Day centres for older people: attender characteristics, access routes and outcomes of regular attendance. Finding of exploratory mixed methods case study research.

BACKGROUND Social prescribing is encouraged to promote well-being, reduce isolation and loneliness. Traditional, generalist day centres for older people could be suggested by social prescribing, but little is known about their clientele or their outcomes. As part of a larger study of the role, outcomes and commissioning of generalist English day centres for older people, the characteristics of attenders at four day centres, their reasons for attendance and outcomes were explored. METHODS This mixed-methods study used qualitative interviews and standardised tools within an embedded multiple-case study design. Semi-structured interviews with older day centre attenders (n=23, 62% of eligible attenders) of four day centres in south-east England, recruited purposively to reect organisational differences, were analysed. RESULTS Participants reported non-elective withdrawal from socialisation following health or mobility decline, or losses. Apart from living arrangements and marital status, attenders’ proles differed between centres. Access had been mostly facilitated by others. Day centre attendance enhanced quality of life for this group of socially isolated people with mobility restrictions and at risk of declining independence and wellbeing. The positive impact on attenders' social participation and involvement and on meaningful occupation was signicant (p-value <0.001, 99% CI), with an average ASCOT gain score of 0.18. Nine outcome themes were identied, with a further theme of process outcomes. CONCLUSIONS Outcomes of day centre attendance are those targeted by social care and health policy. Centres were communities that ‘enabled’ and offset loss or isolation, thus supporting ageing in place through wellbeing and contributed something unique to their attenders’ lives. By monitoring attenders’ health and wellbeing and providing practical support, information and facilitating access to other services, centres offered added value. Attendance needs to be set in the context of other social engagement and care provision which may not overlap or duplicate centre support. Professionals may wish to explore the benets of social prescriptions to day centres but should map local centres’ provision, engage with their organisers, and seek information on attenders, who may differ from those in this study. particularly enjoyable were art, craft, cooking, computer classes, charades, discussion groups, memory exercises, rae/tombola, singing and music sessions, sweet shop, trips out, poetry reading, food tasting, table games, listening to background music, reading the paper, sitting in the garden, visiting speakers, therapy dog visits and performances by folk and belly dancers. Quizzes, bingo, card games, and exercise provoked mixed reactions.


Introduction
Day centres for older people are community building-based services that provide care and/or health-related services and/or activities speci cally for older people who are disabled and/or in need. Attendance can be for a whole or part of a day and cover any number of days. Centres offer a wide variety of programmes that may be considered 'preventive' of decline or ill-being. Public funding reductions [1] have led to fewer older people with high needs being eligible for state funded social care in England. Many day centres, particularly those offering low-level support, have closed or been decommissioned [2], despite evidence that some older people would like to attend them. [3] However, about 59,000 older people using local authority provided or commissioned community services attend day centres. [4] Within this context of change and ongoing efforts to integrate health and social care, the future of day centres is uncertain.
Although evidence-based commissioning is encouraged within English social care [5] and the NHS [6], some interventions, particularly, 'preventive' services, have an under-developed evidence base. [7] Evidence generation for day centres is further complicated by their heterogeneity; centres may be owned by different types of provider, operate in different types of building, and may differ in size, target clientele and funding sources. Moreover, national data are di cult to obtain in England as day centres are not regulated. Outcomes data are increasingly relevant, with Outcomes Frameworks for health and social care Much early research about day centres was undertaken in different policy and funding contexts or concerned centres specialising in dementia care [11]; the last detailed study of day care in England and Wales being published in 1989. [12] However, by applying National Institute for Health and Care Excellence (NICE) criteria, English day centres having been judged cost-effective. [13] A review of the 2005-17 UK and non-UK literature [11] undertaken for the present study of four English day centres for older people identi ed a lack of research about day centres as whole services and evidence gaps concerning what they offer, who uses them and why, what outcomes are experienced and how centres are perceived by their various stakeholders and potential users. [11] Ellen et al.'s review of the literature about day centre outcomes called for better understanding of people's characteristics when they start attending a day centre and factors relating to their access. [14]. This might inform funding and referring decisions by, for example, primary care professionals.
