CoChat
Sign Up
Cited by 351Dementia and Cognitive Impairment ResearchHealth disparities and outcomesAging and Gerontology Research

Does loneliness contribute to mild cognitive impairment and dementia? A systematic review and meta-analysis of longitudinal studies

Open access
Lara, Elvira, Martín-María, Natalia, De la Torre-Luque, Alejandro, Koyanagi, Ai, Vancampfort, Davy, Izquierdo, Ana, Miret, Marta·2019·Ageing Research Reviews
DOI 10.1016/j.arr.2019.03.002
Open this paper in CoChat — free accountOpen PDF

Abstract

There is growing evidence that loneliness is associated with mild cognitive impairment (MCI) and dementia. However, the extent of this association remains unclear. A systematic review and meta-analysis of longitudinal studies examining this association was conducted. Six electronic databases were searched from inception to November 15th 2018. A random-effects meta-analysis was performed to obtain pooled estimates and 95% CIs. Studies were also assessed for heterogeneity, methodological quality and publication bias. A total of 4270 hits were retrieved based on the initial search strategy and ten studies met the eligibility criteria involving 37339 individuals (mean age from 64.9 to 83.1 years). Variation between studies was present for the measurement of loneliness as well as for the case ascertainment of MCI and dementia. Loneliness was positively associated with increased risk of dementia (overall RR = 1.26; 95% CI = 1.14, 1.40; n = 8). Due to lack of sufficient data, we could not explore the association between loneliness and risk of MCI through a meta-analysis, but limited evidence suggests a potential effect of loneliness on MCI. A further understanding of the deleterious effects of loneliness on MCI and dementia may assist the design of environmental and psychological interventions to prevent or delay the onset of these neuropsychiatric conditions.

AI snapshot

Written from the paper’s full text.

TL;DR

This systematic review and meta-analysis found that loneliness was associated with increased risk of incident dementia in longitudinal community-based studies, while evidence for mild cognitive impairment (MCI) was limited. The authors conclude that more research is needed, especially with validated loneliness measures and comprehensive MCI/dementia ascertainment.

Key findings

  • In the meta-analysis of 8 independent studies (n=33,555), lonely individuals or those reporting greater loneliness had increased dementia risk: overall RR=1.26 (95% CI=1.14, 1.40), with moderate heterogeneity (I²=23%; 95% CI=0, 65; Cochran Q χ²(7)=9.10, p=0.25).
  • MCI could not be meta-analyzed because only two longitudinal studies examined it; one did not provide effect sizes. In Lobo et al. (2008), 208 MCI cases occurred at follow-up (12.6%), and loneliness was associated with increased MCI risk (OR=2.05; 95% CI=1.31, 4.97).
  • The only study examining MCI and dementia as a combined outcome found no association: HR=1.02 (95% CI=0.66, 1.57) in 1,601 cognitively healthy individuals, with 163 developing MCI-dementia.
  • Sensitivity analyses for dementia yielded pooled RRs from 1.22 to 1.29 across all-cause dementia, validated loneliness questionnaires, follow-up >5 years, depression adjustment, and moderate risk of bias.

Methods

  • Systematic review and meta-analysis following PRISMA-P, registered in PROSPERO (CRD42018102623). Six databases were searched from inception to November 15, 2018 (CINAHL, Cochrane Library, PsycINFO, PubMed, Scopus, Web of Science), plus manual searching.
  • Inclusion criteria included middle-aged and older adults (50+ years) from general community-based populations, loneliness as exposure, incident MCI or dementia, cognitively healthy at baseline, longitudinal design lasting at least one year, quantitative data, and peer-reviewed articles. No time or language restrictions were imposed.
  • The search yielded 4,272 hits; after screening, 11 studies from 10 articles were included in the qualitative synthesis, and 8 studies were included in the dementia meta-analysis. Random-effects inverse-variance meta-analysis was used, with relative risk as the effect size; adjusted estimates were prioritized, and heterogeneity was assessed with Cochran Q and I².

