The impact of COVID-19 lockdown measures on older residents' social connections and everyday wellbeing within housing schemes that provide care and support in England and Wales

The COVID-19 pandemic and the lockdown measures imposed as a result affected the lives of people in all parts of society across the world. In 2020, during the first UK national lockdown, older adults (aged 70 years and over) were told to ‘shield’ within their homes, as they were regarded as being at higher risk of serious COVID-19 infection compared to other age groups. This paper explores older adults' experiences of COVID-19 lockdown measures whilst living in housing with care schemes for older people. The purpose is to examine the impact of the lockdown measures on scheme life including social connections amongst residents and their general everyday wellbeing during this time. We present qualitative findings based on interviews with 72 residents who took part in longitudinal and cross-sectional interviews across 26 housing with care schemes. Data were analysed using a thematic framework approach to examine specifically their experiences of living in housing with care schemes during the 2020 UK lockdown. The paper highlights that COVID-19 restrictions had a detrimental impact on the social connections and interactions of older residents living in housing with care schemes, as well as on their feelings of autonomy and independence. Despite this, residents adapted and coped with self-isolation restrictions and sought out positive ways to maintain social contact with others inside and outside to the scheme. We further highlight the tensions that providers of housing for older adults faced in promoting residents' autonomy and connectedness whilst also trying to provide a safe living environment and protect residents from risk of COVID-19 infection. Our findings apply not only to a pandemic situation but to the broader understanding of how housing with care for older adults must navigate between autonomy and support.


Introduction
The World Health Organisation declared COVID-19 (coronavirus) a pandemic on 11 March 2020. In response, on 23 March 2020, the UK Government announced a "lockdown" on the whole country, imposing a ban on non-essential travel and contact with people outside the household. Adults over 70 years were instructed to 'shield' within their homes, due to being classified as a "clinically vulnerable" group (Government, U. K, 2021), which involved physically distancing from others outside their household. Older adults have been particularly negatively affected by COVID-19, dying in disproportionately higher numbers, particularly those living in long-term care facilities (Miller, 2020). However, treating older adults as a homogeneous group and categorising all people aged 70 and over as vulnerable can reinforce negative ageist messages that older adults are dependent, frail and at risk (Pentaris et al., 2020). Implications of such social distancing and shielding practices have the potential of adverse side effects including increased social isolation, loneliness, experienced ageism, enhanced economic risk, delayed medical treatment and difficulties in getting basic needs met (Miller, 2020). Further, these measures disproportionately affected older people whose social contact is most often outside of the home (Armitage & Nellums, 2020;Richardson et al., 2020), particularly for those living alone.
Initial studies exploring the impacts of the pandemic on older people suggest that social and emotional loneliness, anxiety, depression, and insomnia have increased in older adults (Krendl & Perry, 2021;Wong et al., 2020), and that the biggest challenges relate to social constraints and activity restrictions (Heid, Cartwright, Wilson-Genderson, & Pruchno, 2021). The pandemic has highlighted not only the vulnerabilities of some older people but also of the settings within which they live (Wagner, 2020). In this paper we examine the impact of COVID-19 and associated state-imposed measures on residents living in housing schemes that provide some level of care and support on site in England and Wales. Our research question aimed to explore: what the impacts of COVID-19 lockdown measures were on older residents' social connections and everyday wellbeing within housing schemes that provide care and support services? We argue that the liminal status experienced by scheme residents did not originate in lockdowns but was instead intensified during the implementation of these measures and that residents' invisibility increased through the imposition of blanket restrictions and the ambiguity surrounding lockdown rules for housing with care. However, this ambiguous location did not inhibit residents from adapting to highly restricted social lives or from exercising care and support to those around them in the micro-neighbourhoods of scheme life. We first examine relevant literature on housing with care as a distinct model of "ageing in place" and the emerging literature on the impact of COVID-19 restrictions on older adults' social wellbeing.

Housing with care for older adults
Within the UK, there is a proliferation of housing models that provide varying levels of care and support for older adults, with extra-care housing emerging as one distinct model for delivering more personalised services to primarily older adults (Riseborough, Fletcher, & Gillie, 2015). In 2020, 18.6% of the UK population were over 65 years of age and projections indicate this percentage will rise to nearly a quarter by 2050 (ONS, 2021). The Local Government Association (2017) has argued there is an urgent demand for more age-friendly, high-quality housing for older citizens and a growing need for cost-effective housing for older adults that combines housing and person-centred care in the same setting.
Key characteristics of housing with care, which encompasses extracare housing, sheltered housing, retirement, and independent living, are self-contained accommodation with its own front door, an ethos of supporting independence, flexible care offer, on-site 24-h care or support (if needed), access to activities and social events and various communal facilities that could include a shop, restaurant, and gardens (Housing, 2019). Research evidence has found several benefits to housing with care residents which relate to health and mental cognition, maintaining independence and reducing loneliness (Evans et al., 2017;Evans & Vallelly, 2007;Netten, Darton, Baumker, & Callaghan, 2011). Housing with care provides opportunities for an increased sense of security and safety, along with social interaction that are facilitated by the physical environment and communal facilities provided on site, which are purposively built to meet the needs of older people (Housing, 2019).
To age well, one must maintain the highest level of autonomy, wellbeing, and self-identity in the face of likely loss of competence (Wahl, Iwarsson, & Oswald, 2012). Environmental gerontology perspectives assume that old age is profoundly influenced by the physical environment, which has the potential to impact negatively and impose constraints to ageing well. Initiatives such as new housing solutions can offer opportunities to enhance ageing by supporting declining competencies (Wahl et al., 2012) as well as bolstering social interactions and wellbeing. Studies have highlighted the physical, social and mental health benefits of these forms of housing for older residents (Baumker, Netten, & Darton, 2010;Netten et al., 2011). Evans and Vallelly (2007) found that residents of well-designed housing with care settings tend to experience relatively higher levels of health and wellbeing. Further, in their detailed case study of a UK extra care housing scheme, Baumker et al., (2010) found an increase in positive social care outcomes and improvements in quality of life of residents living in the schemes, while the communal features of these schemes increased older adults' perceived autonomy compared with those living independently in the community (van Bilsen, Hamers, Groot, & Spreeuwenberg, 2008).

