Research · for befriending services & older people's charities

Befriending and Loneliness in Later Life: Small Trials, Mixed Results

Published 2026-11-15

Befriending, a volunteer who visits or phones the same person regularly, is “commonly offered by the voluntary sector”, including to older people who feel lonely, but the trials behind it are fewer and smaller than its reach[1]. Reviews find at most a modest benefit for low mood and distress and no clear pooled effect on loneliness itself; a handful of newer trials, by telephone and in care homes, found loneliness scores fell over weeks to months.

For many befriending services the work is simple: a trained volunteer, a weekly visit or phone call, the same voice each time. Whether it measurably changes how lonely people feel is a harder question. This study sets out what randomised trials and systematic reviews have found, which outcomes each one actually measured (loneliness, depressive symptoms, distress or wellbeing are not the same thing), and what the evidence suggests for the services and volunteers who run befriending schemes.

−0.03the pooled standardised mean difference in loneliness across five befriending comparisons in a 2017 systematic review, a result indistinguishable from no effect; the same review found a small, borderline overall benefit across each trial’s main outcome[2]
2.71points lower on the UCLA loneliness scale after 16 weeks among Australian care-home residents offered befriending, compared with usual care, in a trial of 345 people published in 2026[3]

Befriending comes in several forms, and trials have measured different things

The 2017 systematic review by Siette, Cassidy and Priebe offers a working definition: befriending is “an emotional supportive relationship in which one-to-one companionship is provided on a regular basis by a volunteer”[2]. It covers several models:

  • Face-to-face visiting at home or in a care home (“friendly visiting” in North America).
  • Telephone befriending, sometimes with group calls.
  • Peer befriending by volunteers who share age, language or background.
  • Befriending for family carers.

Outcomes vary as much as models. Many trials targeted depressive symptoms, distress, mental health or quality of life, and measured loneliness only as a secondary outcome or not at all. Those measuring loneliness mostly used the UCLA Loneliness Scale (higher scores mean lonelier) or the De Jong Gierveld scale, which do not always agree. For how these scales work and why their results differ, see our study of how loneliness is measured.

Most trials compare people offered a befriender with people offered nothing extra. A few newer trials used befriending calls as the comparison for a more structured approach, which asks a different question: whether something else helps more.

Reviews find a modest benefit for mood and distress, and no clear effect on loneliness

A 2010 meta-analysis by Mead and colleagues pooled randomised trials of emotional support for adults in the community, not only older people. Its outcome was depressive symptoms and emotional distress, not loneliness. Compared with usual care or no treatment, befriending had “a modest but significant effect” on depressive symptoms: a standardised mean difference (SMD) of −0.27 in the short term (nine studies) and −0.18 in the long term (five studies)[1].

Siette and colleagues widened the net in 2017 to befriending for people with physical and mental health conditions of any age, including carers of people with dementia. They included 14 trials with 2,411 participants, published between 1991 and 2016; 11 were randomised, eight were run in the UK, and seven were judged at low risk of bias[2]. Taking each trial’s own main outcome, befriending was associated with a small overall improvement of borderline significance (SMD 0.18, p=0.05, 95% CI −0.002 to 0.36, seven trials). But when outcomes were pooled one at a time, there was no significant benefit for any of them. For the five comparisons that measured loneliness, the SMD was −0.03 (95% CI −0.18 to 0.12); for depression it was −0.18, with p=0.12[2]. The authors called the evidence “moderate quality” and the effect “rather small”.

A 2023 Campbell systematic review looked only at face-to-face friendly visiting by a volunteer for people aged 60 and over. It found nine randomised and four non-randomised studies, mostly from the United States, with 470 older people in total; every study had 20 to 88 participants, and programmes lasted 6 to 12 weeks, mostly weekly visits by undergraduate students. All had “major shortcomings in design and execution”, and the reviewers rated the evidence on loneliness, social isolation and wellbeing as very low certainty: they could not say whether friendly visiting helps or not[4]. A 2026 umbrella review of 22 reviews concluded that befriending interventions “warrant more rigorous evidence”[5].

