Association between social integration and risk of dementia: A systematic review and meta-analysis of longitudinal studies.
Level 3 - non-randomized controlled study
Systematic review and meta-analysis of observational cohort studies
PubMed 36307921 · doi:10.1111/jgs.18094
What was done
A systematic review and random-effects meta-analysis of five databases evaluated the association between various aspects of social integration and the risk of dementia. Study reporting quality was appraised using the Newcastle-Ottawa Scale. Longitudinal cohort studies assessing social engagement, frequency of social contact, social support, close social contact, loneliness, social isolation, and social network size were synthesized, alongside sensitivity analyses.
What was found
Across 40 publications from 32 studies/databases: - Strong social engagement was associated with lower dementia risk (overall RR = 0.81, 95% CI: 0.74–0.89, p < 0.001). - Frequent social contact was associated with lower dementia risk (overall RR = 0.86, 95% CI: 0.76–0.97, p = 0.018). - Loneliness was associated with higher dementia risk (overall RR = 1.42, 95% CI: 1.26–1.60, p < 0.001). - Larger social network size was associated with lower dementia risk (RR = 0.75, 95% CI: 0.59–0.97, p = 0.028), though based on an insufficient number of studies for full meta-analysis. - Associations for social support (RR = 0.92, 95% CI: 0.80–1.06, p = 0.238), close social contact (RR = 0.74, 95% CI: 0.48–1.13, p = 0.167), and social isolation (RR = 1.58, 95% CI: 0.80–3.12, p = 0.192) were not statistically significant.
Why it matters
The findings differentiate specific domains of social interaction, showing that active engagement, contact frequency, and subjective loneliness correlate most strongly with dementia risk in longitudinal cohorts.
Limits
The included designs are observational cohorts, leaving potential for residual confounding and reverse causation. Heterogeneity among studies was generally high and remained substantial despite sensitivity analyses. Total participant count across cohorts is not reported in the abstract.