Research · for community groups, senior centres & older people's charities

Libraries, Halls and Parks: Places to Meet and Loneliness in Later Life

Published 2026-11-15

Libraries, halls, parks and other shared places are where much of later life’s everyday contact happens, and the World Health Organization now names strengthening this “social infrastructure” as a way to tackle loneliness and isolation[1]. The research on neighbourhoods and loneliness points the same way, but it is mostly a picture of associations rather than proof of cause.

Many of the groups that older people run and attend need a room: a library meeting space for a reading group, a church or village hall for a lunch club, a community centre for a u3a talk, a park bench where neighbours stop to talk. When those places close, cut their hours or become hard to reach, it is the circumstances that change, not the people. This study sets out what official counts and research from several countries show.

43%of the more than 9,000 public library systems in the US in fiscal year 2023 were in rural areas, by the federal Public Libraries Survey[2]
112studies of neighbourhoods and loneliness, synthesised in a 2025 systematic review, linked green space, walkability, local amenities and neighbourly contact with lower loneliness, mostly in cross-sectional designs[3]

WHO now counts libraries, parks and transport as part of the response to loneliness

“Social infrastructure” means the shared spaces and services where people meet without needing an invitation: public libraries, parks, community centres, halls, cafés, places of worship and the transport that gets people to them. In June 2025 the WHO Commission on Social Connection published its report. It says that “inadequate social infrastructure may exacerbate inequities that could increase the risk of social isolation and loneliness”, and among its community strategies it says: “One approach is to strengthen social infrastructure, such as libraries, parks, transport and social services, even when connection is not their primary aim.” The report is candid about how far the research has got: evidence that strengthening social infrastructure works “is increasing” but “remains limited”[1].

The same report says that “estimates suggest that 25–34% of older people are socially isolated”, from studies dated 1990 to 2022[1]. That is an estimate for isolation (few contacts), not for loneliness (the feeling that connections fall short), and the two need to be kept apart. Places to meet can matter for both: they make contact possible for people who are isolated, and what happens there shapes whether people feel connected.

This framing moves the question from individuals to places: not why a person does not go out, but whether there is somewhere worth going, open when people can come, that they can reach. Programmes that run inside these places, such as senior centres and lunch clubs, have their own studies in this library: see senior centres and day centres and lunch clubs.

US libraries have been rebuilding in-person use since the pandemic closures

Of the places to meet we looked at, US public libraries were the most carefully counted. The Institute of Museum and Library Services (IMLS) runs an annual Public Libraries Survey; its May 2026 research brief analysed more than 9,000 library systems in fiscal year 2023: 43% in rural areas, 25% in towns, 26% in suburbs and 6% in cities[2].

Where US public library systems are, by type of area, fiscal year 2023IMLS Public Libraries Survey, FY 2023, over 9,000 library systems: rural 43%, suburban 26%, town 25%, city 6%.0%12.5%25%37.5%50%Rural43%Suburban26%Town25%City6%
Share of the more than 9,000 public library systems in the US analysed for fiscal year 2023, by the locale of the system. A count of library systems, not of buildings or branches, and not a measure of loneliness or isolation. Source: Institute of Museum and Library Services, Public Library Trends Between FY 2018 and FY 2023 (May 2026), from the Public Libraries Survey.

The pandemic showed what happens when these doors shut. IMLS reports that 90% of public libraries physically closed at some point in fiscal year 2020. Median visits per person stood at 0.4 in fiscal year 2021 and rose to 2.8 by fiscal year 2023. Libraries offered a median of 202 programmes in fiscal year 2023, against 76 in fiscal year 2021, and median programme attendance reached 3,087, close to the fiscal year 2019 peak of 3,612[2]. These are medians across library systems, not national totals, and the survey does not report attendance by age, so it cannot say how many older people take part.

The survey does not measure loneliness, and we found no national survey that links library use directly to loneliness among older people. A rise in visits shows that people came back, not how connected they felt.