Older adults' autonomy and independence
COVID-19 restrictions on the use of the physical environment, may have impacted levels of autonomy and independence in older adults living in housing schemes. Hillcoat-Nalletamby (2014) observes how the concept of independence has become central to policy agendas that promote ageing in place. Her research addresses older adults' understandings of "independence" in relation to three different residential settings, including extra-care housing. In extra-care housing, residents portrayed decisional autonomy when they emphasised the importance of being able to do what they wish, but also recognised that within their setting, help is at hand should it be needed. Residents of these settings also exercised executional autonomy when preferring to complete daily living activities for themselves. Furthermore, being able to assist others within the setting, gave respondents a sense of purpose and self-esteem, thus adding to independence and autonomy.
The concepts of independence and autonomy are inter-related but not synonymous. The notion of independence has been traditionally grounded in the ability to be physically self-sufficient in carrying out personal and social tasks (Haak, Fänge, Iwarsson, & Ivanoff, 2007). The construct of independence is complex, and conceptual distinction between independence and autonomy is needed (Haak et al., 2007). According to Haak et al., (2007) during the ageing process, views of independence shift from being independent in carrying out an activity or task without help to making autonomous decisions about life within their own homes. The notion of autonomy thus relates to freedom, independence, self-government and self-determination (Welford, Murphy, Rodgers, & Frauenlob, 2012). The experience of freedom is an important part of a person's dignity and integrity (Riedl, Mantovan, & Them, 2013) and during COVID-19, notions of freedom in older adults who were required to self-isolate were challenged.
Gerontological discourse on distinctions between the third and fourth ages in later life acknowledges these overlapping notions of independence and autonomythe third and fourth ages represent social imaginaries or dominant representations of older people that shape social, cultural and medical understanding of later life. The third age represents older adults who are enjoying independent living and are autonomous agents pursuing a healthy active life while the fourth age signifies a time of increased frailty and dependency and physical and mental decline (Higgs & Gilleard, 2015). The fourth age signifies a loss of independence and agency along with an increased reliance on institutional care,. As the opposite to active ageing, it reiterates a dominant discourse that the third age is aspirational and ideal with the fourth age to be avoided and resisted (Higgs & Gilleard, 2015). Older people experiencing frailty have very few empowering narratives to draw upon (Gilleard & Higgs, 2010).
In their English study of older adults' identities and experiences in extra care housing, West, Shaw, Hagger, and Holland (2017) suggest that residents living with disabilities and health conditions experience "a sense of 'persistent liminality', between the third age and the 'gravitational pull' of the fourth age" (p. 1881). Liminality is reinforced through their status as occupying housing that supports independent living to a large extent while also relying on the care and support of others to maintain this status but also through wider discursive elements embedded in discourse on the function and purpose of extra care housing. Further, Johnson et al., (2020) argue that for extra care residents the boundaries between the third and fourth age are less clearly defined as suggested by their increased reliance on others for care and support.
The concept of liminality emphasises an in-between or 'threshold' space in which older adults are in a space of transitioning between social structures and socially and culturally inscribed identities and roles (Nicolson, Meyer, Flatley, Holman, & Lowton, 2012). Foregrounding liminality as a 'rite of passage ', Turner (1987) discussed how liminal states are culturally inscribed, in-between states in which individuals are transitioning between visible statuses and recognised identities. Individuals in transition experience social and cultural processes of separation from other societal groups and are rendered invisible and outside social and cultural classification, often without any formal status or title (Turner, 1987). Barrett, Hage, and Gauld (2012) have applied the concept of liminality to the state of transition that community-dwelling older adults experience when requiring assistance and support to maintain independence. They argue that older people enter a state of liminality when being assessed for their support needs that reifies a sense of social disconnection from others (family, friends, social community) and can sometimes be exacerbated by continuing to live at home and "age in place". This manifests as a form of social exclusion that has both social and spatial (i.e., potentially physically isolating) dimensions (Barrett et al., 2012). In this paper we are interested in the ways in which state-imposed periods of national lockdown, and the associated regulations impacting older people, amplify states of liminality within housing with care schemes and have the potential to heighten social disconnection from others both within schemes and in the wider community.

