Being alone and feeling lonely sound like the same thing. They aren't.
You can live alone, see friends rarely, and feel perfectly content. Or you can sit in a full room and feel completely alone.
A huge new study of more than 277,000 adults suggests this difference matters for health. It may also matter for what you can do about it.
Healthy habits went a long way for people who were socially isolated. For people who felt lonely, they didn't seem to close the gap.
- Loneliness and isolation are common, and they were linked to years of worse health.
- The link was strongest with mental health, not physical disease.
- Exercise, diet and sleep were tied to a smaller gap for isolation, but not for loneliness.
Two different problems
Isolation is about your life
Social isolation is an objective state. It's about how much contact you actually have with other people.
In this study, people counted as isolated if at least two of these applied:
- They lived alone.
- Friends or family visited less than once a month.
- They did no social activity, like a club or class, at least weekly.
Loneliness is about your feelings
Loneliness is subjective. It's the painful sense that your connections aren't enough.
Here, people counted as lonely if they said they often felt lonely and could confide in someone close less than once a month.
The two often overlap, but not always. In this study, most isolated people didn't report loneliness, and most lonely people weren't isolated.
The study: 277,000 people, 14 years
Who took part
The researchers, from Tulane University in the US and Fudan University in China, used the UK Biobank. It's a huge health study of people aged about 40 to 69 when they joined, between 2006 and 2010.
They included 277,489 adults who had none of seven major diseases at the start. Then they followed their hospital and death records for around 14 years.
| Group | Share of people |
|---|---|
| Neither lonely nor isolated | 89.1% |
| Isolated only | 6.8% |
| Lonely only | 3.3% |
| Both lonely and isolated | 0.8% |
Who tended to be lonely or isolated
The lonely and isolated groups were, on average, poorer and lived in more deprived areas. They were also more likely to smoke.
Among people who were both lonely and isolated, 39% had a household income under £18,000, compared with 14% of everyone else. Nearly 1 in 5 smoked, versus fewer than 1 in 10.
People who were both were also more likely to be men: about 59%, compared with 46% in the rest of the group.
What they tracked
The team looked at seven diseases, split into two groups.
Physical: type 2 diabetes, heart disease and stroke, chronic lung disease, dementia and Parkinson's, and cancer.
Mental: depression and anxiety.
Then they estimated how many years, from age 50, people could expect to live free of these diseases.
Here's what that looks like in practice. A 50-year-old woman who was neither lonely nor isolated could expect about 25.5 more years free of all seven diseases.
For a woman who was both lonely and isolated, it was about 22.6 years. For men, the figures were 21.8 and 19.6 years.
What they found
People who were lonely, isolated or both could expect fewer disease-free years than everyone else. The biggest gaps were in mental health.
| Both lonely and isolated, from age 50 | Women | Men |
|---|---|---|
| Fewer years free of all seven diseases | 3.0 years | 2.2 years |
| Fewer years free of physical disease | 2.4 years | 1.7 years |
| Fewer years free of depression and anxiety | 3.7 years | 5.8 years |
That 5.8-year gap for men stands out. Nearly six fewer years before depression or anxiety showed up in hospital records.
There was also an interesting split between the sexes. For mental health, loneliness was more strongly linked in women, while isolation was more strongly linked in men.
The authors suggest men and women may lean on different kinds of support, but that's speculation for now.
Lonely only, or isolated only
Even one of the two was linked to fewer disease-free years.
For example, women who were lonely but not isolated could expect 3.9 fewer years free of depression and anxiety. Men who were isolated but not lonely could expect 3.0 fewer.
The twist: healthy habits helped one, not the other
Next, the team scored everyone on five habits:
- not smoking
- a healthy body weight
- regular physical activity
- a good diet
- enough sleep
Healthier habits went with more disease-free years for everyone. No surprise there.
The real question was different. Did good habits shrink the gap linked to loneliness or isolation?
For isolation, the gap shrank
Among women, isolation was linked to 2.2 fewer years free of depression and anxiety when their habits were poor.
With the healthiest habits, the gap was just 0.6 years, too small to be sure it was real.
For all seven diseases combined, the isolation gap in women with the healthiest habits disappeared entirely.
In men, the isolation gaps shrank too, though the mental health gap didn't fully close.
For loneliness, it didn't
For loneliness, the gap didn't reliably shrink as habits improved.
In women, the mental health gap linked to loneliness was 3.1 years with poor habits and 4.2 years with the healthiest ones. Better habits didn't close it.