Limitations

  • The paper states that the meta-analysis was limited by the small number of included studies, insufficient literature on loneliness and MCI/dementia, exclusion of grey literature, inability to retrieve needed data from one study (Wilson et al., 2015), possible selective reporting, and no accounting for persistence of loneliness over time.
  • The authors note heterogeneity in MCI/dementia ascertainment, potential misclassification, attribution bias from deaths during follow-up, possible confounding by unobserved variables such as cognitive reserve, and that causality cannot be inferred from longitudinal observational studies; reverse causation also cannot be ruled out.
  • Statistical limitations stated include poor precision of I² with few studies, inability to conduct subgroup or meta-regression analyses, and no statistical testing for funnel-plot asymmetry because fewer than 10 studies were included.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1016/j.arr.2019.03.002

Cites

83 of 83 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

0 of 25 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    There is an important gap in the literature on the longitudinal association between loneliness and subsequent MCI or dementia.

    “A small number of longitudinal observational studies that explored the association between loneliness and subsequent MCI or dementia was identified, demonstrating the existence of an important gap in the literature.”
    • Also stated in: § A C C E P T E D · plain
      “Additionally, no study has summarised the potential association between loneliness and incident MCI to date.”
    Quote in the paperSubstantial claim
  • Findingcitation

    In Lobo et al. (2008), loneliness was significantly associated with an increased risk of MCI (OR=2.05, 95% CI 1.31-4.97).

    “In Lobo and colleagues´ study (2008), 208 MCI cases were identified at follow up (12.6%).A significant effect of loneliness (OR=2.05;95% CI=1.31, 4.97) for increasing risk of MCI was found (Table 1).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Lobo et al., 2008

      Not checked in the free run.

  • Findingcitation

    In Rawtaer et al. (2017), feelings of loneliness were not associated with subsequent MCI-dementia (HR=1.02, 95% CI 0.66-1.57).

    “Feelings of loneliness were not associated with subsequent MCI-dementia (HR=1.02;95% CI=0.66, 1.57).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Rawtaer et al., 2017 source

      Not checked in the free run.

  • Findingown data

    The association between loneliness and dementia was stronger in studies with longer follow-ups.

    “While we found a stronger association between loneliness and dementia with longer follow-ups”
    Quote in the paperNot a substantial claim
  • Findingreasoning

    Incidence rates were consistent among studies.

    “incidence rates were consistent among studies”
    Quote in the paperSubstantial claim
  • May combine two claims. 6
  • The quote only partly states this claim. 1
  • Classified as probably not a substantial claim. 22

  • May combine two claims.

    Findingown data

    The association between loneliness and dementia was robust in sensitivity analyses: the effect did not differ substantially by exposure measurement, follow-up duration, depression adjustment, or risk of bias (RR=1.23 to 1.29).

    “Several sensitivity analyses further revealed the robustness of the main findings: the effect of loneliness on dementia risk did not seem to be largely affected by the exposure measurement, duration of follow-up, depression adjustment or risk of bias (23% to 29% increased risk).”
    Quote in the paperSubstantial claim

  • The quote only partly states this claim.

    Findingown data

    Loneliness (or greater feelings of loneliness) is associated with an increased risk of incident dementia in community-dwelling middle-aged and older adults (pooled RR=1.26, 95% CI 1.14-1.40).

    “Loneliness was positively associated with increased risk of dementia based on a meta-analysis.”
    • Also stated in: § 3.3.2. Loneliness and the risk for dementia · plain
      “Lonely individuals or individuals reporting greater feelings of loneliness had an increased risk of dementia (overall RR=1.26; 95% CI=1.14, 1.40).”
    • Also stated in: § 4.1. Summary of results and comparison with previous work · plain
      “Our study results showed that loneliness was associated with an increased risk of incident dementia based on a meta-analysis.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingreasoning

    Eleven studies from ten articles were included in the qualitative synthesis.

    “Thus, eleven studies derived from ten articles were included in the qualitative synthesis (Chen et al., 2011;He et al., 2000;Holwerda et al., 2014;Lobo et al., 2008;Rafnsson et al., 2017;Rawtaer et al., 2017;Sutin et al., 2018;Wilson et al., 2015;Wilson et al., 2007;Zhou et al., 2018a).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The level of heterogeneity in the dementia meta-analysis was moderate (I2=23%, 95% CI 0-65) and the Cochran Q test was not significant (χ2(7)=9.10, p=0.25).

    “The level of heterogeneity was moderate with wide CIs (I 2 =23%; 95% CI=0, 65) and the Cochran Q test was not significant (χ2(7)=9.10,p=0.25).”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Findingreasoning

    Eight studies were included in the meta-analyses.

