due to financial problems). They explored   Hallewell et al (1994) Before residents moved, physical and psychiatric diagnoses were recorded, as well as age, sex and length of stay. Care home staff were also interviewed using a version of the Crichton Geriatric Behavioural Rating Scale (CGBRS). This scale is used to assess: mobility, orientation, communication, cooperation, restlessness, dressing, feeding and continence.  Although this is primarily a qualitative study, the authors report the health impacts and mortality rate within the (small-scale) sample.

Gascoigne and
Mashhoudy (2011) This study reports mortality rate one year after relocation. Leyland et al (2016) This is another qualitative study, but the research reports participant reflections on physical health outcomes (mobility and amount of personal care).
Glasby et al (2018) This study uses a standardised measure of health and quality of life (EQ-5D), as well as a Likert scale around outcomes that older people say they value about care, at three timepoints: (1) baseline first assessment; (2) 28-day review; and (3) 12 months after to moving to new care home.  Reed et al (2003) This research indicates the need to introduce strategies to support older people in decision making about relocation. This includes promoting the right to choose and be involved in decisions about their care home environment. It shows that resident views and involvement in relocation decisions in practice vary substantially.

Williams et al (2003)
A range of recommendations for good practice are made by residents and relatives identifying how closure process could be improved in the qualitative data analysed: notification, information sharing, help and support, enabling and facilitating visits to new homes, and maintaining standards of care during closures, the move and to support settling in. Williams and Netten (2005) (see also  Large numbers of councils in England had no guidelines for voluntary closures and there was considerable variation between guidelines that did exist. Nevertheless, the available council guidelines offer a useful starting point for more national approaches and for how best to support residents. Williams et al (2007) This study highlights considerable variation in practice: policy/national guidelines should encourage consistency and equal access to support, and make responsibilities/accountability clear. Robinson et al (2013) (see also Glasby et al, 2011) Alongside wider stakeholder engagement and support, the process of closure requires a combination of a number of factors: strong leadership; clear strategic goals; a fair decision-making process; strong evidence of the need for change; and strong, clear communication. Leyland et al (2016) The model proposed in this study explains how the use of a protocol provided opportunities to improve residents' experiences of involuntary relocation and transfer, and to reduce the experiential risk to residents. Glasby et al (2018) This research highlights the importance of meaningful involvement in initial closure decisions; good, principled and ongoing communication; thinking through what older people value about the care home environment in which they live (which might be different to what councils or social workers value); and paying attention to inevitable feelings of loss and sadness.
whether it was possible to predict which residents would be negatively impacted by closures. Before residents moved, age, sex, length of stay and physical and psychiatric diagnoses were recorded. Care home staff were also interviewed using the Crichton Geriatric Behavioural Rating Scale (CGBRS) (a scale used to assess function across eight different activities: mobility, orientation, communication, cooperation, restlessness, dressing, feeding and continence). This was repeated at three months after relocation and then again one year after moving. While relocation had a statistically significant impact in terms of higher restlessness in the short term when compared with a control group, there was no statistically significant difference between the relocation and control groups after one year. Residents with high CGBRS scores before closure were associated with increased mortality (as perhaps might be expected), and requiring assistance getting dressed was also a good indicator of increased likelihood of mortality following relocation. While 12 residents died in the year after closure, the research could not show that closure itself had a statistically significant impact on mortality.
In 2003, Williams et al found that 14 per cent of the residents in their sample died after their care home closed, with 9 per cent dying within three months of moving. The health of a further 2 per cent of residents deteriorated, and 2 per cent were unsettled. A total of 60 per cent of relocated residents reported no continuing health issues, and half were settled in their new homes. However, this was simply a reporting of patterns of mortality (within a small sample), with no attempt to explore issues of causality. A third study has similarly been unable to show that care home relocation had a statistically significant impact on increased mortality one year after closure compared with care homes that had recently closed (Gascoigne and Mashhoudy, 2011). Importantly, in all three of these studies, research was limited by small sample sizes, indicating the methodological challenges that arise when trying to research these issues (Holder and Jolley, 2012).
