Skip to content
HabitSeek
Back to results

Regular social contact

Maintaining meaningful relationships and keeping up regular contact with other people, rather than letting social ties lapse into isolation.

At a glance

  • Evidence: Stronger social relationships are associated with lower all-cause mortality.All-cause mortality · Moderate certainty · Observational association
  • What’s uncertain: The evidence is observational; illness can both reduce social contact and increase mortality, so causation is uncertain.
  • First step: Choose a recurring, calendar-level commitment rather than relying on spontaneous contact, such as a standing weekly call or a regular shared activity with the same people.
  • Before you try it: Discuss persistent loneliness or low mood with a clinician; sudden social withdrawal can be a consequence of illness. All precautions
Read the full takeaway

Health associations

Keeping up regular contact with other people is associated with lower all-cause mortality, lower risk of developing depression and lower dementia risk. The evidence is entirely observational, so it shows association and not cause.

Mortality and reverse causation

The mortality estimates are the most consistent and the best powered, but they are not interchangeable: pooled estimates for social isolation range roughly from 1.14 to 1.50 across different syntheses, populations and exposure definitions, so no single figure should be quoted as the current best estimate. Reverse causation is a serious concern throughout, because illness, disability and early cognitive decline all reduce social contact and independently raise the risk of death, depression and dementia, and no trial has shown that deliberately increasing social contact changes any of these outcomes.

Depression evidence

The depression evidence is the thinnest of the three, resting on eight pooled cohorts and a field dominated by cross-sectional studies.

Friends sharing coffeeDiscussed by4 peopleView statements ↓
Long-term healthEnergyStress & wellbeingVaries · Easy · Free
Reviewed Sep 23, 2026
Save

Scientific evidence

What the research shows

Outcome assessments based on research, including its uncertainties and limitations.

All-cause mortality

Finding: Benefit supportedCertainty: Moderatestudies found a benefit from doing this

Evidence basis: observational studies (association, not proof of cause)

Stronger social relationships are associated with lower all-cause mortality.

Population: Adults across general community and clinical populations, from midlife to older age. Holt-Lunstad 2010 pooled 308,849 participants with a mean age at intake of 63.9 years (49% female, 51% male) followed a mean of 7.5 years. Wang 2023 pooled 2,205,199 adults aged 18 years or older, including subgroups with cardiovascular disease or cancer. Nakou 2025 restricted inclusion to older adults.

Research details

Summary

Multiple independent meta-analyses of prospective cohort studies agree that people with stronger, more frequent social relationships have lower all-cause mortality. The foundational synthesis of 148 studies covering 308,849 participants found a random-effects weighted average effect of OR 1.50 (95% CI 1.42-1.59), indicating a 50% increased likelihood of survival for those with stronger social relationships, and this held across age, sex, initial health status, cause of death and follow-up length. A 2023 synthesis of 90 cohorts in 2,205,199 individuals and a 2025 synthesis of 86 cohorts in older adults reproduced the direction with smaller, more precise estimates, though the 2025 review found substantial heterogeneity (I2 = 84.0%).

Why this certainty

Rated Moderate rather than Strong. Consistency is strong: three independent syntheses using different inclusion criteria, populations and eras converge on a harmful direction for social isolation, and precision is good with narrow confidence intervals in the two largest analyses. Relevance is high because mortality is the hardest clinical endpoint available. Certainty is held below Strong because every contributing study is observational, reverse causation is plausible and largely unaddressed, effect estimates range widely across syntheses (1.14 to 1.50), and heterogeneity is substantial.

Limitations

All evidence is observational, so residual confounding by socioeconomic position, baseline health and health behaviours cannot be excluded, and reverse causation is a serious concern because illness and disability both reduce social contact and raise mortality risk. Social exposure definitions vary enormously across studies, from complex social-integration indices to single binary items such as living alone, and Holt-Lunstad 2010 explicitly coded research design rather than assessing individual study risk of bias. Heterogeneity was substantial in the newer pooled analyses (I2 = 84.0% in Nakou 2025), and meta-regression identified study quality and adjustment for cognitive function as predictors of the estimated effect, implying the pooled estimate is sensitive to study design.

Effect estimate

Holt-Lunstad 2010: random-effects weighted average effect size OR 1.50 (95% CI 1.42 to 1.59) across 148 studies and 308,849 participants; complex measures of social integration OR 1.91 (95% CI 1.63 to 2.23) and binary residential status (living alone versus with others) OR 1.19 (95% CI 0.99 to 1.44). Wang 2023: social isolation and all-cause mortality pooled effect size 1.32 (95% CI 1.26 to 1.39); loneliness and all-cause mortality 1.14 (95% CI 1.08 to 1.20). Nakou 2025: loneliness HR 1.14 (95% CI 1.10-1.18), social isolation HR 1.35 (95% CI 1.27-1.43), living alone HR 1.21 (95% CI 1.13-1.30).

