Social isolation, loneliness, and their joint effects on cognitive decline and incident Alzheimer's disease: Findings from the Chicago health and aging project.

Background: There has been contradictory evidence on the prospective associations between social isolation/loneliness (SI/L) and cognitive decline (CD). There is also a scarcity of large and diverse population-based cohort studies examining SI/L that have confirmed clinical diagnoses of Alzheimer's Disease (AD). Notably, beyond individual associations, whether the effects of SI/L compound and accelerate CD and incident AD are not known.

Objectives: We hypothesized that SI and L, independently, would be associated with CD and incident AD to a similar extent, and the association of SI with CD and incident AD would be higher in lonely older adults.

Design: Prospective cohort study.

Setting: Urban Chicago areas.

Participants: We analyzed data in the Chicago Health and Aging Project (CHAP), which comprised 7,760 biracial community-dwelling older adults [mean age (standard deviation (SD))=72.3 (6.3); 64 % Black & 63 % women; mean (SD) of follow-up=7.9 (4.3) years].

Intervention (if any): NA MEASUREMENTS: Linear mixed and logistic regression models were used to regress CD and incident AD separately on the SI index/L.

Results: SI index and L were significantly associated with CD, with one-point increase of beta estimate (SE, p-value) = -0.002 (0.001,0.022) and -0.012 (0.003,<0.001), respectively. Given that the SI index ranges from 0 to 5 and the L from 0 to 1, they had similar effect sizes. Similarly, there were significant associations between SI index and incident AD, odds ratio (95 % CI, p-value) = 1.183 (1.016-1.379,0.029), and between L and incident AD, 2.117 (1.227-3.655,0.006). When stratified by loneliness status, compared to older adults who were not isolated and not lonely, older adults who reported being socially isolated and not lonely experienced accelerated CD, -0.003 (0.001,0.004), despite no significantly increased odds of incident AD.

Conclusions: SI/L had significant associations with CD and incident AD. Notably, socially isolated older adults who reported not being lonely appeared to be most socially vulnerable to CD. These findings suggest a specific at-risk subgroup of socially vulnerable older adults for future targeted interventions to improve cognitive health.