What Eases Loneliness in Later Life? Reviews Find Small, Uncertain Effects
Published 2026-11-15
When researchers pull together the reviews of programmes meant to ease loneliness, the answer is consistent: many help a little, on average, and talking and psychological approaches show the most promise for loneliness itself. The WHO Commission on Social Connection calls the effects “small-to-moderate” and says the most effective interventions “cannot be identified accurately” yet[1].
Every week, befriending volunteers ring people who would otherwise hear few voices, men meet in sheds, u3a groups learn together, choirs rehearse and lunch clubs lay the tables. Older people run many of these groups themselves, and funders increasingly ask them: does it work? This study is the map for the “What helps” part of our Loneliness in Later Life library: what the broadest reviews show, why design and outcome matter so much, and where each kind of programme stands, linked to its own study.
Reviews of reviews agree: small average benefits, with psychological approaches most promising for loneliness
A 2024 umbrella review in BMC Public Health took only systematic reviews of randomised trials and found 29, 14 of them on older adults. “Meta-analyses indicated small-to-moderate beneficial effects, while narrative synthesis demonstrated mixed or no effect.” Social interventions looked most promising for social isolation and psychological interventions for loneliness, with caution urged because effects were small, varied widely and came from reviews of variable quality[4].
The WHO Commission on Social Connection, reporting in 2025, drew on that review. It groups individual and relationship strategies into skills training, “social engagement facilitation” (providing regular contact directly) and therapeutic and psychological interventions. Its key message: “psychological interventions hold the most promise for addressing loneliness”[1]. These include cognitive behavioural therapy, “reminiscence therapy (often for older people)”, animal-assisted therapy and mindfulness.
Reviews of older people point the same way. A 2024 review of 19 pre-pandemic reviews concluded that “interventions that included animals, psychological therapies, and skill-building activities were more successful than interventions focused on social facilitation or health promotion”, and that programmes with several aims did better than single-aim ones[2]. A 2026 umbrella review of 22 reviews of psychosocial programmes in the community and in care homes found positive effects for social facilitation with real-time contact, psychological therapies, health and social care provision and animal-assisted programmes, while saying leisure and skill development and befriending “warrant more rigorous evidence”[5].
A 2024 meta-analysis of 60 studies of community-living older adults in high-income countries, all using a validated loneliness scale, found moderate-certainty evidence that group-based treatment (small moderated group sessions, mostly built on psychological approaches) was associated with reduced loneliness (standardised mean difference in 11 trials −0.25 in the paper’s results, −0.27 in its abstract) and that internet training was too (−0.22 in 5 trials). Group exercise may bring “very small reductions” (low certainty), and evidence was insufficient to reach conclusions on group-based activities, individual in-person contact, and internet- or telephone-delivered programmes[6]. In plain terms, standardised mean differences of that size are small shifts in average scores; they say nothing about how much any one person benefited.
How a study is designed changes how well a programme seems to work
The most useful question to ask of an evaluation is whether it had a fair comparison group. A 2011 meta-analysis of 50 loneliness-reduction studies, covering children and adults of all ages, sorted them by design: 12 measured one group before and after, 18 had a non-randomised comparison group and 20 were randomised trials. Before-and-after and non-randomised studies “yielded larger mean effect sizes relative to randomized comparison studies”. Among the randomised trials, 14 of the 20 showed no change in loneliness on their own, though together they showed a small average fall (mean effect size −0.198)[3].
Part of the reason is regression to the mean. Programmes usually recruit people because they are lonely at that moment. Loneliness rises and falls, so a group chosen on a high score will, on average, score lower months later even if nothing is done. An uncontrolled before-and-after fall is not evidence that a programme worked; only a comparison group recruited the same way separates the programme’s effect from that natural ebb. Self-selection adds to the problem for clubs: people who join differ in health, income and confidence from those who do not.
A 2018 integrative review of 38 studies of older people found that most programmes “reported some success in reducing social isolation and loneliness, but the quality of evidence was generally weak”; it linked the most effective ones with adaptability, a community development approach and “productive engagement”[7]. In the 2024 review of reviews, authors judged 63% of the 101 programmes they studied effective or partially effective, but only 8% of those primary studies were at minimal risk of bias[2]. The WHO Commission calls the reviews of the evidence “generally of very low quality”[1].