Wye et al. [15] reported that clinical commissioners seek evidence and information mainly from trusted health-related sources (for example, the National Institute for Health and Care Excellence (NICE), the King's Fund and NHS Improving Quality), their own experience or local contacts. Few sources of academic research are consulted as these are considered to lack rich contextual data. Real life examples which provide context for impact make an important contribution to their knowledge and decision-making. Mixed methods are increasingly being used in health services research to provide such evidence to 'capture the experiences, emotions, and motivations of people providing and receiving health care, as well as the objective conditions of care delivery'. [16:2126] This paper seeks to address the limitations that Wye et al. [15] identi ed as present in information available to clinical commissioners by illustrating what can be learned from pro ling the attenders of four day centres and the outcomes they experienced, and contextualising these within current policy and demographic data. It reports mode and reasons for attendance and access in the context of life with needs for care and support. These data contextualise the outcomes experienced, also reported here, which, like other studies [13], indicate the generally positive outcomes for attenders. We argue that such information could be analysed in other areas to improve strategic decisions about service commissioning and referral pathways for older patients with long-term conditions and limited social support.
Methods mean greater validity in measuring the effects of social care services than EQ5D. [37] Cost-effectiveness can be ascertained if service costs are known.
Administration was relatively straightforward, lasting, on average, 15 minutes. This includes nine questions on general health, functional independence, social support, medication use, nutrition, mood, continence, that are supplemented by the clock-drawing test, for cognition, and the Timed Up and Go (TUG) test, for functional performance. This objective screening tool re ects frailty's multidimensional, unstable and heterogeneous nature, [38] including social support. It has been tested with older people and validated as a reliable and feasible tool for use by non-geriatricians. [38] As frailty is associated with adverse health outcomes, ascertaining frailty levels can assist care planning. [39] Higher scores indicate greater frailty.

Practitioner Assessment of Network Type (PANT)
An eight-question objective tool measuring social support networks in populations aged 50 or older [34], PANT covers level and frequency of contact and physical distance. Likelihood of need for formal services can be predicted based on network typology as each is associated with speci c risks.
Short Warwick-Edinburgh Mental Wellbeing Scale (SWEMWBS) This 7-item scale measures feelings and functioning (not illness and disorder) using positively-worded statements. Its use of raw scores  and metric (transformed) scores (7.00-35.00) enable it to be used as an interval scale for psychometric analysis. Higher scores indicate higher wellbeing.

Data analysis
After removing identi ers, data were entered into NVIVO (version 11) [40]; an SPSS syntax le to compute PANT network type; [41] and an ASCOT Microsoft Excel data entry tool (v2.9). [42] Individuals' attributes and validated scale scores were assigned to participants' NVIVO records to enable interrogation of data by day centre and participant characteristics. [43] Analysis of qualitative data was inductive, iterative and thematic. [44] Cross-case analyses of individual participant group data were undertaken; themes were identi ed across the different day centres and participant sub-groups. [45] The rst author undertook coding and analysis (see Additional File 2). Analysis and data saturation were discussed in team meetings with the second and third authors. Pseudonyms are used in this paper.

Patient and Public Involvement
A Study Advisory Group supported the study and met three times. Members, all with experience of day centres, were consulted about study materials [45] and interpretation of ndings. A separate Advisory Group that acts as a critical friend to the researcher's host Unit also provided feedback on study materials and the rst author's interpretation of ndings. Case study site representatives attending a workshop were also consulted about these.

Findings 1: Day Centre Attenders
Most attender participants were widowed, divorced or single, and two-thirds lived alone (see Table 1). Their average age was 83.3 years (range 68-101 years), three-quarters were women and one-quarter were educated beyond secondary school. All identi ed as heterosexual and as not having changed gender. Threequarters held religious beliefs. Indicative of relative deprivation, two-thirds lived in rented homes while just under one-third were owner-occupiers. Although similar numbers received means-tested bene ts as those who did not, almost two-thirds self-funded their centre attendance (n=14). Self-funders included people whose nancial assessment, after being assessed as eligible for services, were required to pay themselves (n=3) and people not mentioning any assessment (n=11). Others reported being fully local authority (LA) (means-tested) funded (n=6), sharing payment with the LA (n=1) or being unaware of who paid (n=2). Ethnic minority groups accounted for a quarter of the total number, but were only in two of the four centres, re ecting local demography.
All reported having health conditions or disabilities that impacted greatly on their life, with half reporting at least two forms of these. All, except two, had some levels of apparent vulnerability or frailty when measuring general health status with the EFS. Levels of health conditions were under-reported by attenders; some did not report certain health conditions in the interview that the researcher was later made aware of in family carer interviews, by centre staff during discussions or by the attenders themselves, for example, that one participant had dementia and another had terminal cancer and had survived a stroke. Threequarters of attenders had average or good wellbeing as measured by the SWEMWBS. Two-thirds were at greater risk of isolation, depression, loneliness and other mental ill-health because of their network type. People with Locally Self-Contained networks (n=8) are more likely to be isolated than people with stronger network types as people rely on neighbours, lead private lives and have little community involvement; they are, however, in contact with family over 50 miles away. Family Dependent (n=3) and Private Restricted (n=3) networks carry greater risks of depression, loneliness and other mental ill-health. People with the former type rely on local family, but also have some neighbour contact and some community group involvement. People with the latter have no local family, no local informal support and little community contact, but people may rely on distant family. Only one-third had strong Locally Integrated social networks; in these, informal help is exchanged between family, friends and neighbours, and people have community group involvement.