A Japanese cohort links local libraries with later-life health, not loneliness

The closest large study we found comes from Japan. Researchers in the Japan Gerontological Evaluation Study followed 73,138 people aged 65 and over in 19 municipalities between 2013 and 2021, for 7.3 years on average. Living in a municipality with more libraries per head was associated with a lower risk of developing functional disability (needing long-term care), with a hazard ratio of 0.52 (95% confidence interval 0.28 to 1.00); more library books per head also went with lower risk (hazard ratio 0.96)[4]. The interval for libraries reaches 1 (no difference), and a second measure, libraries per unit of area, showed no clear link (hazard ratio 0.74, 95% confidence interval 0.44 to 1.25), so that finding is uncertain. The models adjusted for age, sex, education, income, social participation and municipal finances, among other things. The outcome was disability, not loneliness, and with only 19 municipalities, those with more libraries may differ in other ways: this is an adjusted association, not proof of cause.

We found official counts for some kinds of place, and none for most

The official counts of places to meet that we found are patchy. The UK’s Office for National Statistics (ONS) has tracked pubs and bars from the Inter-Departmental Business Register, its list of registered businesses. Its January 2020 analysis found that the number of pubs and bars had fallen in most parts of the UK between 2001 and 2019, before rising by 315 (0.8%) between 2018 and 2019[5]. Those figures pre-date the pandemic: in November 2026 we found no later UK-wide release. A count of businesses is also not a count of places where older people meet.

For community centres, village and church halls and other meeting rooms, we found no comparable national series in the countries we searched. Where there are no regular counts of the places people can gather, it is hard to say with confidence whether provision is growing or shrinking, or where losses fall hardest.

Opening hours, cost and access matter too: a hall that is costly to hire, or a centre with no step-free access or bus route, serves fewer people than a count suggests. On transport, see transport and isolation.

Neighbourhoods with green space, amenities and friendly contact go with lower loneliness

A 2023 systematic review in Health & Place brought together 57 studies of the built environment and loneliness, and concluded that the field still needs “further conceptual and empirical explorations”[6]. A larger 2025 review in the same journal synthesised 112 studies of how physical and social neighbourhood characteristics relate to loneliness[3].

That 2025 review found that physical features, “such as the use of green spaces, walkability, access to amenities, and neighbourhood aesthetics, are associated with lower levels of loneliness”. Social features, such as neighbourhood cohesion, feeling connected to the neighbourhood and talking with neighbours, were “consistently associated with reduced loneliness”. It also found that how people perceive their neighbourhood was generally more strongly associated with loneliness than objective measures, and it named the main weakness: “the predominance of cross-sectional studies”, which measure neighbourhood and loneliness at the same moment[3].

A Dutch study suggests how these links might work for older people specifically. In a survey of 182 adults aged 65 and over in the Netherlands, loneliness was directly related to satisfaction with one’s social network and to attachment to the neighbourhood, and indirectly related to feeling safe and to satisfaction with local amenities and services[7]. Shops, services and safe streets seemed to matter through the belonging and relationships they support. It is one small survey, measured at one time.

Why these associations are hard to read

Three problems run through this evidence. First, selection: people choose where to live, within the limits of income and housing, and healthier, better-off and more sociable people may both live in greener, better-served areas and feel less lonely. Second, perception: a person who feels lonely may also rate their neighbourhood less warmly, which would make perceived measures look more strongly linked to loneliness, as the 2025 review found. Third, direction: poor health and limited mobility can shrink the neighbourhood a person uses, so what looks like an effect of place may partly be an effect of health on how place is used.

None of this means places do not matter. It means a cross-sectional link is a reason to test changes properly, not a measured effect. Studies that follow people over time give firmer answers, and there are fewer of them.

Green space: promising links with loneliness, and mixed findings when areas are compared

Parks and green space are the best-studied kind of social infrastructure. A 2022 systematic review in Science of the Total Environment found 22 quantitative studies of green space and loneliness; most were from high-income countries, 15 (68%) were cross-sectional, and most used general loneliness measures such as the UCLA Loneliness Scale. Of 132 associations tested, 88 (66.6%) pointed towards green space protecting against loneliness, and 44 (33.3%) were statistically significant[8]. So most results leaned the same way, but two in three did not reach statistical significance.

