Answers are grounded in this paper and cite their sources.
Subjective and objective social isolation may be differently associated with depressive symptoms in rural left-behind older adults, while psychological capital may represent an important psychosocial correlate. This study characterized their conditional association structure and explored directed probabilistic dependencies. We conducted a cross-sectional survey of 445 rural left-behind older adults recruited from Zhejiang, Anhui, and Yunnan provinces between July 2023 and October 2024. Participants completed the Perceived Isolation Scale, Lubben Social Network Scale-6, 10-item Center for Epidemiologic Studies Depression Scale, and Positive Psychological Capital Questionnaire. We evaluated Pearson correlation-based EBICglasso networks (γ = 0.50) using strength, expected influence, bridge strength, bridge expected influence, and 2,000-sample bootstrap procedures. Sensitivity analyses addressed loneliness-indicator overlap, spouse-support applicability, and measurement granularity. We estimated a bootstrap-averaged Bayesian network using hill climbing with Gaussian BIC and 1,000 bootstrap samples. The overall network contained 94 non-zero edges. The strongest edge was PPQ1–PPQ3 (0.517), followed by PIS3–CESD9 (0.461), PIS2–CESD9 (0.342), and PIS1–LSNS2 (0.332). CESD9 had the highest strength (1.263), expected influence (1.152), bridge strength (0.968), and bridge expected influence (0.857). Sensitivity networks excluding CESD9 or PIS3 were highly concordant with the primary network (edge-weight r = 0.967 and 0.995, respectively). In the married-only analysis, overall edge concordance remained high ( r = 0.944), although the PIS2–CESD9 edge decreased from 0.342 to 0.151. In the construct-level sensitivity analysis, subjective isolation was positively connected with depression (0.365), whereas depression was negatively connected with psychological capital (−0.501). Bayesian analysis identified several stable adjacencies, although certainty regarding edge orientation varied substantially. Loneliness-related indicators occupied prominent positions in the network linking social isolation and depressive symptoms, whereas psychological capital was inversely associated with depression. These patterns were generally robust across sensitivity analyses, although some associations depended on how we operationalized nodes and on marital status. Longitudinal studies are needed to determine the temporal and causal nature of these relationships.
No takes yet. Share an insight, caveat, or question.
Xu et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: