Socialising

TL;DR

People with strong social ties (family, friends, partner, community) have about 25–35% lower risk of premature death than socially isolated people [1–4].
A 2023 meta-analysis (1.3M people) found that social isolation increases all-cause mortality risk by 33% (HR 1.33) [2].
In a large Chinese cohort, socially isolated and lonely men at 45 lost ~4.6 years of life expectancy, while social isolation alone cost ~3–4 years in both men and women [3].
In US women, high social integration was linked to 10% longer lifespan and a 41% higher chance of reaching 85+ [5].
In KamaLama, we approximate a range from about –4 to –5 years (very low quality time with close others) up to +2–3 years (frequent, meaningful quality time), based on population-level data.

This article is for education only and does not give medical advice or a diagnosis. For personal decisions, please talk to your clinician.

1. Introduction

A large body of research shows that social connection—the quality and frequency of our relationships—is a major determinant of all-cause mortality and life expectancy.

The classic Alameda County study (4,775 adults) found that people with the fewest social and community ties had more than double the risk of death over 9 years compared with those with the most ties [6].
A 2015 meta-analysis by Holt-Lunstad and colleagues reported that both social isolation and loneliness are associated with a 26–32% higher risk of early death [1].
A 2023 meta-analysis of 90 cohorts (1.30M individuals) estimated a pooled hazard ratio (HR) of 1.33 (33% higher risk) for socially isolated vs non-isolated adults [2].

The impact of social connection on mortality is similar in magnitude to several traditional risk factors such as obesity, physical inactivity, and some levels of smoking [1,7].

Despite this, most people—and many clinicians—underestimate how much social relationships affect health and lifespan [7,8]. For KamaLama, social connection is therefore a high-impact, modifiable longevity factor, alongside physical activity, sleep, weight, and other core domains.

2. Methods (how we selected the evidence)

For this factor, we focused on:

Systematic reviews and meta-analyses of social isolation, loneliness, and mortality.
Large prospective cohort studies with all-cause mortality outcomes.
Studies providing:
- Hazard ratios for mortality.
- Where possible, life-years gained or lost (residual life expectancy).

Priority sources:

Holt-Lunstad et al. 2015 meta-analysis on loneliness, social isolation, and mortality [1].
Naito et al. 2023 meta-analysis on social isolation and mortality [2].
Zhao et al. 2025 cohort analysis in China with sex-specific estimates of life expectancy at 45 [3].
Foster et al. 2023 UK Biobank analysis on social connection and mortality [4].
Trudel-Fitzgerald et al. 2020 Nurses’ Health Study analysis on social integration and exceptional longevity in women [5].
Rueda-Salazar et al. 2021 analysis of social participation and life expectancy & healthy years in Chile, Costa Rica, and Spain [9].
Holt-Lunstad 2024 narrative review on social connection as a critical health factor [7].

We prioritised studies that:

Adjusted for age, sex, smoking, socioeconomic status, and baseline health.
Distinguished structural social connection (network size, contact frequency, living alone) from functional aspects (loneliness, perceived support).
Offered data that can be converted into approximate life-years impact for midlife adults.

3. Results

3.1 Magnitude of effect on all-cause mortality

Meta-analyses

Holt-Lunstad et al. (2015, 70+ studies) found that:
- Social isolation, loneliness, and living alone were each associated with 26–32% greater mortality risk, after controlling for major confounders [1].
Naito et al. (2023, 90 cohort studies, 1.30M people) reported:
- HR = 1.33 (95% CI 1.26–1.41) for all-cause mortality in socially isolated vs non-isolated adults [2].

Plain-language interpretation

On average, socially isolated adults are about one-third more likely to die during follow-up than comparable adults who are socially connected.

Comparison with other risk factors

Holt-Lunstad and colleagues show that the effect size for weak social relationships is similar to or larger than:

Some levels of smoking,
Obesity, and
Physical inactivity [1,7].

This does not mean “loneliness is the same as smoking a pack a day”, but it does mean social connection belongs in the same risk conversation as these factors.

3.2 Years of life gained or lost

Life expectancy at age 45: sex-specific estimates (Zhao 2025)

Zhao et al. used Chinese CHARLS data to estimate loss of residual life expectancy at 45 for different combinations of social isolation and loneliness [3]:

Group (at age 45)	Years lost – Men	Years lost – Women
Socially isolated & lonely	4.6 years	Not significant
Socially isolated only	3.8 years	3.1 years
Lonely only	2.8 years	Not significant

Interpretation:

Severe combined deficit (isolated and lonely) in midlife men: around 4–5 years shorter life expectancy vs peers with good social connection.
“Only” socially isolated or “only” lonely: roughly 2–4 years shorter life expectancy.
In this cohort, women were particularly sensitive to structural isolation, while loneliness alone was not statistically significant.