Healthy habits went with a much smaller health gap for socially isolated people. For lonely people, they didn't seem to close the gap at all.
Why might that be?
The study can't say for sure. But the authors offer a few ideas.
Isolation may harm health partly through behaviour. People who are cut off might move less, eat worse or smoke more, and healthy habits push back on exactly that.
Loneliness may work through different routes, like chronic stress or inflammation. A walk and a salad don't touch the feeling of having no one to confide in.
The authors point to earlier research finding that genes linked to loneliness overlap with genes linked to depression. They say these possible pathways still need proper study.
One more detail: the gaps between groups narrowed with age. The differences were largest in middle age and smaller in later life.
How much should you trust this?
Early. It's a huge, careful study, but it shows links, not cause and effect.
What makes it convincing
- It followed more than 277,000 people for around 14 years.
- Diseases came from hospital and death records, not just self-reports.
- It adjusted for age, income, education and deprivation.
- Results held when people who got ill or died in the first two years were left out.
- The code is publicly available, and the authors declare no competing interests.
What makes me cautious
- It's observational. Loneliness might lead to illness, or early illness might lead to loneliness.
- Loneliness was measured once, with two questions. Isolation used three.
- Habits were self-reported, and grouped into just three levels.
- Only people admitted to hospital with depression or anxiety were counted, which misses milder cases.
- UK Biobank volunteers are healthier and wealthier than average, and only about 5.5% of those invited joined.
| This study shows | This study does not show |
|---|---|
| Loneliness and isolation were linked to fewer disease-free years | That loneliness or isolation causes disease |
| The strongest link was with depression and anxiety | That healthy habits can cancel out isolation |
| Healthy habits went with a smaller isolation gap | That habits don't matter for lonely people |
| Loneliness and isolation behaved differently | Which support works best for loneliness |
What this means for you
Healthy habits are worth it for everyone. In this study, they went with more disease-free years whether people were lonely or not.
But if you feel lonely, a gym membership probably isn't the whole answer. The feeling itself may need attention.
The NHS page on dealing with loneliness is a good place to start. It stresses that loneliness is common, that it can pass, and that it's nothing to be ashamed of.
It also explains different kinds of loneliness. You might lack a close confidant, lack friends to go out with, or feel alone even in a crowd.
Knowing which one you're dealing with can point to what might help. That could mean one deeper conversation, rather than more social events.
The NHS page suggests a few simple, low-cost steps:
- Keep in touch. Regular chats with friends and family, even a quick message.
- Share your feelings. Talking openly about loneliness can really help.
- Join a group built around something you enjoy, in person or online.
- Reach out to others. Connecting with someone else who might be lonely can help you both.
Notice that the "share your feelings" step lines up with this study. Loneliness here was partly about having no one to confide in.
If loneliness comes with low mood that won't lift, it's worth talking to a GP.
And it's worth checking in on others too, especially people who live alone or have gone quiet.
This TED talk on the longest-running study of happiness makes a similar point about relationships:
What we still don't know
- Which comes first? Does loneliness lead to depression, or the other way round?
- Do changes matter? People's loneliness and isolation shift over time, but this study measured them once.
- What actually helps lonely people? Trials of specific support are needed.
- Why the sex difference? The split between men and women needs checking in other groups.
- Does it hold elsewhere? UK Biobank isn't typical of the whole population.
- Does reducing isolation help? Trials could test whether group activities or regular visits change these outcomes.
My take: connection isn't a lifestyle box to tick
This study reframes something I think we often get wrong.
We tend to lump loneliness in with everything else we should "fix". Sleep more, move more, see people more.
But this suggests loneliness isn't just another habit. Healthy routines went a long way against isolation, and much less against feeling alone.
So the question isn't only "how many people do I see?" It's also "is there someone I can really talk to?"
Published: Nature Communications, 2026-08-20
Study: Observational cohort (UK Biobank), around 14 years of follow-up
Who: 277,489 UK adults aged about 40 to 69 at the start, free of seven major diseases
Funding: Fudan University and the US National Institutes of Health (NHLBI, NIDDK); the authors declare no competing interests
Evidence: Early — large and long, but associations only, with loneliness measured once
Share this:
Blue Headline Briefing
Enjoyed this? The best stuff lands in your inbox first.
We don’t email on a schedule — we email when something is genuinely worth your time. No filler, no daily blasts, just the sharpest picks from Blue Headline delivered only when they matter.
