The Mediating Role of Social Engagement in the Relationship between Depressive Symptoms and Cognitive Functioning Among Older Adults: The Korea Longitudinal Study on Aging (KLoSA)

Background and Purpose: Cognitive decline is a major public health challenge in aging populations. Depression is a known risk factor for cognitive impairment; however, mechanisms underlying this association require further investigation. Social engagement, often reduced in depression, may independently affect cognitive health and represents a potential pathway linking depressive symptoms to cognitive decline. Understanding these longitudinal relationships is essential for developing effective interventions. Using data from South Korea, experiencing rapid aging and high elder depression rates, this study examined the longitudinal interplay between depressive symptoms, social engagement, and cognitive functioning in older adults. The study hypothesized that: (1) Depressive symptoms negatively predict cognitive function trajectories. (2) Social engagement mediates this relationship.

Methods: A longitudinal design was employed using data from six waves (2010-2020) of the Korea Longitudinal Study on Aging (KLoSA), a representative panel study with biennial interviews. Participants (N=1,719) were adults aged 65 or older at baseline (M_age=71.5 years, SD=5.9; 59.7% female) with longitudinal data through 2020. Measures included the CES-D (depression), K-MMSE (cognition), and frequency of meeting friends (social engagement). Control variables included baseline age, gender, and education. Data were analyzed using Multivariate Latent Growth Curve Modeling within a Structural Equation Modeling (SEM) framework to assess trajectories and relationships over time. Mediation was tested by examining indirect effects using bias-corrected bootstrapping with 95% confidence intervals (C.I.s).

Results: The measurement model demonstrated acceptable fit to the data (RMSEA=.051, CFI=.932, TLI=.935, SRMR=.041). Supporting Hypothesis 1, higher baseline depressive symptoms significantly predicted lower baseline cognitive function, and worsening trajectories of depressive symptoms significantly predicted steeper declines in cognitive function over the 10-year period. Depressive symptoms (both baseline and slope) were also significantly associated with lower levels and declining trajectories of social engagement. Furthermore, higher baseline social engagement predicted better baseline cognitive function, and improving social engagement trajectories predicted less steep cognitive decline. Supporting Hypothesis 2, significant mediation effects were found: baseline social engagement partially mediated the negative relationship between baseline depressive symptoms and baseline cognitive function (Std. Indirect Effect = -0.048, 95% C.I. [-0.080, -0.015]), and also mediated the effect of the depressive symptom slope on baseline cognitive function (Std. Indirect Effect = -0.043, 95% C.I. [-0.077, -0.010]). Mediation pathways involving the slope of cognitive function were not statistically significant.

Conclusions and Implications: This longitudinal study confirms depressive symptoms predict cognitive decline over ten years among older South Koreans. Findings highlight the critical role of social engagement, demonstrating it partially explains how depression impacts cognitive function, particularly baseline cognitive status. Diminished social engagement appears to be an important pathway linking both existing depression and worsening depressive symptoms to poorer cognitive outcomes. For social work practice, these results underscore the importance of integrated care approaches that address both mental health and social needs in older adults. Screening for social isolation alongside depression may identify individuals at higher risk for cognitive decline. Findings strongly suggest that non-pharmacological interventions aimed at enhancing social engagement could serve as a protective strategy to mitigate cognitive risks associated with depression.