Regular phone calls lowered loneliness on one of two scales in a four-week US trial; group calls in England could not find enough volunteers

The best-known recent telephone trial ran in Texas in the summer of 2020, during the pandemic. Researchers randomised 240 clients of Meals on Wheels Central Texas, all homebound and all with at least one chronic condition, to receive calls or no calls. Participants were aged 27 to 101, though 63% were 65 or older; 56% lived alone and 79% were women. Sixteen callers aged 17 to 23, briefly trained in empathetic conversation, each phoned six to nine people: daily for the first five days, then at least twice a week if the person preferred, for four weeks[6].

Loneliness was the primary outcome, measured two ways. On the three-item UCLA scale (range 3 to 9), the called group scored 1.1 points better than the control group after four weeks (95% CI 0.5 to 1.7), an effect size (Cohen’s d) of 0.48. On the six-item De Jong Gierveld scale the difference was smaller and not statistically significant (d 0.17, p=0.06). Depressive symptoms, anxiety and self-rated mental health also improved, while general physical health did not change[6]. More people were lost from the called group (13) than the control group (1).

Effect sizes after four weeks of empathy-focused phone calls, by outcome, Texas 2020Effect size (Cohen's d) for people who received calls compared with people who did not, after four weeks, among 240 Meals on Wheels Central Texas clients: UCLA loneliness scale 0.48; self-rated mental health 0.46; anxiety 0.35; depression 0.31; De Jong Gierveld loneliness scale 0.17 (not statistically significant).00.20.40.6Loneliness (UCLA 3-item)0.48Mental health (SF)0.46Anxiety (GAD)0.35Depression (PHQ)0.31Loneliness (De JongGierveld)0.17
Cohen’s d, larger means a bigger improvement relative to the no-call group. The two loneliness scales disagreed: the De Jong Gierveld difference was not statistically significant (p=0.06). One randomised trial, four weeks, participants aged 27 to 101 (63% aged 65 or over). Source: Kahlon et al., JAMA Psychiatry 2021.

England’s PLINY trial tested a different telephone model with people aged over 74 who lived independently in one city: six weeks of short one-to-one calls from a trained volunteer, then 12 weeks of weekly group calls with up to six participants. Its primary outcome was the mental health score of the SF-36 health questionnaire, not loneliness[7]. Recruiting older people was not the problem (the team informed 9,579 people about the study); keeping volunteers was. Of 10 trained volunteer facilitators, three delivered the group calls; 50 of the 78 people allocated to telephone friendship never received it, and the trial was closed early[8]. Among the 56 people with outcome data, the mental health difference was 7 points unadjusted (95% CI −3 to 16)[7] and 9.5 points after adjustment for age, sex and baseline score (95% CI 4.5 to 14.5)[8]; with most of the intervention group never receiving the calls, these figures are hard to interpret. The authors said a larger trial would need dedicated volunteer management.

The largest recent trial used befriending calls as the comparison. Hong Kong’s HEAL-HOA trial randomised 1,151 people aged 65 or over who lived alone, felt lonely, were digitally excluded and faced financial hardship; their average age was 76.6. All three groups received eight 30-minute phone sessions over four weeks from trained lay counsellors, themselves aged 50 to 70 and people who reported feeling lonely: behavioural activation, mindfulness, or befriending calls as an “attention control”[9][10]. At 12 months, the behavioural activation and mindfulness groups scored lower (less lonely) on the UCLA scale than the befriending group, by 0.73 and 0.72 points[9]. On the De Jong Gierveld scale the picture was mixed: behavioural activation did not differ significantly from befriending up to six months but was 0.13 points lower at 12 months, and the mindfulness group scored higher (lonelier) than the befriending group at two earlier time points[9][10].