Firmer evidence comes from following people over time. A 2024 Australian study followed a nationally representative sample of 3,043 adults of all ages from October 2020, with 2,421 taking part again at four months and 2,123 at 16 months. Loneliness was measured with the 6-item De Jong Gierveld scale, which separates social loneliness (missing a wider circle) from emotional loneliness (missing a close attachment). Among people who were lonely at the start, having spent 1 to 2 hours in nature in the week before the first survey, compared with less than one hour, was associated with higher odds of no longer being socially lonely at four months (odds ratio 1.69) and at 16 months (odds ratio 2.10). For emotional loneliness, the link appeared at 3 to 4 hours a week (odds ratios 1.84 and 1.82). Time in nature was not linked to who became lonely[9].

Odds of no longer being lonely, by weekly time in nature, Australian adults, 2020 to 2022Astell-Burt, Navakatikyan and Feng, 2024. Among Australian adults who were lonely at baseline (October 2020), odds ratios for relief from loneliness compared with less than one hour a week in nature: social loneliness, 1 to 2 hours a week, 1.69 at 4 months and 2.10 at 16 months; emotional loneliness, 3 to 4 hours a week, 1.84 at 4 months and 1.82 at 16 months.00.511.522.5Social loneliness, 1-2h/week, 4 months1.69Social loneliness, 1-2h/week, 16 months2.10Emotional loneliness, 3-4h/week, 4 months1.84Emotional loneliness, 3-4h/week, 16 months1.82
Odds ratios (1 = no difference) for moving out of loneliness, compared with people who spent less than one hour in nature in the week before the first survey, after adjusting for factors that affect both time in nature and loneliness. Loneliness measured with the 6-item De Jong Gierveld scale (general, social and emotional loneliness). A nationally representative sample of 3,043 adults of all ages, not older people only. Only the time bands the study’s summary reports are shown. Time in nature was not linked to who became lonely. Observational: it shows an association, not proof of cause. Source: Astell-Burt T, Navakatikyan MA, Feng X, Environmental Research, 2024.

The study adjusted for confounders, but people well enough to spend time outside may be recovering for other reasons, so this is still observational. The sample was adults of all ages, first surveyed during the pandemic, so the results may not carry over to older people in ordinary times.

When whole areas are compared, the picture is less clear. A 2026 analysis of 22 years of Australian HILDA survey data (people aged 15 and over, not older people only), linked to census and environmental data for local areas, found loneliness clustered in socio-economically disadvantaged areas: the most advantaged fifth of areas had a lower relative risk of loneliness (0.8). Loneliness here meant agreeing with the single statement “I often feel very lonely”. After adjustment, loneliness was not associated with an area’s green space or walkability[10]. The two Australian studies used different measures and designs, so they cannot be set side by side; neither tested whether time spent in green space matters more than how much of it an area has.

Across European countries, taking part in associations goes with less loneliness

Using the 2014 European Social Survey and other sources, researchers compared loneliness across 26 European societies. They found two “necessary conditions” for a society to have a low degree of loneliness: “high internet access and high association participation”, the second meaning many people taking part in clubs and voluntary groups. Most societies with less loneliness also combined welfare support with cultural support[11]. This compares whole societies at one time and is not specific to older people: it describes patterns between countries, not any one place or person.

What the evidence asks of us

The evidence on places to meet is uneven, but it points in a consistent direction for national and local governments, funders and the organisations that use these spaces.

  • Treat places to meet as part of the response to loneliness and isolation. The WHO Commission names strengthening social infrastructure, with libraries, parks and transport as examples, among its community strategies, and says inadequate social infrastructure may worsen inequities that raise the risk of isolation and loneliness[1].
  • Count what exists and what closes. The US Public Libraries Survey shows what yearly counting makes possible[2]. Similar official counts of community centres and halls, with opening hours and accessibility, would show where provision is thinning.
  • Start with disadvantaged areas. Australian data show loneliness clustering in disadvantaged areas, and the study’s authors call for place-based responses and targeted resources in those communities[10].
  • Build for belonging, not only for square metres. Neighbourhood cohesion, feeling connected to the neighbourhood and contact with neighbours are the features the 2025 review found consistently associated with lower loneliness[3], and neighbourhood attachment was directly related to loneliness in the Dutch survey[7]. Seating, toilets, safe crossings, step-free access and regular activities help turn a space into somewhere people actually meet.
  • Evaluate new places properly. Most of the evidence is cross-sectional[3][8], and the WHO Commission calls for further high-quality research on these strategies[1]. Measuring loneliness and isolation before and after a place opens or closes, with a comparison area, would add more than another snapshot.