Sources for this outcome:

Incident depression

Finding: Benefit supportedCertainty: Limitedstudies found a benefit from doing this

Evidence basis: observational studies (association, not proof of cause)

Loneliness and social isolation are associated with a higher risk of developing depression.

Population: General adult populations and older adults in Western and Asian countries. Mann 2022 synthesised 32 longitudinal studies in the general population, of which eight independent cohorts were meta-analysed, with most primary studies addressing depression. Pearce 2023 covered systematic reviews of both clinical and general population samples and reported that depression (including post-natal depression) and psychosis were the outcomes most often studied.

Research details

Summary

Longitudinal syntheses indicate that people who are lonely or socially isolated are more likely to develop depression, but the evidence base is weaker than for mortality. A meta-analysis of eight independent cohorts found a pooled adjusted odds ratio of 2.33 (95% CI 1.62-3.34) for new-onset depression in adults who were often lonely compared with those who were not often lonely, with the authors advising caution because of heterogeneity. An umbrella review of 53 systematic reviews covering 1,657 studies similarly reported associations between loneliness, social isolation, social support and network size and depression, but its authors concluded that findings were limited by low-quality reviews and a predominance of cross-sectional evidence.

Why this certainty

Rated Limited. The direction is consistent across two independent syntheses and one provides a pooled estimate with a confidence interval excluding the null. However, that estimate derives from only eight cohorts with heterogeneity, the field is dominated by cross-sectional designs, the umbrella review explicitly graded the underlying reviews as mostly critically low quality, and reverse causation between depression and social withdrawal cannot be separated from causation. This prevents a Moderate rating despite the consistency.

Limitations

The pooled estimate rests on only eight cohorts and carries substantial unexplained heterogeneity, and effect sizes were weaker in both magnitude and significance in cohort studies than in cross-sectional studies in the related social-support review. Reverse causation is difficult to exclude because subclinical depression reduces social contact, and most of the 1,657 primary studies in the umbrella review were cross-sectional, with only 340 (21%) longitudinal. The umbrella review applied a modified AMSTAR appraisal and rated most included reviews as critically low quality, and it reported a narrative synthesis rather than pooled estimates. The umbrella review of 53 systematic reviews used narrative synthesis for depression and reported no pooled estimate, so no effect estimate from it is recorded here.

Effect estimate

Pooled adjusted odds ratio for new-onset depression in adults who were often lonely versus not often lonely: 2.33 (95% CI 1.62 to 3.34) across eight independent longitudinal cohorts (Mann 2022).

Sources for this outcome:

Dementia and cognitive decline

Finding: Benefit supportedCertainty: Limitedstudies found a benefit from doing this

Evidence basis: observational studies (association, not proof of cause)

Frequent social contact and stronger social engagement are associated with lower dementia risk, while loneliness is associated with higher risk; social isolation itself was not significant.

Population: Middle-aged and older adults in population-based cohorts, predominantly North America, Europe and Asia. Wang 2023 pooled 40 publications from 32 studies or databases. Luchetti 2024 combined ongoing ageing studies and the published literature for 21 samples (N = 608,561) for dementia and 16 samples (N = 103,387) for cognitive impairment. Su 2022 pooled 38 longitudinal studies with 2,154,818 participants at baseline, including 74,700 all-cause dementia cases.

Research details

Summary

Prospective cohort syntheses consistently associate richer social contact with lower dementia risk, with the most direct evidence coming from measures of contact frequency and social engagement. A meta-analysis of 32 longitudinal studies found that frequent social contact was associated with lower dementia risk (RR 0.86, 95% CI 0.76-0.97) and strong social engagement with RR 0.81 (95% CI 0.74-0.89), while loneliness raised risk (RR 1.42, 95% CI 1.26-1.60). A larger meta-analysis of 21 samples and 608,561 individuals found loneliness increased all-cause dementia risk (HR 1.306, 95% CI 1.197-1.426) with associations persisting after adjustment for depression and social isolation. A separate meta-analysis of leisure activities found social activities were associated with lower all-cause dementia incidence (RR 0.93, 95% CI 0.87-0.99), but the same analysis found no significant association for Alzheimer disease from a single study. The exposures pooled in this outcome are not all measures of contact frequency: the loneliness estimate comes from a loneliness meta-analysis and the social-activity estimate from participation in social activities, so the headline names each exposure separately rather than summarising them as social contact.

