Abstract Body: Background: Vascular risk factors, particularly in midlife, are associated with an increased risk of dementia, and smoking has been inversely associated with Parkinson's disease (PD) risk, but the role of these factors in PD-dementia (PDD) is less clear. This study explores whether midlife vascular risk factors are associated with risk of PDD in the community-based ARIC cohort.

Methods: ARIC participants were evaluated for vascular risk factors (hypertension, diabetes, hypercholesterolemia, smoking, and obesity) in 1987-1989 (ages 44-64) and followed through 2016. PD cases were identified using participant medications, self-reported physician diagnosis, hospitalization and death surveillance, and PD diagnostic data provided by participants and physicians. Dementia was defined by in-person and phone-based cognitive assessment, informant interviews, and hospitalization codes. PDD was defined as having both a PD and dementia diagnosis, with the PD diagnosis occurring first. We excluded participants with missing covariates, on neuroleptic medications, or with PD or dementia at baseline. Adjusted Cox proportional hazards models examined the associations between midlife vascular risk factors (combined in one model) and incident PDD, PD/no dementia, and dementia/no PD, vs no PD/no dementia. We explored effect modification by race.

Results: Of 13,875 participants (25% Black, 54% female), 179 developed PD at a mean age of 73.4 yo, 94 devleoped PDD at a mean age of 79.2 yo, and 1,791 developed dementia/no PD at a mean age of 79.7 yo. Midlife current smoking (HR 0.41, 95% CI 0.18-0.95, Figure) was signficantly associated with a lower risk of PDD; other vascular risk factors had nonsignficiant associations. Older age, APOEe4, male sex, and low education were significantly assoiated with an increased risk of PDD. Smoking, diabetes, hypertension, obesity, Black race, age, low education, male sex, and APOEe4 were associated with an increased risk of dementia/no PD. There was effect modification by race for smoking and obesity, which were significant risk factors for dementia/no PD in White but not Black participants (Table).

Conclusions: Smoking in midlife was significantly associated with a lower rate of PDD vs no PD/no dementia. Other vascular risk factors were not associated with PDD, but demographic associations were similar to dementia. Future studies should evaluate these vascular risk factors over the life course and the mechanisms underlying these associations.