Abstract
INTRODUCTION: Depression is an established risk factor of dementia, but a single one-off measure of depressive symptoms could introduce estimation bias. We aimed to examine the roles of dynamic characteristics of depressive symptoms in predicting the risk of incident all-cause dementia (ACD) and Alzheimer’s dementia (AD).
METHODS: Two longitudinal cohorts were included, involving 97,157 participants (mean age = 56.5 years, female = 47 %, follow-up = 14 years) from UK Biobank (UKB) and 464 participants (mean age = 75.8 years, female = 57.1 %, follow-up = 5 years) from Alzheimer’s Disease Neuroimaging Initiative (ADNI). Dynamic features of depressive symptoms were defined as accumulated periods of depression in UKB and symptom trajectory in ADNI. Cox proportional hazard regression models were used to test the associations of these dynamic characteristics with dementia risk independently in two cohorts. Finally, we meta-analyzed the existing evidence concerning the longitudinal relationships between dynamic characteristics of depressive symptoms and dementia risk.
RESULTS: In the UKB cohort, a higher number of depressive periods was linked to an elevated risk of both ACD (hazard ratio [HR] = 1.28, 95 % confidence interval [CI] = 1.07 to 1.53, p = 0.007) and AD (HR = 1.25, 95 % CI = 1.04 to 1.51, p = 0.019). In ADNI, compared to the consistent absence of depressive symptoms, mild and increasing depressive symptoms trajectory, but not stable mild symptoms was associated with an increased risk of AD (HR = 2.33, 95 % CI = 1.33 to 4.82, p = 0.022). Meta-analyses of 16 longitudinal cohorts revealed six trajectories significantly predicated escalated risk of dementia, including stable mild symptoms (relative risk [RR] = 1.43 %), stable high symptoms (RR = 1.74), increasing symptoms (RR = 1.68), adult-onset depression (RR = 1.38), mid-to-late life depression (RR = 1.84), and life-long depression (RR = 1.33).
CONCLUSION: We revealed multiple dynamic features of depressive symptoms associated with risk of dementia, which not only strengthen the causal relationships but also necessitate the importance of dynamic monitoring and management of depressive symptoms in preventing dementia.