Looking at outcomes other than mortality, Glasby et al (2018) evaluated outcomes for older people from a very large care home closure programme in Birmingham, collecting data on health and well-being at three time points (at closure, after 28 days and then after 12 months). After initial distress and anger from older people and care staff in particular, outcomes after 28 days and one year were either the same or slightly better than before, suggesting that some things ultimately stayed the same for some residents and actively improved for others. The authors describe this as a 'game of two halves', where initial distress might be inevitable but where closures (where existing services were less than optimal and relocation is managed well) might not make things worse for many people and might actively improve the situation for some residents. This was supported by local social workers, who felt that some older people were very upset or disorientated during the initial closure process but that it was ultimately the right thing to do to improve outcomes for older people in the longer term. Although some residents had died after the closures, this was thought locally to be consistent with what would have occurred with or without the closures, and not outside the boundaries of usual mortality in a group that, by definition, tends to be very frail and to have a significant number of physical and mental health problems.
A similarly mixed picture emerges from the qualitative research carried out by Leyland et al (2016). Here, some stakeholders believed that mobility improved following closure and that the amount of personal care required lessened. Others believed that there had been some deterioration in the health of residents. Leyland and colleagues additionally briefly considered negative social outcomes associated with the closure of care homes, such as a lack of integration into a new care home (especially where residents moved without friends or in advance of friends). However, detailed analysis of social outcomes was limited throughout the studies reviewedand there seems to be scope for further research in this regard.
Relatedly, none of the studies included in this review consider potential differences in outcome by social divisions/protected characteristics, such as gender or ethnicity. Although Leyland et al (2016: 391) discuss residents 'finding a role within the social structures in their new home', they do not analyse if and how positionality might shape the experiences of closure or the ability to integrate into a new care home. Indeed, most of the research reviewed is very broad-brush in terms of the distinctions drawn between different groups, and the research reviewed does not examine differential outcomes between social groups. 2 As is discussed in more detail later, there is also very little research that tries to understand the issues at stake from multiple perspectives (such as older people, families, care staff, social workers and senior leaders) (for rare exceptions, see Glasby et al, 2011;2018;Robinson et al, 2013).
The process of care home closure Table 5 provides an overview of insights into potential good practice during the process of closing care homes. In the early 2000s, for example, researchers at the Personal Social Services Research Unit (PSSRU) at the University of Kent produced a series of linked outputs on the care home market. This included an attempt to collect and analyse any written guidelines/local protocols and to identify good practice 2005). Information was collected from 55 of 150 local councils in England. Nearly two thirds of responding authorities reported having a closure protocol, though document scope varied greatly. While we do not know about arrangements in councils that did not respond, this still suggests that a significant number of local authorities at this time did not have policies in place to guide local closures. From the policies that were able to be analysed, good practice was felt to include taking residents' social and personal needs into account (for example, existing friendships with residents, preferred geographical location and ability of family and friends to visit) and maximising residents' ability to make informed choices. However, it was noted in several policies that enabling choice was likely to be difficult to achieve in practice. A quarter of the protocols referred to the length of notice care home owners should or ideally would provide (typically, these recommended notice of around one month or as much notice as possible). Considerable importance was also placed on involving care staff in the closure process, respecting their friendships with residents and their likely concern for residents' future welfare.
