Latent profile analysis and determinants of social isolation among older adults in China

Abstract

Objective This study aimed to identify social isolation profiles among older adults in China using latent profile analysis and to explore their associations with demographic and health factors. Methods A cross-sectional study was conducted between July and December 2023 in the Chaoyang and Dongcheng districts of Beijing, China. Participants aged 60 years or older were recruited through convenience sampling from six communities. Latent profile analysis was used to classify social isolation patterns, and multivariate logistic regression models to assess associated factors. Results 529 participants were included. Three social isolation profiles were identified: the mild social isolation subgroup (n = 239, 45.2%), the moderate social isolation subgroup (n = 213, 40.3%), and the severe social isolation subgroup (n = 77, 14.5%). Comparing mild to severe social isolation, multivariate analysis revealed that older adults living alone without caregiver had significantly higher odds of severe social isolation compared to those living with children (OR = 0.099, 95% CI: 0.034–0.377, P < 0.001). Frailty (OR = 0.141, 95% CI: 0.039–0.588, P = 0.033), severe ADL dependence (OR = 0.249, 95% CI: 0.145–0.454, P = 0.048), and moderate to severe IADL dependence (OR = 0.448 to 0.307, P = 0.046, P = 0.017) were also significantly associated with severe social isolation. Similarly, in the comparison between the moderate and severe subgroups, compared to those living with children, older adults living alone without caregiver were more likely to experience severe social isolation (OR = 0.199, 95% CI: 0.078–0.328, P = 0.001). In terms of functional status, severe ADL dependence (OR = 0.152, 95% CI: 0.061–0.505, P = 0.015) and moderate to severe IADL dependence (OR = 0.231 to 0.128, P = 0.045, P = 0.033) were significantly associated with severe social isolation. Conclusion Older adults living alone without caregiver, with frailty, and with functional dependence are at significantly higher risk of severe social isolation. Family network integrity emerged as the primary dimension differentiating severe social isolation from mild-to-moderate isolation, which may be linked to a quantitative gradient of overall network restriction. Understanding the specific factors contributing to different profiles could inform tailored support strategies to enhance social engagement and quality of life in this population.