These are among the clearest life-years estimates available and support using a –2 to –5 years range for very poor social connection in midlife adults.

Social integration, lifespan, and exceptional longevity in women

In the Nurses’ Health Study (72,322 women, ~22 years of follow-up), high social integration (marriage/partnership, regular contact with close others, and community participation) was associated with [5]:

10% longer lifespan, and
41% higher odds of surviving to age 85+.

If remaining life expectancy at 60 is approximately 25 years, a 10% increase suggests roughly +2.5 additional years for women with high social integration compared to those with very low integration.

Social participation and healthy life expectancy

Rueda-Salazar et al. (2021) examined older adults in Chile, Costa Rica, and Spain and found that social participation (clubs, community groups, volunteering) contributed to [9]:

Higher total life expectancy, and
More years lived in good health (healthy life expectancy).

Differences between low vs high social participation often corresponded to about 1–3 extra healthy years.

3.3 Patterns and ranges of social connection

Most studies distinguish:

Structural social connection
- Living alone vs with others,
- Frequency of in-person or phone contact with friends/family,
- Group/community participation.
Functional social connection
- Loneliness,
- Perceived emotional support,
- Having someone to confide in.

Both structural and functional aspects independently predict mortality, and combined deficits carry the highest risk [1,4,7].

Threshold patterns (UK Biobank)

People with less than monthly visits from friends/family and other markers of structural isolation (e.g., living alone) show some of the highest mortality risk [4].
Adding loneliness on top of structural isolation further increases risk, suggesting a cumulative effect.

Dose–response patterns

Meta-analyses show a graded association: as the number and diversity of social ties increase, mortality risk decreases [1,7].
Moving from severe isolation → moderate connection → high integration appears to reduce risk stepwise, with diminishing returns at the highest levels.

3.4 Who is most affected?

Age: Social deficits often have stronger effects in midlife and early older age; in very old age, frailty and comorbidities may dominate risk [7].
Sex: Several studies (including Zhao 2025) suggest stronger effects of combined isolation + loneliness in men, and strong effects of structural isolation in women, although not all cohorts agree [3].
Cultural context: Across US, Europe, Asia, and Latin America, better social connection consistently predicts lower mortality, but the most protective types of ties (family vs community vs friends) vary by cultural norms and living arrangements [3,4,6,9].

3.5 Mechanisms: how do relationships influence longevity?

The pathways linking social connection to longer life are likely multifactorial [1,7,10]:

Behavioural mechanisms
- Socially connected people are more likely to:
  - Be physically active,
  - Smoke less and drink alcohol more moderately,
  - Eat more balanced diets,
  - Attend preventive screenings and adhere to treatment.
Psychological mechanisms
- Social support buffers the effects of stress,
- Reduces depression and anxiety,
- Increases sense of purpose, belonging, and optimism,
- All of which are linked to better survival.
Biological mechanisms
- Social isolation and loneliness are associated with:
  - Higher low-grade inflammation,
  - Immune dysregulation,
  - Adverse cardiometabolic markers,
  - Potentially accelerated biological ageing (e.g., telomere dynamics in older adults) [5,10,11].

Most evidence is observational, but the consistency of associations, dose–response patterns, and biological plausibility support a largely causal interpretation.

4. From evidence to KamaLama scoring

4.1 KamaLama social-connection question

To capture a strong, actionable signal in a simple way, KamaLama uses this question:

How many times per month do you have quality time with people close to you?

Count meaningful moments with your partner, family, friends, or colleagues like having dinner together, going for a walk, watching a movie, playing a game, or spending a day out.
Don’t count quick check-ins, chores, or daily routines.

Answer options:

Once a month or less
A few times a month
A few times a week or more

This item is a pragmatic proxy for both:

Structural connection – having people you regularly meet and spend time with.
Functional connection – sharing emotionally meaningful experiences, not only logistics.