Age-friendly city and community programmes take a similar whole-place approach: see age-friendly communities. All the studies in this series are collected at Loneliness in Later Life.

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

Community groups, senior centres, lunch clubs and older people’s charities depend on rooms, halls, libraries and parks to do their work, and they are often the first to notice when a venue closes or a bus route is cut. We have gathered the official counts and research from the US, Australia, Japan, the Netherlands, the UK and a comparison of 26 European societies so that the people making the case for these places can cite the evidence, including its limits.

How to read this data

Counts. The US library figures count library systems, with medians for visits and programmes; they are not counts of buildings and say nothing about loneliness. The UK pubs figures count registered businesses and stop in 2019. Several of the studies are Australian, which reflects where this research has been done.

Measures. The studies use different measures of loneliness: the UCLA Loneliness Scale in most green-space studies, the 6-item De Jong Gierveld scale (which separates social and emotional loneliness) in the 2024 Australian study, a single agree-or-disagree statement in the Australian area analysis, and a variety of scales in the neighbourhood reviews; results from different measures are not directly comparable. The Japanese library study measured functional disability, not loneliness, and the WHO Commission’s 25–34% is an estimate of social isolation, not loneliness.

Designs. Most neighbourhood and green-space studies are cross-sectional: they show associations at one moment and cannot rule out selection (who lives where), reverse direction (health shaping how places are used) or perception bias. The Australian nature study and the Japanese library study follow people over time, which is stronger, but remain observational. The Australian area analysis and the European comparison describe areas or countries, not individuals. We found no randomised trials of opening or closing a place to meet. Most of the evidence comes from high-income countries.

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.

Suggested citation: Understanding Aging (2026). Libraries, Halls and Parks: Places to Meet and Loneliness in Later Life. https://understandingaging.org/research/social-infrastructure-places-to-meet

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

  1. WHO Commission on Social Connection. From loneliness to social connection: charting a path to healthier societies (report of the WHO Commission on Social Connection). World Health Organization, 2025. who.int/publications/i/item/978240112360
  2. Institute of Museum and Library Services. Public Library Trends Between FY 2018 and FY 2023: From Pandemic Disruption to Steady Recovery (research brief, Public Libraries Survey). IMLS, 2026. imls.gov/sites/default/files/2026-05/FY2023-Research-Brief-508_05132026.pdf
  3. Schepers W, Gryp D, Häussermann F, Heylen L, Van Regenmortel T, De Witte J. The relation between physical and social neighbourhood characteristics and loneliness: a systematic review. Health & Place, 2025. europepmc.org/article/MED/40472474
  4. Otani S, Sato K, Kondo N. Public libraries and functional disability: a cohort study of Japanese older adults. SSM – Population Health, 2025. europepmc.org/article/PMC/PMC11851285
  5. Office for National Statistics. Economies of ale: changes in the UK pubs and bars sector, 2001 to 2019. ONS, 2020. ons.gov.uk/…/changesintheukpubsandbarssector2001to2019
  6. Bower M, Kent J, Patulny R, et al. The impact of the built environment on loneliness: a systematic review and narrative synthesis. Health & Place, 2023. europepmc.org/article/MED/36623467
  7. Kemperman A, van den Berg P, Weijs-Perrée M, Uijtdewillegen K. Loneliness of older adults: social network and the living environment. International Journal of Environmental Research and Public Health, 2019. europepmc.org/article/MED/30708985
  8. Astell-Burt T, Hartig T, Putra IGNE, et al. Green space and loneliness: a systematic review with theoretical and methodological guidance for future research. Science of the Total Environment, 2022. europepmc.org/article/MED/35878853
  9. Astell-Burt T, Navakatikyan MA, Feng X. Contact with nature may be a remedy for loneliness: a nationally representative longitudinal cohort study. Environmental Research, 2024. europepmc.org/article/MED/39284494
  10. Earnest A, Lim MH, Engel L, et al. Geographic variation in loneliness and social isolation in Australia: socio-demographic and healthcare utilisation determinants. Healthcare (Basel), 2026. europepmc.org/article/PMC/PMC13205294
  11. Swader CS, Moraru AV. Social infrastructure and the alleviation of loneliness in Europe. Kölner Zeitschrift für Soziologie und Sozialpsychologie, 2023. europepmc.org/article/MED/37360992