Apart from marital status and living arrangements, the pro les of attenders varied between centres. DCLA was most age-diverse. DCHA was the least genderdiverse. People who could be classi ed as having Severe Frailty were highest in number at DCHA, despite its younger pro le; levels of Apparent Vulnerability were matched across the remaining three centres. DCHA and DCLA attenders had the highest levels of mobility di culties and more long-standing health conditions, while deafness was most prevalent at DCV1 and DCV2. Owner-occupiers prevailed at DCV1 and DCV2 and renters at DCLA and DCHA. Circumstances when starting to attend a day centre mainly related to loss and a desire for something different in life. These were classi ed into six themes: 1) social isolation (mainly due to bereavement or having lost existing social networks), 2) loss of mobility (declining physical health, sometimes suddenly, or no longer driving their car), 3) activity-related (stopping attending another day centre or club due to closure or changed entry criteria, stopping volunteering or retirement, wanting 'something to do' for stimulation or a change, or 'somewhere to go'), 4) mental health or emotional problems (feeling depressed or very low, lonely, having lost con dence or reporting a diagnosed anxiety disorder), 5) feeling 'stuck' at home or not getting out enough, and 6) carer-related (recognising the need for family to have a break, feeling isolated as a spousal carer, accompanying cared-for to a day centre).
Principal motivations for starting to attend a day centre re ected these circumstances in that participants had wanted social contact, something to do, to get out of their home or to improve their mental health. Additional motivations were to meet goals for better physical health through exercise and meals, to improve mental health and to accompany a partner for whom the participant provided care. Although length of attendance ranged from a few months to decades, these motivations are likely to re ect attenders' marital status, living arrangements and health. Behind one principal motivation for attendance were different clusters of circumstances which interacted and overlapped, often triggered by an event or a series of interlinked events. Box 1 illustrates the complexity of two attenders' circumstances when they started thinking about attending a day centre. Almost half (n=11) reported starting attendance following contact with social care or health professionals (social workers=7, district nurse=1, hospital rehabilitation service=1); the tenth was unsure of the professional's identity and the eleventh had spoken to a GP, then a social worker. Six had found out about centres from family (n=4), from its manager, an acquaintance (n=1), and by another attender's recommendation (n=1). Some paths were less straightforward. Two had been told by a local councillor (politician) and a GP, and had subsequently had social work involvement. The former received a social worker assessment after starting to attend. The latter, after a 'bad' rst experience at another centre she had been referred to by her GP, had asked her family to nd an alternative centre. In the event, the LA-provided home care worker had linked them to a social worker who arranged the current centre place. Two had proactively approached social workers about centres. Four participants were unsure how they found out about their centres.

Contexts of current day centre attendance
Average weekly centre attendance was for 1.8 days. Overall, just over half (n=13, 56%) attended once weekly, one-third (n=8, 35%) for two (n=4) or three (n=4) days and two (9%) for four or ve days a week. It was the only weekly outing for 5 participants (22%). Participants had been attending their centres for anything between six months and 32 years. The following paragraphs provide an overview of the whole sample's contexts.
In addition to their day centre days, the number of days attenders left their homes each week ranged from none to three (average 1.3 days), plus monthly outings. Three-quarters saw family at least once weekly. While two saw family fortnightly or monthly, three saw them irregularly, not at all or had no family. A handful reported at least twice weekly telephone conversations with adult children. Two-thirds had no regular unstructured non-familial social events, such as seeing friends. Those who met with friends relied on friends or family providing transport. Two-thirds had no regular structured non-familial social events. The one-third who did either also attended another day centre, a social club or bingo session weekly, or went on monthly outings or 'tea parties' run by the day centre provider or a national charity. Two were enrolled at a skills centre for the visually-impaired. One-third undertook weekly or fortnightly food shopping outings, mostly with support from family, friends, dial-a-ride (mobility service for disabled people) or a voluntary organisation's support worker. Two attended church regularly, and two received weekly or monthly Holy Communion at home from a visiting priest. One visited a gym weekly.
