Holly Elser1, Erzsébet Horváth-Puhó2, Jaimie Gradus3, Meghan Smith4, Timothy Lash5, Maria Glymour6, Henrik Sørensen2, Victor Henderson7
1Neurology, Hospital of the University of Pennsylvania, 2Department of Clinical Epidemiology, Aarhus University Hospital, 3Epidemiology, Boston University School of Public Health, 4Boston University School of Public Health, 5Rollings School of Public Health, 6UCSF, 7Stanford University School of Medicine
Objective:
Examine the relation between depression and incident dementia in Denmark.
Background:
Depressive symptoms often precede onset of dementia, but prior research reports diminished or null associations with increasing time since depression diagnosis. The absence of an association between remote depression diagnosis and incident dementia suggests depression is an early symptom rather than a cause of dementia.
Design/Methods:
This nationwide population-based cohort study leverages data collected routinely and prospectively from Danish population-based registries from 1978–2018. Individuals with depression ages ≥18 years were matched with members of the general population on sex, birth year, and date of depression diagnosis. Using Cox proportional hazards regression models, we estimated hazard ratios (HR) for the association between depression and dementia. To characterize the relationship between depression and dementia, all analyses were stratified by time elapsed between diagnoses (1–10 years; >10–20 years; >20–39 years). Because risk of depression and dementia may differ by age and sex, we conducted separate analyses for men and women and by age category.
Results:
Among 246,499 persons with depressive disorders and 1,190,302 matched persons, depression was associated with more than double the hazard of dementia (HR=2.43; 95% confidence interval [CI]: 2.37, 2.49). This HR was largest within the first 10 years after depression diagnosis (HR=2.60, 95%CI: 2.53, 2.68), but persisted >10–20 years (HR=1.89, 95%CI: 1.79, 2.01) and >20–39 years (HR=1.80, 95%CI: 1.59, 2.04). The association between depression and dementia was stronger among men (HR=3.02, 95%CI: 2.88, 3.16) than women (HR=2.22, 95%CI: 2.16, 2.29), and strongest among individuals diagnosed with depression between ages 18–44 (HR=3.09, 95%CI: 2.66, 3.60).
Conclusions:
This study demonstrates a robust association between incident depression and subsequent dementia when the diagnoses are separated by more than two decades, and thus suggests a role for depression as a major risk factor.