AIM

This study aims to assess the association between trajectories of depressive symptoms and the risk of dementia, and to compare the predictive ability of trajectories using multiple data points with depressive symptoms at a single data point.

METHODS

We included 5306 older adults from the Health and Retirement Study. We assessed depressive symptoms using the Center for Epidemiology Depression Scale (CES-D), and identified its 8- year trajectories (2002-2010) using latent class trajectory modeling. We calculated hazard ratios (HR) using Cox proportional hazards models. The concordance index (C-index) was used to compare the discriminative power of the models.

RESULTS

We identified two trajectories of depressive symptoms, characterized by maintaining low CES-D scores, and moderate starting scores that steadily increased throughout the follow-up period. During 40,199 person-years, compared to the low trajectory, the increasing trajectory of depressive symptoms was associated with a higher risk of dementia (HR = 1.35; 95% CI: 1.09-1.67) (C-index = 0.759). For every point increase in the degree of depressive symptoms (CES-D scores) in 2010, the risk of dementia increased by 7% (95% CI: 1.03-1.12) (C-index = 0.760). The presence of depressive symptoms (CES-D scores ≥3) in 2010 was not associated with an increased risk of dementia (HR = 1.18; 95% CI: 0.98-1.43) (C-index = 0.759). The C-index values of cox models showed similar discriminative power.

CONCLUSIONS

The increasing trajectory of depressive symptoms at multiple data points and the degree of depressive symptoms at a single data point were associated with an increased risk of subsequent dementia among older adults.