Two-thirds managed their personal care without paid support, including a handful who had weekly or monthly hairdresser home visits but no other personal care help. Most of the one-third who had personal care received this every day. Just over half of attenders had no paid home care. One-third received once weekly help with housework from home care workers, cleaners or neighbours and, for two, this was daily. Two attenders' family members undertook housework for them. A handful of attenders had regular medical appointments.
Some participants provided further context about their lives more broadly and their characters. Home maintenance and self-care had become, or was becoming, increasingly effortful for many. While some talked about how independent they had always been, one was still mourning her recent loss of independence. There was a sense of resignation to the situations in which they found themselves, in that some had adjusted to these. A small number mentioned how helpful neighbours were. While some considered themselves to be 'joiners' of activities, others said they were not . Findings 2: Outcomes Of Day Centre Attendance As data are complementary, outcomes gained are presented thematically, with verbatim quotations illustrating commonly-expressed or contrasting perspectives (underlining indicates participants' own emphasis), together with ASCOT scores where relevant.
Ten outcome areas arising, and experiences contributing to these, were reported: 1) social participation and companionship, 2) the way time was spent, 3) getting out of the house, 4) improved mental wellbeing and health, 5) practical support, information and access to other services, 6) physical wellbeing and safety, 7) having a meal, 8) accommodation cleanliness and comfort and 9) personal cleanliness and comfort. Certain aspects of attenders' experiences and outcomes contributed to a tenth theme of process outcomes which were identi ed from both qualitative and quantitative data.
In qualitative interviews, attenders reported bene ting from attending their day centres. Attendance had added something unique to the lives of all but one attender who later added that the cost was worth it for the change of environment. For one, it had not just 'added' something to her life, it had changed it.
Qualitative outcomes themes (1-7) and process outcomes were reported across socio-demographic and health characteristics, social network types and day centres and across both self-funding and publicly-funded source sub-groups.
Completion of ASCOT INT4, by 22 attenders, also indicated that quality of life improvements were directly attributable to day centre attendance, covering themes 1, 2, 4, 6 and 7-9. Figure 1 shows average ASCOT current and expected SCRQoL score in each domain as a percentage of the total possible score (unweighted), gain in each domain and numbers of attenders saying that centres made a difference to their lives in each domain).

<INSERT FIGURE 1>
Overall average preference-weighted SCRQoL scores were 0.88 (current) and 0.70 (expected), with a resulting overall quality of life gain of 0.18 resulting from day centre attendance. Average gain varied between centres (0.13, 0.15, 0.16, 0.24) and between individuals (individual gain score range 0.00-0.62). Gain scores may have undervalued attendance's actual impact on some participants' quality of life. For instance, a small number of attenders' answers did not always correspond with their qualitative interviews, and clari cation of questions' meaning was needed in some cases. Notwithstanding socio-demographic and health variations between centres, in three centres, overall expected SCRQoL, in the absence of day centre attendance, was 0.69 and 0.75 in the fourth.
Average gain was higher for publicly-funded attenders (0.24) than for those self-funding (0.15). The 2016-17 Adult Social Care Outcomes Framework (ASCOF) in England reports an average current SCRQoL score of 18.9 across England (18.95 across the four day centre areas) among ASCOF service-using respondents aged 65 and older [46]. Study participants' average score of 19.4 compares favourably when converted to be comparable. [37] Theme 1: Day centre attendance provided access to social participation and companionship Centre attendance was said to address the problem of not meeting people when physically unable to get out of one's home. 'Social participation and involvement' was one of two ASCOT domains in which centre attendance made a signi cant impact (p-value <0.001, 99% CI) on participants as a group. There were different facets to this most talked-about outcome which was one of the unique contributions that participants reported centre attendance made to their lives.
Many valued having social contact which contrasted with usually being alone: 'It's like, if somebody is married and they are not happy in their marriage, they look for a way out. Well I am not happy being at home on my own and so that's my way out.' (Tina) 'I get conversation instead of talking to myself… And I'm mixing with human beings.' (Nellie) Attenders enjoyed doing things in company: 'We sit together and play together, like cards or any other games or … the memory class, and of course, the exercise.' (Mariana) The group environment enabled the opportunity for laughter or fun. There was a good deal of banter in some centres, sometimes group-based, or one-to-one between attenders or one-to-one with staff or volunteers: 'I think that's one of the things that I like about it. You have heard [female attender] and I roar with laughter before now, haven't you? …I can make her laugh so easily. I love it. I know when she laughed her head off one day when one of the questions was what did Richard II lose in the bushes and I called out, "his virginity." She said, "for God's sake." I don't know. I just like laughing anyhow.' (Kaye) Participants highlighted how they looked forward to regular contact with day centre friends, which was interesting as there was mostly no contact between attenders outside centres on non-attendance days; some considered it unnecessary as they saw each other regularly at centres: 'You look forward to seeing friends again, you know.' (Elizabeth) However, around half of the participants mentioned that increasing proportions of cognitively impaired attenders, either due to dementia or a learning disability, impacted negatively on levels and quality of connection possible. A small number also commented upon 'annoying' or disruptive behaviours and conversational faux pas sometimes made.
