Lunch Clubs, Shared Meals and Day Centres: What the Evidence Shows
Published 2026-11-15
In the largest evaluation we found, older Americans who ate at congregate meal sites were more satisfied with their chances to spend time with other people than similar older people who did not, and fewer met the threshold on a short depression screener. Their average loneliness scores, though, were the same. That is the honest shape of the evidence on lunch clubs, shared meals and day centres: real signs of benefit, measured mostly in satisfaction, wellbeing and social contact, and very little rigorous research on loneliness itself.
Lunch clubs, community meals and day centres are some of the most ordinary ways that older people come together: a hot meal in a church hall, a village canteen, a centre where staff notice when someone seems unwell. This study sets out who uses these services in several countries, what cohort studies say about eating alone, what evaluations have found, and where the evidence runs out.
Shared meals and day centres reach large numbers of older people, and many of them live alone
The biggest public programme of shared meals for older people that we found is in the United States, where the Older Americans Act funds meals at senior centres and other community sites. In 2015, more than 76 million congregate meals were served to 1,554,658 people. A nationally representative survey of participants found that more than two thirds were women, about 60 percent lived alone, and their average age was 77. Many came often: 43% ate at their meal site five or more days a week, 39% three to four days and 18% one to two days[1]. Living alone is a household fact, not a measure of loneliness; our study of older people living alone sets out the difference.
In the United States, the National Center for Health Statistics also counts adult day services centres, which it defines as “a community-based center, generally open on weekdays, that provides long-term care services”. An estimated 237,400 participants were enrolled in 5,500 of these centres in 2020. They serve adults with care needs of all ages, so this is not a count of older people only: 69% of participants in for-profit centres and 56% in nonprofit centres were aged 65 and over[3].
In England, official counts of day care arranged by local councils were falling in the early 2010s. The number of people aged 65 and over who received such day care during the year went from about 95,000 in 2010-11 to 83,000 in 2011-12; the statisticians warn that the 2009-10 figure was recorded differently and is not comparable[4]. That counts council-arranged care, not every centre or lunch club, and people using direct payments are counted separately. Researchers in England wrote in 2021 that “assumptions that they are an outdated or costly service model have resulted in many centre closures in England”[5]. We found no up-to-date national count of day centres or lunch clubs in the UK, or in most other countries, so it is hard to say how many remain.
Elsewhere, shared meals are being expanded rather than cut. In China, the government has been piloting canteen services for older adults, including free meals in rural villages[6]. Japan’s local community salons for older people, and the long-term research on them, are covered in our study of senior centres, so we do not repeat them here.
Eating alone is linked with poorer health in Japanese studies, but the link is tangled up with social networks
Part of the case for shared meals rests on research about eating alone. The strongest study we found is the Japan Gerontological Evaluation Study (JAGES), which followed 33,083 men and 38,698 women aged 65 and over from 2010 to 2013. Over three years, 3,217 died. Compared with men who ate and lived with others, men who ate alone despite living with others had a hazard ratio for death of 1.48 (95% confidence interval 1.26–1.74), and men who ate and lived alone had a hazard ratio of 1.19 (1.01–1.41), after adjusting for age and health status. For women the equivalent figures were 1.18 and 1.10, and both confidence intervals included no difference[7]. The authors suggested that “a setting in which older adults eat together may be protective”, but this is an observational association: people who eat alone may differ in ways the adjustment did not capture.
Other studies show how hard it is to separate eating alone from everything around it. In a study of 710 older people living in the community, by researchers in Tokyo, both eating alone and having a small social network were linked to more depressive symptoms (measured with the Zung Self-Rating Depression Scale). Among those who lived alone, however, “poor social network size, and not eating alone, was associated with greater depression symptoms”[8]. A cross-sectional analysis of 2022 JAGES data on 14,467 people aged 65 and over found that men receiving public assistance had less varied diets than other men. Its authors concluded that these men “were more likely to increase their food diversity by eating together”, though the statistical test behind that (an interaction, p = 0.07) fell short of the usual threshold[9]. It links shared meals to income as well as company (see our study of old-age poverty and health).