COVID-19 and the impact on older adults
An increasing number of research studies have begun to examine the impact of the COVID-19 pandemic, with many focusing on the lives of older adults. In the initial months, Heid et al., (2021) found that older individuals reported feeling most affected by the lack of in-person contact with others and the need to change their activity routines. Brown et al., (2021) conducted one of the first research studies to explore the impact of lockdown on the lives of community-dwelling older people in the United Kingdom. Older adults that reported challenges included absence of social relationships, managing activities of daily living, lifestyle and wellbeing priorities, and managing health and wellbeing. In contrast some participants reported positive aspects to life in lockdown, which included an increased sense of community and social engagement with people in their immediate vicinity, including neighbours; a break from routine; and life now being simpler, slower, and easier because of lockdown (Brown et al., 2021). Despite being only a snapshot of experiences of older adults from one area in England, it does suggest a broadly positive picture of how they adapted to lockdown measures.
Adding to this positive picture, Greenwood-Hickman et al., (2021) found that maintenance of a positive attitude and perspective gained through past hardships was a valuable coping strategy for many older adults and pointed to the resilience of the older population, their adaptability to new technologies and their ability to maintain a positive outlook. Similarly, in a UK and Republic of Ireland study, Brooke and Clark (2020) found that people over 70 years adapted to household isolation, social distancing and shielding through use of social media and neighbourhood resources. Participants made plans for the immediate future, which motivated them by giving structure to their days through household tasks whilst use of online social media enabled contact with friends and family.
COVID-19 had a particularly negative impact on care facilities for older people globally (Thompson et al., 2020). Several international studies have focused on the impact of lockdowns in long-term care settings Van der Roest et al., 2020). In efforts to protect the lives of their residents, many long-term care settings imposed extended periods of lockdown, prohibiting visitors and activities, and depriving residents of in person contact with significant others, which has been found to be detrimental to residents living in these settings . Several studies considered the tensions and challenges that staff members working in long-term homes and facilities faced. A study of Dutch nursing homes explored the dilemmas experienced by elderly care physicians (ECPs), in needing to balance safety for all through infection prevention measures versus the quality of life for individual residents and their loved ones (Sizoo, Monnier, Bloemen, Hertogh, & Smalbrugge, 2020). Sunner, Giles, Parker, Kable, and Foureur (2021) explored the role of care facility managers and noted how staff were particularly concerned that whilst keeping residents safe was the main priority, they may have also been causing residents harm in doing so.
The tensions between staff-imposed restrictions and residents' independence and quality of life were also reported by Ayalon and Avidor (2021). They recognised the power imbalances that emerged during lockdown which included staff not telling residents that another resident in the scheme had passed away. In addition, residents described feeling emotions of despair, depression, and anger, which was intensified when the rest of society returned to a new routine, while they were still in lockdown .
Although limited, there have been some international studies that have explored the impact of COVID-19 on older adults living in housing schemes such as retirement villages, sheltered housing and independent living. An Australian study found that older adults living in retirement villages recognised physical activity (including walking, gardening, housework) as important to maintain health and fitness and evoke positive experiences. Participants appeared to be resourceful and coped well through a variety of strategies, despite their mental wellbeing and social connectedness being negatively affected (Ng, Hill, & Burton, 2021). In an Israeli study, Shuv-Ami, Alon, and Bareket-Bojmel (2021) sought to compare the attitudes and feelings amongst older adults living in sheltered housing with those in independent housing during the pandemic. They found that respondents in sheltered housing were more satisfied with their lives and were more optimistic than those living independently during the COVID-19 pandemic.
Measures introduced to mitigate COVID-19 forced people to spend most of their time at home. As a result, attention has been drawn to older adults' accommodations and housing solutions and the impact they can have on older people's physical and mental health. In this paper we examine older residents' experiences of COVID-19 lockdown measures, social interactions and rule compliance within housing schemes that provide care and support services in the UK. Housing with care schemes provide a unique 'micro-neighbourhood' for generating a deeper understanding of the impact of lockdown restrictions on older residents' social wellbeing.

Research design and methods
The findings presented here originate from a mixed-methods study that explored the social inclusion of older people from socially diverse backgrounds living in housing with care schemes in England and Wales (2019-2021). The DICE (Diversity in Care Environments) Study captured longitudinal and cross-sectional (qualitative and quantitative) data across three providers of housing for older adults in England and Wales. Qualitative and survey data were gathered from residents and staff across 121 schemes (interviews with 72 residents across 26 schemes; a survey of 741 residents across 95 schemes). The aim was to develop a better understanding of the ways in which housing providers seek to promote residents' human rights and social participation within their schemes and to identify good practices for making residents feel included.
During the qualitative fieldwork, the first national UK lockdown was announced and subsequently, the research team were unable to visit schemes in person. Research methods and tools were modified to adapt to the new restrictions, and the topic of COVID-19 and experiences of living in housing with care schemes during lockdowns were added to interview schedules. In this paper, we focus on the accounts of 56 residents from 24 housing with care and support schemes who took part in longitudinal and cross-sectional interviews across. We examine specifically their experiences of living in these schemes during the UK lockdowns and how they maintained social connections and wellbeing in the context of COVID-19. Some cross-sectional interviews took place prior to lockdown occurring and so were not relevant to this focus on lockdown experiences. As data were collected between March 2020 and January 2021, we do not have data on participants' experiences of the third UK lockdown that took place between January and April 2021. The study received ethical approval from the Faculty of Social Sciences and Law, University of Bristol. Participants in the sample varied by social minority characteristics. Table 1 provides details about the sample's characteristics.