Two of the ve male participants expressed a preference for higher numbers of men since they maintained that men and women chat about different things. Attenders enjoyed being occupied or they enjoyed speci c activities they did at their centres. In most centres, activities were varied although not all attenders' (mainly sensory impairment) needs were always catered for. Joining in these was said to be enjoyable, stimulating, and, in some cases, satisfying: 'I like to be doing something.' (Lenny) 'I enjoy the art. I enjoy the singing. I love to sing. I don't mind whether there is one or two singing.' (Wilma' emphasis added in bold).
Activities cited as particularly enjoyable were art, craft, cooking, computer classes, charades, discussion groups, memory exercises, ra e/tombola, singing and music sessions, sweet shop, trips out, poetry reading, food tasting, table games, listening to background music, reading the paper, sitting in the garden, visiting speakers, therapy dog visits and performances by folk and belly dancers. Quizzes, bingo, card games, and exercise provoked mixed reactions. Theme 3: Day centres provided the opportunity to go out and have a change of environment A qualitative theme, being enabled to 'get out of' their home, another of centre attendance's unique contributions to their attenders' lives, was framed in two ways.
Firstly, it was tantamount to escaping from their home in which they felt they were 'stuck', or even imprisoned: 'It's like being a prisoner in my house now … That's how it feels now and again, because you don't see nobody there now.' (Olive) Others did not feel imprisoned and were more concerned with having a change of scene: 'Well, it gets me out of my four walls for a start.' (Nellie) 'Well, it gets me out once a week, which I wouldn't do otherwise.' (Ruby) Secondly, the day centre was somewhere to go when you had 'nowhere else to go' (Nellie). It was also a place to go and gather; saying hello to an acquaintance in passing was 'not the same as actually going to a function with the people' (Bob).
There was much stoicism and acceptance of current situations in which some attenders said they would never have imagined themselves. For some, centres appeared to be a good substitute for what they may have preferred to do had their abilities been different: 'I used to like going here, there and everywhere. Now I can't do that so I don't mind coming here…I am happy with it…It gives me a chance to come out. (…) if I could go round and do things that I would like to do and so I'd go shopping and maybe walk around.' (Dorothy) Theme 4: Improved mental wellbeing and health Participants reported improved mental wellbeing and health as a further unique contribution of attending their centres. Not only did participants enjoy certain aspects of what was provided by centres, but many enjoyed the whole experience. They gained a sense of purpose, felt like they belonged, felt in control or more independent and 'felt better' generally. They had something to look forward to that they enjoyed -and some found fun. Attendance also counteracted boredom and life's monotony, helped participants gain a better perspective of their own situations and feel more relaxed, less lonely or depressed or more con dent, mentally stimulated or energised. Centres were also referred to as a 'lifeline'. Furthermore, participants felt valued and respected as individuals which re ected centre staff's and volunteers' character, behaviour and their delivery of the service. ASCOT domains of dignity and control, the second and third highest scoring, fall within this theme.
Theme 5: Practical support, information and access to other services A qualitative theme, practical support, information and access to other services were either provided as part of the day centre service, via occasional or regular visiting professionals, speakers or other centre visitors, or were other services offered by the day centre providers. Most mentioned were the supply of hearing aid batteries or maintenance, and useful talks. Other examples provided were lunch clubs, holidays, shopping trips, hairdressing, ngernail ling and painting, help with arranging health or other appointments, referrals to, for example, occupational therapy to get a shower installed at home, or to the local authority to get a personal falls alarm installed, help to claim taxi vouchers (discounted taxi fares) and visiting chiropodists or massage therapists or clothes-sellers.