Around a decade later, Robinson et al (2013) placed the issue of care home closures in the broader context of health and social care disinvestment, rationing and decommissioning. They highlight a lack of national guidance and a tendency for local areas to develop their own local approaches, often in isolation from each other and without sharing lessons learned outside their own authority. In particular, the linked studies by Robinson et al (2013) and Glasby et al (2011) place a significant emphasis on meaningful, honest and ongoing communication between local authorities and residents, families and care staff. While local closure processes try to prioritise communication, it is crucial that key stakeholders (especially older people and their families) feel they have had a chance to contribute to initial decisions to close services and feel that their voices have been heard -even if they do not agree with the actual decision that has been taken. Communication should also be ongoing, particularly when there are delays or plans change. It should also be honest, even if this leads to difficult conversations or people to admit that they simply do not know the answer to a particular question yet. Robinson et al also found that care staff were a crucial source of information for older residents. No matter how much effort had been put into communicating, many older people could not really remember the information they had received from the council, tending to get the bulk of their information and updates from chatting to care staff. In situations where care staff did not feel sufficiently involved in the closure process, therefore, this sense of dissatisfaction and a sense of frustration at not knowing what might happen next could easily be transmitted to residents. This points to the need for really good communication with care staff/ providers. However, this study also found that basing social work assessors in homes that are closing helped these practitioners to get to know residents, families and care staff better, potentially helping with the communication.
When reflecting on their experience of care home closures, directors of adult social services stressed the importance of having sufficient time to undertake such complex and sensitive work at the pace of individual older people (Glasby et al, 2011). Often, there can be a balance to be struck between working in a person-centred way and trying to conduct the closure in a timely manner so as not to extend uncertainty. Nevertheless, it was felt that once the decision to close a care home had been made, it was crucial to stick to established deadlines and timescales. Williams et al (2007) highlight the types of time-intensive tasks involved in the closure process (such as paperwork, face-to-face interactions and calls), which also indicates the importance of adequate time frames when conducting closures. Sufficient time for residents to talk to residents and staff in care homes prior to relocation is also shown to be salient (Reed et al, 2003).
Other outputs have evaluated the use of closure guidelines during real closures, highlighting what was and was not perceived to have worked in practice. For example, Williams et al (2007) analysed care managers' experiences of care home closures and associated good-practice recommendations. Interestingly, care management staff from four councils in England interviewed in this research reported that they did not use local council closure guidelines and their belief in the value of guidelines varied. In particular, staff noted dilemmas in the closure process, including the need to balance the availability of places at different homes, a desire to take into account the perspectives of residents and relatives, and a recognition of the importance of promoting rights and freedom, and not being seen to 'recommend' specific homes, which might be interpreted negatively by providers. These potential tensions could appear incompatible at times or result in serious compromises.
Similarly, Leyland et al (2016) explored whether a protocol designed by a local council was adhered to in practice and the influence it had on the experiences of residents relocated between care homes. The research considered the perspectives of multiple stakeholders. In terms of protocol adherence and translation into practice, the authors articulate four themes, including: (1) involvement, (2) staff approaches, (3) consistency/familiarity and (4) preparation. First, in terms of involvement, visits were offered to relocating residents, though they were not always taken up, and residents also felt that these visits were short. The presence of choice was emphasised by stakeholders, including of care home and even of bedroom. Family members were able to be part of decision making relating to relocation, though residents had a choice about how involved relatives could be. Provision of information and consultation were identified as important, and it was said that this occurred throughout the process for the majority. Independent advocacy was said to be useful for residents who lacked capacity and did not have relatives to support them.
Second, staff were supported and offered reassurance throughout the process, and this was facilitated by existing positive relationships. The process of closure was felt to be person centred and particularly successful in pacing moves and being flexible in relation to individual needs. Staff were able to separate their own concerns about moving jobs from the needs of residents -and this was achieved because of being well informed about closures by managers.
