Is the ‘loneliness epidemic’ real?

Summary

More or Less Podcast Summary

Episode: “Is the ‘loneliness epidemic’ real?”

Host: Tim Harford (A), BBC Radio 4
Guest Contributor: Tom Coles (B)
Guest Expert: Professor Melody Ding (C), Epidemiologist, University of Sydney
Date: May 23, 2026

---

Overview

This episode delves into whether the often-cited “epidemic of loneliness” is supported by data. With claims from major institutions like the US Surgeon General and the World Health Organization about rising loneliness, host Tim Harford and reporter Tom Coles scrutinize the numbers behind these assertions. They unpack what “epidemic” means, explore global data, and consult expert epidemiologist Professor Melody Ding to clarify confusing trends in loneliness statistics.

---

Key Discussion Points and Insights

1. What Does “Epidemic” Mean in the Context of Loneliness?

[00:00–01:37]

• Tim Harford sets the stage by drawing attention to the use of the word “epidemic.” Traditionally, an epidemic involves a rapidly spreading disease.
• “But what if the thing you’re talking about is not a virus, but a feeling?” (A, 00:23)
• The term “loneliness epidemic” gained traction in the US in the 2010s and persisted through official warnings, such as the US Surgeon General’s 2023 report.

2. Is There Evidence of Rising Loneliness Globally?

[01:37–03:24]

• Tom Coles points out the lack of unambiguous evidence:
  “I haven’t been able to find any graphs like that. It’s a bit of an odd one, to be honest.” (B, 01:50)
• Gallup’s World Poll (2023): 23% globally “felt lonely a lot of the day yesterday.”
• WHO Commission: Estimated global loneliness rate at 16%.
• “Somewhere around one in five people experiencing regular loneliness is potentially in the right ballpark.” (B, 02:49)
• Harford notes the ambiguity and numerous caveats in these numbers.

3. Are Loneliness Rates Increasing Over Time?

[03:24–04:21]

• Comparing older US data:
  • 1990 Gallup survey: 36% of US adults “sometimes or frequently” felt lonely.
  • 1981 poll: 42% felt “sometimes or frequently very lonely.”
• Tim: “[It’s] hardly evidence of a sudden recent outbreak of loneliness, which does raise the question of whether the word epidemic is really the right way of describing what’s going on.” (A, 04:07)

4. Loneliness Varies Through Life Stages

[04:21–05:03]

• Loneliness is not constant: Peaks in adolescence and older age; declines in middle age.
• Summarizing population data without accounting for these cycles can be misleading.

5. High-Quality Scientific Data: What Do We Know?

[05:03–06:11]

• Prof. Melody Ding analyzed US adults 50+ (1996-2018) and found: “We actually didn’t observe an increase. In fact, we saw a small decrease in loneliness in this age group alone.” (C, 05:41)
• European data is “noisy”: Some increases, some decreases, some stable depending on country, measure, and group.

6. Is There an Epidemic in Any Group?

[06:19–07:10]

• In adults: No consistent evidence of an epidemic.
• In adolescents: Yes, a clear and consistent increase in loneliness across “different continents and countries” over the last 20 years. “We are observing consistent increasing trends in young people in the last 20 years.” (C, 06:32)
• “We already know that there’s a worrying general decline in the mental health of young people around the world. And if young people are more anxious and more depressed, it’s not surprising they’re also more lonely.” (B, 06:44)

7. Why Bundle Different Age Groups as One “Epidemic”?

[07:10–07:46]

• The drivers of loneliness differ vastly between teenagers and the elderly.
• Tim: “The causes and associated problems of teenage loneliness must be very different to those for older people.” (A, 06:58)
• Research on teenagers and older adults comes from different datasets, and aggregating them confuses the picture.

