Relational Health in the Modern Age
Relationships are not a soft topic. They show up in the data.
What the research actually says about connection and health — including the parts that are still uncertain.
How we cite. Every number on this page is quoted from a peer-reviewed study or an official public health body, with a DOI or official link in the references. Where a popular claim does not hold up, we say so.
Connection shows up in how long people live.
The largest review of this question pooled 148 studies following more than 300,000 people for an average of seven and a half years.
50%
increased likelihood of survival for people with stronger social relationships
Odds ratio 1.50 (95% CI 1.42–1.59). Holt-Lunstad, Smith & Layton, 2010 — 148 studies, 308,849 participants, average 7.5-year follow-up. [1]
Being alone and feeling alone are two different things.
Researchers separate social isolation — how much contact a person actually has — from loneliness, which is how a person feels about their connections. Someone can be surrounded by people and still feel lonely. Someone else can live quietly and feel well connected.
This matters more than it might seem. The two do not always behave the same way in the data, and much of the popular coverage blurs them together. We keep them apart on this page.
The links to physical health are measurable.
Each of these comes from a peer-reviewed study. Each measures something slightly different, so we have kept them separate rather than adding them together.
Increased likelihood of mortality
After controlling for other risk factors, pooled across 70 prospective studies: living alone 32%, social isolation 29%, loneliness 26%. Holt-Lunstad, Smith, Baker, Harris & Stephenson, 2015. [2]
+29%
risk of coronary heart disease
Pooled relative risk 1.29 (95% CI 1.04–1.59). Valtorta et al., 2016. [3]
4.2×
more likely to catch a cold
Healthy adults given a cold virus under quarantine, fewest versus most social ties. Cohen et al., 1997. [4]
2.3–2.8×
higher nine-year mortality, most isolated vs most connected
2.3 for men, 2.8 for women. Berkman & Syme, 1979, 6,928 adults. [5]
A caveat we keep in plain view: both coronary and stroke confidence intervals nearly touch 1.0 (lower bound 1.04). The link is real, but the precision is modest. Presenting 29% and 32% as hard numbers would overstate what the study supports.
Connection and mental health run together.
2.33× the odds of developing depression — adults who often feel lonely versus those who rarely do. Adjusted odds ratio 2.33 (95% CI 1.62–3.34). Mann et al., 2022. [6]
+31% risk of dementia — hazard ratio 1.31 (95% CI 1.20–1.43), pooled across 21 samples and 608,561 people. Luchetti et al., 2024. [7]
72% lower odds of depression in adulthood — for adults who had 6–7 positive childhood experiences versus 0–2. Odds ratio 0.28 (95% CI 0.21–0.39). Bethell et al., 2019. [11]
Where the evidence gets complicated — and we think that is worth showing.
The largest single study on this question followed 462,619 adults for nearly twelve years. It found that social isolation predicted dementia (hazard ratio 1.26), but loneliness did not once depression was accounted for (hazard ratio 1.04, not statistically significant). Roughly 75% of the loneliness link ran through depressive symptoms. [8]
On anxiety, the strongest review found a positive association but produced no pooled figure at all. Anyone quoting a percentage for loneliness and anxiety is quoting something the evidence does not yet contain.
It starts early, and it never stops mattering.
Hard childhoods leave marks in adult health. So do good ones.
4–12× increased health risks in adulthood — for people reporting four or more categories of adverse childhood experience versus none, for alcoholism, drug use, depression, and suicide attempt. Felitti et al., 1998, 9,508 adults. [10]
24% of Americans 65 and older are socially isolated — among community-dwelling older adults. National Academies of Sciences, Engineering, and Medicine, 2020. [13]
National Scientific Council on the Developing Child, Harvard University, Working Paper 13 (2015). A consensus statement from an expert council, not a measured percentage. [12]
Connection is human, so this is a global story.
In 2023 the World Health Organization created a Commission on Social Connection. Its first full report landed in June 2025.
1 in 6 people worldwide affected by loneliness. 871,000 deaths a year linked to loneliness — a modelled WHO estimate; deaths linked to loneliness, not deaths counted as caused by it. World Health Organization, 2025. [15]
Who actually feels lonely, by age
Share reporting they feel "very or fairly lonely," across 142 countries and territories: ages 19–29 27%, ages 65+ 17%. Meta–Gallup, 2023. [16]
Loneliness by country income: low-income countries 24%, high-income countries 11%. World Health Organization Commission on Social Connection, 2025. [15]
This turns a common story upside down.
We tend to picture loneliness as something that happens to older people. In global data it is highest among young adults and lowest among people 65 and older. Isolation and loneliness are measured differently and move differently, which is exactly why we keep the two words apart.
Care looks different from place to place.
Support is not offered the same way everywhere. One review found that people in some cultures are far less likely to ask openly for help, and benefit more from support that does not require spelling out the problem. A 2025 study of 202,898 people across 22 countries found a shared pattern in what builds strong adult relationships — and real variation between countries in how strongly each factor mattered. [17] [18]
We are careful here. The big mortality reviews report no clear difference across world regions, but their data come overwhelmingly from North America and Western Europe. It is honest to say the link looks similar wherever it has been tested. It is not honest to say it has been proven the same in every culture.
