What '61% of Americans Are Lonely' Actually Measured

Cigna’s 2020 workplace report put a number on American loneliness that has since traveled well beyond its original context: 61%. It appears in policy testimony, op-eds, and conference slides, almost always stripped of its source population and its instrument. It is worth asking what that figure actually measured, and what happens when it is set beside the other headline figures circulating in the same conversation.

The population was not “Americans”

The 61% comes from Cigna’s Loneliness and the Workplace report, and the sample was U.S. workers, not the general adult population. The report also found that 73% of workers aged 18 to 22 reported loneliness, and over 80% of employed Gen Z workers did. These are striking numbers, but they describe an employed, largely working-age cohort surveyed through a corporate wellness lens, with loneliness tied explicitly to workplace outcomes: Cigna reports lonely workers missing work twice as often due to illness and five times as often due to stress. That is a specific claim about a specific population and its productivity costs. It is not a claim about the loneliness of children, retirees, or the unemployed, groups the survey did not cover.

Compare this with the AARP Foundation’s 2018 survey of adults 45 and older, which found one in three reporting loneliness. That survey used the 20-item UCLA Loneliness Scale, a validated academic instrument with decades of use, rather than a bespoke corporate questionnaire. The AARP figure is lower than Cigna’s, in an older population that might be expected, on some assumptions, to show more isolation. The two numbers are not measuring the same thing, and the gap between them is not evidence that one is wrong. It is evidence that “percentage of people who are lonely” is not a single, stable quantity waiting to be discovered. It depends entirely on who was asked, what instrument they were given, and what threshold counted as “lonely.”

Isolation and loneliness keep getting collapsed into each other

A parallel confusion runs through the National Academies’ 2020 consensus report, which states that roughly one quarter of adults 65 and older are socially isolated. That is a different measurement from loneliness entirely. Isolation, as the National Academies use the term, describes an objective property of a person’s network: how many social ties they have, how often those ties are activated, whether they live alone. Loneliness is the subjective experience of feeling that one’s social needs are unmet, regardless of how many people are actually present.

Cacioppo’s 2008 account of loneliness makes the distinction sharply: a person can be surrounded by family and still register the aversive signal of loneliness, while another living alone by choice may not. The two measures are correlated but not interchangeable, and Holt-Lunstad’s 2015 meta-analysis in Perspectives on Psychological Science found that social isolation, loneliness, and living alone each independently predicted early mortality, with odds ratios of 1.29, 1.26, and 1.32 respectively. If Cigna’s 61% figure is being cited to argue that a quarter of the population is isolated in the National Academies’ sense, that is a category error, however common it has become in secondary reporting.

Instrument choice moves the number more than the underlying reality does

The Harvard Making Caring Common survey, published in February 2021, found 36% of American adults reporting serious loneliness, and 61% among adults aged 18 to 25 specifically. That last figure happens to match Cigna’s overall workforce number almost exactly, but the two describe different populations measured at different times with different questions, and the resemblance is coincidental rather than corroborating. The Harvard report also found that about half of lonely young adults said no one had taken more than a few minutes recently to ask, in a way that felt genuine, how they were doing. That is a texture of finding that a single prevalence percentage cannot carry: it says something about the quality of available contact, not just its quantity.

None of this means the underlying phenomenon these surveys are circling is not real or not serious. The AARP survey’s finding that 61% of people who have never spoken with a neighbor report loneliness, against 33% of those who have, points to something concrete and structural about proximity and contact. But headline percentages travel faster than their methodology, and by the time a figure like Cigna’s 61% reaches a policy brief or a news segment, the population, the instrument, and the definition attached to it have usually fallen away.

What would settle this

A defensible national loneliness estimate would use one validated instrument, such as the UCLA scale, applied to a probability sample of the full adult population, repeated on a fixed schedule so that year-over-year change is measurable rather than an artifact of a different survey house asking a different question. It would also report isolation and loneliness as separate figures rather than folding them into one number. Until something like that exists for the United States, on the model the UK government adopted when it began embedding loneliness measurement into the Office for National Statistics in 2018, comparisons across these surveys should be treated as illustrative rather than as evidence of trend.