Third, in terms of consistency and familiarity, friendships between residents were maintained when moving, which helped with residents acclimatising to their new home. Additionally, similarities in room and building layout were thought to be important, as well as residents being able to bring furniture with them. Finally, in the theme of preparation, the protocol and its translation into practice meant that relevant paperwork was used consistently across care homes, which helped with the transfer of documentation. Additionally, clear leadership/communication/ continuity resulted from a single care manager working full time on the process, which resulted in positive working relationships and accessibility. The logistics of relocation were also felt to be important and given close attention, including transport, enhanced staffing levels, the scheduling of moves and the consideration of impacts on residents who were the first or last to move (particularly as the moves were near to Christmas). Overall, Leyland et al (2016) provide salient qualitative evidence about what aspects of planning and preparation work in practice on a small scale at a local level.
In a similar manner, Reed et al (2003), as part of a broader project on care home relocation, have discussed the experiences and perspectives of older people moving between care homes. Although choice and involvement in the closure process is typically recognised as good practice in most accounts, Reed and colleagues found that the extent to which resident views were considered and residents were involved in relocation decisions varied substantially. They suggest that residents' experiences of involvement in relocation can be described under four types or categories (which could vary for the same person across different moves). People in the category of 'residents reluctant to move' often had little involvement in decision making around their move, which was based on an assessment of their needs, rather than their preferences. Where residents were 'passive', it was, however, unclear whether they had been offered a range of possibilities. The authors indicate that this category was defined by a lack of autonomy and choice. They suggest that a continuum of involvement in relocation decisions is evident in their data (reflecting certain personal and institutional conditions). Thus, Reed et al (2003) highlight a potential disjuncture between choice and involvement as good-practice recommendations and what is actually feasible or occurs in reality. In contrast, Glasby et al (2018) argue that even apparently well-conducted closures might not be able to prevent anger and distress en route, even if longer-term outcomes are positive for some. As they suggest, 'success can often look like failure part way through' (Glasby et al, 2018) and it might sometimes be a question of social care leaders holding their nerve -understanding and acknowledging the anxiety and distress that closures cause, whilst also working to improve longer-term outcomes.

The voices heard during the closure process
There are a range of stakeholders involved in, or impacted by, care home closures. Table 6 sets out the types of stakeholders that existing research has included as participants, revealing the limited nature of the evidence from different perspectives. Crucially, only four of the reviewed outputs directly include the experiences of older people (Reed et al, 2003;Leyland et al, 2016;Glasby et al, 2018). Indeed, Hallewell et al (1994), rather than asking older people directly about mobility, restlessness and other outcomes, instead asked care staff for these data. There are even fewer outputs (three) analysing evidence collected from relatives/family members. Importantly, Glasby et al (2018) asked older people and family members about the experiences of the closure, while  sought relatives' reflections on good practice during various dimensions and stages of the process. The other output, however, analyses data from only two (n = 2) family/relative participants (Leyland et al, 2016).
The experiences of care staff are similarly poorly represented in research, with their voices found in only four outputs. In the research by Hallewell and colleagues (1994), although care staff participants were asked to contribute outcomes evidence relating to older people, their own experiences and the outcomes for care staff themselves are neglected. Glasby et al (2018) and Reed et al (2003) do analyse the experiences of residents, but as in the research by Leyland et al (2016), the focus is on maximising the experience of, and outcomes for, older people, and again, such research does not examine the outcomes of closure for care staff themselves (for example, in terms of employment/financial outcomes, or stress and anxiety).
Despite the considerable potential impacts of closure on older people, their families and care staff, our review shows that there are very few examples of research engaging with the experiences of these stakeholders. The evidence that does exist is also now quite dated (with only two studies including the views of older people, their families and/or care staff in the last 15 years). This might suggest that limited value is attached to the experiences of older people, their relatives and care staff, and that such stakeholders have limited power to influence care home closure processes. However, there have additionally only been a limited number of attempts to understand the experiences of the more powerful groups within adult social care (relatively speaking) who are leading the difficult work of care home closure, including social workers and social care leaders/commissioners (see, for example, Glasby et al, 2011;2018;Robinson et al, 2013). In this regard, a dearth of research attention is evident across the breadth of stakeholders who are influencing or are impacted by care home closures.

