Abstract

BACKGROUND: Increasing evidence associates air pollution with dementia, but some pollutants and susceptible groups are understudied. METHODS: We followed all Danish residents aged ≥60 years as of 1-1-2000, without prior dementia, until 12-31-2018 for dementia incidence identified via hospital contact or prescription. We assessed annual mean levels of fine particulate matter (PM2·5), nitrogen dioxide (NO2), and black carbon (BC) in 2010 utilizing European-wide hybrid land-use regression models, at baseline (2000) residential addresses. We examined the associations between air pollution exposure and dementia incidence with Cox proportional hazard models, accounting for individual- and area-level socio-demographic covariates and whether the effects were modified by age, sex, income level, education attainment, employment status, and the presence of comorbid conditions, including cardio-metabolic, respiratory diseases, and depression. FINDINGS: Among 934,792 individuals, 81,731 developed dementia over a mean follow-up of 11·6 years. Mean levels of PM2·5 and NO2, and BC were 12·5 and 20·6 µg/m3, and 1·0 × 10-5/m respectively. We detected strong associations between these pollutants and dementia incidence, with hazard ratios (HR) [95 % confidence intervals (CIs)] of 1·14 (1·12, 1·16), 1·25 (1·22, 1·28), and 1·23 (1·20, 1·26) per interquartile range increase of 1·9 μg/m3 for PM2·5, 10·2 μg/m3 for NO2, and 0·5 × 10-5/m for BC, respectively. Stronger associations were observed in elderly (≥75 years), those with stroke, the unemployed, and those with lower income or education levels than corresponding groups. DISCUSSION: Even low levels of air pollution in Denmark were associated with dementia development, especially among certain susceptible groups, emphasizing the need for targeted intervention strategies.