Older Adults Report Less Loneliness Than the Young — What the Numbers Actually Show

Most public discussion of loneliness treats old age as its natural home. The image is a widow living alone, a retiree with no visitors. Gallup’s 2023 global survey, covering roughly 142 countries and about 1,000 respondents aged 15 and older in each, found the opposite age pattern for loneliness itself: adults 65 and older reported the lowest rate of loneliness of any age group, at 17%. Young adults aged 19 to 29 reported the highest, at 27%. That is not a rounding artifact. It is one of the more consistent findings in the recent survey literature, and it sits uneasily next to decades of aging-and-isolation policy language.

The finding, stated plainly

Gallup’s figure comes from the State of Social Connections survey, fielded with Meta between June 2022 and February 2023, covering roughly 3.2 billion people in the 142 countries surveyed (the sample excludes China, so it covers about 77% of the world’s adult population). The follow-up analysis, “How Strong Are the World’s Social Connections?”, breaks the same data down by age and confirms the gradient: loneliness declines fairly steadily from young adulthood into old age. Harvard’s Making Caring Common project found a parallel pattern in the United States in 2021: 61% of adults aged 18-25 reported serious loneliness, against much lower rates among older cohorts. Gallup’s later 2024 survey, reporting that just over one in five adults worldwide feel lonely a lot of the previous day, again found the pattern skewed toward younger respondents.

This is not a new observation once one looks past the aging-and-isolation framing. The AARP Foundation’s 2018 survey of adults 45 and older, using the 20-item UCLA Loneliness Scale rather than a bespoke instrument, found that a third of that population is lonely — a substantial figure, but the survey’s own age breakdowns showed loneliness trending downward within the 45-plus range as respondents got older, not upward.

Why the policy narrative points the other way

The disconnect is not really a disagreement about loneliness. It is a conflation of loneliness with isolation, two constructs that the literature treats as related but distinct. Isolation describes an objective property of a person’s network — how many people they see, how often, how large their household is. Loneliness is the subjective gap between the connection a person wants and the connection they have. The National Academies’ 2020 consensus report on older adults reported that roughly a quarter of Americans 65 and older are socially isolated by structural measures: living alone, having few social ties, limited contact with family. That figure is genuinely alarming and genuinely concentrated in old age, for reasons that are mechanical rather than psychological — retirement removes a daily social structure, widowhood removes a primary companion, mobility and health limits reduce the radius of contact.

But isolation and loneliness do not move together in a simple way. A person can be structurally isolated and not feel lonely, because expectations for contact have adjusted, or because the relationships that remain are experienced as sufficient. A 2024 study in Scientific Reports examined this interplay directly and found that the relationship between isolation and loneliness varies by age rather than staying constant — younger people report more loneliness relative to their objective isolation, older people less. Japan’s JACSIS study, a genuinely longitudinal dataset covering 2020 to 2021 rather than a single cross-sectional snapshot, found isolation and loneliness diverging over the pandemic in ways a one-time survey would have missed entirely.

Holt-Lunstad’s 2015 meta-analysis in Perspectives on Psychological Science is worth returning to here, because it is often cited as evidence that loneliness is an older-adult problem when it says something more specific. The odds ratios for loneliness and isolation as mortality risk factors were 1.26 and 1.29 respectively, but the paper reported that social deficits were more predictive of death in samples averaging under 65, not over. That is a mortality-risk finding, not a prevalence finding, and the two get run together in secondary coverage more often than they should.

What this does not mean

None of this argues that loneliness among older adults is unimportant. The 2023 Surgeon General’s advisory put loneliness at roughly half of U.S. adults across all ages and drew its mortality comparison — risk comparable to smoking up to fifteen cigarettes a day — from a pooled literature that includes older cohorts prominently, because the health consequences of loneliness compound with the health vulnerabilities that already concentrate in old age. A 65-year-old who is lonely and has a chronic condition is a different clinical case than a 22-year-old who is lonely and healthy, even if the 22-year-old is statistically more likely to report the feeling. The National Academies report’s call for the health care system to screen for isolation is a response to that compounding risk, not a claim about prevalence rank.

The more useful reading is that youth loneliness and elder isolation are two separate problems that happen to share a word. Treating them as one continuous phenomenon, worse the older a person gets, obscures both. It understates the acute subjective loneliness reported by young adults — a pattern AARP’s own data and Gallup’s cross-national data agree on — and it risks treating older adults’ objective isolation as though it should manifest as felt loneliness, when a meaningful share of isolated older adults do not report feeling that way at all.

A study designed to separate these cleanly would need to measure network structure and subjective loneliness in the same respondents over time, across the full adult age range, using one consistent instrument rather than comparing surveys that were never built to be compared. Little of the current literature does that. Most of what exists measures one construct well and infers the other.