A 2021 scoping review included 98 of 989 articles on eating alone or together among older people living in the community, and called for “future research designs allowing identification of causal relationships, using refined ways of measuring meals alone or together”[10]. The three Japanese studies above measured death, depressive symptoms and diet, not loneliness. Eating alone is a fact about meals, as living alone is a fact about households; neither tells us how a person feels about their relationships.
The best-controlled evaluation found more satisfaction with social life and fewer positive depression screens, but no difference in loneliness
The US Administration for Community Living commissioned Mathematica to evaluate the Older Americans Act nutrition programme, one of whose stated purposes is to “promote socialization of older individuals”. The evaluators surveyed 1,137 congregate and home-delivered meal participants and 1,164 nonparticipants at baseline in 2015–2016, with follow-up surveys of 754 participants and 939 nonparticipants. Nonparticipants were chosen to resemble participants, using Medicare records and geography, and the analysis used propensity-score matching and regression to account for observed differences[1].
Loneliness was measured with the three-item Revised UCLA Loneliness Scale, which asks how often people feel they lack companionship, feel left out and feel isolated from others. On this scale, congregate meal participants and nonparticipants both averaged 4.1: no difference. Other measures did differ. 94.0% of congregate participants were satisfied with their opportunities to spend time with other people, against 85.8% of nonparticipants, and 67.5% were very satisfied, against 55.5%. On the two-question PHQ-2 depression screener, 2.3% of participants met the evaluation’s threshold (labelled “affirmed 4 out of 6”), compared with 6.5% of nonparticipants. All three differences were statistically significant[1].
Matching can only allow for what was measured, so people who choose to go to a meal site may still differ from those who do not. And satisfaction with social opportunities is not the same as loneliness: a lunch club can add company to someone’s week without changing how they feel about their closest relationships. The finding suggests congregate meals do something people value socially, and it is a reason to measure loneliness carefully rather than assume the effect. (Meals delivered to people’s homes, which the same evaluation also covered, are discussed in our study of older people who are housebound.)
Reviews of shared meals and day centres find promising signs, mostly from studies that cannot show cause and effect
A 2022 scoping review of structured shared meal programmes for older people living in the community searched nine databases and screened 5,996 studies, but could include only 18, published between 1980 and 2021. Most came from the United States and most were cross-sectional, comparing people at one point in time; one followed people over time. The review reported links between shared meal programmes and better dietary intake, “minimal improvements” in physical health measures, and “a positive impact on attendees’ social networks and perceived well-being”. It concluded that more work is needed “to understand how these programs work”[2].
Studies elsewhere share the same limits. In Jinhua, in China’s Zhejiang province, a cross-sectional study compared 140 people aged 75 to 98 in seven villages with free canteen services and 144 in seven villages without. The canteen villages’ older residents had better overall mental health (on the 12-item General Health Questionnaire), life satisfaction and social capital, though no difference in nutritional status. The authors concluded that canteens could improve these outcomes, “but this still needs to be future evidenced”[6]. Loneliness was not measured.
Day centres have a wider evidence base, though still a thin one. A systematically conducted scoping review of day centres for older people without dementia, covering 2005 to 2017, found 77 relevant papers, mostly from outside the UK. “The largest body of evidence concerned social and preventive outcomes”, and attendance was reported to have positive effects on mental health, social contacts, physical function and quality of life. Its authors also found that such centres were “under-researched generally, particularly in the UK”[11].
The same team then studied four day centres in south-east England, interviewing 23 regular attenders (62% of those eligible) and using the Adult Social Care Outcomes Toolkit (ASCOT). Attenders described a “non-elective withdrawal from socialisation following health or mobility decline, or losses”: they had not chosen to become isolated. The gain attributed to attendance in social participation and involvement, and in meaningful occupation, was statistically significant, with an average ASCOT gain score of 0.18. The study is small and has no comparison group[12]. In an ethnographic study of four day centres in Denmark and Norway in 2018 and 2019, activities were valued most when they brought “an enjoyable social dimension with ‘a touch of fresh news’”: something new to talk about and take home[13].