Mapping KamaLama answers to risk and life-years

Using:

High-risk patterns from UK Biobank (e.g., less than monthly contact associated with elevated mortality) [4],
Life-expectancy gaps from Zhao 2025 (up to 4.6 years lost at 45 for isolated + lonely men, 3.1–3.8 years for isolated men and women) [3],
Longevity gains associated with high social integration in women (+10% lifespan, +41% odds to reach 85+) [5],

we derive approximate ranges for midlife adults (assuming otherwise average risk profile):

KamaLama answer option	Interpreted level of social connection	Approx. mortality risk vs “best” group	Approx. life-years impact*	How we interpret this pattern in real life
Once a month or less	Very low connection	About 30–60% higher risk of premature death vs those with quality time several times per week	–3 to –5 years of life expectancy	You rarely have shared meals, walks, or days out. This resembles “< monthly contact + structural isolation” profiles, which show the highest mortality and ~3–5 years shorter life expectancy in midlife adults.
A few times a month	Low–moderate connection	Roughly 10–30% higher risk vs the best-connected group	–1 to –3 years	You do have quality time, but it is irregular or limited. This sits between clear isolation and strong integration: better than high-risk profiles, but still below the protective range.
A few times a week or more	High connection	Used as reference (best) group with lowest risk	0 to +2 years (vs an “average” person; treated as the protective zone)	You enjoy frequent, meaningful contact most weeks—similar to “high social integration” profiles associated with longer life and higher odds of reaching older ages.

*Life-years values are population-level estimates, not promises for any one person. They reflect typical shifts observed in large cohorts when moving between different social-connection profiles.

For the KamaLama engine, a conservative working range is:

From about –4 to –5 years (very low quality time: “once a month or less”),
Up to about +2 to +3 years (frequent quality time: “a few times a week or more”, ideally combined with other positive social factors when available).

Future versions of KamaLama can refine this factor by adding:

Explicit loneliness items,
Living situation (alone vs with others),
Community participation (clubs, volunteering, religious services, group activities),

to avoid double-counting and better capture the full social-connection profile.

5. Claims and evidence summary

Claim	Evidence strength	Reasoning	Key references
Strong social connections reduce all-cause mortality risk by ~25–35%.	Strong (10/10)	Multiple meta-analyses and large cohorts show HRs around 0.70–0.80 (25–35% lower risk) for socially connected vs isolated individuals.	Holt-Lunstad 2015 [1]; Naito 2023 [2]; Berkman & Syme 1979 [6]; Foster 2023 [4]; Holt-Lunstad 2024 [7]
Severe social isolation/loneliness can reduce life expectancy by ~3–5 years in midlife.	Strong (8/10)	Zhao 2025 provides explicit life-years lost at age 45; other cohort data support multi-year gaps by social integration level.	Zhao 2025 [3]; Trudel-Fitzgerald 2020 [5]; Berkman & Syme 1979 [6]
Both structural (e.g., living alone) and functional (loneliness) deficits independently increase mortality.	Strong (9/10)	UK Biobank and other cohorts show additive and independent effects of structural isolation and loneliness on mortality.	Foster 2023 [4]; Holt-Lunstad 2015 [1]; Holt-Lunstad 2024 [7]
The effect size is comparable to some traditional risk factors such as obesity and physical inactivity.	Strong (8/10)	Comparative epidemiological work indicates similar or larger HRs for social isolation vs several metabolic and lifestyle risks.	Holt-Lunstad 2015 [1]; Holt-Lunstad 2024 [7]; Naito 2023 [2]
Social connection also improves healthy life expectancy, not just total years.	Moderate–strong (7/10)	Social participation is associated with more years lived in good health in multiple countries.	Rueda-Salazar 2021 [9]; Trudel-Fitzgerald 2020 [5]; Holt-Lunstad 2024 [7]
Public underestimates the role of social connection in health and longevity.	Moderate (5/10)	Survey data show people rank diet, exercise, smoking above relationships, despite similar effect sizes.	Proctor & Holt-Lunstad 2025 [8]; Holt-Lunstad 2024 [7]

6. Research gaps

Even with strong evidence, key gaps remain:

Area	Current status	Gap
Precise life-year estimates across populations	Good estimates from China (Zhao 2025) and some high-income cohorts.	Need more life-table style estimates across regions, socioeconomic groups, and age bands.
Intervention trials	Evidence is mostly observational; a few small “social prescribing” and group-programme trials exist.	Need large, long-term trials to test whether improving social connection reduces mortality and major cardiovascular events.
Biological mechanisms	Strong observational links with inflammatory markers, immune function, and some ageing biomarkers.	Need more studies linking changes in social connection to changes in biological ageing markers (e.g., epigenetic clocks) over time.
Digital/online connection	Historical cohorts mostly pre-social-media; newer data are mixed.	Need clearer data on when online relationships substitute vs complement in-person contact.
Cultural differences	Data from multiple regions, but coverage is uneven (limited low- and middle-income settings).	Need more studies in under-represented populations and different family/community structures.