8. Should We Call It an ‘Epidemic’?

[07:46–08:36]

• Even scientists use “epidemic” for urgency, but some, like Prof. Ding, prefer “an important public health problem.”
• “This is a term I usually try to stay away from. As a scientist myself, I usually would define loneliness as an important public health problem.” (C, 07:51)
• On the term “social epidemic”:
  "It also pathologized loneliness in a way, which is actually quite a common human experience. So I think social epidemic also kind of creates that negative connotation and make us feel like loneliness is a disease. Which it is not.” (C, 08:11)

---

Notable Quotes and Memorable Moments

• “I haven’t been able to find any graphs like that. It’s a bit of an odd one, to be honest.”
  — Tom Coles, on evidence for a loneliness “epidemic” ([01:50])
• “Somewhere around one in five people experiencing regular loneliness is potentially in the right ballpark.”
  — Tom Coles, summarizing global data ([02:49])
• “We actually didn’t observe an increase. In fact, we saw a small decrease in loneliness in this age group alone.”
  — Prof. Melody Ding, on older US adults ([05:41])
• “We are observing consistent increasing trends in young people in the last 20 years, and that’s been quite consistent across different continents and countries.”
  — Prof. Melody Ding ([06:32])
• “This is a term I usually try to stay away from. As a scientist myself, I usually would define loneliness as an important public health problem.”
  — Prof. Melody Ding ([07:51])
• “It also pathologized loneliness in a way, which is actually quite a common human experience.”
  — Prof. Melody Ding, challenging the “epidemic” framing ([08:11])

---

Key Timestamps

• 00:00–01:36 – Introduction; the meaning of “epidemic”
• 01:37–03:24 – Evidence (or lack thereof) for rising loneliness globally
• 03:24–04:21 – Contradictory historical survey data
• 04:21–05:03 – Life-stage variation in loneliness
• 05:03–06:11 – Prof. Ding on US and European data
• 06:19–07:10 – Clear rise in youth loneliness
• 07:10–07:46 – Differing causes of loneliness by age
• 07:46–08:36 – Language and framing: “epidemic” vs. public health problem

---

Conclusion

• There is no strong evidence of a sudden or significant increase in loneliness among the general adult population worldwide or in the US and Europe.
• There is clear evidence of a rising trend in loneliness among young people, consistent across continents.
• The “epidemic” framing may oversimplify and mislead, especially when aggregating trends across very different age groups.
• Experts urge caution against pathologizing loneliness, emphasizing instead that it is a complex, multifaceted public health issue needing nuanced understanding and solutions.

---

For statistics skepticism and public health nuance, this episode provides a gold-standard example of the importance of context in interpreting social data.

Transcript

[A]

Hello and thanks for downloading the More or Less podcast with a program that looks at the numbers in the news and in life and in mysterious epidemics. I'm Tim Harford. What does it mean if you say that something's an epidemic? In the case of a virus, it usually means that it's spreading rapidly and that more and more people are getting infected. When a disease isn't on the rise, but is there in a population at a reasonably steady level, we tend to say that the disease is endemic. But what if the thing you're talking about is not a virus, but a feeling? In 2023, the US Surgeon General launched a report called Our Epidemic of Loneliness and Isolation, warning of the health harms of being lonely and socially isolated. The idea that there's an epidemic of loneliness didn't start there. The term was already in use in the US in the 2010s, and it's a phrase that's still going strong, popping up in news stories on a regular basis. After that warning from the US Surgeon General, the World Health Organization launched the Commission on Social Connection with their Director General, warning that more and more people are finding themselves isolated and lonely. But is it true that loneliness rates are increasing? Is it right to say we're in the midst of an epidemic of loneliness? Tom Coles has been looking into this one. Hello, Tom.

[B]

Hi, Tim.

[A]

So what is the evidence that we've got a loneliness epidemic on our hands? You know, the kind of thing sharply rising line on a graph. Clear long term trend of significant increase. Evidence that more and more people are lonely.

[B]

Well, okay, look, I haven't been able to find any graphs like that. Oh, yeah, it's a bit of an odd one, to be honest. I mean, there are some graphs showing an upward trend and others showing a slight downward trend, depending on where in the world you're looking and which groups you're looking at.

[A]

It doesn't sound like an epidemic, but just run me through what you've worked out.

[B]

Okay, so, definitions first. Loneliness is a subjective experience, an unpleasant feeling of being isolated or alone. One way to understand what proportion of the pop is experiencing that unpleasant feeling is to ask them. And I want to be clear. Surveys show that hundreds of millions of people describe themselves as regularly feeling lonely. The numbers slightly depend on how you ask the question and who you're asking. But for example, polling company Gallup's World Poll found that in 2023, 23% of people globally said they felt lonely a lot of the day Yesterday, all the

[A]

lonely people, where do they all come from?

