The Association Between Exposure to Air Pollution and Dementia Incidence: The Modifying Effect of Smoking.

Background: The influence of overall air pollution on dementia risk and the potential effect modification by other risk factors remain to be clarified.

Methods: We included 459 844 UK residents who were free of dementia and had data on the exposure to particulate matter (PM)2.5, PM2.5-10, PM10, NO2, and NOx during baseline recruitment. The combined exposure to various PMs and NOx was estimated by using an air pollution score. Hazard ratios (HRs) and 95% confidence intervals (CIs) for incident dementia were estimated by multivariable Cox models.

Results: During a median 11.7 years follow-up, 5 905 incident cases of all-cause dementia were identified. With the exception of PM2.5-10, all other air pollutants were separately associated with a higher risk of all-cause dementia (all p-trend < .001) with generally similar associations for dementia subtypes. An increasing air pollution score was associated with higher risks of all-cause as well as individual dementia outcomes, with adjusted HRs (95% CI) of 1.27 (1.18, 1.37) for all-cause dementia, 1.27 (1.14, 1.43) for Alzheimer's disease, and 1.35 (1.16, 1.57) for vascular dementia when comparing the highest with the lowest quartile of the score (all p-trend < .001). These associations of air pollution score with dementia and its subtypes were observed among never and former smokers but not among current smokers (all p-interaction 
< .030).

Conclusion: Air pollution was associated with a higher risk of dementia among nonsmokers but not current smokers. Additional studies are required to confirm our findings and to explore the potential mechanisms underlying the possible effect modification by smoking status.