Sampling and interview methods
Residents who identified from social minority groups (for example LGBT, Black, Asian and other minority ethnic) were recruited for longitudinal interviews through purposive sampling using a recruitment call placed in a large survey that was distributed to residents across the three housing providers. Residents from social minority groups were purposively sampled through targeted recruitment activity due to the aims of the wider study being the exploration of diversity and inclusion of older people in environments with care. Some respondents participated in three sequential interviews (one every four months). In addition to this, recruitment flyers were sent out to specific resident association groups aimed at those with social minority characteristics, for example LGBT residents' groups. The research team contacted respondents who met the inclusion criteria and arranged to visit them for an interview.
For the cross-sectional interviews, eight housing sites were purposefully selected from the three providers, reflecting differences in geographic location (rural, urban) and type of scheme (sheltered, independent living, extra care). Single semi-structured interviews with residents at selected sites were carried out to generate in-depth data on residents' current and recent experiences of inclusion within the scheme and more broadly in the local community. Following the lockdown, the research team worked with scheme managers to distribute research flyers to all residents' apartments inviting them to take part in telephone or online interviews. In July 2020, we also returned to participants who were interviewed in person before the first lockdown and invited them to take part in follow-up telephone interviews to explore their experiences of life in the scheme during the first lockdown. Twelve participants took part in follow-up interviews.
Participants were asked to complete an informed consent form prior to telephone calls and to indicate this verbally over the telephone before the commencement of interviews. All participants were provided with a list of local wellbeing support services at the conclusion of interviews, including local COVID-related support services. Interviews were conducted either by telephone or via online platforms approved by the university such as Zoom or Teams. Interviews lasted between 45 and 90 min over and followed a pre-designed interview schedule including questions on: experiences of lockdown within the scheme; contact with neighbours and significant others; use of technology during lockdown; health and wellbeing concerns; types and forms of activities within the scheme, and sources of help and support internally and externally. Participants received a £20 high-street store voucher as a token of appreciation.

Data analysis
Through an interpretivist lens, the chief focus of our thematic analysis was on residents' understanding of their everyday social world, identities and expectations about current and future housing and care needs. We were interested in the subjective meanings attached to daily experiences of social dynamics and social encounters in scheme life. All interviews were recorded and transcribed verbatim. The data were imported and categorised into a framework matrix in NVivo 12, using the framework approach to analysis and data management (Gale, Heath, Cameron, Rashid, & Redwood, 2013). Four members of the team read a small sample of transcripts and then two of these members developed initial coding frameworks, with separate frameworks for the longitudinal and cross-sectional data. The frameworks included a priori categories as well as categories arising inductively. An additional category and sub-categories based on COVID-19 data was added to the two frameworks following the completion of interviews. Once the analytical frameworks were confirmed, two of the authors charted the data from the transcripts across the devised frameworks. This categorical data was then thematically analysed using an iterative process of moving between initial coding across categories and defining and naming recurrent themes across the dataset (Braun & Clarke, 2006). Four themes are presented belowplease see Table 2 for list of themes and subthemes. Participant numerical codes are used to protect anonymity.

General experiences of lockdown
There were differences across the three housing providers in how participants had experienced lockdown(s).
The themes presented below focus on issues pertaining to residents' experiences of maintaining social connections, disruption to their social connections, and the social dynamics surrounding rule compliance during the lockdowns.

