Marital Status and Cognitive Decline in a Prospective Northern Manhattan Cohort
Courtney Gao1, Jackson Roberts2, Huaizhi Ge3, Hannah Gardener4, Tatjana Rundek4, Mitchell Elkind5, Sarah Tom5, Jose Gutierrez3
1Columbia University Irving Medical Center, 2Massachusetts General Brigham, 3Columbia University, 4University of Miami, 5Columbia University Mailman School of Public Health
Objective:

To determine whether marital status is associated with cognitive decline in a population-based cohort.

Background:

Being unmarried has been associated with clinical markers of cerebral vasculopathy and adverse neurovascular events. It remains unknown whether marital status is associated with cognitive decline.

Design/Methods:
In the Northern Manhattan Study, a prospective and diverse community-dwelling cohort, we evaluated cognition of the participants who received MRI, all of whom were > 55 years of age and stroke-free. The sample was divided into married and unmarried participants, and the latter were subdivided into divorced/separated, widowed, or single. During 2 to 3 visits over 9±4 years, we measured their memory, processing speed, language, and executive function, and computed a global score that averaged these four outcomes. We used longitudinal analyses to investigate associations between marital status and cognitive decline. Covariates included sociodemographic variables — age, ethnicity, and sex — and the following neurovascular risk factors: hypertension, diabetes, hypercholesterolemia, smoking, APOE-ε4 status, total cranial volume, presence of covert brain infarct, and body mass index.
Results:
Of 970 participants, 43% were married, 15% were single, 14% were widowed, and 27% were divorced/separated. Compared to married participants, unmarried participants had a greater decline in memory (β=-0.14; 95% CI [-0.26, -0.02]), processing speed (β =-0.24, 95% CI [-0.36, -0.12]), and global cognition (β=-0.13, 95% CI [-0.22, -0.04]). Decline in processing speed was most pronounced in single participants (β=-0.32, 95% CI [-0.49, -0.15]) and widowed participants (β= -0.29, 95% CI [-0.47, -0.11]). Marital status did not predict a decline in executive function.

Conclusions:
Processing speed, memory, and global cognition scores decreased more for unmarried participants compared to their married counterparts. These results link marital history with cognitive sequelae. Future studies should explore how health behaviors vary by partnership status, and whether supporting healthy relationships can improve neurocognitive outcomes.
10.1212/WNL.0000000000216007
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.