Abstract for the 2018 Annual Conference
Health effects of air pollution components, noise and socio-economic status (“HERMES”)
Ole Raaschou-Nielsen1,3, Theis Lange2, Matthias Ketzel3, Ulla Hvidtfeldt1, Henrik Brønnum-Hansen2, Thomas Münzel4, Ole Hertel3, Jørgen Brandt3, Mette Sørensen1,5
1Danish Cancer Society Research Center, Copenhagen, Denmark; 2University of Copenhagen, Copenhagen, Denmark; 3Aarhus University, Roskilde Denmark; 4Johannes Gutenberg University, Mainz, Germany; 5Roskilde University, Roskilde, Denmark
Background. Traffic-related air pollution (TRAP), traffic noise and low socio-economic status (SES) impair health, including CVD and diabetes. However, knowledge gaps still remain including identification of the causal agent(s) in the complex TRAP, the most relevant timing of exposure, the degree of confounding or possible interaction between TRAP and traffic noise, and how SES and individual susceptibility interplay in this equation. HEI has funded the HERMES study to address these questions.
Objectives. To identify the specific traffic-related air pollutants most strongly associated with myocardial infarction (MI), stroke and diabetes. To disentangle how TRAP and road traffic noise interact in relation to these endpoints. To investigate how socio-economic status (SES), green spaces, co-morbidity and stress confound/interact with the associations between TRAP and road traffic noise and risk of MI, stroke and diabetes. To investigate effects of TRAP and road traffic noise in relation to a cardiovascular and metabolic biomarkers.
Experimental design. We will take advantage of an “administrative” cohort covering the entire Danish population (DKPOP) (n=5.5 million), the Diet, Cancer & Health (DCH) cohort (n=57,053), and the DCH Next Generations (DCH-NG) cohort (n=50,000). We will link each individual to the unique and reliable nationwide Danish registries with information on residential address history, prevalent and incident MI, stroke and diabetes, vital status, indicators of stress and SES. Both DCH and DCH-NG have information on individual lifestyle, and DCH-NG furthermore has measurements of cardiovascular and metabolic biomarkers. We will use state-of-the-science models to calculate traffic-related air pollutants (NO2, NOx, black carbon, ultrafine particles, PM2.5, PMcoarse and PM10) and road traffic noise for all present and past residential addresses for each cohort participant at the exact time of living there.
We will develop new statistical methods for multipollutant analyses based on random forest methodology and apply these to identify the traffic-related air pollutants strongest related to MI, stroke and diabetes. The statistical analyses will estimate associations expressed both as relative and absolute risk. We will strive to separate effects of long- (years, decades) and short-term (days, weeks) exposure to TRAP by including both measures in the same statistical model. We will focus on effects of recent exposures (health endpoints from 2005 onwards) and we will assess the source-specific contributions to exposure (traffic/non-traffic; tail pipe/non-tail pipe TRAP). Further, we will develop an index for SES at neighborhood level and describe spatial associations between TRAP, noise, SES, co-morbidity and stress-markers. We will investigate how these factors interact with associations between TRAP/noise and the health endpoints.
Discussion. It has been a challenge for previous studies to separate effects of single traffic-related air pollutants on health. We will address that challenge by 1) development and application of new statistical tools based on random forest methodology and 2) use data for the entire Danish population providing excellent statistical power. It has also been a challenge to investigate separate health effects of and interactions between air pollution and noise, which we will address by using large populations and applying state-of-the-science exposure models at similar geographical level and with similarly precise input data.