Men experiencing both loneliness and social isolation were estimated to have 5.8 fewer years free of depression and anxiety from age 50 than men experiencing neither, according to a Nature Communications study of 277,489 UK Biobank adults by Xuan Wang and colleagues. Published on 20 August 2026, the research followed people initially free of seven major chronic diseases through hospital and death records for a median of nearly 14 years.
The lifestyle findings highlight the difference between having few social connections and feeling lonely. Healthier habits were linked to a smaller gap in years lived without disease for socially isolated people. For lonely people, the study found no clear statistical evidence that this gap became smaller.
Two words that aren’t the same
The researchers counted someone as socially isolated if at least two of three things were true: they lived alone, saw friends or family less than once a month, and took part in no weekly social activity. Loneliness came from two questions, “Do you often feel lonely?” and “How often are you able to confide in someone close to you?” Participants counted as lonely only if they often felt lonely and could confide in someone less than once a month.
One measure describes the structure of a social life; the other describes the experience of it. A person with few contacts may feel well supported, while someone surrounded by people may have no one to speak to openly. Calling loneliness subjective does not make it imaginary or chosen. It names a mismatch between the connection someone has and the connection they need, which a full address book cannot settle.
What the estimated years measure
“Mental disorder” here means depression and anxiety only. The 5.8-year figure (95% confidence interval 4.5 to 7.1 years) is a modelled difference between groups, calculated from age 50, not a countdown for any individual man. The gap in years free of physical disease was smaller, at 1.7 years for men. For women, the figures were 3.7 years free of depression and anxiety and 2.4 years free of physical disease.
The sexes also differed. Among women, loneliness alone was linked to a larger loss of years free of depression and anxiety than isolation alone, 3.9 years against 2.0; among men, the two were almost equal, at 2.9 and 3.0 years. The authors suggest, as a possibility rather than a finding, that men may depend more on practical support and women on emotional support.
The lifestyle split
The lifestyle score gave one point each for healthy levels of smoking, body mass index, physical activity, diet and sleep, with 0 to 2 counting as unhealthy, 3 as intermediate, and 4 or 5 as healthy. Among women, the estimated gaps in years free of depression and anxiety looked like this:
| Condition | Unhealthy lifestyle | Intermediate lifestyle | Healthy lifestyle |
|---|---|---|---|
| Social isolation | 2.2 years | 1.1 years | 0.6 years |
| Loneliness | 3.1 years | 2.6 years | 4.2 years |
The isolation gap narrowed until, in the healthiest group, its confidence interval included zero. The loneliness gap showed no statistically significant narrowing.
The 4.2 figure does not mean good habits make loneliness worse. Healthier habits were linked to more years free of depression and anxiety for lonely and non-lonely women alike; what stayed open was the distance between the two groups. The table compares different people at different lifestyle levels, not the same people before and after a change. And because so few participants were both lonely and isolated, each condition was analysed separately here, adjusting for the other, so the table does not speak directly to the 5.8 years.
Arendt’s three kinds of being alone
In the concluding chapter of The Origins of Totalitarianism, Hannah Arendt distinguished isolation, solitude and loneliness. She was examining political domination, not measuring health, so her categories illuminate the study’s split without matching its questionnaire scores.
Isolation, in her political vocabulary, meant lacking others with whom to act. Solitude allowed a person to be alone while remaining in dialogue with themselves, the condition she associated with thinking. Drawing on Epictetus, she distinguished that self-company from abandonment.
Loneliness reached further. It meant feeling deserted by human companionship and losing a secure place in a shared world, and it could be felt among other people.
Her account makes the important difference a matter of quality. Counting contacts cannot capture whether someone feels recognised, can trust others, or experiences their life as belonging to a common world. The study’s separate measures leave room for that distinction, even though they cannot test Arendt’s philosophy.
Why habits might reach isolation differently
The study’s authors point to a 2018 study by Lindsay Kobayashi and Andrew Steptoe as one possible explanation. Following 3,392 English adults aged 52 and over for ten years, it found that social isolation was associated with several health behaviours, including less consistent physical activity and more smoking. Loneliness showed no comparable pattern after adjustment, although lonely smokers were less likely to quit.
That makes a behavioural pathway plausible. Social ties can provide routines, practical help and encouragement around meals or activity, and healthier habits might make up for some of what a sparse network fails to provide. Loneliness may run along other routes. The authors suggest psychological stress and inflammation in the brain as candidates that still need study.
This remains a hypothesis. A lifestyle score may capture the routines that company provides more readily than the experience of having nobody who understands you.
What the study cannot establish
The study is observational, so it cannot show that loneliness or isolation caused the lost years, or that changing habits would recover them; the authors say plainly that they cannot draw causal conclusions. Loneliness and isolation were assessed once, through brief questions, and UK Biobank’s volunteers, recruited at a 5.5% response rate, are not representative of the population. Only 0.8% of participants met both classifications, and a statistically nonsignificant result is not proof of zero effect. A 2013 study of 6,500 adults aged 52 and over found that isolation, rather than loneliness, remained linked to mortality after adjustment. Which of the two looks worse depends partly on whether researchers measure death, physical disease or years free of depression and anxiety.
Alone at a full table
Think of someone who is surrounded by people but still has nobody they can talk to openly. That person may feel lonely without being socially isolated. The study found no clear statistical evidence that healthier habits narrowed the gap in disease-free years linked to loneliness.
For Arendt, loneliness meant feeling that you had no place among others, even when people were around. The research cannot tell us how to give someone that sense of belonging. But it highlights a simple difference: having company and feeling understood are two different things, and it is easy to imagine more contact leaving the second untouched.