Loneliness is a measurable health risk, not just a mood: the landmark 2015 meta-analysis by Julianne Holt-Lunstad and colleagues in Perspectives on Psychological Science found that loneliness was associated with a 26 percent increase in the likelihood of early death, and social isolation with 29 percent — effect sizes comparable to well-established behavioral risk factors such as smoking up to 15 cigarettes a day or obesity in midlife. The findings have reshaped how public-health bodies treat social connection.
This article publishes information, not medical advice. It explains what loneliness is, what the research links it to, why women report it more, and which interventions have evidence. If loneliness has slid into persistent low mood, hopelessness, or loss of function, a clinician can help — it may be depression, which is treatable.
What is the difference between loneliness and isolation?
Social isolation is an objective condition: few contacts, little contact. Loneliness is subjective — a felt gap between the connection you have and the connection you want. The two overlap but separate cleanly in research: a person can be isolated and content, or surrounded by people and lonely, and the 2015 meta-analysis found both carried elevated mortality risk independently.
The distinction matters for solutions. Isolation responds to opportunities — transportation, clubs, safer neighborhoods. Loneliness responds to relationship quality — which is harder to legislate and easier to mistake for a personality flaw.
What does the evidence link loneliness to?
The strongest associations, all observational and drawn from large cohorts and meta-analyses, run through cardiometabolic territory. Per the Centers for Disease Control and Prevention, loneliness and social isolation are associated with roughly a 29 percent higher risk of heart disease and a 32 percent higher risk of stroke. Proposed mechanisms include chronically elevated stress hormones, higher blood pressure, worse sleep, and increased inflammation — with higher rates of smoking and inactivity traveling alongside.
The National Academies of Sciences, Engineering, and Medicine, in its 2020 report on adults 50 and older, catalogued further associations: higher rates of depression, dementia risk, and worse self-rated health among isolated older adults. Two calibrations apply throughout. First, association is not proof of causation — lonely people differ in many ways, and reverse causation (poor health shrinking the social world) is real. Second, the effect sizes, while meaningful, are averages across wildly different studies.
Who is most affected?
By self-report, women consistently score higher on loneliness scales in most Western surveys, though older men are more often objectively isolated. The pattern by life stage is less intuitive than expected: per large national surveys, loneliness peaks not in old age but in young adulthood and again in the 40s and 50s — the years many women report the specific texture of loneliness that arrives while socially surrounded, sometimes called being lonely in a crowd.
Specific groups carry elevated risk, per the National Academies' 2020 report: adults with chronic illness or disability, caregivers, immigrants, LGBTQ adults, people with low income, and the recently bereaved or divorced. Motherhood deserves particular mention — the research on postpartum and caregiving loneliness shows that being constantly with a small child and being socially connected are different things, and mothers describe loneliness as one of the least-discussed parts of the job.
Related stories: Strength Training and Long-Term Health in Women: What the Cohorts Show · The Mental Load: What Research Actually Says About the Work Behind the Work.
Do lonely people just need to get out more?
Not exactly, and the intervention literature explains why. A 2020 umbrella review of loneliness interventions in Perspectives on Psychological Science and earlier meta-analyses found that simply providing social contact produces weak, inconsistent effects. What works better targets the internal side: cognitive approaches that address the hypervigilant social thinking loneliness generates — expecting rejection, reading neutral faces as disapproving — produce moderate, more reliable reductions in loneliness.
| Intervention type | Evidence status |
|---|---|
| CBT-based programs targeting social thinking patterns | Moderate evidence; most consistent effect on loneliness itself |
| Social prescribing (referral to community activities) | Promising, largely UK-based observational and early trial data; US evidence young |
| Group activities and clubs alone | Weak, inconsistent effects on measured loneliness |
| Technology-based contact for isolated older adults | Mixed; some benefit for video contact, less for passive use |
| Volunteering | Observational associations with reduced loneliness; trials limited |
Public-health framing has shifted accordingly. The US Surgeon General issued a formal advisory on loneliness and isolation in May 2023, framing social connection as a health priority — recognition of the research above rather than a new finding.
Is there a screening test for loneliness?
There are validated measures — the UCLA Loneliness Scale and the three-question campaign tool used in primary care settings — but no professional body currently recommends routine loneliness screening for all patients; the evidence that screening plus intervention changes health outcomes has not been established, per the USPSTF's framework for evaluating such services. Practically: you can self-assess, and answering honestly — I feel left out, I feel isolated, I lack companionship — is itself useful information to bring to a clinician.
When to talk to a clinician
Bring loneliness to a healthcare visit when it is persistent — most days for weeks — or when it arrives with low mood, loss of pleasure, sleep change, increased drinking, or thoughts that life is not worth living; that last one is a reason to seek help urgently, the same day. A clinician can distinguish loneliness from depression and anxiety, both of which both mimic and amplify it, and can point to evidence-based psychological therapy rather than another activity list. Caregivers and the recently bereaved should also raise it proactively — both groups are known high-risk and rarely asked.
Frequently asked questions
Is loneliness as harmful as smoking 15 cigarettes a day? The comparison comes from Holt-Lunstad's 2015 meta-analysis and refers to effect sizes, not biology: the 26 percent mortality increase associated with loneliness was comparable in magnitude to risks attributed to heavy smoking or obesity in the same analysis. It is a communication shorthand, not a claim that loneliness works like tobacco in the lungs — and it is an association, not proof of causation.
Does social media make loneliness worse? The research is mixed rather than dramatic. How you use it matters more than how much, per reviews cited by the National Academies: passive scrolling correlates with worse mood in some studies, while active contact with existing close ties can supplement real connection. No high-quality trial shows social media reliably increases or decreases loneliness across populations — honest summaries say the evidence is unsettled.
Can loneliness be treated like a medical problem? Partially. There is no pill, but psychological therapies targeting social cognition have moderate trial evidence, per the 2020 umbrella review, and treating co-occurring depression or anxiety often loosens loneliness's grip. Public-health tools — transport, community programs, social prescribing — address the isolation side. The combination of clinical care and structural connection is where the evidence currently points.