Bene ts gained included, for instance, saving or having more money, feeling safer or having peace of mind, enjoying trips out or feeling more settled after a group holiday: 'Before that I was buying the batteries because… I could get them free from the hospital but I'd have to take a taxi to the hospital to get them. So I used to buy them from Boots [pharmacist]. But [manager] said "oh no, don't buy them. We'll give them to you." … That's another thing that's been a great help. (…) I can clean part of it but I can't take the things apart and clean it properly.' (Francine) 'And then eventually heard about the holiday (…) I made myself go and it did me the world of good because since then, my dark side seems to have lifted. Although I've physically got all these problems, mentally I'm ne now, really.' (Ruth) Co-located facilities were a bonus. At one centre, the short, midweek religious service was attended by some participants. At another, the library was appreciated by a keen reader who also enjoyed occasional contact with babies at the mother and baby group, as was an advice service. First, attenders reported bene ting from informal health and wellbeing monitoring and follow-up undertaken by day centre personnel, such as being asked how they were or if something was the matter, which they appreciated. Staff, and volunteers, listened to attenders talk about continence or pain, for example, and spoke to named relatives about health concerns: 'They come around asking "Are you alright? What's the matter?"' (Thomasina).
They also measured blood pressure, made GP appointments, reported safeguarding matters (e.g. about possible elder abuse) to the local authority and replaced a screw in one attender's reading glasses.
Second, exercise was felt to help maintain mobility and alleviate depression: Although a minor outcome theme arising from qualitative ndings, almost half the participants said that day centre attendance made a difference to them in the ASCOT domain of food and drink, and most shared their, mainly positive, views of the meals provided. Reasons for reporting having a meal as an outcome included being unable to stand for long periods when cooking, closure of a lunch club, wanting 'a meal put down in front of me without having to cook it myself' (Isobel). The opportunity for conversation over lunch was welcomed. Negative comments, concerning one centre only, included long waiting time, lukewarm food, feeling rushed, poor variety and disliking the meals. Meals are categorised separately from physical wellbeing and health since people would be eating lunch at home on non-day centre days.
Theme 8: Accommodation cleanliness and comfort A theme emerging from quantitative data only, centre attendance was reported to make a difference to just under one-third of participants in this ASCOT domain, but the average gain score was very small.
Theme 9: Personal cleanliness and comfort Another theme emerging from quantitative data only, centre attendance was reported to make a difference to just under one-third of participants' quality of life in the personal cleanliness and comfort ASCOT domain, but average gain was zero. However, while responding to the tool's questions, two participants implied that attendance did impact positively on them: one commented that attendance affected how clean he felt since he bathed and wore his best shirt on attendance days and the other said she took care of her appearance as she knew men would be present.
Two of the day centres had bathing facilities (suitable for people with disabilities), but managers reported attenders using these only in emergencies.
Theme 10: Process outcomes While certain experiences contribute to the overall centre attending 'experience' (e.g. activities offered), others contribute towards 'process outcomes' which are those pertaining to the way services are accessed and delivered; they may include feeling valued and respected, being treated as an individual, having a degree of control over the way a service is delivered, the extent to which a service ts with other support received and value for money. [47] Some of these outcomes (emerging from both datasets) have already been alluded to, particularly under Theme 4, Improved mental wellbeing and health.
Overall, reported experiences and feelings about day centres indicated that attenders experienced mainly positive process outcomes. All planned to continue attending and would recommend their centre to friends, family or somebody in the same situation as themselves. Many considered their centre offered good value for money. Although a very small number of attenders reported mixed feelings on centre attendance mornings and making themselves get ready as knew they would enjoy it once there, feelings of positivity were widespread, with many enjoying the whole experience, looking forward to or loving it.
'I think it's the best thing they have done, [local authority], make this place (…). they do a wonderful job here. I don't think I'd rather be anywhere else but here. I really do enjoy being here (…) I am glad I come.' (Isobel) 'All I can say is that, anyone who doesn't go there is missing out on something. It is likely that less positive feelings, a judgement that attendance may not have been such good value for money and, perhaps, fewer attenders planning to continue attending may be been more apparent had attenders not found themselves feeling valued, respected, treated as an individual or with a degree of control over service delivery.
Comments on centre personnel were overwhelmingly positive. The few criticisms related to attenders exhibiting unpleasant or disruptive behaviour not being dealt with, certain staff very occasionally being a bit domineering or lacking understanding of sight loss. In once centre, three attenders but were less enthusiastic about one staff member than others but did not 'dislike' her.

Principal ndings
This study provides new understandings of the circumstances behind people's motivations to attend a day centre and of attenders' lives, going beyond their socio-demographic details, and how they accessed centres. Individual contexts, which were previously largely unexplored, are useful for situating attenders' outcomes. Circumstances behind centre attendance were, speci cally, decreased mobility that was health-related or socially constructed due to having stopped driving, bereavement, living alone, closure of another service, stopping a volunteering role, retiring, having lost con dence, feeling low, lonely, depressed or being isolated as a carer. Participants exhibited advanced age, declining health and mobility, sensory loss, bereavement and were retired, all of which are risk factors for social isolation [48], and their social network types put many at risk of mental ill-health. [34] Attenders' pro le places them within the National Institute for Health and Care Excellence's (NICE) category of 'vulnerable older people' who are 'most at risk of a decline in their independence and mental wellbeing'.[49:para 1.5.3] There was low awareness of the existence of day centres prior to attending one and acess to day centres had mostly been facilitated for attenders by social care or health professionals or family, but few by GPs.
