An Examination of the Physiological Mediators Linking Social Isolation, Loneliness, and Cognitive Function in Older Adults with Diabetes

Background: Diabetes is an independent risk factor for cognitive impairment and dementia. Social isolation and loneliness are modifiable yet understudied factors that may exacerbate the effect of diabetes on cognitive impairment. Furthermore, evidence of physiological mediators (systemic inflammation and sleep disturbance) in the relationship between social isolation, loneliness, and cognitive function is limited, highlighting the need for further research. Purpose: The aims of this study were 1) to examine the effects of social isolation and loneliness on the initial levels of cognitive function and rates of cognitive decline in older adults with diabetes, 2) to investigate the mediating effect of systemic inflammation on the longitudinal association between social isolation and cognitive function, and 3) to determine the mediating effects of physiological mediators on the associations between social isolation, loneliness, cognitive function, and to identify whether these association vary by the presence of diabetes. Methods: A secondary analysis of existing data from the University of Michigan Health and Retirement Study (HRS) was conducted. Data from participants aged 50 and over with diabetes were analyzed. This study utilized data from the Steptoe Social Isolation Index, the UCLA loneliness scale, C reactive protein results, and the Jenkins Sleep Scale to measure social isolation, loneliness, systemic inflammation, and sleep disturbances, respectively. Cognitive status was measured using data from the telephone interview. Latent growth modeling, a cross lagged panel model, and multigroup analysis within a structural equation modeling framework were used. Results: For Aim 1, social isolation and loneliness negatively affected the initial levels of cognitive function, but were not associated with a faster decline in cognitive function in older adults with diabetes over the ten year timeframe. For Aim 2, this study found reciprocal negative longitudinal associations between social isolation and cognitive function when controlling for sociodemographic covariates. However, these associations attenuated after adjusting for health-related covariates. C reactive protein did not mediate the longitudinal association between social isolation and cognitive function. For Aim 3, social isolation directly influenced poorer cognitive function in older adults, while loneliness did not. Sleep disturbance mediated the association between social isolation, loneliness, and cognitive function. Systemic inflammation mediated the association between loneliness and cognitive function but did not mediate the association between social isolation and cognitive function. When comparing the nondiabetes and diabetes groups, there was a significant difference in the pathway between systemic inflammation and cognitive function. Conclusions: These findings provide new insights into the effects of social isolation and loneliness on cognitive function, and the mediating roles of systemic inflammation and sleep disturbances in older adults with diabetes. Physiological mediators may act as pivotal targets for developing interventions aiming at mitigating the pathways between social isolation, loneliness, and cognitive function. However, current evidence remains limited, and further empirical studies are needed. Implications: These findings could support the integration of comprehensive assessments of social isolation, loneliness, sleep disturbance, and inflammatory biomarkers into standard evaluation protocols for evaluating cognitive impairment and dementia. Such an inclusion could enhance early detection and enable timely interventions to prevent the progression of cognitive decline among older adults.