CoChat

Social Isolation and Memory Decline in Later-life

Open access
Read, Sanna, Comas-Herrera, Adelina, Grundy, Emily · 2020 · The Journals of Gerontology: Series B

The DOI matches its Crossref record; 45 of the 46 references it cites were found in a registry or an index; 1 of the 27 sources its claims cite were checked for support.

Abstract

Objectives: To investigate associations between level and changes in social isolation and in memory in older men and women. Methods: The sample included 6,123 women and 5,110 men aged 50+ from the English Longitudinal Study of Aging (ELSA). Extended latent change score models from six measurement occasions every 2 years from 2002 were used to investigate associations between social isolation and memory. Models were adjusted for age, socioeconomic position, and health. Results: Social isolation increased and memory decreased over time. Among men an initially high level of social isolation was associated with a somewhat greater decrease in memory. Among women a greater increase in social isolation predicted a greater decrease in memory and a larger change in social isolation was associated with further larger changes in isolation, although when social isolation reached a higher level it subsequently decreased. Conclusions: Results suggest that the association between social isolation and memory decline arises because social isolation is associated with increased memory decline rather than poor memory leading to increases in social isolation. Men with high levels of social isolation and women with accumulated social isolation over time are especially affected as these patterns of isolation were associated with more profound memory decline.

Overview

CoChat Summary

TL;DR

Using six waves of the English Longitudinal Study of Aging, this study finds that social isolation increased and memory decreased over time, and that the association is better explained by social isolation predicting memory decline than by poor memory leading to isolation. Patterns differed by gender: high initial isolation mattered more in men, while accumulating isolation over time mattered more in women.

Key findings

  • In the ELSA sample, social isolation increased and memory decreased over the follow-up period.
  • Among men, an initially high level of social isolation was associated with a somewhat greater decrease in memory: a high social isolation score (4+) was associated with a 0.9-point (18%) memory decline over 2 years, compared with a 0.3-point (6%) decline among men with an average isolation score of one.
  • Among women, a greater increase in social isolation predicted a greater decrease in memory: one increase in the social isolation scale involved a decrement of 2.6 on the memory score in the subsequent follow-up window, on average a 0.5-point (9%) decrease over 2 years.
  • Women who experienced larger changes in social isolation were more likely to experience further changes in isolation, although when social isolation reached a higher level it subsequently decreased.

Methods

  • The sample included 6,123 women and 5,110 men aged 50+ from the English Longitudinal Study of Aging (ELSA), using six measurement occasions every 2 years from 2002.
  • Extended latent change score models were estimated in Mplus 7.3, separately for men and women, and adjusted for age, socioeconomic position, and health-related covariates.
  • Memory was measured with immediate and delayed word-list recall; social isolation was measured with a 0–5 index based on living alone, contact frequency with children/other family/friends, and organizational membership.

Limitations

  • The paper states that the sample may be initially selected, that ELSA does not cover people in institutional settings at baseline so results generalize only to community-dwelling older people, and that attrition bias may remain despite FIML.
  • The paper also states that gender differences were not formally tested due to the complexity of the latent change score modeling, that the social isolation score does not capture perceived support or contact quality, and that 2-year intervals may miss shorter-term changes.
  • Evident from the design: this is an observational longitudinal study, so the reported associations cannot by themselves establish causal effects of social isolation on memory.