Changing how people think about connection, and doing several things at once, did best in trials
The 2011 meta-analysis sorted programmes into four strategies: “(a) improving social skills, (b) enhancing social support, (c) increasing opportunities for social contact, and (d) addressing maladaptive social cognition”. The last is close to the WHO Commission’s therapeutic and psychological category, which aims to help people “change negative thoughts about self-worth and how the self is perceived by others”[1]. Among randomised trials, the four that addressed social cognition did far better than the rest. Programmes offering more social support had a small effect, and the few trials offering social skills training or more chances to meet showed no clear effect, though each of those rested on only two trials[3]. The authors also found that “neither group-based format nor the use of technology” changed the effect size.
This does not mean social contact is unimportant: loneliness concerns relationships as people experience them, so company alone may not reach it. The WHO Commission reports that meta-analyses found small effects of psychoeducation on loneliness in older populations, and that “multi-component interventions are effective in older populations, with a small effect on social isolation and a moderate effect on loneliness”[1].
A 2022 meta-analysis in JAMA Network Open took in 44 randomised trials that measured loneliness in older adults, 33 in the community and 11 in long-term care, some with as few as 8 participants. “Most interventions had a small effect size.” Its largest effects, in long-term care, varied widely between studies, and it rated the overall quality of evidence very low[8]. It is also summarised on our page on older people living alone.
The programmes groups run: where each stands
Most of the programmes that community groups and charities actually run sit in the “social” families above, where the reviews find the thinnest or most mixed trial evidence. They are hard to test in randomised trials and do much more than target loneliness. The detail and figures are in each linked study.
- Befriending. Reviews find at most a modest effect on low mood and no clear pooled effect on loneliness, though some newer telephone and care-home trials did find loneliness fell. See befriending and loneliness in later life.
- Social prescribing. It has spread fast, but evidence maps rate much of the review literature critically low in quality, and loneliness results are mixed. See social prescribing in later life.
- Getting online. Being online is linked with slightly less loneliness, and computer and internet training shows modest trial effects, which have not always lasted. See the digital divide and loneliness.
- Singing and the arts. One large US trial of senior-centre choirs found loneliness eased; most other arts trials measured mood or wellbeing instead, and reviews rate the evidence weak. See singing, arts and loneliness.
- Group exercise. Physical activity, especially in groups, brings small gains in social functioning, but too few trials measured loneliness to pool in one review, and a 2024 meta-analysis found at most very small reductions, with low certainty[6]. See group exercise in later life.
- Men’s Sheds. Most research is small, qualitative or based on members’ surveys; Ireland’s Sheds for Life evaluation, which had a waiting-list comparison group, found no difference in loneliness; members reported little loneliness at the start. See Men’s Sheds and the evidence.
- u3a. Many members say in the movement’s own surveys that taking part has eased loneliness or isolation, but these are self-selected respondents, and we found no controlled trial that measured loneliness. See the u3a movement worldwide.
- Learning in later life. Informal group learning goes with better wellbeing and, in a large Japanese survey, lower odds of loneliness; these are associations, and learners differ from non-learners before they start. See learning in later life.
- Senior centres. Frequent attenders report less loneliness, but most studies compare users with non-users at one point in time and cannot show cause. See senior centres and loneliness.
- Lunch clubs, shared meals and day centres. In a US national evaluation, congregate meal users were more satisfied with their chances to socialise than matched non-users, yet loneliness scores were the same. See lunch clubs, shared meals and day centres.
- Volunteering. Older volunteers report less loneliness in cohort studies, but they start out healthier and better off. See volunteering in later life.
- Intergenerational programmes. A Campbell systematic review found its randomised trials all at high risk of bias and could not say whether these programmes change older people’s loneliness. See intergenerational programmes and loneliness.
Read together: where a group programme has been tested against a comparison group with a loneliness scale, effects are usually small or unclear; where large benefits are reported, they mostly come from members’ surveys or before-and-after designs. That describes the evidence, not the groups’ worth: most were never set up as loneliness treatments, and they offer friendship, purpose, learning and food that no loneliness score captures.
What “effective” means depends on what was measured, for whom and for how long
Loneliness, isolation or something else? Loneliness is a feeling; social isolation is an objective lack of contact, and a programme can change one without the other. The WHO Commission notes that “more evidence is available on how interventions affect loneliness than on how interventions affect social isolation or forms of social connection”, and that one review found no evidence that these programmes improved social support, although it reported small reductions in depression[1]. Many befriending, arts and exercise trials measured depressive symptoms, distress or wellbeing rather than loneliness; a result on one is not a result on the other, and different loneliness scales can disagree about the same people.
For how long? Most trials follow people for weeks or months, not years. The WHO Commission lists the “lack of evidence on the longer-term effects” among the field’s main limitations, and notes that WHO’s dementia risk-reduction guidelines found “insufficient evidence for promotion of social activity as an intervention for wider health outcomes”[1].