[B]

Well, the highest rates were in Africa, though it's possible that people might have been interpreting that as time spent on their own physically, rather than experiencing the feeling of loneliness. The World Health Organisation Commission, you mentioned earlier, also analyzed the global data and suggested that the global rate of loneliness was more like 16%. Again, these are using different methods and measuring all kinds of slightly different things, but I think somewhere around one in five people experiencing reg loneliness is potentially in the right ballpark.

[A]

Well, that's quite a run of caveats on that one, Tom. So that's very roughly how many people might be regularly feeling lonely at the moment. Can we see a long term trend? Is that number higher than in the past?

[B]

I'm not entirely sure.

[A]

Oh, yes.

[B]

So if we look at the US, when I asked Gallup what polling they had pre 2000, they said that a 1990 survey found that 36% of US adults reported sometimes or frequently feeling lonely. In 1981, a different poll with different wording found that 42% of respondents said they were sometimes or frequently very lonely.

[A]

So those two data points aren't straightforward to compare, but it's hardly evidence of a sudden recent outbreak of loneliness, which does raise the question of whether the word is epidemic is really the right way of describing what's going on.

[B]

No, the thing with loneliness is that it changes quite a lot through people's lives. Adolescents seem to experience a lot of loneliness and then this declines as people get jobs and partners and families and a network of friends and all that, then often rises again towards the end of people's lives as they stop working, their physical health gets worse and their social circles shrink. So if you just take a snapshot of the whole population, you're bundling people at all those ages together. You might just be measuring a change in demographics rather than a change in loneliness per se.

[A]

This sounds like a case for social science and there must be some better data out there based on established systems of measurements and some good statistics.

[B]

Yeah, there is. But this evidence is also a very mixed bag. And if we're talking about adults, there's no evidence of an epidemic. Professor Melody Ding is an epidemiologist from the University of Sydney. She's done loads of studies on this subject, one of which analysed loneliness in older adults in the US using more rigorously controlled survey data.

[C]

So we use the data called Health and Retirement Study. It's from the US, particularly on the middle age and older population, generally 50 plus. And looking at this particular group from 1996 to 2018. We actually didn't observe an increase. In fact, we saw a small decrease in loneliness in this age group alone.

[B]

Melody has also looked at the data from Europe.

[C]

The picture is really quite noisy. It really depends on the measure, depends on the population, depends on the country. Some studies suggest increase, others suggest stable, and others suggest decrease. So that's why I hesitate to come to any conclusions when it comes to adults.

[A]

Okay, so the good quality scientific research finds no evidence of anything you'd call an epidemic in adults.

[B]

No, but remember the pattern that we talked about earlier, with two peaks, one in adolescence and one in older age? As far as Melody is concerned, there is a clear trend in the younger group.

[C]

We are observing consistent increasing trends in young people in the last 20 years, and that's been quite consistent across different continents and countries.

[B]

That's not perhaps all that surprising. We already know that there's a worrying general decline in the mental health of young people around the world. And if young people are more anxious and more depressed, it's not surprising they're also more lonely.

[A]

It sounds like this is one of those stats where two stories are really being forced into one to the detriment of both. The causes and associated problems of teenage loneliness must be very different to those for older people.

[B]

Absolutely. For older people, loneliness can be a consequence of retirement, chronic illness, or friends and family dying. Lonely teenagers aren't usually lonely. For those reasons, even the data sets are disconnected. The studies Melody is analyzing on teenagers don't join up with the data sets on older people.

[A]

So if you see a loneliness epidemic being used to talk about the population as a whole, that's not really a thing. If you see it used for young people in isolation, it might be, but it's likely to cause confusion. So why do even scientists in the field use the phrase?

[B]

I'm not totally sure. Melody tries to avoid it, although that can be hard in her field.

[C]

This is a term I usually try to stay away from. As a scientist myself, I usually would define loneliness as an important public health problem.

[B]

The term social epidemic is sometimes used by scientists to describe loneliness, to draw a distinction with an infectious disease. But even then, Melody isn't sure it's a good strategy.

[C]

So I can see the point of using the word epidemic because we really want to call for action, and then that really highlights the urgency. But it also pathologized loneliness in a way, which is actually quite a common human experience. So I think social epidemic also kind of creates that negative connotation and make us feel like loneliness is a disease. Which it is not.

[A]

Thank you, Tom. And thanks to Melody. Ding. That's it for today. If you've seen a number in the news you think we should take a look at, email more or lessbc.co.uk. we'll be back next week and until then, goodbye.