Disruptions to social routines and everyday connections
Daily life and the sense of community within schemes altered substantially, and lockdown curtailed everyday social interactions that would have normally occurred and been a big part of scheme life. Shielding and social distancing requirements disrupted any sense of community within schemes for some participants.
During the national lockdowns, all activities within schemes, including coffee mornings and social activities, were cancelled and communal areas closed off. Resident relationships were impacted and  Participants described both similarities and variations in their everyday social routines. The most difficult change was the prohibition of visitors to schemes including family and friends, as well as not being able to attend social activities and events. These changes undermined the residents' sense of independence and autonomy. Anxiety about risk of infection had changed participants' shopping behaviours with most turning to online shopping (some with assistance from family members) and others limiting the number of visits to shops. In contrast, some participants described moments of enjoyment in the solitude of selfisolation and used it as an opportunity to get around to doing things they had thus far put off doing such as writing a book.
Lockdown preparedness. Some participants, mainly those with disabilities or long-term social anxieties, found lockdown resulted in a minimal change to their routines. One participant believed that living with a disability had made him mentally stronger and better able to cope with lockdown.
I think people who are without disabilities have found that. There has been much more emphasis on people with depression and who haven't been able to live a normal life. Suddenly, they find themselves in a situation where they've got to stay at home. It has had a big impact on them. So I think that actually disabled people, because they put up with more of an impact on their lives because of their disabilities, tend to be stronger. [Participant 19, male, 62 years, identifies as gay, disability (registered blind), independent living].
Similarly, those who had been living alone for a significant part of their life spoke of feeling more prepared for the lockdown. For this participant, a daily routine of occupying the home on their own and having established routines were not disrupted by self-isolation requirements but rather supported and endorsed as older people were expected to stay at home.
Residents as active contributors to scheme life. A prevailing assumption underpinning self-isolation requirements for older people is that all older adults are vulnerable and in need of care and support (Pentaris et al., 2020). This denotes a passive and homogenising understanding of old age despite many older adults maintaining paid employment and unpaid social contributions such as volunteering, including some within the participating schemes. In contrast to this public discourse, some residents reported actively supporting other residents or staff as a way to maintain their own mental wellbeing. For example, helping other residents with shopping and distribution of food: And while we've been in lockdown I have been helping to organise free meals. Within the scheme, we deliver free meals, so I take orders and things like that. Another participant, supported the scheme manager to deliver newspapers and the post: Yes. I don't like to see things… It just seemed stupid to me that newspapers, which is why I started asking, weren't being delivered until 4:00 o'clock in the afternoon. I said, "Look, surely I can do that." [Participant 7, male, 77 years, identifies as gay, extra care housing].
This particular participant was used to being active externally to the scheme prior to lockdown and supporting the scheme manager in this way provided a daily structure and kept him active and out of his flat for a short while.

Overcoming restrictions and maintaining social connections
Although lockdown impacted on all aspects of scheme life, residents and staff adopted new ways to stay connected with others and overcome the detrimental effect of imposed restrictions. Residents actively engaged with technology to maintain connections, and some did so with the support of staff.
Keeping connected: resident-led engagement with technology. At the beginning of the first lockdown in 2020 visitors to schemes were restricted to staff, carers, managers, and nurses only. Unsurprisingly virtual and smart technology became even more important to residents in staying connected with significant others. Most participants spoke of staying connected with family and friends via telephone and digital platforms such as Zoom, WhatsApp, Facebook and Messenger. Most participants used smart technology either by themselves or with some assistance.
Every day, I got six, seven, eight WhatsApps as soon as I woke up.
[Laughter) Things like that, and people sending photographs and people sending jokes, this type of jokes that go on in the social media. Yes, I think in the first three weeks of isolation there was more online activity than normal. [Participant 12, female, 70 years, minority ethnic background, identifies as lesbian, retirement living].
Online video platforms were not for everyone. For some residents messaging platforms were more straightforward to engage in.
In some schemes, staff members and carers on site would support residents to use digital technology where necessary, which included providing information about using digital technology such as smartphones or lending their own phones for residents to speak with their family. Participants also told us that religious/faith organisations supported members of their congregation through telephone calls, while councils and voluntary services helped with food deliveries.
Keeping connected: staff-initiated activities. Participants' spoke of the importance of having staff based onsite for daily welfare checks and for helping to prevent social isolation. Welfare calls would be made by staff via telephone or using the intercom to check residents were well, and for some, these calls from staff became a "lifeline". One participant spoke of the importance of having staff onsite and referred to the carers as "absolute bricks". Schemes where there were supportive general managers were spoken about most favourably. However, there were divergent views on the quality of support provided by staff on site. One participant felt that support was tokenistic and not provided equally to all residents.
At the height of restrictions, staff in some schemes would send around quizzes to individuals to undertake together or attempt to carry out doorstep quizzes or exercises. Not only did they facilitate social contact between residents and with the wider organisation, but residents indicated they felt more supported, secure and safer by having staff on site. Seeing staff in person helped to tackle isolation amongst residents. . I think that's why I think it's important that we do have managers on site, is because they do look out for people and they do notice if somebody hasn't been around or has been out of sight or hasn't been very well. They know about it because they do morning calls. Some people opt into this, "Oh, do you want a morning call?" They will speak to people every day, except at the weekends, just to make sure they are up and around and okay. [Participant 15, female, 67 years, identifies as lesbian, retirement living].
Across a few schemes, care workers, caretakers and managers that were based on site throughout the lockdown had been particularly supportive and praised by participants. Being based on site, these staff members had built up good relationships with residents, and this facilitated a sense of connection within the scheme:.
We When restrictions started to ease staff worked hard to facilitate social connections amongst residents, where government rules allowed, for example, by arranging socially distanced coffee mornings (depending on number limitations) and social distanced events such as pizza or fish and chip nights.

