What the data actually shows
The broadest recent figure comes from a Meta-Gallup global survey, the State of Social Connections (2023), which found that roughly one in four people worldwide aged 15 and older — about 24% — reported feeling very or fairly lonely. Notably, the data did not show loneliness rising neatly with age; younger adults frequently reported feeling lonely at least as often as, and sometimes more than, older adults.
Public-health bodies have treated this as serious. In 2023 the U.S. Surgeon General, Vivek Murthy, issued an advisory titled 'Our Epidemic of Loneliness and Isolation,' framing weak social connection as a population-level health issue rather than a purely personal one. The advisory drew heavily on research linking social connection to physical health outcomes.
The most-cited evidence on those outcomes is a 2010 meta-analysis by Holt-Lunstad and colleagues in PLoS Medicine, which pooled many studies and found that stronger social relationships were associated with substantially lower mortality risk — an effect the authors compared in magnitude to well-established risks like smoking. This is an association drawn from observational data, so it should be read as a strong correlation rather than simple proof of cause, but it is among the more robust findings in the field.
Why this feels different from how it actually is
Loneliness tends to feel like a private, individual failing, which is part of what keeps it hidden. Because people rarely admit to it, each lonely person can assume they are unusually isolated while everyone around them appears connected. The survey data tells a different story: at roughly one in four adults, loneliness is a majority-adjacent experience that simply does not get said out loud.
It also runs against the intuition that loneliness is mainly an older-person's problem. Several datasets, including the Meta-Gallup survey, find young adults reporting high levels of loneliness — which can make a young person's loneliness feel especially anomalous and shameful, when in fact it is statistically common for their age group.
And loneliness is decoupled from how much company you actually have. Because it is a subjective gap between the connection you want and the connection you feel, it does not track neatly with the number of people around you. That is why someone with a full calendar can feel deeply lonely, and why 'just see more people' so often misses the point — the feeling is about perceived closeness, not headcount.
What the research says to do about it
The research points more toward the quality and felt closeness of connections than their sheer number. Because loneliness is the gap between desired and perceived connection, interventions that deepen existing relationships — more honest, more regular contact with people who already matter — tend to be better supported than simply increasing social volume.
Some of the better-evidenced approaches target the thinking patterns around connection rather than only the behaviour. Reviews of loneliness interventions have suggested that addressing the negative expectations lonely people often hold — the anticipation of rejection or of being a burden — can be as important as creating opportunities to socialise, because those expectations otherwise undermine the contact that does happen.
At a population level, the Surgeon General's advisory frames connection as something supported by structure, not willpower alone — community spaces, routines, and institutions that create repeated, low-effort contact. For an individual, the practical translation is favouring recurring, built-in social contact over one-off effort, since consistency is what the evidence repeatedly rewards.
What the research says does not help
Treating loneliness as a simple headcount problem — more parties, more contacts, more activity — often does not resolve it, because the feeling is about perceived closeness rather than quantity. People can add social events and still feel just as lonely if those events do not produce a sense of being known.
Self-blame tends to make it worse. Because the data shows loneliness is common across ages and regions, framing it as a personal defect is both inaccurate and counterproductive; the negative self-narrative is one of the patterns that research suggests actually sustains loneliness over time.
Leaning harder on passive, scrolling-style social media use is, at best, an unreliable substitute. The evidence here is genuinely mixed and contested, but the consistent thread is that passively consuming others' curated lives does little to close the gap that defines loneliness, and may widen the sense that everyone else is more connected than you.
Real numbers in context
The headline figure to hold onto is that roughly 24% — about one in four adults worldwide aged 15+ — reported feeling very or fairly lonely (Meta-Gallup, 2023). That makes loneliness common rather than exceptional, and notably it is not concentrated only among the old; young adults often report it at least as much. The U.S. Surgeon General's 2023 advisory treated this as a public-health concern at the population level.
On the health side, the Holt-Lunstad et al. (2010) meta-analysis found weak social connection associated with markedly higher mortality risk, an effect often likened to smoking — though this is an association from observational studies, not proof of cause. It is also worth flagging that loneliness measurement varies between surveys: widely quoted figures from sources such as the Cigna U.S. Loneliness Index (2018 and 2021) have produced high numbers, but the methods and definitions differ, so exact percentages are debated even where the overall direction is clear.