For whom, and where? “Older people are the group that has been studied the most in terms of interventions,” the WHO Commission says, but most of the evidence is from high-income countries, with none for low-income countries[1]. The Campbell Collaboration’s evidence and gap map of digital programmes for older adults classed 72% of the systematic reviews it found as critically low quality and only 2% as high quality[9]. Most programmes target people already identified as lonely or isolated, which the Commission warns “entails a risk of stigmatizing those who are eligible to take part”. Open, universal activities avoid that label but are studied less.
What the evidence asks of us
- Fund the groups for what they do, and evaluate them fairly. Thin trial evidence is not evidence of no effect. Funders who want loneliness outcomes proved should pay for evaluation that can answer the question, with a comparison group and a named scale, not before-and-after surveys that regression to the mean flatters. The WHO Commission calls for “high-quality syntheses” and a standard way of classifying interventions so results can be compared[1].
- Offer several kinds of help, matched to the person. Canada’s first clinical guidelines on social isolation and loneliness in older adults recommend “an individualized approach to interventions with shared decision-making” and list social prescribing, social and physical activity, psychological therapies, animal-assisted approaches, leisure skills and technology among the options[10]. Programmes with several aims did better than single-aim ones in the older-people review of reviews[2].
- Make psychological support part of the offer. Since psychological approaches show the most promise for loneliness itself, health services can make talking therapies accessible to older people and link them with the community groups that provide company, purpose and a reason to go out[4][1].
- Close the gaps. Research is needed in low- and middle-income countries, on isolation and connection as well as loneliness, and on longer-term effects[1].
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
The people who run befriending schemes, sheds, u3a groups, centres and lunch clubs are often judged on whether their work “reduces loneliness”, against standards the research itself rarely meets. We set out what the broadest reviews say, in their own terms, so groups, funders and commissioners can argue from the same evidence, honest about how small many measured effects are and about how hard these programmes are to study.
How to read this data
This study draws on umbrella reviews, meta-analyses, an evidence and gap map, an integrative review and national guidelines; single programmes are covered in the linked studies. Effect sizes are standardised mean differences or similar: the average difference between groups in units of spread on a loneliness scale, saying nothing about any individual. Trials used validated scales such as the UCLA and De Jong Gierveld scales, so we report average changes, not how many people stopped feeling lonely. The 2011 meta-analysis covered all ages. Sources report falls in loneliness as negative numbers; our charts show the size of the fall. Most evidence comes from high-income countries, and many reviews were rated low or critically low in quality. Gaps described here are gaps in what we found, not proof that nothing works.
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
- WHO Commission on Social Connection. From loneliness to social connection: charting a path to healthier societies. World Health Organization, 2025. who.int/publications/i/item/978240112360
- Patil U, Braun KL. Interventions for loneliness in older adults: a systematic review of reviews. Frontiers in Public Health, 2024. europepmc.org/article/MED/39091527
- Masi CM, Chen HY, Hawkley LC, Cacioppo JT. A meta-analysis of interventions to reduce loneliness. Personality and Social Psychology Review, 2011. pmc.ncbi.nlm.nih.gov/articles/PMC3865701
- Hansen T, Nes RB, Hynek K, et al. Tackling social disconnection: an umbrella review of RCT-based interventions targeting social isolation and loneliness. BMC Public Health, 2024. europepmc.org/article/MED/39020331
- Oosterhouse K, Atakro CA, Chan JHM, et al. Systematic reviews of psychosocial interventions for loneliness among older adults in community and residential care settings: an umbrella review. Journal of Advanced Nursing, 2026. europepmc.org/article/MED/41517834
- Shekelle PG, Miake-Lye IM, Begashaw MM, et al. Interventions to reduce loneliness in community-living older adults: a systematic review and meta-analysis. Journal of General Internal Medicine, 2024. europepmc.org/article/MED/38200279
- Gardiner C, Geldenhuys G, Gott M. Interventions to reduce social isolation and loneliness among older people: an integrative review. Health & Social Care in the Community, 2018. europepmc.org/article/MED/27413007
- Hoang P, King JA, Moore S, et al. Interventions associated with reduced loneliness and social isolation in older adults: a systematic review and meta-analysis. JAMA Network Open, 2022. europepmc.org/article/MED/36251294
- Welch V, Ghogomu ET, Barbeau VI, et al. Digital interventions to reduce social isolation and loneliness in older adults: an evidence and gap map. Campbell Systematic Reviews, 2023. europepmc.org/article/MED/38024780
- Conn D, et al. Canadian clinical guidelines on social isolation and loneliness in older adults. Canadian Geriatrics Journal, 2024. europepmc.org/article/MED/39619376