This study also provides new understandings of what are 'the outcomes that matter' for cognitively intact older people attending day centres. Attendance enhanced quality of life, sometimes signi cantly, and made a unique contribution to their 'vulnerable' attenders' lives [49] thus demonstrating their policyrelevance. Day centres emerged as life-enriching gateways to companionship, activities, the outside world, practical support, information, other services, the community and enjoyment for people who had experienced loss, were socially isolated and unable to go out without support. They appear to promote elements of successful ageing, by enabling attendees to maintain social connections, compensate for mobility problems and social isolation, and optimising opportunities for company and chosen activities. [50] A major contribution to participants' mental wellbeing was through centres supplying a source of enjoyment or fun which signi cantly contributes to wellbeing [51] with laughter having a positive psychophysiological impact. [52] Centres were communities of choice that 'enabled' and offset loss or isolation, in the absence of other suitable options, and community resources that supported people in adverse circumstances, thus supporting ageing in place through wellbeing. Centres' two 'safety-net' outcomes-monitoring attenders' health and wellbeing and providing practical support, information and facilitating access to other services -meant they offered added value beyond the purposes for which they were commissioned or funded, beyond what may be assumed to be covered by a service aim of improving quality of life or supporting people to remain at home and beyond what attenders may have wanted or expected, given their reasons for attending. These added value outcomes emphasise centres' underlying nature of being long-term maintenance and monitoring services rather than services that deliver speci ed improvements after which people are discharged.
Importance has been placed on single entry points to minimise people 'falling through' gaps between services. [53] Notably, outcomes were achieved despite three-quarters of participating attenders using centres for only one or two days a week (i.e. 4.5-12 hours excluding travelling time), something perhaps attributable to the fact that many outcomes fall within the higher order categories of human need. [54] Comparison with other work Attenders' characteristics cannot be benchmarked against an existing dataset due to the lack of national data about English day centres. This study found attenders were people whose health and mobility had already declined due to loss (e.g. of health) or had resulted in loss (e.g. of social contact). This pro le contrasts sharply with that of attenders in studies undertaken in the 1970s-80s, more of whom was younger and more active; [55,56] during these decades, publicly-funded social care eligibility criteria covered people with low and moderate needs and more 'low-level' day centres operated. [12,57] In the 2000s, increasing disability among English day centre and lunch club users was noted [58] and, in the 2005-17 literature, centres were reported to be mainly used by people whose health had begun to decline. [11] Further corroborating this nding was the exclusion from the present study of 41 per cent of observed attenders due to cognitive limitations, a proportion double the 19 per cent of publicly-funded day centre attenders with dementia reported in the 2013-14 annual survey of Councils with Adult Social Services Responsibilities [4] and is despite the present study including older people not referred by a LA or being funded to do so, and that some participants reportedly had a diagnosis of dementia. This present study also indicates that generalist day centres -those without the identity of speci c nationality or ethnicity-focussed provision -may be acceptable to people from black and minority ethnic backgrounds.
Day centres remain a gendered service. The literature has persistently failed to note the over-representation of women among attenders. [59] In this study, women accounted for three-quarters of participants (and attenders overall). Although women account for 60 per cent of people in England aged 80 and older and 55 per cent of people aged 68 and older [youngest study participant's age] [60], morbidity rates are higher among older women than men. Also remaining unchanged is their attendance by widowed/single people and people living alone. No studies have been identi ed that report attenders' religious a liation, sexual orientation or gender reassignment. [11] Feeling the need to get out of the house for a change of environment was a motivation for attendance not identi ed in the recent literature. [11] The contextual data about day centre attenders' lives beyond the centres are missing from the literature and are a unique contribution of this study.