Why this certainty

Rated Limited. Three independent syntheses point in the same protective direction and the direct measure of interest, frequent social contact, has a pooled estimate whose confidence interval excludes the null. Certainty is limited because the largest and most direct synthesis explicitly called its inference tentative because of high heterogeneity, another found publication bias, subgroup estimates for social activity rested on very few studies, and reverse causation is especially plausible for dementia, where social withdrawal is an early manifestation. No randomised evidence exists for incident dementia.

Limitations

Heterogeneity was generally high in Wang 2023 even after sensitivity analysis, and the authors described the inference as tentative; social isolation itself was not significantly associated with dementia in that analysis (RR 1.58, 95% CI 0.80-3.12), which undercuts a simple dose-response story. Su 2022 found evidence of publication bias on Egger testing for all-cause dementia and Alzheimer disease and relied on only four studies for the social-activity subgroup, with a single study for the Alzheimer disease social-activity estimate. Dementia diagnosis and reverse causation are major concerns because preclinical cognitive decline reduces social activity years before diagnosis, and Luchetti 2024 reported that large heterogeneity was partly due to differences in loneliness measures and cognitive-status ascertainment.

Effect estimate

Wang 2023: frequent social contact RR 0.86 (95% CI 0.76-0.97, p = 0.018); strong social engagement RR 0.81 (95% CI 0.74-0.89, p < 0.001); loneliness RR 1.42 (95% CI 1.26-1.60, p < 0.001); social support RR 0.92 (95% CI 0.80-1.06, p = 0.238); social isolation RR 1.58 (95% CI 0.80-3.12, p = 0.192). Luchetti 2024: loneliness and all-cause dementia HR 1.306 (95% CI 1.197-1.426); Alzheimer disease HR 1.393 (95% CI 1.290-1.504); vascular dementia HR 1.735 (95% CI 1.483-2.029); cognitive impairment HR 1.150 (95% CI 1.113-1.189). Su 2022: social activities and all-cause dementia RR 0.93 (95% CI 0.87-0.99, I2 = 12.7%); social activities and Alzheimer disease RR 0.89 (95% CI 0.63-1.26, from 1 study).

Sources for this outcome:

What people say

Who discusses this?

Public statements, with positions specific to each source and topic. These views do not determine the scientific evidence assessment.

What the research shows ↑
Show more (1)

Practical steps and precautions

How to try it

  1. Choose a recurring, calendar-level commitment rather than relying on spontaneous contact, such as a standing weekly call or a regular shared activity with the same people.
  2. Prioritise contact frequency and a stable network over the size of the network; the dementia synthesis found frequent social contact and strong social engagement associated with lower risk, and network size was promising but too thinly studied to pool.
  3. If contact has lapsed, restart with one relationship rather than several at once. No dose has been validated by trials, so treat any specific frequency as an editorial starting point rather than an evidence-based threshold.
  4. Treat social contact as additive to, not a substitute for, other health behaviours; the cohort evidence adjusts for lifestyle factors rather than showing that social contact replaces them.

Precautions

  • These associations are observational. Reduced social contact can be an early consequence of illness, disability or cognitive decline rather than a cause of it, so new or sudden social withdrawal warrants clinical assessment rather than being treated only as a habit to fix.
  • Persistent loneliness or low mood should be discussed with a clinician. This habit is not a treatment for depression, and the depression evidence here is the weakest of the three outcomes.
  • People with diagnosed mental health conditions, or who are grieving or socially anxious, may need structured support to rebuild contact; a self-directed habit change is not always sufficient or appropriate.
  • Social contact is not a substitute for medical care, and no study in this record shows that increasing contact reduces mortality, prevents depression or prevents dementia.

References

Social Relationships and Mortality Risk: A Meta-analytic Review

Holt-Lunstad J, Smith TB, Layton JB · PLoS Medicine · 2010

Systematic review and meta-analysis of prospective cohort studiesVerified from full text
Open original source

A systematic review and meta-analysis of 90 cohort studies of social isolation, loneliness and mortality

Wang F, Gao Y, Han Z, Yu Y, Long Z, Jiang X, Wu Y, Pei B, Cao Y, Ye J, Wang M, Zhao Y · Nature Human Behaviour · 2023

Systematic review and meta-analysis of prospective cohort studiesVerified from abstract
Open original source

Loneliness, social isolation, and living alone: a comprehensive systematic review, meta-analysis, and meta-regression of mortality risks in older adults

Nakou A, Dragioti E, Bastas NS, Zagorianakou N, Kakaidi V, Tsartsalis D, Mantzoukas S, Tatsis F, Veronese N, Solmi M, Gouva M · Aging Clinical and Experimental Research · 2025