Discussion
In this review, we have analysed the existing research evidence about the outcomes and processes of care home closure. We have also explored the extent to which the experiences of the breadth of relevant stakeholders are represented in existing studies. In this regard, we found limited available evidence across each of these three areas. First, in the available research reviewed, a mixed picture across closure outcomes, such as mortality rates, health and well-being, was evident. The evidence base also suffers from considerable limitations (such as small sample sizes, restricting statistical power). Importantly, however, in the context of existing services operating less than optimally and when a relocation is managed well, some of the research does show that closure does not necessarily make things worse for many people and may actively improve the situation for some residents (Glasby et al, 2018).
Furthermore, when considering good practice in the process of closure, due to the limited nature of the evidence and the lack of consistency and consensus in the understandings and processes analysed in research, it was difficult to draw out thematic similarities or patterns from the available processual evidence. However, our analysis does show salient (if limited) insights about effective communication strategies and the significance of time when conducting closures, as well as evaluative research examining the difficulties of translating local guidelines into practice and the differential involvement of residents in decision making about relocation.
Finally, across the breadth of relevant stakeholders influencing or impacted by care home closure, our review again highlights limited empirical analysis of the experiences of all relevant stakeholders in the existing research. Even the experiences and insights of the most powerful actors in adult social care have been minimally analysed in studies of the processes and outcomes of care home closure. This, of course, also means that we know little about the experiences of the people most impacted by the closure of care homes: older people, their relatives and care staff.

Conclusion
Overall, particularly for such a salient policy and practice issue, there is currently little underpinning research evidence that might be drawn on when establishing what works in care home closures and how to improve outcomes for older people. The limited research evidence is likely to be a significant factor in the lack of evidencebased national guidelines to shape care home closure (and even the evidence that does exist was gathered in prior political and fiscal contexts). It is true that care home closures are sensitive and there are layers of complexity. Emergency closures represent relocation in a crisis, while planned closures may attract political and media debate, and associated controversy. In both contexts, it can be difficult for those conducting the work of closure to extract or disseminate important learning points, and local authorities are thus forced to adopt locally based and often unevaluated procedures. Although research on this topic may not be simple to conduct, this point, reinforced by the findings of our review, supports the need for well-planned further empirical research on the processes and outcomes of care home closure.
Our review has demonstrated, first, that further research is required on a range of outcomes when care homes close. This includes analysis of mortality but must also include other outcomes, including health and well-being, as well as social outcomes. In this regard, researchers also need to analyse the importance of social divisions (such as gender and ethnicity) in terms of the experience of the processes and outcomes of closure. Although maximising outcomes for older people is of paramount importance, other groups, such as care staff, are also impacted by care home closure, and research should seek to understand outcomes for them.
Future research should additionally seek to expand the processual evidence base through the further inclusion of professional expertise and the identification of good practice in the work of care staff and social workers conducting the complicated work of closure. Perspectives from senior social care leaders are similarly lacking. Above all, however, the lived experiences of older people and their families should be centre stage when analysing what and how closure processes work, as well as asking for whom processes work. We believe that the increasingly fraught and precarious fiscal context, alongside an enhanced political and popular visibility for the adult social care sector in the aftermath of the COVID-19 pandemic, provides an important backdrop in which to begin such a crucial programme of research.
Notes 1 As Hudson (2021: 28) explains, it is accurate to describe adult social care as a 'quasimarket', 'where provision is effectively privatised but much of the purchasing (or commissioning) of care still lies with the public sector'. 2 Glasby et al (2018) do, however, draw attention to high levels of mental health problems and reflect on the experiences of people specifically with dementia.

Funding
This study/project is funded by the National Institute for Health and Care Research (NIHR) under its Programme Grants for Applied Research Programme (NIHR201585). The views expressed are those of the authors and not necessarily those of the NIHR or the Department of Health and Social Care.