For KamaLama, this means:

Treat social connection as a major, modifiable factor, but
Use conservative life-year ranges, and
Keep the model update-ready as new evidence appears.

7. Everyday actions to strengthen social connection

(Educational information only; not medical advice.)

Check your starting point
- Are you currently in the “once a month or less” group?
- Do you often feel lonely, even if you see people?
- Do you have at least one person you can confide in?
Aim to move up one level
- From once a month or less → a few times a month:
  - Schedule two specific social activities next month (dinner, walk, cinema, day out).
- From a few times a month → a few times a week or more:
  - Create one recurring ritual (e.g., weekly walk, Sunday lunch, regular game night).
Combine social and health habits
- Walk or exercise with a friend.
- Join a class, club, or team.
- Cook and eat healthier meals together.
Protect key relationships
- Invest in communication, conflict resolution, and repair.
- High conflict or toxic dynamics can cancel out some benefits of being “socially active”.
Seek help if loneliness is severe or persistent
- Persistent loneliness is linked to depression, anxiety, and higher mortality.
- Consider speaking to a mental health professional, joining support groups, or using community resources.

8. References

Holt-Lunstad, J., Smith, T. B., Baker, M., Harris, T., & Stephenson, D. (2015). Loneliness and social isolation as risk factors for mortality: A meta-analytic review. Perspectives on Psychological Science, 10(2), 227–237. https://doi.org/10.1177/1745691614568352
Naito, R., McKee, M., Leong, D. P., Bangdiwala, S. I., Rangarajan, S., Islam, S., & Yusuf, S. (2023). Social isolation as a risk factor for all-cause mortality: Systematic review and meta-analysis of cohort studies. PLOS ONE, 18(1), e0280308. https://doi.org/10.1371/journal.pone.0280308
Zhao, M., Huo, X., Zhang, H., Wu, C., Peng, S., Liu, Z., Sha, S., Li, M., & Wang, K. (2025). Sex-specific associations of social isolation and loneliness with residual life expectancy at age 45 years among middle-aged and older adults in China. BMC Public Health, 25. https://doi.org/10.1186/s12889-025-23708-x
Foster, H. M. E., Gill, J. M. R., Mair, F. S., Celis-Morales, C. A., Jani, B. D., Nicholl, B. I., Lee, D., & O’Donnell, C. A. (2023). Social connection and mortality in UK Biobank: A prospective cohort analysis. BMC Medicine, 21, 384. https://doi.org/10.1186/s12916-023-03055-7
Trudel-Fitzgerald, C., Zevon, E. S., Kawachi, I., Tucker-Seeley, R. D., Grodstein, F., & Kubzansky, L. D. (2020). The prospective association of social integration with life span and exceptional longevity in women. The Journals of Gerontology: Series B, 75(10), 2132–2141. https://doi.org/10.1093/geronb/gbz116
Berkman, L. F., & Syme, S. L. (1979). Social networks, host resistance, and mortality: A nine-year follow-up study of Alameda County residents. American Journal of Epidemiology, 109(2), 186–204. https://doi.org/10.1093/oxfordjournals.aje.a112674
Holt-Lunstad, J. (2024). Social connection as a critical factor for mental and physical health: Evidence, trends, challenges, and future implications. World Psychiatry, 23(3), 312–332. https://doi.org/10.1002/wps.21224
Proctor, A., & Holt-Lunstad, J. (2025). Blind spots in health perception: The underestimated role of social connection for health outcomes. BMC Public Health, 25.
Rueda-Salazar, S., Spijker, J., Devolder, D., & Albala, C. (2021). The contribution of social participation to differences in life expectancy and healthy years among the older population: A comparison between Chile, Costa Rica and Spain. PLOS ONE, 16(3), e0248179. https://doi.org/10.1371/journal.pone.0248179
Boen, C. E., Barrow, D. A., Bensen, J. T., Farnan, L., Gerstel, A., Hendrix, L. H., & Yang, Y. C. (2018). Social relationships, inflammation, and cancer survival. Cancer Epidemiology, Biomarkers & Prevention, 27(5), 541–549. https://doi.org/10.1158/1055-9965.EPI-17-0836
Gan, D., Baylin, A., Peterson, K. E., Rosero-Bixby, L., & Ruiz-Narváez, E. A. (2025). Social connections, leukocyte telomere length, and all-cause mortality in older adults from Costa Rica: The Costa Rican Longevity and Healthy Aging Study (CRELES). Journal of Aging and Health. https://doi.org/10.1177/08982643251313923