Long-term exposure to ambient air pollution and risk of dementia: results of the prospective Three-City study

Abstract

BackgroundEmerging epidemiological evidence suggests a relationship between exposure to various air pollutants (AP) (fine particles (PM2.5), ozone (O3), nitrogen dioxide (NO2)) and dementia. However, most of existing studies relied on population-based health administrative databases to obtain a dementia diagnosis. In a large French population-based cohort (the Three-City Study), we aimed at assessing the effects of different AP on dementia risk using reliable diagnosis tools.MethodsParticipants aged ≥65 years were recruited between 1999-2001 and followed for 12 years. At baseline and every 2 years, dementia diagnoses were preliminary made using DSM-IV criteria and validated by an adjudication committee. NO2, O3, Black Carbon and PM2.5 levels were obtained at the residential addresses of participants using land-use regression models (ELAPSE project). For each year of follow-up (FU), we estimated a 10-year moving window of mean past exposure to each pollutant for each subject. We used Cox Proportional Hazard models where exposure was included as a time-varying variable. Analyses were adjusted for individual (age, sex, education, APOE4 genotype, vascular risk factors, respiratory diseases) and contextual (neighborhood’s deprivation index) level confounders.ResultsThe mean (SD) age of the 8031 participants was 74.0 (5.4) years-old. Sixty-one percent of participants were women. The mean time of FU was 9.2 (3.6) years. The mean annual PM2.5 levels ranged from 15 to 33 µg/m3 and 920 participants developed dementia during the FU. We observed a positive association between PM2.5 levels and dementia risk [HR=1.16, 95%CI (1.03-1.32) for a 5µg/m3 increase in PM2.5]. We did not observe any relationship between the other AP exposures and dementia risk.ConclusionIn this large cohort with reliable diagnosis of dementia and individual estimates of AP exposure, long-term exposure to PM2.5 was associated with higher dementia incidence. These results suggest that PM2.5 exposure might be a modifiable risk factor of dementia.