People who report feeling lonely are at higher risk of developing dementia in most cohort studies, but the evidence is observational and cannot show that loneliness causes dementia. A 2024 Nature Mental Health meta-analysis of longitudinal data from 608561 individuals found a hazard ratio of 1.306 (95% CI 1.197-1.426) for all-cause dementia, and the link held after accounting for depression and social isolation.
What the studies found
The strongest single source is a 2024 meta-analysis in Nature Mental Health that pooled longitudinal data from 608561 individuals. Loneliness was linked to a 31% higher risk of all-cause dementia (HR 1.306, 95% CI 1.197-1.426). The risk was higher for Alzheimer's disease (HR 1.393, 95% CI 1.290-1.504, 5 samples) and vascular dementia (HR 1.735, 95% CI 1.483-2.029, 3 samples), and there was a smaller but significant increase in general cognitive impairment (HR 1.150, 95% CI 1.113-1.189, 16 samples). These associations held after statistically controlling for depression and social isolation, which suggests loneliness has an effect distinct from simply lacking social contact.
- Study
- A Meta-analysis of Loneliness and Risk of Dementia using Longitudinal Data from >600,000 Individuals
Being alone and feeling alone are different measurements, and the studies often separate them. In a 2022 Neurology analysis of 462,619 UK Biobank participants (mean age at baseline 57.0 years), social isolation was associated with a 1.26-fold increased risk of dementia (95% CI 1.15-1.37) independently of various risk factors including loneliness and depression. In the Amsterdam Study of the Elderly, which followed 2173 non-demented community-living older people for 3 years, feelings of loneliness were associated with incident dementia (OR 1.64, 95% CI 1.05-2.56) but social isolation, defined as living alone, being unmarried or lacking social support, was not associated with a higher dementia risk in the multivariate analysis.
- Studies
- Associations of Social Isolation and Loneliness With Later Dementia
- Feelings of loneliness, but not social isolation, predict dementia onset: results from the Amsterdam Study of the Elderly (AMSTEL)
Earlier cohorts found the same pattern. In a Chicago-area cohort, 823 older persons free of dementia at enrollment were followed for up to 4 years, and 76 developed clinical Alzheimer's disease. The risk of AD was more than doubled in lonely persons (score 3.2, 90th percentile) compared with persons who were not lonely (score 1.4, 10th percentile), and controlling for indicators of social isolation did not affect the finding. In the Health and Retirement Study (N = 12,030), cognitive status was assessed at baseline and every 2 years over a 10-year follow-up, and loneliness was associated with a 40% increased risk of dementia.
Social contact, rather than the feeling of loneliness, was studied in the Whitehall II cohort of 10,228 people with 28 years of follow-up. More frequent social contact at age 60 was associated with lower dementia risk (hazard ratio for each SD higher social contact frequency = 0.88, 95% CI 0.79-0.98). The effect sizes for social contact at 50 or 70 years were similar (0.92 [95% CI 0.83-1.02] and 0.91 [95% CI 0.78-1.06]) but not statistically significant.
The timing and persistence of loneliness appear to matter. In the Framingham Heart Study (n = 2880), persistent mid-life loneliness was associated with a higher risk of dementia onset (HR 1.91, 95% CI 1.25-2.90) and transient loneliness with a lower risk (HR 0.34, 95% CI 0.14-0.84), compared with no loneliness. In the HUNT cohort (9389 participants), persistent loneliness from midlife into older age, as well as becoming lonely, was associated with increased odds of dementia, whereas transient loneliness in midlife was not. The Rotterdam Study (4,514 participants followed up to 14 years) and the Swedish SNAC-K cohort (2,112 participants) found that loneliness, not social support, was associated with cognitive decline and dementia, and tested whether depressive symptoms confound the association.
- Studies
- Associations of loneliness with risk of Alzheimer's disease dementia in the Framingham Heart Study
- Loneliness trajectories and dementia risk: Insights from the HUNT cohort study
- Loneliness, Not Social Support, Is Associated with Cognitive Decline and Dementia Across Two Longitudinal Population-Based Cohorts
A UK Biobank analysis of 155 070 participants (mean age 64.1 years) added genetic risk. Overall, 8.6% of participants reported being socially isolated and 5.5% were lonely, and during a mean follow-up of 8.8 years (1.36 million person years), 1444 (0.9% of the total sample) were diagnosed with dementia. The study asked whether the risk of dementia is modified or confounded by genetic risk, using the polygenic risk score for Alzheimer's disease.
