Loneliness: a real health risk – and what helps

Loneliness is often treated as a private matter. Yet it is common, it is by no means confined to old age, and in large studies it is linked to heart disease, stroke, depression and earlier death. How solid is that evidence – and what actually helps?

Lonely or isolated – an important difference

Loneliness is a feeling: the painful sense that one's relationships are not what one wants them to be. Social isolation is an objective state: few contacts, little exchange, a small network. Someone who lives alone can feel well connected; someone surrounded by family and colleagues can be deeply lonely. The difference matters in practice. Isolation responds to opportunities for contact. Loneliness often does not, when shame, mistrust or the expectation of rejection stand in the way.

How common is it?

German figures come from the federal family ministry's Einsamkeitsbarometer ("loneliness barometer"). Its first edition (May 2024, based on the German Socio-Economic Panel) shows that 7.6% of adults had elevated loneliness in 2017, 28.2% in the first pandemic year 2020 and 11.3% in 2021. By 2021 people over 75 were roughly back at pre-pandemic levels; young adults were not: among 18- to 29-year-olds the share was 14.1% (2017: 8.6%). The latest edition (June 2026, focused on people with disabilities, 2023 data) confirms the pattern: those aged 18 to 40 and over 75 are most affected. In December 2023 the federal cabinet had already adopted a Strategy against Loneliness with 111 measures.

The WHO Commission on Social Connection estimated in June 2025 that about one in six people worldwide feel lonely, and 17–21% of those aged 13 to 29. It linked loneliness to more than 871,000 deaths a year. That is a model estimate assuming causality, not a count.

What loneliness does to health

Mortality. A 2010 meta-analysis by Julianne Holt-Lunstad (148 studies, 308,849 participants) found that people with stronger social relationships had 50% higher odds of survival (odds ratio 1.50). Put concretely: by the time half of 100 people have died, about five more are still alive in the group with stronger relationships. A follow-up analysis (2015) found 29% higher odds of death with social isolation, 26% with loneliness and 32% with living alone. A meta-analysis of 90 cohort studies with 2.2 million people (Wang et al., 2023) confirmed the pattern, with a stronger signal for isolation (hazard ratio 1.32) than for loneliness (1.14).

Heart and blood vessels. In a meta-analysis of longitudinal studies (Valtorta et al., 2016), poor social relationships were associated with a 29% higher risk of coronary heart disease and a 32% higher risk of stroke – with considerable variation between studies and signs of publication bias.

Brain and mind. The Lancet Commission lists social isolation in later life among 14 modifiable risk factors for dementia (see our article on dementia prevention). According to the WHO, lonely people are twice as likely to become depressed. It works both ways: depression makes people withdraw, and withdrawal deepens depression.

Cause or consequence?

All of these data come from observational studies; nobody can randomise people to loneliness. First, confounding: poverty, unemployment, smoking or disability make people both lonely and ill. Second, reverse causation: people who are chronically ill, hard of hearing or in pain withdraw. Baseline health status influenced the results in Holt-Lunstad's analysis. The consistency of the association across countries and age groups suggests it is at least partly causal; how large that part is, nobody knows.

What helps – and how good is the evidence?

A classic meta-analysis (Masi et al., 2011) compared four approaches: improving social skills, enhancing support, creating opportunities for contact, and addressing maladaptive social cognition – such as expecting to be rejected anyway. Uncontrolled and non-randomised studies showed larger effects than randomised ones. In randomised trials the overall effect was small, and approaches targeting these thought patterns, often with methods from cognitive behavioural therapy, did best.

For older adults, a meta-analysis of randomised trials (Hoang et al., 2022) found mostly small effects, from exercise and group activities to psychotherapy; a few larger effects rest on small, heterogeneous studies, and the overall quality of evidence was rated very low. For befriending visits by volunteers, a Campbell review (2023) found the effect uncertain. Volunteering oneself was linked to less loneliness in five of six studies in one review; whether volunteering is the cause remains open. Social prescribing, in which UK general practices refer patients to groups or courses, is popular – but in 2025 the Swedish assessment agency SBU found only reviews at high risk of bias (for people under 60) and drew no conclusions.

An often overlooked lever is hearing: people who can no longer follow conversations withdraw out of exhaustion. A hearing test is sensible, even though the effect of hearing aids on loneliness itself is barely studied (see dementia prevention).

What this means for you

• If you feel lonely: it is not a failure but a signal – like hunger, it points to a need. Small, regular contacts count. Doing something together – a choir, a sports group – is often easier than "meeting people".
• Watch your thoughts. Long loneliness makes people wary; neutral signals start to feel like rejection. That protective reflex is exactly where the most effective approaches work.
• If you notice it in someone else: make concrete invitations without lecturing ("Would you like to come along on Saturday?" rather than "You need to get out more"). Stay reliable; one refusal is not the end. With older relatives, look for practical barriers: hearing, mobility, fear of falling.
• When professional help makes sense: if low mood, loss of interest, poor sleep or hopelessness persist for weeks, depression may be involved – and it is treatable. In Germany, the family doctor is the first port of call; the 116117 appointment service arranges an initial psychotherapy consultation, and no referral is needed.

This article is general information and not individual medical advice. Decisions about tests and treatment belong in a conversation with your own physician.