The association among social participation, loneliness and depression in stroke survivors: A longitudinal...

To examine longitudinal associations among depressive symptoms, loneliness, and social participation in stroke survivors, and the temporal role of loneliness between social participation and depressive symptoms.

277 stroke survivors were conveniently sampled from two tertiary hospitals in Yangzhou, China. Depressive symptoms, loneliness, and social participation were assessed using the Patient Health Questionnaire-9, UCLA Loneliness Scale, and Utrecht Scale for Evaluation of Rehabilitation-Participation, respectively. Data were collected at baseline (T1), 3 months (T2), and 6 months (T3) post-stroke, with a cross-lagged panel model used to examine the reciprocal relationships among variables.

Cross-lagged panel analysis revealed: T1 depressive symptoms positively predicted T2 loneliness (β = 0.133, P = 0.046)；T1 social participation negatively predicted T2 loneliness (β = -0.226, P < 0.001) and depressive symptoms (β = -0.256, P < 0.001)；T2 depressive symptoms negatively predicted T3 social participation (β = -0.098, P = 0.027)；T2 loneliness positively predicted T3 depressive symptoms (β = 0.171, P = 0.006)；T2 social participation negatively predicted T3 loneliness (β = -0.174, P = 0.001). Mediation analysis showed T2 loneliness mediated T1 social participation's effect on T3 depressive symptoms (indirect effect = -0.037, 95% CI [-0.082, -0.010]).

Alternating cross-lagged effects were observed among depressive symptoms, loneliness, and social participation in stroke survivors, with loneliness potentially acting as a mediator between social participation and subsequent depressive symptoms. These findings suggest hypotheses for phase-specific interventions, which should be tested in future studies.