Associations of social isolation with memory and cognitive function in middle-aged and older Chinese adults

Abstract

Background

Although social isolation has been identified as a risk factor for cognitive impairment, its potential impact relative to other documented risk factors has not been comprehensively quantified, leading to its underestimation in public health strategies. We aimed to address this gap by quantifying the contribution of social isolation to cognitive decline in the context of other risk factors.

Methods

Social isolation was evaluated using a modified Social Network Index (SNI) and cognitive function through the Delayed Word Recall Test (DWRT) and the Mini-Mental State Examination (MMSE). Linear and logistic regression models were employed to analyze the associations between social isolation and cognitive outcomes, adjusting for demographic and health-related factors. Additionally, the XGBoost algorithm with SHapley Additive exPlanations (SHAP) was used to quantify the relative importance of predictors.

Results

A total of 25,981 participants were recruited from 2003 to 2008. The mean age was 62.0 years, with 28.4% being men. Higher social isolation was significantly associated with lower DWRT (β=-0.15; 95% CI: -0.21 to -0.09) and MMSE scores (β=-0.34; 95% CI: -0.48 to -0.19), and higher odds of memory impairment (OR = 1.27; 95% CI: 1.15 to 1.40) and poor cognitive function (OR = 1.56; 95% CI: 1.23 to 1.99). XGBoost analysis ranked social isolation as the fifth most important predictor for MMSE scores (SHAP value = 0.175) and the eighth for memory impairment (SHAP value = 0.0133). Subgroup analyses indicated stronger associations among older adults, and individuals with lower education or manual occupation.

Conclusion

Our findings showed that social isolation is an important risk factor for cognitive outcomes. This underscores the urgent need for targeted public health interventions addressing social isolation, alongside other key risk factors, to preserve cognitive health.