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

Integrity signals

Registry

Crossref · 10.1093/geronb/gbz152

Cites

45 of 45 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1. Full bivariate latent change score model for men (N = 5,110). The diagram shows that the level of social isolation predicts the change in memory over the six waves. Only significant level-to-change paths are shown (see Table 1 for the estimates of all paths), paths with no coefficient are fixed to 1, fully adjusted models. The model fit: −2LL = 140,599.01, df = 469, AIC = 282,136, BIC = 285,202. Iso = social isolation, mem = memory. *p < .05, **p < .01, ***p < .001.
Figure 1. Full bivariate latent change score model for men (N = 5,110). The diagram shows that the level of social isolation predicts the change in memory over the six waves. Only significant level-to-change paths are shown (see Table 1 for the estimates of all paths), paths with no coefficient are fixed to 1, fully adjusted models. The model fit: −2LL = 140,599.01, df = 469, AIC = 282,136, BIC = 285,202. Iso = social isolation, mem = memory. *p < .05, **p < .01, ***p < .001.CC BY · PubMed Central
Figure 2. Full bivariate latent change score model for women (N = 6,123). The diagram shows that the level and change of social isolation predicts the change in social isolation and the change in social isolation predicts memory over the six waves. Only significant level-to-change and change-to-change paths are shown (see Table 1 for the estimates of all paths), paths with no coefficient are fixed to 1, fully adjusted models. The model fit: −2LL = 176,095.99, df = 473, AIC = 353,138, BIC = 356,316. Iso = social isolation, mem = memory, *p < .05, **p < .01, ***p < .001.
Figure 2. Full bivariate latent change score model for women (N = 6,123). The diagram shows that the level and change of social isolation predicts the change in social isolation and the change in social isolation predicts memory over the six waves. Only significant level-to-change and change-to-change paths are shown (see Table 1 for the estimates of all paths), paths with no coefficient are fixed to 1, fully adjusted models. The model fit: −2LL = 176,095.99, df = 473, AIC = 353,138, BIC = 356,316. Iso = social isolation, mem = memory, *p < .05, **p < .01, ***p < .001.CC BY · PubMed Central

Verification

Claims the paper makes

0 Valid · 1 Not valid · 16 Not checked · 45 Own results and interpretations

28 more claims

Create a free account to see all 62 claims and to verify the 26 references still open.

See all claims — free account
  • It may state two separate claims as one. 17
  • Probably not a substantial claim. 15

Create a free account to see these claims.

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?

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.

1 of 1

Citations

Cites

Research works printed but not linked: 2

  1. Hubbert, 2006, Cognitive function., 217
  2. de Oliveira, 2010, Health risk and health protective biological measures in later life., 275

Cited by

See all 135

Metadata

Authors
Read, S., Comas-Herrera, A., & Grundy, E.
Affiliations
  • London School of Economics and Political Science (GB)
  • Care Policy and Evaluation Centre, London School of Economics and Political Science, UK
  • University of Essex (GB)
Corresponding author
Read, Sanna
Venue
The Journals of Gerontology: Series B
Published
2019-11-29
Volume
75
Issue
2
Pages
367-376
ISSN
  • 1079-5014 (ISSN-L)
  • 1758-5368
OpenAlex
W2990762867
Cited by
135
References
46
Citations by year
2026: 13 · 2025: 13 · 2024: 25 · 2023: 29 · 2022: 21 · 2021: 22 · 2020: 12
FWCI
13.35
Type
journal article
Language
en
Open access
Locations
  • The Journals of Gerontology Series B (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Open Access at Essex (University of Essex) (acceptedVersion)
  • London School of Economics and Political Science Research Online (London School of Economics and Political Science) (submittedVersion)
  • London School of Economics and Political Science Theses Online (London School of Economics and Political Science) (submittedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
Funders
  • UK Economic and Social Research Council (1025561/2/3)
  • National Institute for Health Research (ES/L001896/1)
Topics
  • Social Sciences › Health › Health disparities and outcomes
  • Social Sciences › Psychology › Developmental and Educational Psychology › Aging and Gerontology Research
  • Health Sciences › Medicine › Psychiatry and Mental health › Dementia and Cognitive Impairment Research
Keywords
social isolation, memory decline, older adults, aging, sex differences, longitudinal change, latent change score models, English Longitudinal Study of Ageing
MeSH terms
Latent Class Analysis, Age Factors, Aged, Female, Health Status, Humans, Longitudinal Studies, Male, Memory Disorders, Middle Aged, Social Isolation, Socioeconomic Factors

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.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-05, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-05 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (qwen3.8-omni-flash).

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

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.