The Missing Cost Estimates in Loneliness Research

The World Health Organization’s Commission on Social Connection reported on 30 June 2025 that loneliness is linked to an estimated 871,000 deaths annually worldwide — roughly 100 an hour. That figure travels well. It is precise, alarming, and easy to repeat. What does not travel with it is any comparably rigorous estimate of what loneliness costs in economic terms — health system spending, lost productivity, or reduced GDP. The mortality evidence in this field is unusually strong. The cost evidence is not, and the gap between the two is worth stating plainly rather than papering over.

What the mortality evidence actually supports

Julianne Holt-Lunstad’s 2010 meta-analysis, covering 148 studies and more than 308,000 participants, found that stronger social relationships were associated with a 50% increased likelihood of survival — an effect size the authors compared to established risk factors like smoking cessation. Her 2015 follow-up put the mortality risk more precisely: social isolation carried an odds ratio of 1.29, loneliness 1.26, living alone 1.32, with the effects holding after adjustment for baseline health. The American Heart Association’s 2022 scientific statement, led by Crystal Cene, translated this into cardiovascular terms: roughly 29% increased risk of heart attack or death from heart disease, 32% increased risk of stroke. The 2023 Surgeon General’s advisory compared the mortality risk of disconnection to smoking up to 15 cigarettes a day.

These are causal-sounding numbers built on a genuinely large evidence base, even though most of the underlying studies are observational rather than experimental. The AHA statement is candid about this: it identifies the absence of intervention evidence as the central research gap. That candor is the relevant precedent for what follows, because the same caveat applies with far more force to economic figures, and it is far less often stated.

Where a dollar figure would have to come from

No source reviewed here produces a rigorous, bottom-up estimate of the economic cost of loneliness comparable to what exists for, say, smoking-attributable healthcare spending. The WHO report quantifies deaths, not dollars. The Surgeon General’s advisory frames loneliness as a public health priority and outlines a six-pillar national strategy, but its numbers are prevalence and mortality-risk figures, not cost accounting. The CDC’s 2024 surveillance report, using 2022 data, links loneliness to heart disease, stroke, dementia, type 2 diabetes, depression, and anxiety — a list of cost-generating conditions — but stops at the epidemiological association. Turning “loneliness raises the odds of stroke” into “loneliness costs the health system $X billion” requires attributable-fraction modeling: isolating the share of stroke incidence attributable to loneliness specifically, net of confounding, and then costing the resulting care. None of the sources reviewed here do this work, and the Center is not aware of one that does at a standard comparable to the mortality meta-analyses.

This matters because cost estimates are what move budgets. A mortality odds ratio persuades a researcher. A number with a dollar sign in front of it persuades a finance ministry. The absence of that number is not a footnote; it is arguably the reason loneliness policy has moved slower than the health evidence would justify.

The intervention literature makes the gap worse, not better

Even if the attributable cost of loneliness were established, translating that into a case for intervention spending requires knowing what interventions cost and what they return. Here the evidence is thinner still. Two 2021 systematic reviews of social prescribing — one in the International Journal of Environmental Research and Public Health, one in Perspectives in Public Health — report improvements in self-esteem, confidence, and in three of nine studies, reduced use of GP, emergency, and inpatient services. That reduced-utilization finding is the closest thing to a cost signal in this literature, and it comes from three studies out of nine, in a review that itself flags heterogeneity across programs as a limitation.

The 2024 HEAL-HOA trial in The Lancet Healthy Longevity is notable mainly for what it is: one of the only randomized controlled trials of a loneliness intervention, testing volunteering against a control among lonely older adults in Hong Kong. A 2025 medRxiv protocol for a systematic review of social prescribing in older adults states outright that only one peer-reviewed randomized controlled trial exists in this specific area, despite growing adoption of social prescribing programs across health systems. Governments are spending on interventions whose cost-effectiveness has essentially not been measured by trial-grade evidence. The National Academies’ 2020 consensus report calls on health systems to routinely screen for isolation among older adults, a recommendation made on the strength of the mortality and morbidity evidence, not on any demonstrated return on the screening itself.

What would actually close the gap

A credible cost estimate would need three things this literature currently lacks in combination: a population attributable fraction for loneliness across specific conditions, drawn from longitudinal rather than cross-sectional cohorts; standardized costing of the resulting care across health systems; and randomized trial data on interventions, of which HEAL-HOA is nearly alone. Until those exist, the honest position is that the mortality case for treating loneliness as a public health priority is strong, and the economic case, however plausible it sounds, is currently an inference resting on numbers that were never designed to support it.