Social participation related improvement was more widespread compared with the only other identi ed day centres study employing ASCOT, [13] perhaps re ecting this study's older and less mobile participant pro le and because the strongest effects of social activities are felt amongst those least physically active. [61] Day centres have not previously been conceived as gateways, as sources of enjoyment or as offering added value; with only one Bahranian study [62] referring to aiming to provide 'fun'. Although people have been reported to enjoy attending centres and to laugh there, this has previously been conceived as contributing to overall life satisfaction [63,64] and enjoyment of activities as contributing to whether attending has made a difference to people's lives. [65] Findings con rmed two previous English studies reporting day centres' high emphasis on process outcomes [66] and attenders' high satisfaction with their relationships with day centre workers, workers' behaviour and their work. [13] Meaning of the study and implications for clinicians and policymakers In line with Outcomes Frameworks, this study identi ed mainly positive experiences and evidence that outcomes of day centre attendance address those targeted by health and social care policy. Centres supported their mainly socially isolated and housebound attenders to age in place by focusing on their wellbeing and by preventing deterioration, and acted on any safeguarding or health concerns. Since centre attendance was linked with non-elective withdrawal from social participation, for various reasons, and centres facilitated valued social contact, activities and interventions that improved quality of life, there is potential for day centres to feature on social, or wellbeing prescriptions, non-medical interventions whereby primary care professionals refer people with social, emotional or practical needs to non-clinical services with a goal of promoting improved outcomes. [67] Outcomes are of consequence due to the rising numbers of older people living alone [68] and because age in uences social exclusion, with people aged 85 or older at greatest risk. [69] Getting out is something older people with high support needs want to do and value. [70] Feeling 'stuck' at home [71:159] due to mobility restrictions may lead to social isolation which has a negative health impact. [25] Very old people (≥85 years) spend an average of 80 per cent of their time at home. [23] Staying at home for most of the time due to health constraints is signi cantly associated with poor wellbeing. [24] Promoting positive wellbeing makes a real impact on older people's lives and contributes to reducing future demands on health and social care services, families and communities [72]; attenders with a similar pro le to this study's participants may have needed publicly-funded care and support if their centre ceased operating.

Strengths and weaknesses of the study
The study's strengths lie in its focus on generalist day centres, on entire experiences and outcomes of day centre attendance rather than of speci c interventions in centres or experiences, and its in-depth nature. Combining a quantitative measure with qualitative interviews added richness and depth, enabling quanti cation of some of the outcomes emerging using both methods, and providing insight into factors contributing to outcomes experienced and the signi cance of these to attenders. Data collected using validated scales are comparable. Rigour was maximised by lay scrutiny of interview questions, undertaking regular visits which habituated participants to the researcher and led to a trusting rapport, recording and transcription of interviews and taking a systematic approach throughout. The interviewer kept a re ective diary. Findings' transparency and trustworthiness are reinforced by feedback from the Study Advisory Group and representatives from each case study site. Their validity is enhanced by the diversity of participating centres despite these not re ecting all typologies. Centre diversity and the emergence of common themes across centres compensated for its limitations which relate to the low number of participating centres in one English region and small sample sizes. Frailer and less cognitively able attenders may be under-represented, although participation was high among those eligible, and, in conversation during eldwork, many non-participants expressed similar views and experiences to the participants. A risk of bias is that poor-quality day centres may not have agreed to participate.

Conclusions and further research
Generalist, traditional day centres are no longer the 'low-level' services they once were. While centre attenders share some characteristics, they are not homogeneous, although many are socially isolated, due to their circumstances, and are unable to go out without support. Loss and a resulting desire for something different in their lives were driving factors for attendance. Day centre access was mostly facilitated for older people due to their lack of awareness of the existence of day centres. As resources to address the risks of loneliness and isolation [25], these are promising interventions that are acceptable to some and accessible. By supplementing or replacing inadequate or insu cient informal care, support, social networks or other opportunities for vulnerable people who are housebound but not necessarily frail, centre attendance counterbalances some of the potentially negative consequences of ageing in place with mobility restrictions and improves quality of life. This suggests the value of promoting professionals' ability to recommend consideration of day centre attendance in the context of increased interest in social prescribing. [6] This research highlights the value and complementarity of mixing qualitative and quantitative methods. With policy encouragement of integrating health and social care commissioning in England, further research could usefully explore NHS professionals' and commissioners' awareness of and views on generalist day centres, their purpose and potential, for example about centres' health and wellbeing monitoring role. Further research might also explore the costs of day centres and any expenditure-related bene ts linked with their use, for example the relationship between health and wellbeing monitoring at day centres and the use of primary or secondary care.

Declarations
Ethics approval and consent to participate The Health Research Authority's Social Care Research Ethics Committee (15/IEC08/033) awarded ethical approval. Local authority Research Governance approvals were subsequently granted. All participants gave informed consent before taking part.

Availability of data and material
The authors declare that all data supporting the ndings of this study are available within the article; the data are not publicly available. This manuscript is an honest, accurate, and transparent account of the study being reported; no important aspects of the study have been omitted. Figure 1 Average unweighted current/expected scores, gain scores and attenders centres made a difference to.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.