Taken together, this evidence rests mostly on cross-sectional comparisons, participant surveys and interviews. We found no randomised trial of a lunch club, congregate meal programme or day centre that measured loneliness with a validated scale. That is a gap in the research, not proof that these services do not help. For how these findings compare with other approaches, see our overview of what reduces loneliness in later life and the rest of our research on loneliness in later life.
What the evidence asks of us
The evidence does not support grand claims, nor closing what people who use these services value. It points to five things.
- Count what exists. We could find no current national count of day centres or lunch clubs in the UK, and the English figures on council-arranged day care used here are more than a decade old[4]. Governments and funders cannot protect, or plan, services they do not count.
- Treat places to meet and eat as social infrastructure. The WHO Commission on Social Connection, reporting in 2025, proposed “strengthening social infrastructure (e.g., parks, libraries, cafés)”, and its co-chair Chido Mpemba said that “social connection must be integrated into all policies”[14]. Lunch clubs, community canteens and day centres belong on that list.
- Value the outcomes people report. Satisfaction with social life, social participation and lower rates on a depression screener are outcomes in their own right[1][12]. Researchers in England have urged those who shape local care to recognise the “congregate nature” that attenders value, rather than treating group services as outdated[5].
- Fund better evaluation. Commissioners and researchers should measure loneliness with a named, validated scale, include a comparison group, follow people over time and report results that show no change as well as those that do[2][10].
- Keep the route in open. In the English study of four day centres, most attenders had been helped to get there by other people[12]. Transport, referral and affordable meals decide who can take part.
Language: this piece says "older people" and treats loneliness as a common human experience that anyone can have, not a personal failing or an illness. It keeps apart three things that are often mixed up: loneliness (how people feel about their relationships), social isolation (how few social contacts they have) and living alone (a household arrangement). Where a study uses another term, such as "seniors" or "older adults", we keep it when quoting that study.
Why we're publishing this
Lunch clubs and day centres are often run by volunteers, small charities and churches on uncertain funding, and are asked again and again to prove their worth. We publish this so that the groups doing the work can see the evidence plainly: what it shows, where it is thin, and why a lack of trials is a reason to fund research rather than to cut services. The case for places where older people can meet and eat together is theirs to make; we hope these figures help.
How to read this data
What was measured. Only one study here measured loneliness directly: the US congregate meals evaluation, which used the three-item Revised UCLA Loneliness Scale (questions on lacking companionship, feeling left out and feeling isolated) and compared average scores rather than a share above a cut-off[1]. We took its figures from the evaluators’ 2017 conference slides, because we could not open the full report. The Japanese studies measured eating alone, living alone, social network size, depressive symptoms, diet or death; the Chinese canteen study measured general mental health, life satisfaction and social capital. Living alone and eating alone are facts about households and meals, not measures of loneliness.
Study designs. The JAGES mortality study is a large cohort that followed people over three years, and its hazard ratios are associations adjusted for age and health status, not proof of cause[7]. The US evaluation compared participants with matched nonparticipants; the other studies are cross-sectional comparisons, qualitative interviews or reviews of such studies. People who choose to attend may differ from people who do not.
Dates and places. The US meal figures refer to 2015 and 2015–2016, the US adult day services figures to 2020, the English day care figures to 2010-11 and 2011-12, and the Japanese cohort to 2010–2013. Most of the shared meal research comes from the United States, and most day centre research from outside the UK. We found little published research from low- and middle-income countries other than China.
Use this data
Free to cite with attribution to Understanding Aging. u3a groups, men's sheds, senior centres, lunch clubs, befriending services, charities, campaigners and journalists are welcome to reuse these figures and charts. For the underlying figures as a spreadsheet, get in touch.