Social dynamics across communal areas: shared and contested spaces
Restrictions across schemes resulted in communal areas becoming both shared and contested spaces whereby social dynamics amongst residents were influenced by shared experience and connection as well as fears and anxieties.
Communal areas as essential shared spaces. There were different perspectives on access to and the use of communal areas during the lockdown amongst residents. During the earlier part of the first lockdown, fear of contracting the virus resulted in residents not wanting to walk through shared corridors: I look out for people here to talk, but many of the people would be too frightened even to talk on the corridor, even to pass each other or something. But, I mean, it was quieter, and sometimes I felt, I wished I could talk to someone. [G2, female, 84, ethnic minority background, retirement housing].
This participant expresses feelings of social loneliness through their desire to speak to other residents in communal areasa contrast from the busy social life of the same scheme prior to the lockdown.
However, as the lockdown continued internal and external communal spaces such as corridors, shared gardens, and apartment patios and balconies, became essential for staying connected with other residents as residents' confidence in using these spaces grew. Residents could speak to others in these spaces at a safe physical distance: Generally, communal areas of housing schemes, such as corridors, laundry rooms and gardens, became essential spaces for maintaining social (while physically distanced) contact with others. For some participants, being able to do this during the lockdowns was a 'saviour' and helped them cope with isolation: We asked them [staff], because we can't go out… We've got an atrium here, it's a glass roof on it. We asked, "Could we go out there?" They [scheme staff] said, "As long as you self-isolate." […] We just go out, again, an hour, have a chat, have a cuppa. At this scheme, such was the significance of these social opportunities that two male residents, who had not previously been socially connected with other residents, joined in regular conversations with a small group of women who had an established friendship.
On the other hand, there were some residents who maintained they would not use or walk-through communal areas at all and did not agree with doorstep activities or residents meeting up at a distance outside: Yes, so, I didn't want to get mixed up with anybody out there. You know, they were having things going on like dancing and all of this sort of thing and I thought, "Well, there are people who have died in here, have a bit of respect," and whatever. I thought, "Why were they able to do it? Why were they able to do it?" because of the lockdown, we were told to stay in our flats, but that's something I don't really know. [A12, female, 67 years, disability, extra care housing].
This reflects differing anxieties about risk of infection within communal areas due to the different levels of clinical vulnerability across different residents.
Communal areas as contested spaces. The use of shared spaces to interact with neighbours generated frictions between residents at times, as evident in the previous quote. In the scheme described above, meeting others in communal spaces was not welcomed by all residents, and some discussed their anxieties and the perceived risks related to other residents meeting in communal spaces: They've shut off all the communal areas. I thought, and [Person 3] thought that the atrium outside [our doors], the glass, we thought that should be shut off. But there are about five people… The people that are causing a lot of aggravation here, they're in there every day, and it has been worrying us. That's why we don't walk up through the atrium. Because there are people here that have died of the virus, quite a few have died here of the virus. We just feel that there might be germs in the atrium. So if we do go out, […], we go the back way. We go out through our patio doors and we walk around the back and go through the back gate. [A9, female, 65 years, ethnic minority background, extra care housing].
In some instances, this led to other residents having to use alternative entrances/ exits to the scheme and avoiding certain shared spaces due to the fear of infection where others are meeting up. The actions of more socially active residents were a source of resentment to othersthese subgroups of residents were sometimes framed as rule-breakers, a source of disruption and disrespect towards the wellbeing of other residents of the same schemes.

Staying compliant versus bending and breaking the rules
Tensions between autonomy and compliance. The translation of national government rules and the implementation of restrictions within schemes were contentious amongst residents across some schemes. These restrictions were founded on providers' understanding of the organisation's duty of care towards residents' wellbeing. Given the aims of housing with care to maintain older people's independence and autonomy, the COVID-19 lockdowns generated new challenges. Compliance with self-isolation requirements sometimes varied, generating tensions between neighbours. While welcomed by some residents, others questioned the imposition of restrictions and rules, which raised questions for them about the type of scheme they were living in. Some compared their experiences to being more like a 'care home' than independent living: No, they [staff] wouldn't let us. Like I said, this isn't a nursing home. We have asked the manager of the carers, he just said, "Well you're being treated as a nursing home." I wonder why [Company 1] have done that. Really, speaking, we're 42 individual flats. It's like a normal street, so why have they done that? […] Like I say, I think that's the most heart-breaking thing about it. Like I said, I know if I'd been living in a house, I would've had to isolate but I would've been able to see my family outside, wouldn't I? [A6, female, 69 years, extra care housing].
In schemes where these tensions arose, rules, and changes in rules that were implemented were often down to the interpretation of scheme managers and led to ambiguities about what was permissible.
Changes to practice were communicated to residents via letters posted through letterboxes, notices on notice boards in communal areas and via telephone calls. Signs reminding residents and staff of mask wearing were displayed around housing schemes. Changes to rules were not always communicated in a clear manner, which caused some confusion. In addition, echoing Ayalon and Avidor (2021) finding, some residents spoke about how the housing provider did not directly inform them when neighbours had died during lockdowns. This caused anxieties as residents were not sure if the death was the result of COVID-19 or another cause.
Nobody was allowed to visit me because we had the virus here, unfortunately. Yes, we had quite a few people who had the virus. Unfortunately, a couple died of it, who we know of, there are a couple more who died, but we don't know whether they died of the virus or just health issues. We're not told anything, this is the thing, we've only found out from other residents, sort of thing, who have passed it on… [A12, female, 67 years, disability, extra care housing].
This highlights a particularly harrowing tension for housing providers between maintaining individual resident confidentiality, including respecting family members' privacy, and not breaching data protection requirements whilst balancing the ethos of community and the needs of neighbours to be informed and have opportunities to grieve collectively.
Challenges with rule compliance. Ambiguity in the way rules were implemented led to individual residents interpreting what they could and could not do in the scheme differently. Again, this generated tensions amongst residents: She [prior neighbour upstairs] has three university grandchildren and a daughter, and I said, "You go every day with them, obviously you cannot be all day in doors with a mask." She says to me, "Look, I chose this. It is my conscious decision. What is more important to me is being with them, so if I die from COVID I die from COVID, but that is what I want from life." Which is okay, except that in this situation it is not okay because it might affect others, do you see what I mean? [Participant 12, female, 70 years, minority ethnic background, identifies as lesbian, retirement living].
Participants described how, as the lockdowns progressed, some residents were not compliant with the rules and were perceived as taking risks such as having visits from family members when scheme rules and public health regulations stipulated otherwise. Whilst housing providers addressed these breaches, residents did not always comply thereafter.
Overall, most residents were compliant with mask wearing; however, mask wearing created barriers to communication and inclusion for those who had hearing or sight difficulties. Some residents who were shielding due to clinical risk, found contravening rules a challenge which caused distress, and they expressed anger with the lack of compliance around mask wearing in the wider community. The lack of compliance around COVID-19 safety rules was put down to generational differences around compliance rooted in ageism. For example, some referred to younger people and their lack of consideration for others when it came to following COVID-19 rules and supporting those who may be more vulnerable, such as older adults. This contributed to anxieties experienced by participants about re-entering community life.