    “Of them, eight studies were included in the meta-analyses (Chen et al., 2011;He et al., 2000;Holwerda et al., 2014;Lobo et al., 2008;Rafnsson et al., 2017;Sutin et al., 2018;Wilson et al., 2007;Zhou et al., 2018a).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingnone given

    Most included studies were recent population-based studies.

    “Still, most studies were recent population-based studies”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Major methodological flaws were found for exposure assessment (4/10), length of follow-up (7/10), and attrition (4/10).

    “Major methodological flaws were found for the exposure assessment (4/10), the length of follow-up for the outcome to occur (7/10) and attrition (i.e.follow-up rate or description of individuals that were lost to follow-up) (4/10).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    The study protocol was registered at PROSPERO (CRD42018102623).

    “The protocol of this study has been previously registered at the International Prospective Register of Systematic Reviews (PROSPERO) (registration number: CRD42018102623).”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Boss et al. (2015) systematic review concluded that greater loneliness was associated with poorer global cognition and declines in specific cognitive domains including processing speed, immediate recall, and delayed recall.

    “Boss and colleagues (2015) conducted a systematic review on loneliness and cognitive function in the ageing population and concluded that greater loneliness was associated with poorer global cognition and declines in specific cognitive domains including processing speed, immediate recall, and delayed recall.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Boss et al., 2015 source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisenone given

    The association between loneliness and MCI and dementia remains unclear.

    “The association between loneliness and mild cognitive impairment and dementia remains unclear.”
    • Also stated in: § A C C E P T E D M A N U S C R I P T · plain
      “Yet, the understanding regarding the association between loneliness and the risk for MCI and dementia remains unclear.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    The authors performed a series of sensitivity analyses to support the robustness of the main pooled estimates.

    “we performed a series of sensitivity analyses to support the robustness of the main pooled estimates.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    The study was conducted following PRISMA-P guidelines.

    “This study was conducted following the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA-P) guidelines (Moher et al., 2015).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Moher et al., 2015

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Potentially unobserved variables, such as CR, could have confounded the findings.

    “potentially unobserved variables, such as CR, could have also confounded the findings of this study.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    The authors were unable to retrieve necessary information from one study with insufficient data (Wilson et al., 2015) despite contacting the authors.

    “we were not able to retrieve the necessary information from one study with insufficient data (Wilson et al., 2015), despite having contacted the authors.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Wilson et al., 2015 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    The systematic search was comprehensive, including multiple databases, manual searching, broad search terms, and coverage of all languages and years.

    “First, we performed a comprehensive systematic search including a variety of databases combined with manual searching, using a broad range of search terms and covering all languages and years.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Statistical testing for asymmetry was precluded because these tests are often underpowered and may yield misleading results when less than 10 studies are included.

    “statistical testing for asymmetry was precluded as these tests are often underpowered and may yield misleading results when less than 10 studies are included (Higgins and Green, 2008).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Higgins and Green, 2008

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable).

    Study design and conductown data

    The conclusion on the level of between-study heterogeneity should be interpreted cautiously.

    “the conclusion on the level of between-study heterogeneity should be interpreted cautiously.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    Publication bias was assessed by visual inspection of a contour-enhanced funnel plot.

    “Visual inspection of potential publication bias was conducted via the contourenhanced funnel plot (Peters et al., 2008).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Peters et al., 2008 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Unsatisfying social relationships have important implications for physical and mental health.

    “unsatisfying social relationships have important implications for physical and mental health (Heinrich and Gullone, 2006).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Heinrich and Gullone, 2006 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Independent reviewers completed study selection with a high level of inter-rater agreement.

    “Second, independent reviewers completed study selection, with a high level of inter-rater agreement.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Loneliness may appear either as a behavioural manifestation of cognitive deterioration or as an early sign of an underlying pathology contributing to MCI and dementia.

    “loneliness may appear either as a behavioural manifestation of cognitive deterioration or as an early sign of an underlying pathology contributing to MCI and dementia (Hawkley and Cacioppo, 2010;Wilson and Bennett, 2017).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Hawkley and Cacioppo, 2010 source

      Not checked in the free run.

    • cites Wilson and Bennett, 2017

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    Quality of included studies was assessed using an adapted version of the Newcastle-Ottawa Scale (NOS).

    “Quality of study assessment was conducted using an adapted version of the Newcastle-Ottawa Scale (NOS) for assessing risk of bias (Wells et al., 2000).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Wells et al., 2000

      Not checked in the free run.

  • May combine two claims.

