Air pollution effects on cognition in the CONSTANCES cohort – PoCoMo

We will assess air pollution exposure individually, at the home address, using three complementary maps of exposure. First, for classic pollutants, we will use dispersion models that give annual concentrations of PM10, PM2.5, NO2, SO2, C6H6 and O3, and a Europe-wide LUR model that gives annual concentrations of NO2, PM10 and PM2.5. Then we will assess exposure to airborne metals using the data from the BRAMM network (moss bio-monitoring of atmospheric metal deposition), a moss-biomonitoring database over France. Moss-based data are not available for urban participants; we plan a moss sampling campaign in two major cities in CONSTANCES: Paris and Lyon.
We will analyse the cross-sectional associations between air pollution exposure and cognitive functions at baseline for participants aged 45+ who took a comprehensive set of cognitive tests (93,200 participants). Once they will have taken these tests again, 4 years after their inclusion, we will analyse these associations longitudinally (we expect 35,000 participants). We planned to quantify cognitive functions either by individual test, or by computing a global score, and either as continuous or by threshold. We will use multilevel models adjusted for at least age, sex, education, taking account of recruitment city; we will test for interactions with age, sex, and education.
Four trans-disciplinary partners are involved in the project, including researchers with a strong background in environmental epidemiology, in cognition, in cohort management, and in ecology and bio-monitoring.

Exposure assessment: collaborators from the Basel University provided exposure to PM2.5, NO2 and O3 for 87,760 participants (exposure assessed for 2010), which we extrapolated between 2012-2015 taking account of relocations. At baseline, we found average exposures of 16.9±3.2, 25.5±12.0, and 67.8±7.4 µg.m-3 for PM2.5, NO2 and O3, respectively.
With our partner MNHN, we assessed the exposure to 13 airborne metals (including cadmium and lead), estimated in 2011 but attributed to the baseline address for all rural and suburban participants. We found average exposures of 0.21±0.09 and 4.6±2.2 µg.g-1(moss) of cadmium and lead, respectively.
Plus, we collected mosses in 186 sites over the areas of Greater Paris and Greater Lyon. They were sorted and prepared for chemical analyses, which will allow us to generate airborne metal exposure maps for these two metropolises. We will get all exposure data in 2019 and start the epidemiological analyses.
Study population: the CONSTANCES cohort recruited its 200,000th participant in February 2019. Now complete, CONSTANCES makes it possible to include ca. 93,200 participants aged 45+ for our cross-sectional analysis.

Shortly, we will start estimating the associations between air pollution exposure (LUR models) and cognition; we will replicate these analyses with the dispersion-model-based data as soon as available (late 2019). We will soon generate the airborne metal exposure maps in Paris and Lyon; in the meantime, we will analyse the associations between airborne metals and cognition within rural and suburban participants. In another project, and to assess longer term exposure to air pollution we developed an online tool (currently being deployed) for CONSTANCES participants to track and geocode home addresses before their inclusion.
In conclusion, we will assess air pollution effects on cognitive function in a large cohort of adults in whom we performed cognitive tests from the age of 45, an earlier age than of the majority of other cohorts, allowing to identify accelerated decline in cognitive functions in early stages. Furthermore we will have three different fine scale models of air pollution exposure, all covering all the French territory, allowing for exposure assessment at individual level at the residential address – with great novelty for moss-based data.
Our project will contribute to the knowledge of cognition impairment and its environmental determinants. Such knowledge can benefit public health and the society in general, by providing facts to policy makers to handle air pollution and trying to reduce population exposure to the air pollutants highlighted by our proposal.

Oral presentation at the 2018 ISEE-ISES conference in Ottawa. The scientific production based on PoCoMo was planned from 2020 onwards.

The evidence of the effects of air pollution on cognitive function in adults is scare. Furthermore, most of the previous studies did not have individual measures of air pollution, using aggregated data, did not have a comprehensive set of confounding variables, did not have standardized cognitive tests but used registry-based diagnosis of disease, only contained one measurement of cognitive function, only included older adults and had small sample size. In summary, the study of the effects of air pollution on cognition still has many questions to answer and presents several methodological challenges, but we will overcome most of them in this project.
Our project aims to evaluate the association between air pollution exposure and cognitive function in a large cohort of adults living all over France. Our overarching hypothesis is that air pollution, even at low levels of exposure, is a potential determinant for cognitive outcomes in adults. A secondary objective is to assess if exposure to heavy metals (HM) from air pollution, and which ones, are associated with cognitive outcomes in adults.
Our project is based on the CONSTANCES cohort which is a large population-based cohort launched in late 2012, aiming to include 200,000 participants. The cohort is designed as a randomly selected sample of French adults aged 18-69 years at inception, living in different regions throughout France. At inclusion, the selected subjects are invited to fill a questionnaire and to attend a comprehensive health examination. The follow-up includes an annual questionnaire, a health examination every 5 years and the linkage to health national databases (“SNIIRAM” and the national mortality database). A large range of data is collected on social and demographic characteristics, socioeconomic status, life events, behaviours, and occupational factors; the health data cover a wide spectrum. A unique feature of CONSTANCES is the inclusion of a comprehensive set of cognitive tests exploring global cognitive performances, attention and executive functioning, verbal fluency, memory, and psychomotor speed starting as young as 45 years, earlier in life than most of the available population-based cohorts studying cognitive impairment or decline.
Air pollution exposure will be assessed individually at the home address using three complementary maps of exposure. First, for classic pollutants, we will use a dispersion models that gives annual concentrations of PM10, PM2.5, NO2, SO2, C6H6 and O3 and a Europe-wide LUR model that gives annual concentrations of NO2, PM10 and PM2.5. Then we will assess exposure to atmospheric heavy metals using the data from the BRAMM network, a moss-biomonitoring database over France.
Four trans-disciplinary partners are involved in the project, including researchers with a strong background in environmental epidemiology, in cohorts management, in cognition and in ecology and bio-monitoring.
In conclusion, we will assess air pollution effects on cognitive function in a large cohort of adults in whom we performed cognitive tests from the age of 45, an earlier age than of the majority of other cohorts, allowing to identify accelerated decline in cognitive functions in early stages. Furthermore we will have three different fine scale models of air pollution exposure, all covering all the French territory, allowing for exposure assessment at individual level at the residential address. Two models will give exposure for classic pollutants and one very novel as it uses mosses that will allow assessing exposure to atmospheric HM.
Our project will contribute to the knowledge of cognition impairment and its environmental determinants. Such knowledge can benefit public health and the society in general, by providing facts to policy makers to handle air pollution and trying to reduce population exposure to the air pollutants highlighted by our proposal.