Discussion
The findings presented give a valuable insight into the social dynamics, connections and tensions generated from living in housing with care as micro-neighbourhoods during early COVID-19 restrictions. A central premise underpinning housing with care models is to provide a supportive living environment that facilitates independence, autonomy, social interaction, and ageing well in place. However, COVID-19 lockdowns and the restrictions in place within schemes challenge this premise and compromise the notions of autonomy and independence that schemes pursue.
The findings provide a unique window into how residents adapted and coped with self-isolation requirements and maintained social contact internally and externally to the scheme. Furthermore, the findings illustrate inherent tensions for housing providers in promoting both community connectedness and residential autonomy while also maintaining rule compliance and upholding a duty of care in seeking to protect residents from risk of infection and ill-health. These tensions are not new, and we would argue that the finely weighed balance between promoting resident autonomy and providing a safe and supportive living community was exacerbated, rather than originated, under self-isolation requirements. Nonetheless, such restrictions also resulted in new points of conflict and disagreement between neighbours. Housing staff are required to weigh up similar competing priorities between keeping residents connected while keeping residents safe, echoing tensions noted in Dutch nursing homes where there is less emphasis on independent living (Sizoo et al., 2020). Residents' capacity to exercise decisional autonomy was severely limited during the 2020 lockdowns and, there was arguably ageist imposition of self-isolation requirements for older people. However, this did not inhibit participants from maintaining social connections with significant others and with those within schemes. As noted from the wider study (Beach et al., 2022), there is evidence that pre-COVID-19, residents of housing with care have lower levels of loneliness than they would if they were living in nonspecialist private housing. Although this paper does not aim to compare housing with care with private housing in the community, we have highlighted some positive experiences in relation to social connections during a time of COVID-19 restrictions.
Our research demonstrates that both staff and residents employed a variety of practices to maintain social connections and counter social isolation under highly challenging circumstances. While restrictions were placed on residents' movements and forms of contact with others the findings illustrate residents' exercising agency and re-establishing their autonomy -in some cases in the form of rule bending and breaking. Residents reasserted their autonomy external to their apartment (for example, through maintaining socially distanced contact in communal areas) and internally (for example, maintaining family contact through available technology or in person through rule violation). Housing staff and carers on site also played an important role in bolstering residents' autonomy, through supporting residents in using smart technology for social contact and through supporting residents with daily tasks such as grocery shopping. In this sense interdependency in daily support was key to facilitating ongoing autonomy. Staff who worked on site during the pandemic or were in frequent telephone contact with residents were instrumental in supporting residents' wellbeing and sustaining social interaction within schemes. Residents valued welfare checks and contact with staff during the height of lockdown, and this ensured participants felt valued by the housing provider.
The findings point to how residents quickly adapt to new ways of engaging with significant others under lockdown restrictions. Similar to Brooke and Clark (2020) study of community dwelling older adults we found residents adapted to household isolation, social distancing and shielding through the uptake and use of social media and neighbourhood resourcesneighbours in this sense being in the receipt of and offering to help others in the micro-neighbourhood of scheme life. Exercising care to others was demonstrated by some residents through daily helping tasksa duty of care as a moral position adopted by both residents and staff that exceeds legal and state-imposed requirements. This is critical to recognise at a time when national government-imposed restrictions provided little guidance to housing with care providers on how to set and maintain self-isolation requirements within schemes. In terms of policy implications, this also indicates of a wider lack of understanding about housing with care provision across government in public health and social care policy.
Adhering to self-isolation requirements represented an intensified liminal space that within scheme settings had potential to be highly marginalising due to the risk of social isolation within the home and physical impairments such as frailty and disabilities that limit mobility outside the home. Applying Turner (1987) theorisation of liminality, residents first experience an unfamiliar separation from others which can then, position them at the social margins and render them invisible to others. Residents' liminal status is intensified by the ambiguity of national government rules on lockdowns, which were developed for nursing homes in mind rather than the more independent context of housing with care schemes which are not as clearly defined or recognised (Dutton, 2021). We see the trickledown effect of this ambiguity through residents' experiences of rules swiftly changing and being unevenly applied and regular violations of scheme rules. This is not to undermine the collective efforts of scheme staff to support residents' autonomy during this phase of ambiguity and uncertainthe findings also highlight their efforts in ensuring the welfare of residents and facilitating social opportunities and activities under challenging conditions.
The notion of definitional ambiguity (Turner, 1987) associated with occupying a liminal state also applies as residents experience a sharp definitional shift from independent home occupiers to vulnerable persons requiring care and protection. Similar to Ayalon and Avidor (2021) study, participants in our study note an institutional creep in their housing neighbourhoods as rules on visitors and movements within the scheme were introduced. These rules assumed the shape of more routinised patterns of living managed by other organisational actors such as housing staff and carers. For some residents, this raised a fundamental question about the type of scheme they had elected to move into and their original intentions for maintaining independent living as lockdowns threatened the central ethos of housing with care models. The challenge for providers as pandemic measures were increasingly rolled back was to re-establish the independent ethos of housing schemes while also instilling residents' confidence in interacting with each other within and external to their apartments.
Finally, as a distinct subgroup, disabled participants in our study point to their preparedness for solitude and the opportune moments this opened for solitary activities. Disabled participants highlighted how the pandemic provided an opportunity for others to experience how isolated they are in their everyday lives and therefore they experienced little difference in their lives during the pandemic and were more psychologically prepared for lockdown. This suggests that residents accustomed to high levels of movement within and external to the scheme may be more impacted by new and unfamiliar restrictions and therefore require greater emotional support with this adjustment.

