Objective:

To investigate the role of traffic-related air pollution (TRAP) on risk of dementia. 

Background:

While age remains the most important risk factor for dementia, there has been growing evidence implicating modifiable risk factors such as lifestyle and environmental exposures. Many studies have investigated the role of air pollution, especially particulate matter (PM_2.5), on dementia risk. However, the role of the different types of air pollutants on dementia risk remains understudied. 

Design/Methods:

We conducted a nationwide, population-based, geographical study of 20,789,037 Medicare beneficiaries (aged 67+), 756,321 with incident dementia. We estimated exposure to TRAP, specifically on-road nitrogen oxides (NOx), volatile organic compounds (VOCs), and PM_2.5 emissions for 2017 from Neighborhood Emission Mapping Operation (NEMO), and average annual daily traffic (AADT) for 1997-2010. We used logistic regression models to test the nationwide and region-specific associations between zip+4 level TRAP and dementia. We adjusted for age, sex, race, prevalent cardiometabolic disease, neighborhood-level income, and educational attainment as potential confounders. We used Geographic Weighted Regression to map nationwide county-level associations between TRAP and dementia to inform subregion analyses.

Results:

TRAP was associated with a dose-dependent greater risk of dementia nationwide. The strongest association between TRAP and dementia risk was found in the Upper Midwest.  Beneficiaries who lived in Upper Midwest neighborhoods with the highest decile of exposure had greater odds of dementia compared to the lowest decile for AADT (OR=22% (95% CI: 17%, 26%)), on-road VOCs (OR=28% (95% CI: 22%, 34%)), PM_2.5 (OR=26% (95% CI: 20%, 31%)), and NO_x, (OR=23% (95% CI: 17%, 28%)). Results from our sensitivity analysis using incident Alzheimer’s disease corroborated our primary analyses using incident dementia.

Conclusions: