Loneliness: what it is and what the evidence says

Short answer

Loneliness — the distressing gap between the relationships a person has and those they want — is common, and it is linked with depression and with a higher risk of earlier death [C2][C3]. Trials show that psychological programmes, social support and some digital formats can reduce it, though most effects are small and evidence quality is often low [C4][C8].

What it is and how it shows up

Loneliness is the distress that comes from a gap between the relationships a person has and the relationships they want. It is subjective: two people can live with the same number of contacts, and only one of them feels lonely. Researchers separate this feeling from social isolation, an objective lack of contact and participation, and the two do not overlap completely. Both states can occur together, and each can exist without the other. In prospective studies, actual and perceived social isolation are both associated with increased risk for early mortality, and a meta-analysis found no differences between measures of objective and subjective social isolation [C9].

Because loneliness is subjective, it is measured by asking people how they feel, most often with short questionnaires; different scales and cut-offs produce different figures, which is one reason estimates vary between surveys [C1].

How common is it? A systematic review and meta-analysis of nationally representative samples found prevalence data for 113 countries or territories [C1]. Coverage varies by age: data for adolescents were available in 77 countries or territories, and for older adults in 40 countries [C1]. Among adolescents, the pooled prevalence of loneliness ranged from 9.2% in South-East Asia to 14.4% in the Eastern Mediterranean region [C1]. For adults, the meta-analysis could be pooled for the European region only, where a consistent geographical pattern appeared: prevalence was lowest in northern European countries and highest in eastern European countries, reaching 21.3% among older adults in eastern Europe [C1].

Loneliness also travels with health problems. In a meta-analysis of prospective studies, loneliness was associated with 1.26 times the odds of earlier death, social isolation with 1.29 times the odds and living alone with 1.32 times the odds — a magnitude the authors compare with well-established mortality risk factors [C2]. Separately, a meta-analysis of 88 studies with 40,068 individuals found that loneliness had a moderately significant effect on depression [C3]. These are associations: the pooled studies are mostly correlational, so they cannot settle which condition feeds the other.

What research shows by approach

Psychological and cognitive approaches

The most studied group of interventions targets how people think about and respond to social situations. Cognitive approaches work on expectations: if a person reads a missed message as rejection, the programme teaches them to test that reading against the evidence. Distress-tolerance approaches target the discomfort of meeting people; behavioural-activation and social-skills formats build contact step by step.

In a meta-analysis of thirty-five randomized trials with 5291 older adults testing such mechanistic interventions, group interventions, longer and more frequent sessions, and cognitive or distress-tolerance approaches had larger effect sizes [C7]. The same review cautions that heterogeneity between studies was high and that there was a trend toward publication bias [C7]. The pattern that emerges is that the format matters as much as the content: interventions delivered in groups, and those with more session time, tended to do better than brief individual formats [C7].

Social support and group-based activities

Programmes that add or restructure contact — group activities, peer support, counselling formats — have been tested mainly with older adults. In a meta-analysis of nineteen randomized trials in community-dwelling older adults, social support interventions effectively reduced loneliness [C8]. In long-term care the picture is less certain: in a systematic review and meta-analysis of seventy randomized clinical trials with 8259 participants aged 65 years or older, most intervention types had a small effect size [C4]. Animal therapy in long-term care had the largest effect size on loneliness reduction in that review, but the authors advise caution in interpreting its magnitude because of high heterogeneity, and the overall quality of evidence was rated as very low [C4].

Digital interventions

Digital formats matter because they reach people who cannot attend in person, including homebound older adults and people in remote areas. In a meta-analysis of 40 randomized trials of digital interventions, psychological programmes with group or social components, group-based activities and robotic pets were effective in reducing loneliness, while social contact interventions, self-guided individual activities and conversational robots showed limited impact [C5]. What separates the formats that worked from those that did not, the review suggests, is whether the technology creates a genuine social connection: a robot pet or an online group with a human facilitator can, while a one-way broadcast or a chatbot conversation tends not to. Social media reduction interventions suggested potential benefits, though the results were not statistically significant [C5]. The authors of that review note moderate to high risk of bias, heterogeneity and limited long-term follow-up [C5].

Young adults

Young adulthood is a period of particular vulnerability, because social networks are reorganised around study, work and moving. In a meta-analysis of social connection interventions in young adults, six studies showed an overall mean reduction in depression, but the pooled reduction in loneliness was not statistically significant [C6]. The authors of that review call for more high-quality research across diverse settings [C6]. What this means practically is that evidence for this age group is thinner than for older adults, and adaptations of group and psychological formats remain the best-supported starting point [C7].

Limitations and uncertainty

The evidence has real limits. Loneliness is usually measured with self-report scales, so prevalence figures depend on the instrument and its cut-offs; the global prevalence review could pool adult estimates for the European region only and describes its findings as limited by data scarcity and methodological heterogeneity [C1]. Intervention trials are often small, with few studies per intervention type, high heterogeneity and short follow-up; in the older-adults review the overall quality of evidence was very low, and the largest effects came from long-term-care settings where the authors advise caution in interpreting their magnitude [C4]. Digital trials showed moderate to high risk of bias [C5]. The links with depression and mortality come mostly from observational designs, so reverse causation and confounding cannot be ruled out [C2][C3]. Finally, many trials compare a programme with no treatment rather than with an established alternative, which makes it hard to know whether one approach works better than another.

When to see a professional

Feeling lonely from time to time is part of life. It warrants more attention when it persists for weeks, comes with persistent low mood, loss of interest, changes in sleep or appetite, or when a person withdraws from contact they previously valued. For people who became lonely after a move, bereavement or retirement, structured group activities with a shared task — singing, exercise, volunteering, classes — are often a more natural first step than individual self-help, and they match the formats that trials favour [C7].

Loneliness that accompanies depressed mood, anxiety or thoughts of self-harm is a reason to talk to a professional rather than to keep trying self-help alone. A clinician can help distinguish loneliness from a depressive episode and discuss what may help, including structured psychological support of the kind tested in trials [C7].

FAQ

What is the difference between loneliness and social isolation?

Social isolation is an objective lack of contact; loneliness is the subjective feeling that your relationships fall short of what you want. In prospective studies both are associated with increased risk for early mortality, and a meta-analysis found no differences between measures of objective and subjective social isolation [C9].

Is loneliness bad for your health?

On average, in observational and prospective studies, yes. Loneliness was associated with 1.26 times the odds of earlier death in a meta-analysis [C2], and across 88 studies loneliness had a moderately significant effect on depression [C3]. These are associations, not proof that loneliness causes disease.

Can loneliness be reduced?

In trials, several intervention types have reduced it: psychological programmes, social support interventions and some digital formats [C4][C5][C8]. Most effects in the older-adults review were small, and evidence quality was rated as very low [C4].

Do digital tools help with loneliness?

Some do and some do not. In a meta-analysis of 40 randomized trials, psychological digital programmes with group or social components, group-based activities and robotic pets were effective in reducing loneliness, while social contact interventions, self-guided individual activities and conversational robots showed limited impact [C5]. Social media reduction interventions suggested potential benefits, though the results were not statistically significant [C5].

Which interventions work best in older adults?

The consistent pattern is that format matters. In a meta-analysis of mechanistic interventions in older adults, group interventions, longer and more frequent sessions, and cognitive or distress-tolerance approaches had larger effect sizes [C7]. Social support interventions in community settings effectively reduced loneliness in nineteen randomized trials [C8].

Self-help information does not replace a clinician; people with severe or persistent symptoms, or thoughts of self-harm, should contact a professional or their local emergency services.