Limitations to the current study
As this research was conducted at the height of the first and second UK lockdowns, we do not have data on how scheme providers managed the re-introduction of the third lockdown in January 2021. This no doubt presented further challenges for residents' autonomy and independence and ways in which scheme providers supported residents' wellbeing. The shift from in-person to remote fieldwork resulted in fewer residents taking part from each scheme and may have excluded residents who either did not have access to digital platforms or who did not want to participate in telephone interviews. However, we sought to partly address this through follow-up interviews with those who had taken part in in-person interviews pre-lockdown. We have purposively captured some diversity amongst residents identifying with social minority groups, but the numbers of participants from black and ethnic minoritised groups are small. Recruiting residents from these groups remained a challenge, particularly when we were unable to visit schemes and talk to potential participants in person. We also purposively recruited from schemes in more urban areas, in an attempt to capture the diversity of experiences based on the design and infrastructure of buildings and the surrounding areas.

Policy implications
Our findings have several implications for housing providers and policy. Housing providers and their senior management faced much difficulty during the lockdowns in mitigating infection risk and maintaining residents' sense of independence and autonomy and these housing settings should be recognised by national government policy as distinct settings that require attention and their own set of regulations. Such challenges included practical issues such as accessing sufficient supplies of personal protective equipment for staff to more aspirational elements like ensuring residents' wellbeing and dignity. As a specialist form of housing to support older adults, similar to but different from traditional residential care settings, many housing with care schemes found themselves in a grey area with respect to national policies and initiatives during lockdowns. In certain respects, they were expected to act as if they were nursing homes but without the same levels or types of support delivered to the formal care sector. Moving forward, national government policy should recognise these housing settings as distinct settings that require attention and their own set of regulations, as was recommended in the final report by the Communities and Local Government Select Committee Inquiry on Housing for Older People in 2018.
In addition, our findings highlight that some of the challenges for housing providers in strengthening residents' sense of autonomy and independence were exacerbatedbut not createdby the pandemic. This underscores the relevance of the pandemic experience for ongoing strategies, i.e., the lessons are not restricted to the crisis situation of the pandemic. Housing providers should therefore continue their efforts to provide social activities for residents, reach out to those who may be at higher risk of isolation (e.g., due to social minority status), and encourage the use of new opportunities for connections such as those available through digital technologies. Listening to what residents want and how service delivery directly affects them is crucial in this respect and housing associations and local authorities should take into account resident views when making policy changes. Greater attention may be needed to address potential conflicts between residents who hold different ideas of what is expected, to enhance inclusive approaches to the use of shared spaces, and to ensure that no barriers exist to participation by residents with unique circumstances or interests. Well-trained and supportive staff will be key to achieving this. Finally, in the context of the liminal position of residents, awareness and sensitivity will need to increase, to help residents navigate their own transitions across difference levels of capacity and independence.
This paper sought to qualitatively investigate the impact of COVID-19 on older adults living in housing with care schemes, noting the tensions that housing providers faced in maintaining residents' autonomy and independence whilst also highlighting the positive ways in which older adults adapted to lockdown measures. Our findings add to the growing literature that examines the impact of COVID-19, uniquely focusing on older adults living in housing with care in England and Wales, and further pointing to the need for older adults living in such housing settings to receive greater policy attention.

Funding
This work was supported by the UKRI Economic and Social Research Council [grant reference ES/R008604/1].

Data availability
Data will be made available on request.