Emerging yet contrasting evidence associates air pollution with incident dementia, and the potential role of cardiovascular disease (CVD) in this association is unclear. To investigate the association between long-term exposure to air pollution and dementia and to assess the role of CVD in that association. Data for this cohort study were extracted from the ongoing Swedish National Study on Aging and Care in Kungsholmen (SNAC-K), a longitudinal population-based study with baseline assessments from March 21, 2001, through August 30, 2004. Of the 5111 randomly selected residents in the Kungsholmen district of Stockholm 60 years or older and living at home or in institutions, 521 were not eligible (eg, due to death before the start of the study or no contact information). Among the remaining 4590 individuals, 3363 (73.3%) were assessed. For the current analysis, 2927 participants who did not have dementia at baseline were examined, with follow-up to 2013 (mean [SD] follow-up time, 6.01 [2.56] years). Follow-up was completed February 18, 2013, and data were analyzed from June 26, 2018, through June 20, 2019. Two major air pollutants (particulate matter ≤2.5 μm [PM2.5] and nitrogen oxide [NOx]) were assessed yearly from 1990, using dispersion models for outdoor levels at residential addresses. The hazard of dementia was estimated using Cox proportional hazards regression models. The potential of CVD (ie, atrial fibrillation, ischemic heart disease, heart failure, and stroke) to modify and mediate the association between long-term exposure to air pollution and dementia was tested using stratified analyses and generalized structural equation modeling. At baseline, the mean (SD) age of the 2927 participants was 74.1 (10.7) years, and 1845 (63.0%) were female. Three hundred sixty-four participants with incident dementia were identified. The hazard of dementia increased by as much as 50% per interquartile range difference in mean pollutant levels during the previous 5 years at the residential address (hazard ratio [HR] for difference of 0.88 μg/m3 PM2.5, 1.54 [95% CI, 1.33-1.78]; HR for difference of 8.35 μg/m3 NOx, 1.14 [95% CI, 1.01-1.29]). Heart failure (HR for PM2.5, 1.93 [95% CI, 1.54-2.43]; HR for NOx, 1.43 [95% CI, 1.17-1.75]) and ischemic heart disease (HR for PM2.5, 1.67 [95% CI, 1.32-2.12]; HR for NOx, 1.36 [95% CI, 1.07-1.71]) enhanced the dementia risk, whereas stroke appeared to be the most important intermediate condition, explaining 49.4% of air pollution-related dementia cases. This study found that long-term exposure to air pollution was associated with a higher risk of dementia. Heart failure and ischemic heart disease appeared to enhance the association between air pollution and dementia, whereas stroke seemed to be an important intermediate condition between the association of air pollution exposure with dementia.