Loneliness and social isolation and mortality risk

A meta-analysis of 70 studies shows that loneliness, social isolation, and living alone are associated with a higher risk of premature death.

Loneliness and social isolation and mortality risk

Table of contents

    Why social relationships matter for health

    Loneliness and social isolation are often treated as emotional or quality-of-life issues. This meta-analysis shows something stronger: social deficits are associated with a measurable increase in mortality risk.

    The authors distinguish three related but separate concepts:

    • Social isolation — objectively having few social contacts, relationships, or social participation.
    • Loneliness — the subjective feeling of being disconnected or lacking meaningful relationships.
    • Living alone — a simple structural indicator that may, but does not have to, reflect isolation.

    This distinction matters. A person can be surrounded by others and still feel lonely. Conversely, someone can live alone and feel socially fulfilled. From a health perspective, both the objective and subjective dimensions appear to matter.

    Study details

    • Publication title: Loneliness and Social Isolation as Risk Factors for Mortality: A Meta-Analytic Review
    • Authors: Julianne Holt-Lunstad, Timothy B. Smith, Mark Baker, Tyler Harris, David Stephenson
    • Year of publication: 2015
    • Journal: Perspectives on Psychological Science
    • Identifiers: DOI: 10.1177/1745691614568352
    • Full-text source: SAGE Journals
    • Study type: meta-analysis of prospective observational studies
    • Population: 70 independent studies, 3,407,134 participants; mean age at baseline ~66 years
    • Follow-up: average of 7.1 years
    • Exposure: loneliness, social isolation, or living alone
    • Primary outcome: mortality during follow-up
    • Funding: grants from Brigham Young University
    • Conflicts of interest: none reported

    Key findings

    After adjusting for multiple confounders, the authors found a significant increase in mortality risk associated with social deficits:

    • Social isolation: OR = 1.29 (~29% higher risk of death)
    • Loneliness: OR = 1.26 (~26% higher risk)
    • Living alone: OR = 1.32 (~32% higher risk)

    Effect sizes were smaller in fully adjusted models compared to unadjusted analyses but remained statistically significant. This strengthens the conclusion that the association is not solely driven by poorer baseline health among isolated individuals.

    Loneliness and isolation are not the same

    One of the key insights is that loneliness and social isolation should not be treated as interchangeable.

    Social isolation reflects the external structure of relationships: how many contacts a person has and how often they engage socially. Loneliness reflects the internal experience: whether those relationships meet emotional needs.

    The meta-analysis found no clear difference in mortality risk between these dimensions. This suggests that both should be assessed independently.

    Importantly:

    • Increasing social contact does not automatically reduce loneliness.
    • Addressing loneliness without improving actual social connections may also be insufficient.

    Effective interventions likely need to target both dimensions simultaneously.

    Age and mortality risk

    Contrary to common assumptions, the association between social deficits and mortality was not strongest in the oldest populations.

    The effect was actually stronger in studies with participants younger than 65 years. In fully adjusted analyses:

    • <65 years: OR = 1.57
    • 65–75 years: OR = 1.25
    • >75 years: OR = 1.14

    Possible explanations include differences in lifestyle, health behaviors, social expectations, and survivorship bias. The findings suggest that loneliness and isolation in midlife may be particularly important from a longevity perspective.

    Potential biological and behavioral mechanisms

    Loneliness and social isolation likely affect health through multiple pathways.

    Behaviorally, socially disconnected individuals may be more likely to:

    • have poorer sleep,
    • engage in less physical activity,
    • smoke more frequently,
    • show lower adherence to medical care.

    Biologically, these factors are associated with:

    • higher blood pressure,
    • elevated inflammatory markers (e.g., CRP),
    • unfavorable lipid profiles,
    • impaired immune function,
    • increased chronic stress load.

    This does not prove causality, but it supports the idea that social relationships are part of a broader regulatory system influencing long-term health.

    Limitations

    This is a meta-analysis of observational studies, so causality cannot be definitively established.

    The authors attempted to address reverse causality by focusing on prospective studies and using fully adjusted models. Still, residual confounding cannot be excluded.

    Another limitation is that most studies assessed only one dimension (loneliness, isolation, or living alone), making direct comparisons difficult.

    Additionally, “living alone” is an imperfect proxy and does not capture the complexity of social relationships.

    What this means for longevity

    From a longevity perspective, social relationships should not be seen as optional or secondary. They may represent a core pillar of health, alongside sleep, nutrition, and physical activity.

    The key takeaway is simple: loneliness and social isolation are not just psychological states—they are measurable risk factors.

    Maintaining meaningful relationships, staying socially engaged, and fostering a sense of belonging may play a direct role in long-term health and survival.

    Sources