Text

Objectives and questions: Osteoarthritis (OA) and depressive symptoms frequently co-occur in middle-aged and older adults, yet most studies focus on disorder-level associations and rarely examine symptom-specific mechanisms or underlying molecular correlates. Understanding how OA connects to the depressive symptom may help identify targeted intervention pathways.

Material and methods: Using data from 15,721 adults aged ≥45 years in CHARLS 2018, hierarchical regression and mixed graphical modeling were used to assess population-level associations between OA and ten CES-D symptoms and to estimate the symptom network. Separate EBICglasso networks were constructed for OA and non-OA groups, and structural differences were evaluated using the Network Comparison Test. Three GEO datasets were integrated to identify shared differentially expressed genes, hub genes, enriched pathways, and machine-learning-derived cross-disease feature genes.

Results: After full adjustment, OA remained significantly associated with higher depressive symptom levels (β = .10, p < .001). Network analysis identified depressed mood, lethargy, loneliness, and restless sleep as central symptoms, whereas “restless sleep” and “everything is an effort” emerged as key bridge symptoms linking OA with depressive symptoms. Several affective-motivational and social pathways were stronger in the OA group. Bioinformatic analyses identified 177 shared OA-depression genes and two robust feature genes (TGFB2, ENO2), enriched in ECM remodeling, PI3K-Akt signaling, focal adhesion, and metabolic stress-related pathways.

Discussion and conclusions: OA is associated with distinct depressive symptoms and bridge mechanisms involving sleep disturbance, reduced energy, and diminished positive affect. Multilevel convergence across symptom networks and molecular pathways highlights potential screening targets and supports early detection and personalized intervention for OA-depression comorbidity.