Abstract
OBJECTIVES: Loneliness in the aging population is associated with decreased cognitive function and increased neuropathology; less is understood about the association of loneliness and cognitive resilience (CR), defined as the discordance between a person's actual and expected cognition given their neuropathology. Here we assess the effect of loneliness and change in loneliness on CR at end of life and across older adulthood. METHODS: Data were combined from 2 longitudinal studies of older adults. CR proximate to death (CRlast_level) and across time (CRslope) was obtained by independently regressing global cognition and change in cognition onto multiple neuropathology indicators and extracting the resulting residuals. We used a series of simple linear regression models to assess the effect of loneliness level and change on CRlast_level and CRslope. RESULTS: Higher baseline loneliness was associated with lower CRlast_level (β = -0.11, 95% confidence interval [95% CI; -0.18, -0.04], p < .01); higher baseline loneliness and increasing loneliness over time was associated with lower CRslope (β = -0.13, 95% CI [-0.22, -0.05], p < .01 and β = -0.12, 95% CI [-0.20, -0.04], p < .01, respectively). Results were robust to covariate inclusion and independent of objective social isolation. DISCUSSION: Higher and increasing loneliness was associated with lower CR in the face of neuropathology. These results suggest that some individuals are less resilient to the accumulation of neuropathology than others, and experiencing high/increasing loneliness is a key factor putting some at risk. Interventions aimed at optimizing cognitive function across older adults should include loneliness reduction as a potential area of focus.