Associations of Physical Exercise With Specific Symptoms of Depression in Older Adults: A Network Analysis Following Propensity Score Matching

OBJECTIVES: To reveal pathways through which physical exercise is associated with the specific depressive symptoms in older adults from the perspective of network analysis. METHODS: 8,931 participants were selected from the 2018 Chinese Longitudinal Healthy Longevity Survey database. Participants were divided into exercise and nonexercise groups using propensity score matching to minimize confounding effects. The 10-item Center for Epidemiologic Studies Depression Scale was used to assess depression in this cross-sectional survey. Treating each item of 10-item Center for Epidemiologic Studies Depression Scale as a node, depressive symptom networks were constructed for both groups. Network centrality, stability, accuracy, and intergroup differences were analyzed. RESULTS: Core symptoms in the exercise group were D3 (sad mood), D8 (felt lonely), and D1 (bothered by little things), whereas core symptoms in the nonexercise group were D3 (sad mood), D9 (cannot get going), and D6 (felt fearful). The expected influence of the D9 (cannot get going) node was higher in the nonexercise group compared with the exercise group; no differences in symptom network structure between the two groups were found. CONCLUSIONS: Depressive symptom networks varied between older adults who participated in physical exercise and those who did not. Physical exercise was associated with the local features (centrality indexes) of the depressive network, rather than the global features (network structure invariance and overall network strength). CLINICAL IMPLICATIONS: The D9 (cannot get going) is crucial for the beneficial effects of exercise, providing a potential target for interventions aimed at efficiently alleviating depressive symptoms in older adults.