A smaller US trial pointed the same way. Among 89 homebound older people, average age 74, who received home-delivered meals and all reported loneliness, five weekly video sessions of behavioural activation from a lay coach reduced loneliness (measured with the PROMIS Social Isolation Scale), depressive symptoms and disability more than five weekly “friendly visits” by video[11]. These trials had no group without calls, so they do not show that befriending does nothing; they suggest brief structured approaches from trained lay people can add something beyond friendly conversation.

Befriending helped care-home residents in one large trial, but did not help carers who were offered it

The strongest recent trial of befriending itself comes from Australian residential aged care. In a pragmatic randomised trial of 345 residents, people who received befriending had fewer depressive symptoms than those receiving usual care, 0.76 points lower on the 15-item Geriatric Depression Scale at 8 weeks and 0.95 lower at 16 weeks (Cohen’s d −0.458). Loneliness, measured with the UCLA scale, was 2.39 units lower than in the control group at 8 weeks and 2.71 lower at 16 weeks (d −0.481). There was no significant difference in anxiety or perceived social support. The authors described the gains as a “small improvement” in depressive symptoms and loneliness[3]. The abstract we read does not say which UCLA version was used, how befriending was delivered or how often. See also our study of loneliness in care homes.

Carers tell a different story. The BECCA trial in East Anglia and London offered 236 family carers of people with dementia access to a befriending scheme, with weekly home visits from a trained volunteer expected for at least six months. Its primary outcome was carers’ mood, on the depression scale of the Hospital Anxiety and Depression Scale; health-related quality of life and loneliness were among the secondary outcomes. After 15 months there was no benefit on any of them[12].

The trial’s own explanation is uptake. Just 48% of carers offered a befriender asked to be matched, and only 32% received the intended six months of befriending before follow-up. The intended hour a week was rarely achieved because many carers were too busy, and befriending was more likely to be used by carers with no local family and little contact with family, friends or neighbours[12]. A trial of an offer measures the offer, not the befriending relationship.

How many carers offered a befriender took it up, BECCA trial, EnglandAmong family carers of people with dementia offered access to a befriending scheme in the BECCA trial: 48% asked to be matched with a befriender; 32% received the intended minimum six months of befriending before the 15-month follow-up.0%25%50%75%100%Asked for a befriender48%Matched for 6+ months32%
Share of carers in the intervention group (offered access to a befriender facilitator). The trial measured carers’ depressive symptoms (HADS) and quality of life, and found no benefit on these or on secondary outcomes, which included loneliness. Source: Charlesworth et al., BMJ 2008.

Peer befriending looks promising in small studies, and how a scheme is run seems to matter

In Canada, a small randomised trial assigned 60 older Chinese immigrants aged 65 and over to an eight-week peer support programme or a control group, 30 in each. Twenty-four volunteers aged 48 to 76 worked in pairs, visiting and phoning each participant. The intervention group showed a statistically significant fall in loneliness compared with the control group, along with fewer depressive symptoms[13]. The assessors knew which group each person was in, which can bias results.

In San Francisco, a two-year study followed 74 low-income older people recruited through a senior center, 86% of whom lived alone, as they received visits from peers matched by background; on average there were 43 contacts in the first year. Scores on the three-item UCLA loneliness scale fell by 0.8 points on average over 24 months, and the share screening positive for depression on the two-item PHQ-2 fell from 38% to 16%[14]. There was no comparison group. People who join when struggling tend to improve on average anyway (regression to the mean), so this is encouraging but not proof.

Qualitative studies help explain why some matches work. A realist evaluation of five befriending programmes in Northern Ireland, based on 46 interviews, found that reciprocity was most likely when befriender and older person shared characteristics, the befriender was a volunteer and they met in person, and that a sense of autonomy grew in long-term relationships without a fixed agenda[15]. In 40 interviews with people using telephone befriending at eight UK project sites in 2007–2008, older people described the service helping them “gain confidence, re-engage with the community and become socially active again”[16]. In an Ontario student phone-call programme, 58.3% of 60 feedback-survey respondents (out of 220 active participants) agreed they felt less lonely; there was no control group[17].

Not everyone wants a befriender. In interviews with 28 people aged 65 and over in England who felt lonely, participants generally thought group activities built around a shared interest preferable to one-to-one befriending, and that support labelled as being for loneliness, or as being for older people, put them off[18]. Befriending is one option among many; our study of what reduces loneliness in later life maps the others.

What the evidence asks of us

In June 2025 the World Health Organization called on all member states to make social connection a public health priority, citing a World Health Assembly resolution that urges them to “develop and implement evidence-based policies, programmes and strategies”[19]. For befriending, the trials point to several things funders, commissioners and services can do:

  • Fund the volunteer infrastructure, not just the volunteers. PLINY closed early because too few volunteers stayed, and BECCA’s befriending rarely reached its intended dose[8][12].
  • Reach the people with the fewest other connections. In BECCA, carers with no local family and little other contact were likelier to use befriending[12].
  • Offer choice. Some older people prefer shared-interest groups to one-to-one support, and dislike being labelled lonely[18].
  • Consider structured elements. In Hong Kong and the US, lay-delivered behavioural activation produced larger falls in loneliness than befriending calls or friendly video visits[9][11]. This may be worth testing.
  • Evaluate with a comparison group and name the outcome. The reviews’ main complaint is small, short, poorly controlled studies[4][5]. Services need support to measure loneliness with a validated scale, separately from mood, and for longer than a few weeks.

None of this means befriending should be cut. Trials measure average questionnaire changes over weeks; they cannot capture everything a reliable weekly visit can mean. The evidence asks for honest expectations, sustained funding and better evaluation. More studies of what helps are collected in our loneliness in later life library, and our study of loneliness and depression explains why the two outcomes are linked but distinct.

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

Befriending services are run largely by volunteers and small charities, often on short-term grants, and they are regularly asked to prove their impact. We wanted to give them, and the people who fund them, a fair account of the trial evidence: where it is encouraging, where it is uncertain, and where the studies measured something other than loneliness. The volunteers who make it happen deserve evidence presented accurately, neither oversold nor dismissed.

How to read this data

Most figures here come from randomised controlled trials and systematic reviews; a few come from uncontrolled before-and-after studies, a feedback survey and qualitative interviews, and we say which. Standardised mean differences (SMD) and Cohen’s d express group differences in standard deviations; raw point differences only make sense on the scale used. The UCLA and De Jong Gierveld loneliness scales sometimes disagreed within one trial, and several trials measured mood or quality of life rather than loneliness; we name the outcome each time. Trials that randomise people to an offer of befriending measure the effect of the offer, which many people did not take up. Most studies come from high-income countries, and we found few trials following people beyond a year. Where we could read only an abstract, we used only what it said.

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). Befriending and Loneliness in Later Life: Small Trials, Mixed Results. https://understandingaging.org/research/befriending-evidence

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. Mead N, Lester H, Chew-Graham C, Gask L, Bower P. Effects of befriending on depressive symptoms and distress: systematic review and meta-analysis. British Journal of Psychiatry, 2010. europepmc.org/article/MED/20118451
  2. Siette J, Cassidy M, Priebe S. Effectiveness of befriending interventions: a systematic review and meta-analysis. BMJ Open, 2017. europepmc.org/article/PMC/PMC5594212
  3. Fearn M, Harper R, Major G, et al. Randomized controlled trial on the impact of befriending on depression, anxiety, loneliness, and social support in older people in aged care. Clinical Gerontologist, 2026. doi.org/10.1080/07317115.2025.2579846
  4. Laermans J, Scheers H, Vandekerckhove P, De Buck E. Friendly visiting by a volunteer for reducing loneliness or social isolation in older adults: a systematic review. Campbell Systematic Reviews, 2023. europepmc.org/article/PMC/PMC10688573
  5. 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/PMC/PMC13460981
  6. Kahlon MK, Aksan N, Aubrey R, et al. Effect of layperson-delivered, empathy-focused program of telephone calls on loneliness, depression, and anxiety among adults during the COVID-19 pandemic: a randomized clinical trial. JAMA Psychiatry, 2021. europepmc.org/article/PMC/PMC7903319
  7. Mountain GA, Hind D, Gossage-Worrall R, et al. ‘Putting Life in Years’ (PLINY) telephone friendship groups research study: pilot randomised controlled trial. Trials, 2014. europepmc.org/article/PMC/PMC4022155
  8. Hind D, Mountain G, Gossage-Worrall R, et al. Putting Life in Years (PLINY): a randomised controlled trial and mixed-methods process evaluation of a telephone friendship intervention to improve mental well-being in independently living older people. Public Health Research (NIHR Journals Library), 2014. doi.org/10.3310/phr02070
  9. Tang VFY, Jiang D, Kwok JYY, et al. Behavioral activation and mindfulness interventions in reducing loneliness and improving well-being in older adults: the HEAL-HOA randomized clinical trial. JAMA Network Open, 2026. europepmc.org/article/PMC/PMC12873770
  10. Jiang D, Kwok JYY, Yeung DY, et al. Six-month outcomes of layperson-delivered, telephone-based behavioural activation and mindfulness interventions on loneliness among older adults during the COVID-19 pandemic: the HEAL-HOA dual randomised controlled trial. Age and Ageing, 2025. europepmc.org/article/PMC/PMC12315544
  11. Choi NG, Pepin R, Marti CN, Stevens CJ, Bruce ML. Improving social connectedness for homebound older adults: randomized controlled trial of tele-delivered behavioral activation versus tele-delivered friendly visits. American Journal of Geriatric Psychiatry, 2020. europepmc.org/article/PMC/PMC8767809
  12. Charlesworth G, Shepstone L, Wilson E, et al. Befriending carers of people with dementia: randomised controlled trial. BMJ, 2008. europepmc.org/article/PMC/PMC2413394
  13. Lai DWL, Li J, Ou X, Li CYP. Effectiveness of a peer-based intervention on loneliness and social isolation of older Chinese immigrants in Canada: a randomized controlled trial. BMC Geriatrics, 2020. europepmc.org/article/PMC/PMC7507625
  14. Kotwal AA, Fuller SM, Myers JJ, et al. A peer intervention reduces loneliness and improves social well-being in low-income older adults: a mixed-methods study. Journal of the American Geriatrics Society, 2021. europepmc.org/article/PMC/PMC8648986
  15. Fakoya OA, McCorry NK, Donnelly M. How do befriending interventions alleviate loneliness and social isolation among older people? A realist evaluation study. PLoS One, 2021. europepmc.org/article/PMC/PMC8428774
  16. Cattan M, Kime N, Bagnall AM. The use of telephone befriending in low level support for socially isolated older people: an evaluation. Health & Social Care in the Community, 2011. europepmc.org/article/MED/21114564
  17. Ahuja M, Lewis K, Owais S, et al. Evaluation of a student-older adult telephone befriending program to reduce social isolation during the COVID-19 pandemic. Canadian Geriatrics Journal, 2026. europepmc.org/article/PMC/PMC12931409
  18. Kharicha K, Iliffe S, Manthorpe J, et al. What do older people experiencing loneliness think about primary care or community based interventions to reduce loneliness? A qualitative study in England. Health & Social Care in the Community, 2017. europepmc.org/article/PMC/PMC5655770
  19. World Health Organization. Social connection linked to improved health and reduced risk of early death (news release). WHO, 30 June 2025. who.int/news/item/30-06-2025-social-connection-linked-to-improved-heath-and-reduced-risk-of-early-death