    Study design and conductown data

    A meta-analysis for MCI could not be conducted because only two longitudinal studies existed and one did not provide effect sizes.

    “We were unable to conduct meta-analyses with MCI as the outcome since there were only two longitudinal studies of which one did not provide effect sizes (Lobo et al., 2008;Wilson et al., 2015), but limited evidence suggests a potential effect of loneliness on MCI.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Lobo et al., 2008

      Not checked in the free run.

    • cites Wilson et al., 2015 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning

    Some included studies further adjusted for participants' cognitive status at baseline in statistical models.

    “some of them further adjusted for participants' cognitive status at baseline in the statistical models.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Premisecitation

    These are common statistical limitations of most meta-analyses.

    “Altogether, these are common statistical limitations of most meta-analyses (Davey et al., 2011;Ioannidis et al., 2007).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Davey et al., 2011 source

      Not checked in the free run.

    • cites Ioannidis et al., 2007 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductreasoning

    Selective reporting may have occurred in the literature.

    “selective reporting may have occurred”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    The limited number of included studies (n=8) did not support statistical testing for funnel plot asymmetry.

    “The limited number of included studies (n=8) did not support statistical testing for asymmetry.”
    Quote in the paperNot a substantial claim
27 of 27

  • The paper relies on this without support.

    Individuals, as social beings, have a vital need to belong.

    “Individuals, as social beings, have a vital need to belong.”

  • The paper relies on this without support.

    Loneliness may vary over time in some individuals.

    “Although loneliness may vary over time in some individuals”
2 of 2

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • Consistentχ²(7) = 9.1reported p = .25 · recomputed p = .246

    “χ2(7)=9.10,p = .25”

1 of 1

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Degrees of freedom do not match the studies listed for risk-of-bias analysis

    Minor

    The risk-of-bias sensitivity analysis lists five study codes, for which Cochran's Q has four degrees of freedom, but reports five. The reported degrees of freedom would require six studies.

    Reported 5 · recomputed 4

  • Sutin study is classified as moderate risk despite its score of 8

    Minor

    The quality assessment calls Sutin et al. a moderate-risk study, but Table 2 assigns it a score of 8. Under the paper's stated scoring rule, scores above 7 are high quality.

2 of 2

Cites

Flagged first, then by citations

1 of the 84 references couldn’t be matched to a known paper.

Research works printed but not linked: 1

  1. Lobo, 2008, Non-cognitive psychopathological symptoms associated with incident mild cognitive impairment and dementia, Alzheimer’s Type. Neurotoxicity Research, 14, 10

Cited by

Sorted by citations

Related

Authors
Lara, E., Martín-María, N., De la Torre-Luque, A., Koyanagi, A., Vancampfort, D., Izquierdo, A., & Miret, M.
Affiliations
  • Instituto de Salud Carlos III (ES)
  • KU Leuven (BE)
Corresponding author
Lara, Elvira
ORCID
Venue
Ageing Research Reviews
Publisher
Elsevier BV
Published
2019-07
Volume
52
Pages
7-16
ISSN
1568-1637 (ISSN-L)
DOI
10.1016/j.arr.2019.03.002
OpenAlex
W2928245690
Semantic Scholar
848240af3570b38ae743188d1b748aec1446261c
Cited by
351
References
84
Citations by year
2026: 44 · 2025: 61 · 2024: 68 · 2023: 54 · 2022: 52 · 2021: 39 · 2020: 24 · 2019: 8
FWCI
5.17
Type
journal article
Language
en
Open access
Licence
Locations
  • Ageing Research Reviews (publishedVersion)
  • PubMed (publishedVersion)
  • Biblos-e Archivo (Universidad Autónoma de Madrid) (submittedVersion, cc-by-nc-nd)
  • Hispana (submittedVersion)
  • Lirias (submittedVersion)
Topics
  • Health Sciences › Medicine › Psychiatry and Mental health › Dementia and Cognitive Impairment Research
  • Social Sciences › Health › Health disparities and outcomes
  • Social Sciences › Psychology › Neuropsychology and Physiological Psychology › Aging and Gerontology Research
Keywords
loneliness, mild cognitive impairment, dementia, dementia risk, cognitive decline, older adults, social isolation, neuropsychiatric disorders
MeSH terms
Aged, Dementia, Humans, Loneliness, Longitudinal Studies, Risk Factors, Cognitive Dysfunction

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-09.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-09.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review