Important: This article summarises published research and statistics for general information. It is not medical or care advice and cannot be used to diagnose, treat or predict anything about an individual. If loneliness, low mood or isolation is affecting your health or someone else’s, a doctor or another health professional can help, and local services for older people can often put you in touch with groups and support nearby.
Sources
- Mabli J, Gearan E, Menne H. Findings from the Administration on Aging Title III-C Nutrition Services Program Outcomes Evaluation (conference presentation, 2017 National Home and Community Based Services Conference). Administration for Community Living and Mathematica Policy Research, 2017. advancingstates.org/sites/default/files/Final OAA Title III-C.pdf
- Middleton G, Patterson KA, Muir-Cochrane E, Velardo S, McCorry F, Coveney J. The Health and Well-being Impacts of Community Shared Meal Programs for Older Populations: A Scoping Review. Innovation in Aging, 2022. europepmc.org/article/MED/36588625
- Singh P, Lu Z, Sengupta M. Variation in Adult Day Services Center Participant Characteristics by Center Ownership: United States, 2020. NCHS Data Brief No. 453. National Center for Health Statistics, 2022. cdc.gov/nchs/products/databriefs/db453.htm
- NHS Digital. Community Care Statistics, Social Services Activity, England - 2011-12, Final release (tables and charts, Table 3.1). NHS Digital, 2013. digital.nhs.uk/data-and-information/publications/statistical/adult-social-care-activity-and-finance-report/community-care-statistics-social-services-activity-england-2011-12-final-release
- Orellana K, Manthorpe J, Tinker A. Choice, control and person-centredness in day centres for older people. Journal of Social Work, 2021. europepmc.org/article/MED/34629957
- Wang X, Liu M, Li Y, Guo C, Yeh CH. Community canteen services for the rural elderly: determining impacts on general mental health, nutritional status, satisfaction with life, and social capital. BMC Public Health, 2020. europepmc.org/article/MED/32059652
- Tani Y, Kondo N, Noma H, Miyaguni Y, Saito M, Kondo K. Eating Alone Yet Living With Others Is Associated With Mortality in Older Men: The JAGES Cohort Survey. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 2018. europepmc.org/article/MED/28093448
- Sakurai R, Kawai H, Suzuki H, Kim H, Watanabe Y, Hirano H, Ihara K, Obuchi S, Fujiwara Y. Association of Eating Alone With Depression Among Older Adults Living Alone: Role of Poor Social Networks. Journal of Epidemiology, 2021. jstage.jst.go.jp/article/jea/31/4/31_JE20190217/_article/-char/en
- Tanaka K, Nishioka D, Nakagomi A, Ueno K, Ide K, Kino S, Murayama N, Kondo K. Public assistance program and food diversity among older people: a cross-sectional study using the Japan Gerontological Evaluation Study data. International Journal for Equity in Health, 2025. europepmc.org/article/MED/40355882
- Björnwall A, Mattsson Sydner Y, Koochek A, Neuman N. Eating Alone or Together among Community-Living Older People: A Scoping Review. International Journal of Environmental Research and Public Health, 2021. europepmc.org/article/MED/33801775
- Orellana K, Manthorpe J, Tinker A. Day centres for older people: a systematically conducted scoping review of literature about their benefits, purposes and how they are perceived. Ageing and Society, 2020. europepmc.org/article/MED/31798195
- Orellana K, Manthorpe J, Tinker A. Day centres for older people - attender characteristics, access routes and outcomes of regular attendance: findings of exploratory mixed methods case study research. BMC Geriatrics, 2020. europepmc.org/article/MED/32366223
- Øye C. Fresh news as meaningful? A multi-site ethnographic analysis of meaningful activities at four day centres in Denmark and Norway. Health & Social Care in the Community, 2022. europepmc.org/article/MED/35695064
- World Health Organization. Social connection linked to improved health and reduced risk of early death (news release on the report of the WHO Commission on Social Connection). World Health Organization, 2025. who.int/news/item/30-06-2025-social-connection-linked-to-improved-heath-and-reduced-risk-of-early-death