Where this science is still thin.
A research institute should be as clear about what it does not know as what it does. These are the honest gaps.
Is loneliness "as bad as smoking 15 cigarettes a day"?
Not as it is usually told. The original comparison was about lacking social connection measured broadly — not loneliness on its own — and referred to smoking up to 15 cigarettes daily. The researcher behind it has published a clarification, because the claim keeps growing in the retelling. We do not repeat it as a fact.
Does this research prove that relationships cause better health?
No, and we will not say it does. Nearly all of this work is observational. It shows that connection and health travel together. It does not prove one causes the other. That honesty is the point.
Doesn't loneliness kill?
In two large, well-regarded studies, being isolated predicted worse outcomes while feeling lonely did not, once other factors were accounted for. Isolation and loneliness are not the same thing, and any page that says "loneliness kills" is skipping over that.
Is the evidence the same everywhere in the world?
We cannot claim that. Most participants in the major reviews are from North America and Western Europe. Africa, Latin America, and much of Asia are thinly represented. The link looks similar wherever it has been tested, but "tested everywhere" is not the same as "true everywhere."
What about loneliness and anxiety?
The strongest review of loneliness and new mental health problems produced a solid figure for depression and none for anxiety. If you see a specific percentage for loneliness and anxiety, it is not coming from the best available evidence.
These gaps are where our grants go.
Upside offers small research grants to scholars and professionals studying the connection between relationships and health. We are especially interested in the thin spots above — work outside North America and Western Europe, studies that separate isolation from loneliness, and research that tests whether strengthening relationships actually changes health outcomes.
References
- Holt-Lunstad, J., Smith, T. B., & Layton, J. B. (2010). Social relationships and mortality risk: A meta-analytic review. PLoS Medicine, 7(7), e1000316. doi:10.1371/journal.pmed.1000316
- Holt-Lunstad, J., Smith, T. B., Baker, M., Harris, T., & Stephenson, D. (2015). Loneliness and social isolation as risk factors for mortality. Perspectives on Psychological Science, 10(2), 227–237. doi:10.1177/1745691614568352
- Valtorta, N. K., Kanaan, M., Gilbody, S., Ronzi, S., & Hanratty, B. (2016). Loneliness and social isolation as risk factors for coronary heart disease and stroke. Heart, 102(13), 1009–1016. doi:10.1136/heartjnl-2015-308790
- Cohen, S., Doyle, W. J., Skoner, D. P., Rabin, B. S., & Gwaltney, J. M. (1997). Social ties and susceptibility to the common cold. JAMA, 277(24), 1940–1944. doi:10.1001/jama.1997.03540480040036
- Berkman, L. F., & Syme, S. L. (1979). Social networks, host resistance, and mortality. American Journal of Epidemiology, 109(2), 186–204. doi:10.1093/oxfordjournals.aje.a112674
- Mann, F., Wang, J., Pearce, E., et al. (2022). Loneliness and the onset of new mental health problems in the general population. Social Psychiatry and Psychiatric Epidemiology, 57(11), 2161–2178. doi:10.1007/s00127-022-02261-7
- Luchetti, M., Aschwanden, D., Sesker, A. A., et al. (2024). A meta-analysis of loneliness and risk of dementia. Nature Mental Health. doi:10.1038/s44220-024-00328-9
- Shen, C., Rolls, E. T., Cheng, W., et al. (2022). Associations of social isolation and loneliness with later dementia. Neurology, 99(2), e164–e175. doi:10.1212/WNL.0000000000200583
- Steptoe, A., Shankar, A., Demakakos, P., & Wardle, J. (2013). Social isolation, loneliness, and all-cause mortality in older men and women. PNAS, 110(15), 5797–5801. doi:10.1073/pnas.1219686110
- Felitti, V. J., Anda, R. F., Nordenberg, D., et al. (1998). Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. American Journal of Preventive Medicine, 14(4), 245–258. doi:10.1016/S0749-3797(98)00017-8
- Bethell, C., Jones, J., Gombojav, N., Linkenbach, J., & Sege, R. (2019). Positive childhood experiences and adult mental and relational health. JAMA Pediatrics, 173(11), e193007. doi:10.1001/jamapediatrics.2019.3007
- National Scientific Council on the Developing Child (2015). Supportive relationships and active skill-building strengthen the foundations of resilience (Working Paper 13). Center on the Developing Child, Harvard University.
- National Academies of Sciences, Engineering, and Medicine (2020). Social isolation and loneliness in older adults. National Academies Press. doi:10.17226/25663
- Office of the Surgeon General (2023). Our epidemic of loneliness and isolation. U.S. Department of Health and Human Services.
- World Health Organization (2025). From loneliness to social connection: Charting a path to healthier societies. Report of the WHO Commission on Social Connection. Geneva: WHO.
- Meta & Gallup (2023). The global state of social connections.
- Kim, H. S., Sherman, D. K., & Taylor, S. E. (2008). Culture and social support. American Psychologist, 63(6), 518–526. PMID: 18793039
- Wilkinson, R., Shiba, K., Gibson, C. B., et al. (2025). Life course insights into social relationship quality: A cross-national analysis of 22 countries. Scientific Reports, 15, 12096. doi:10.1038/s41598-025-86246-x