Premium
Depressive symptoms and risk of dementia in non‐depressed older adults
Author(s) -
Fitzpatrick Annette L.,
Lopez Oscar L.,
Leary Cindy S.,
Semmens Erin O.,
Adam Claire,
Park Christina,
Senter Camilla,
Hajat Anjum
Publication year - 2020
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.044035
Subject(s) - dementia , depression (economics) , comorbidity , medicine , psychiatry , longitudinal study , vascular dementia , proportional hazards model , receiver operating characteristic , epidemiology , psychology , disease , pathology , economics , macroeconomics
Background The relationship between depression and dementia is complex. Whether depression serves as a risk factor, prodromal symptom, or comorbidity of dementia remains unclear. Longitudinal studies that establish temporality of these conditions in late life may shed light on their relationship. Method We analyzed data from the Ginkgo Evaluation of Memory Study (GEMS) initiated in 2000 and followed for 8 years to determine incident dementia and its subtypes Alzheimer’s disease (AD) and vascular dementia (VaD). Inclusion criteria for enrollment in GEMS required participants to have a Center for Epidemiology Studies Depression score (CES‐D) <16 or to be on treatment for depression. Dementia was determined by consensus of study neurologists based on standardized criteria using neuropsychiatric testing, MRI imaging, and comorbidities. Cox Proportional Hazards Regression was used to evaluate the associations of dementia (n=515), AD (without VaD, n=348) and VaD (with or without AD, n=145) with number of depressive symptoms at baseline. We determined the optimum cutpoint for depressive symptoms to predict dementia using Receiver Operating Characteristic (ROC) curves. Adjustments for age, gender, race, education, area level income, and study site were included in models. Result 3,042 participants mean age 78.6 (SD 3.3), 46% women, mean follow‐up time 5.5 (SD 1.6) person‐years, were included in analyses. Mean number of CES‐D depressive symptoms at baseline was 3.5 (SD 3.2) with 525 participants reporting none. Number of depressive symptoms was associated with an increased risk of dementia (HR:1.09, 95% CI:1.06‐1.12, p<0.001), AD (HR:1.10, 95% CI:1.07‐1.13, p<0.001) and VaD (HR:1.05, 95% CI:1.00‐1.10, p=0.04). The optimum cutpoint based on the ROC curve was 2.6 with sensitivity of 0.64 and specificity of 0.49. Participants who reported > 3 depressive symptoms had a 45% increased risk of dementia compared to those with fewer symptoms (HR: 1.45, 95% CI:1.04‐2.04, p=0.03). Effect modification by presence of mild cognitive impairment (MCI) was found (p interaction = 0.008). Results were only significant for those without MCI. Conclusion Reporting just a few depressive symptoms in late life may indicate future onset of dementia. These results suggest the need for close clinical follow‐up of cognitive impairment in these non‐depressed older adults.