Social isolation and cognitive function in middle aged and older adults in the Wisconsin Alzheimer’s Disease Research Center: Implications for the pandemic

Background Social isolation and loneliness contribute to cognitive decline, and social isolation represents a key modifiable risk factor for dementia. Substantial interruptions in social engagement due to the COVID‐19 pandemic have heightened awareness of this issue, but common later‐life events may also reduce social connectedness. The neurocognitive impacts of such disruptions are unclear. Using the validated NIH Emotion Toolbox, we examined perceived social support and changes in perceived support as predictors of cognitive function in a sample of middle aged and older adults. Method 672 participants from the Wisconsin Alzheimer’s Disease Research Center (ADRC) completed cognitive tests and Emotion Toolbox questionnaires at 2+ visits between 2017 and 2021 (mean interim = 1.7 years). Predictors comprised t‐scores on three perceived social support scales: Emotional Support, Instrumental Support, and Loneliness. Cross‐sectional analyses examined associations between social support and cognitive outcomes including memory (RAVLT Immediate and Delayed Recall) and processing speed/executive function (Trails A&B times). Using median t scores from the first and last available visit to categorize social support as high or low (for Loneliness, low loneliness was categorized as “high” social support), support over time was classified as stable‐high, stable‐low, or “change” (high‐low or low‐high). Mixed‐effects regression models examined the demographic‐adjusted predictor‐outcome relationships between social support and cognition across visits. Result In this ADRC sample (mean age 65.2, SD = 9.2; Table 1), all three social support scores associated cross‐sectionally (Table 2) with Trails A and B performance. Relative to stable‐high participants, those reporting either stable‐low or declines from high to low social support exhibited poorer performance on measures of processing speed/executive function. Conversely, participants in the low‐high group performed comparably on cognitive tests to those in the stable‐high group (Table 3; Figure 1). Conclusion Social support associated with processing speed/executive function in this sample. Participants who reported a change in social support performed comparably to those with either stable high or stable low support, suggesting that processing speed/executive function associated with social support may be modifiable in the shorter term. These findings underscore the cognitive cost of social isolation and highlight the substantial benefits of maintaining ‐ and improving ‐ social connections among middle aged and older adults.