Source of perceived social support and cognitive change: an 8-year prospective cohort study

Objectives:This study explored the longitudinal association between overall and individual sources of social support and global cognitive function in older adults. Methods: Data were drawn from three waves (2006, 2010 and 2014) of the Health and Retirement Study (HRS). The analytic sample included participants aged 65 and above who were married and had at least one child (N= 1319). Global cognitive function was measured through a 35-pointTelephone Interview Cognitive Screen (TICS). Perceived social support was measured via questions across four sources of support (spouse, child, otherfamily members, and friends). Analyses were conducted using Hierarchical Linear Modeling (HLM) with the addition of a cross-lagged panel model (CLPM). Results: There was no significant association between perceived overall social support and global cognitive function overtime (beta= -0.02, p =0.19). However, changes in perceived support from children were positively associated with changes in global cognition (beta= 0.05,p <0.01). In contrast, changes in perceived support from other family members were negatively associated with changes in cognitive function (beta= -0.07,p < 0.01) . Conclusions: Our findings highlight the importance of examining perceived social support from specific elements of one's social networks, rather than in an aggregated variable. Although the positive effect of perceived support from children may be small to moderate, the findings could provide a target for interventions to protect the cognitive function of older adults.