What the Gallup Loneliness Figures Do Not Cost Out

Gallup’s release on October 24 put a precise number on a familiar pattern: 24% of people surveyed across 142 countries report feeling very or fairly lonely, and the rate is highest among 19-to-29-year-olds, at 27%, against 17% among adults 65 and older. The underlying data come from the Gallup-Meta State of Social Connections survey, roughly 1,000 respondents per country, fielded between June 2022 and February 2023. It is one of the largest and most methodologically transparent loneliness datasets available, and it confirms something the U.S. Surgeon General’s May 2023 advisory and Harvard’s 2021 survey work had already suggested domestically: young adulthood, not old age, is where loneliness now concentrates.

What Gallup’s release does not do, and does not claim to do, is put a dollar figure on any of this. That gap is worth examining directly, because the cost estimates that do circulate in policy discussion are almost always borrowed from adjacent, narrower studies and then generalized well past what those studies support.

The mortality evidence is comparatively strong

Start with what is well established. Holt-Lunstad’s 2015 meta-analysis in Perspectives on Psychological Science found an odds ratio of 1.26 for loneliness and mortality, 1.29 for social isolation, drawn from a large pooled sample and holding after adjustment for baseline health. The American Heart Association’s 2022 scientific statement, synthesizing cardiovascular evidence specifically, put the increased risk of heart attack or death from heart disease at roughly 29%, and stroke risk at 32%. The Surgeon General’s advisory translated this into a comparison that has since become the standard shorthand: mortality risk from disconnection comparable to smoking up to 15 cigarettes a day.

These are real, replicated, biologically plausible effect sizes. They are also mortality and morbidity risk ratios, not costs. Converting a hazard ratio into a health care expenditure figure requires a chain of further assumptions — about which conditions the excess risk manifests as, what those conditions cost to treat in a given system, and how much of the excess is attributable to loneliness specifically versus the isolation, income, and comorbidity that often travel with it. The AHA statement itself is candid about this: it names the absence of intervention evidence as the central research gap in the field, which is a different problem than lacking prevalence data, but a related one. If nobody has run a trial showing that reducing isolation reduces downstream cardiovascular events, there is no reliable basis for costing out what fixing it would save.

Workplace numbers are the most cited and the least generalizable

Cigna’s 2020 workplace survey is probably the most frequently quoted “cost” of loneliness in circulation: lonely workers miss work twice as often due to illness, five times as often due to stress, and 73% of workers aged 18-22 report loneliness. These are striking numbers, and they are genuinely useful for an employer thinking about absenteeism. But they describe one country, one survey instrument, and self-reported absence rather than payroll records or productivity audits. Applying a Cigna-derived absenteeism multiplier to Gallup’s global 27% youth loneliness figure would mean applying a US workplace finding to informal workers, students, and people outside employment altogether, groups the Cigna survey did not sample and the Gallup release does not disaggregate by employment status.

The National Academies’ 2020 report on older adults, by contrast, is explicit that it could not offer a system-wide cost estimate, despite finding that roughly a quarter of adults 65 and older are socially isolated. It calls for routine clinical assessment of isolation partly because the health care system does not currently generate the billing and outcomes data that would let anyone compute the cost of not intervening.

Why the instrument gap matters for costing, not just prevalence

Gallup measures loneliness with a direct self-report question, feeling very, fairly, not much, or not at all lonely. This is different from the UCLA Loneliness Scale used in AARP’s 45-and-older survey, and different again from the structural isolation measures — network size, contact frequency — that underlie much of the mortality literature. A single global “lonely” percentage, however well sampled, cannot be directly plugged into cost models built on a different instrument’s definition of the same word. The BMC Public Health review of the field this year flags exactly this: inconsistent measurement across studies is a structural barrier to comparison, and costing exercises inherit that inconsistency whether or not they acknowledge it.

None of this means the economic burden of youth loneliness is small. A 27% global rate among 19-to-29-year-olds, if the mortality and cardiovascular risk ratios generalize even partially to that age group over a working lifetime, implies a substantial aggregate health and productivity cost. But “implies” is doing real work in that sentence. No source in the current literature runs that calculation with defensible assumptions at global scale, and the pieces that would need to be assembled — age-stratified risk ratios, country-specific health system costs, attributable-fraction estimates that separate loneliness from the poverty and disability that often accompany it — do not yet exist in combination.

What would strengthen this is not another prevalence survey; Gallup and the Surgeon General’s advisory have made the scale of the problem clear enough. It is a costing study that starts from a single consistent loneliness instrument, tracks a cohort longitudinally rather than cross-sectionally, and reports health care utilization and labor market outcomes for the same people whose loneliness was measured, rather than borrowing a mortality risk ratio from one population and a workplace absenteeism rate from another.