Systematic review, meta-analysis and meta-regression of prospective cohort or longitudinal studiesVerified from abstract
Open original source

Loneliness and the onset of new mental health problems in the general population

Mann F, Wang J, Pearce E, Ma R, Schlief M, Lloyd-Evans B, Ikhtabi S, Johnson S · Social Psychiatry and Psychiatric Epidemiology · 2022

Systematic review of longitudinal studies with meta-analysis of independent cohortsVerified from abstract
Open original source

Associations between constructs related to social relationships and mental health conditions and symptoms: an umbrella review

Pearce E, Birken M, Pais S, Tamworth M, Ng Y, Wang J, Chipp B, Crane E, Schlief M, Yang J, Stamos A, Cheng LK, Condon M, Lloyd-Evans B, Kirkbride JB, Osborn D, Pitman A, Johnson S · BMC Psychiatry · 2023

Umbrella review (systematic review of systematic reviews) of observational studies, with narrative synthesisVerified from full text
Open original source

Association between social integration and risk of dementia: A systematic review and meta-analysis of longitudinal studies

Wang S, Molassiotis A, Guo C, Leung ISH, Leung AYM · Journal of the American Geriatrics Society · 2023

Systematic review and meta-analysis of longitudinal cohort studiesVerified from abstract
Open original source

A Meta-analysis of Loneliness and Risk of Dementia using Longitudinal Data from >600,000 Individuals

Luchetti M, Aschwanden D, Sesker AA, Zhu X, O Suilleabhain PS, Stephan Y, Terracciano A, Sutin AR · Nature Mental Health · 2024

Meta-analysis of longitudinal ageing studies and published literatureVerified from abstract
Open original source

Leisure Activities and the Risk of Dementia: A Systematic Review and Meta-analysis

Su S, Shi L, Zheng Y, Sun Y, Huang X, Zhang A, Que J, Sun X, Shi J, Bao Y, Deng J, Lu L · Neurology · 2022

Systematic review and meta-analysis of longitudinal studies (37 cohort studies and 1 nested case-control study)Verified from full text
Open original source

Community

Discussion

Have you tried this? What helped or made it difficult?

Reader experiences are separate from the scientific evidence. Explore all discussions

No discussions yet.

Be the first to ask a question or share your experience.

Sign in to vote or join the discussion.

Related habits

Explore other habits in a shared category. Each has its own evidence and precautions.

Editorial history and review method

Revision 4 · Approved for publication · Codex (publication approval explicitly authorized by site owner; category migration)

Prepared and checked with AI. Publication approval does not imply independent expert review. Request a correction.

Audit correction pass (group 7). (1) dementia-and-cognitive-decline headline replaced: the previous headline claimed that more frequent social contact is associated with lower dementia risk, but three of the four effect-estimate families in that outcome measure loneliness or social-activity participation rather than contact, and the entry's own limitations already flagged that social isolation itself was not significant (RR 1.58, 95% CI 0.80 to 3.12). The headline and an added sentence now name each exposure separately. (2) incident-depression effectEstimate replaced: the field previously held the sentence 'No pooled effect estimate for depression was retrievable from Pearce 2023, which used narrative synthesis', which occupied the slot reserved for a quantitative estimate; it now carries the Mann 2022 pooled estimate that the outcome already cited, and the narrative-synthesis point moved into limitations. No number changed: the audit verified all eight sources and found every stored fact reproduced verbatim, and it recorded three wrong DOIs that arose from its own lookup attempts rather than from this entry, so no canonicalUrl was altered. A second reviewer must re-check: (1) that the revised headline is not read as support for increasing contact, since all evidence remains observational and no trial increased social contact; (2) that the dementia outcome's mixture of contact, loneliness and activity exposures is acceptable or should be split; and (3) the audit's open item — that the two largest syntheses (wang-2023-social-isolation-mortality and nakou-2025) are verified at abstract level only and their full texts should be checked for adjustment for baseline health and for the meta-regression findings. Publication release 2026-10-04: site owner authorized a bounded AI audit and release of this exact revision. See docs/release-audit-2026-10-04.md for source-check findings and limits. No independent health-professional or peer review is claimed. Category migration 2026-10-04: removed the separate goals field; categories: Long-term health, Energy → Long-term health, Energy, Stress & wellbeing. Research assessments, sources, guidance, precautions, and evidence review dates are unchanged. This revision awaits publication approval; no new evidence review is claimed. Publication approval 2026-10-04: site owner explicitly accepted all current category-migration revisions for deployment and requested that remaining limitations be retained in docs/content-review-backlog.md. Approval applies to this exact content revision. No new evidence review, independent expert review, or clinical validation is claimed; existing precautions and disclosures remain in effect.