Could dementia be causing the loneliness?
The direction of the link is not settled. In a 2016 JAMA Psychiatry study of 79 cognitively normal older adults (mean age 76.4 years), higher cortical amyloid burden was significantly associated with greater loneliness after controlling for age, sex, APOEε4, socioeconomic status, depression, anxiety and social network. Those in the amyloid-positive group were 7.5-fold (95% CI 1.7-fold to 34.0-fold) more likely to be classified as lonely than nonlonely. The authors describe this as a novel association of loneliness with cortical amyloid burden.
- Study
- Association of Higher Cortical Amyloid Burden With Loneliness in Cognitively Normal Older Adults
Loneliness is also common among the people who look after someone with dementia. A 2024 meta-analysis of 27 research papers involving 11,134 informal carers from 17 different countries found a pooled prevalence of loneliness of 50.8% (95% CI 41.8-59.8%) and of social isolation of 37.1% (95% CI 26.7-47.6%). Its limitations include very high heterogeneity between studies, different instruments and cut-offs, and cross-sectional data that cannot show whether caregiving causes loneliness.
What remains uncertain
Depression is the main complication. In the HUNT cohort the odds ratio for persistent loneliness was 1.47 (95% CI 1.10-1.95) in the fully adjusted model, but it attenuated to 1.28 (95% CI 0.95, 1.72) when depression was adjusted for, and the 95% CI then included no effect. The 2024 meta-analysis reports that associations held after controlling for depression, so the studies do not agree on how much of the link depression explains.
- Studies
- Loneliness trajectories and dementia risk: Insights from the HUNT cohort study
- A Meta-analysis of Loneliness and Risk of Dementia using Longitudinal Data from >600,000 Individuals
The pooled analysis states that loneliness is self-reported and subjective, which can introduce measurement variability, and that as an observational meta-analysis it cannot establish that reducing loneliness would causally lower dementia risk. The Whitehall II limitations include dementia from electronic records that may miss milder cases, a predominantly white civil-servant sample and self-reported social contact. The UK Biobank isolation study notes that volunteers are healthier and less isolated than the population and that isolation was measured once at baseline.
- Studies
- A Meta-analysis of Loneliness and Risk of Dementia using Longitudinal Data from >600,000 Individuals
- Association of social contact with dementia and cognition: 28-year follow-up of the Whitehall II cohort study
- Associations of Social Isolation and Loneliness With Later Dementia
Reducing loneliness has not been shown to prevent dementia in the records in this library. A 2020 review of why loneliness interventions are unsuccessful concluded that current interventions fail because they do not treat loneliness as a heterogeneous construct and are not targeted at individuals' needs and contexts.
This page describes what published studies in alzhub's library report. It is not medical advice and does not apply to any one person. Where a decision would follow, a clinician who knows the person is the one who can say what a result or a choice means for them.
Questions
- Is loneliness the same as social isolation?
- No. In the Amsterdam Study of the Elderly, feelings of loneliness were associated with incident dementia (OR 1.64) but social isolation, defined as living alone, being unmarried or lacking social support, was not in the multivariate analysis. A UK Biobank analysis of 462,619 people found social isolation was associated with a 1.26-fold increased risk of dementia independently of loneliness and depression.
- How much does loneliness raise dementia risk?
- A 2024 meta-analysis of longitudinal data from 608561 individuals found a hazard ratio of 1.306 (95% CI 1.197-1.426) for all-cause dementia, about 31% higher, and 1.393 for Alzheimer's disease. Individual cohorts range from an OR of 1.64 in the Amsterdam study to a 40% increased risk in the Health and Retirement Study.
- Will treating loneliness prevent dementia?
- That has not been shown in these studies. The meta-analysis states that it cannot establish that reducing loneliness would causally lower dementia risk, and a review of loneliness interventions concluded that current interventions are not targeted at individuals' needs and contexts.
- Do family carers of people with dementia feel lonely?
- Often. A 2024 meta-analysis of 11,134 informal carers found a pooled prevalence of loneliness of 50.8% (95% CI 41.8-59.8%) and of social isolation of 37.1% (95% CI 26.7-47.6%), though the data are cross-sectional and cannot show whether caregiving causes loneliness.
Read on
- Depression and dementia: the studies
- Caregiver support: the studies
- Exercise and dementia: the studies
- Does air pollution increase dementia risk?
- All answer pages
Related questions
Last updated