20‐year depressive symptoms, dementia, and structural neuropathology in older women

Abstract INTRODUCTION The course of depressive symptoms and dementia risk is unclear, as are potential structural neuropathological common causes. METHODS Utilizing joint latent class mixture models, we identified longitudinal trajectories of annually assessed depressive symptoms and dementia risk over 21 years in 957 older women (baseline age 72.7 years old) from the Women's Health Initiative Memory Study. In a subsample of 569 women who underwent structural magnetic resonance imaging, we examined whether estimates of cerebrovascular disease and Alzheimer's disease (AD)‐related neurodegeneration were associated with identified trajectories. RESULTS Five trajectories of depressive symptoms and dementia risk were identified. Compared to women with minimal symptoms, women who reported mild and stable and emerging depressive symptoms were at the highest risk of developing dementia and had more cerebrovascular disease and AD‐related neurodegeneration. DISCUSSION There are heterogeneous profiles of depressive symptoms and dementia risk. Common neuropathological factors may contribute to both depression and dementia. Highlights The progression of depressive symptoms and concurrent dementia risk is heterogeneous. Emerging depressive symptoms may be a prodromal symptom of dementia. Cerebrovascular disease and AD are potentially shared neuropathological factors.


Table of Contents
Page 2: Supplemental Table S1: Model fit statistics of the joint latent class mixture models with competing risk of incident dementia and nondementia death Page 3: Supplemental Table S2.S3: Covariate effects on the level of depressive symptoms and risk of dementia from the five-class Joint Latent Class Mixture Model (N = 957) Page 5: Supplemental Table S4: Model fit statistics of the sensitivity analyses estimating the joint latent class mixture models with competing risk of incident dementia and nondementia death with using chronological age as time instead of study years as time.
Page 6: Supplemental Figure S1: Graph of the estimated mean score of the 15-item Geriatric Depression Scale over time for each joint latent class (Panel A) and cumulative incidence of dementia (Panel B) respective for each latent class of depressive symptoms when chronological age was used as time in the mixed model regression.
Page 7: Supplemental Table S5: Sensitivity analysis examining the risk of dementia over the Women's Health Initiative Memory Study Epidemiology of Cognitive Health Outcomes (WHIMS-ECHO) study period by Identified Joint Latent Class Relative when chronological age is modeled as time compared to women with minimal symptoms throughout the Study Period (N = 957).
Page 8: Supplemental Table S6: Sensitivity analyses examining the risk of dementia over the Women's Health Initiative Memory Study Epidemiology of Cognitive Health Outcomes (WHIMS-ECHO) study period by identified joint latent class relative to women with minimal symptoms throughout the study period after excluding the 75 women who self-reported a history of depression before study baseline (N = 882).
Page 9: Supplemental Figure S2: Graph of the estimated mean score of the 15-item Geriatric Depression Scale over time for each joint latent class (Panel A) and cumulative incidence of dementia (Panel B) respective for each latent class of depressive symptoms when omitting the 75 women who self-reported a history of depression before the study baseline.
Page 10: Supplemental Table S7  Supplemental Table S7 : Class-specific estimated latent depressive symptoms (in z-score standardized units) at the Women's Health Initiative Memory Study of the Epidemiology of Cognitive Health Outcomes (WHIMS-ECHO) baseline and average linear change during the Women's Health Initiative Study of Cognitive Aging (WHISCA) and WHIMS-ECHO study periods (N = 957) Page 4: Supplemental Table for education, race/ethnicity, region of residence, and household income b -2LL = negative log likelihood from respective model c Parameters = number of parameters in each respective model d BIC = Bayesian Information Criterion Supplement Figure S1.Graph of the estimated mean score of the 15-item Geriatric Depression Scale over time for each joint latent class (Panel A) and cumulative incidence of dementia (Panel B) respective for each latent class of depressive symptoms when chronological age was used as time in the mixed model regression.Supplemental Figure S2.Graph of the estimated mean score of the 15-item Geriatric Depression Scale over time for each joint latent class (Panel A) and cumulative incidence of dementia (Panel B) respective for each latent class of depressive symptoms when omitting the 75 women who self-reported a history of depression before the study baseline.

Table S1 .
: Weighted* Multivariable Multinomial Logistic Regressions to Examine the Effect of White Matter Small Vessel Ischemic Disease (WM-SVID) and Alzheimer's Disease like Neurodegeneration (AD-PS) on Probability of Being Classified into Respective Joint Latent Class † (N=526).Model fit statistics of the joint latent class mixture models with competing risk of incident dementia and nondementia death a a all models adjust for age at initial WHISCA assessment, education, race/ethnicity, region of residence, and household income b -2LL = negative log likelihood from respective model c Parameters = number of parameters in each respective model d BIC = Bayesian Information Criterion Supplemental

Table S2 .
Class-specific estimated latent depressive symptoms (in z-score standardized units) at the Women's Health Initiative Memory Study of the Epidemiology of Cognitive Health Outcomes (WHIMS-ECHO) baseline and average linear change during the Women's Health Initiative Study of Cognitive Aging a The effect estimates are adjusted for age at WHISCA baseline, race/ethnicity, region of residence, household income, and education.Bolded estimated denote p<.05 Supplemental

Table S3 .
Covariate effects on the level of depressive symptoms and risk of dementia from the five-class Joint Latent Class

Table S4 .
Model fit statistics of the sensitivity analyses estimating the joint latent class mixture models with competing risk of incident dementia and nondementia death with using chronological age as time instead of study years as time

.
Weighted* Multivariable Multinomial Logistic Regressions to Examine the Effect of White Matter Small Vessel Ischemic Disease (WM-SVID) and Alzheimer's Disease like Neurodegeneration (AD-PS) on Probability of Being Classified into Respective Joint Latent Class † (N=526).To account for the uncertainty in latent class membership, the posterior probability of latent class membership was included as a weight in the multivariable multinomial logistic regression.†Groupmembership derived from the joint latent class model examining trajectories of depressive symptoms and competing risks of dementia and nondementia mortality.‡The effect estimates are adjusted for age at WHISCA baseline, education, race/ethnicity, region of residence, employment status, household income, smoking, alcohol use, physical activity, cardiovascular disease, hypertension, hypercholesterolemia, diabetes, hormone use, and hormone therapy assignment. *