6-Year trajectories of depressive symptoms and incident stroke in older adults: Results from the Health and Retirement Study

Background: Depressive symptoms (DS) can increase the risk of stroke, but it is unclear whether long-term DS trajectories are associated with incident stroke. This study aimed to explore the association of long-term DS trajectories with incident stroke. Methods: This prospective cohort study included 11,002 adults aged 50 and older from the Health and Retirement Study during 1994-2018. DS was assessed using the 8-item version of the Center for Epidemiologic Studies Depression Scale. Stroke was obtained through self-report of doctors' diagnosis. The group-based trajectory model was used to determine DS trajectories from 1994 to 2000. Cox proportional hazard model was applied to explore the correlation of DS trajectories with incident stroke from 2000 to 2018. Results: We identified five distinct 6-year DS trajectories. Compared with the persistent no DS trajectory, the fulladjusted HRs (95% CIs) for the persistent mild, improving, worsening, and persistent high DS trajectories were 1.15 (1.01, 1.30), 1.27 (0.88, 1.84), 1.41 (1.17, 1.71), and 1.61 (1.21, 2.16), respectively. In addition, the persistent mild DS trajectories had the largest population attributable risk percent (PAR%). Limitations: There was a lack of information on stroke subtypes. Conclusions: This study suggests that compared with persistent no DS, persistent mild, worsening, and persistent high DS trajectories increase the risk of stroke in the elderly. Considering that the PAR% of stroke events in the persistent mild DS trajectory is the largest, we should pay attention not only to individuals with DS, but also to those being chronically close to the cut-off value of DS.