The Association Between Social Isolation and Cognitive Function in Older Adults in Urban, Fringe, and Rural Regions of Lebanon

Abstract

Background: The ongoing rectangularization of the global age structure signals that one out of four persons will be 65+ by 2050. Studies have shown that altered cognitive function (e.g. cognitive decline) and social status (e.g. social isolation) are significant determinants of health and quality of life in the older adult. Although significant public health and epidemiologic efforts are invested in examining the relationship between these two notions, but it remains poorly understood. The aim of this study is to evaluate the link between social isolation and cognitive decline in the Lebanese older adult and to investigate whether this association is affected by geographical residence. Methods: Statistical analysis methods are applied to secondary data derived from two national cross-sectional studies that assessed the prevalence of dementia in older adults. Out of a 744 cohort, comprised of older adults from urban, fringe and rural regions of Lebanon, 728 participants were included in this study. Living in social isolation was measured via an index score for 5-items: marital status, employment, social networking with children and relatives, social networking with friends and neighbors, and social participation. Living with cognitive decline was assessed via the screening tool, A-IQCODE: a 16-item questionnaire that assesses several cognitive domains. Univariate and multivariate logistic regression analysis evaluated the association between social isolation and cognitive decline while accounting for key confounders and the moderator, area of residence. To examine the effects of the latter on the former association, regression models were stratified by area of residence. All models were adjusted to cluster effect, the sampling technique. Results: The prevalence of cognitive decline and social isolation among older adults was 21% and 19%, respectively. After adjustment for age, gender, education, income, chronic diseases, mental illness and area of residence, the social isolation index score was significantly associated with cognitive decline (aOR 1.45, 95% CI 1.17 – 1.81). Only employment (aOR 3.04, 95% CI 1.19 – 7.75) and social participation (aOR 1.85, 95% CI 1.07 – 3.20) showed significant association with cognitive decline on separate regression models. When all indicators were included in a single regression model, social participation lost significance (aOR 1.63, 95% CI 0.84 – 3.17). All interaction terms for area of residence were not significant, thus we observed no effect modification by area of residence. Upon stratification of the multivariate regression analysis, the social isolation index score was significantly associated with cognitive decline in urban (p 0.03) and rural (p 0.02) areas but not in the fringe (p 0.31). 1 Conclusion: Our findings reveal much higher prevalence of cognitive decline in Lebanese older adults as compared to previous reports. Along with the notable prevalence of social isolation and the finding of a significant differential association with cognitive decline across geographical regions, our study is the first to describe two burdening public health issues in a large and representative cohort of older adults in Lebanon. Future research into the causal relationship and the socio-environmental risk factors for social isolation and cognitive decline could lay the groundwork for a much-needed policy and public health interventions.