O2‐11‐06: Reciprocal Relations of Loneliness and Cognitive Function in Older U.s. Adults

Nancy J. Donovan,Qiong Wu,Dorene M. Rentz,Reisa A. Sperling,Gad A. Marshall,M. Maria Glymour

DOI: https://doi.org/10.1016/j.jalz.2015.07.194

2015-01-01

Alzheimer s & Dementia

Abstract:In older adults, loneliness has been associated with functional decline and increased progression to dementia. The objective of this study was to examine the reciprocal relations of loneliness and cognitive function in older adults, adjusting for sociodemographic and health-related factors, social network and depression. The sample comprised 8,311 adults, ages 65 and older, participating in the U.S. Health and Retirement Study from 1998 to 2010. Participants underwent biennial assessments of loneliness (measured using a single binary question from the Center for Epidemiologic Studies Depression Scale (CES-D), 8-item version), depression (based on the remaining seven CES-D questions and classified as no, low (1-2) or high (3-7) depressive symptoms), global cognition, (using a composite score derived from the Telephone Interview for Cognitive Status, the Informant Questionnaire on Cognitive Decline in the Elderly and a word list memory task), health status and sociodemographic and social network characteristics. Repeated measures analyses were used to examine the reciprocal relations of loneliness and cognitive function over 12 years. In separate sets of regression models for baseline loneliness predicting cognitive function and for baseline cognition predicting loneliness, we controlled sequentially and cumulatively for (1) sociodemographic factors, (2) social network, (3) health conditions and (4) depression. Greater baseline loneliness was associated with accelerated cognitive decline over 12 years independent of sociodemographic factors, social network, health conditions and baseline depression (for loneliness interacting with time: b=-0.02, p=0.002, indicating a rate of decline approximately 25% faster in the lonely compared to the non-lonely). In post-hoc models including interaction terms for depression and time, both low and high depression categories were significantly associated with faster cognitive decline (b=-0.01, p=0.01 and b=-0.02, p=0.003 respectively) and the estimated effect of loneliness on rate of cognitive decline became marginally significant (b=-0.01, p=0.08). Reciprocally, lower baseline cognitive score was associated with greater odds of loneliness at baseline and across time (OR=1.12, p=0.005), but these associations were not significant when controlling for baseline depression. Loneliness is a risk factor for accelerated cognitive decline in older individuals. Although loneliness and depression appear closely linked, loneliness may have independent effects on cognitive decline.