Living alone and risk of dementia: A systematic review and meta-analysis

To systematically review longitudinal studies on living alone and incident dementia, to pool the results in a meta-analysis and calculate the population risk.

Embase, Medline and PsycInfo were searched from inception to August 2019 for longitudinal cohort studies of people living alone and risk of dementia. Relative risks (RR) were extracted and effect sizes pooled, with a sensitivity analysis for risk of bias (QUIPS quality rating tool). Population Attributable Fraction (PAF) was calculated, with prevalence of living alone calculated from UK Census data.

Twelve studies were identified for inclusion, nine of which had low risk of bias. The pooled effect size indicated an elevated risk of incident dementia when living alone (all studies RR = 1.30; 95 % CI: 1.15-1.46; low risk of bias studies (RR = 1.31; 95 % CI: 1.13-1.51). The PAF for living alone was 8.9 %.

Social isolation is a more important risk factor for dementia than previously identified, with living alone associated with greater population risk than physical inactivity, hypertension, diabetes and obesity.