Citation

Ragettli MS, Phuleria HC, Tsai MY, Schindler C, de Nazelle A, Ducret-Stich RE, Ineichen A, Perez L, Braun-Fahrländer C, Probst-Hensch N, Künzli N. J. Expo. Sci. Environ. Epidemiol. 2014; 25(5): 474-481.

Affiliation

1] Swiss Tropical and Public Health Institute, Basel, Switzerland [2] University of Basel, Basel, Switzerland.

Copyright

(Copyright © 2014, Nature Publishing Group)

DOI

10.1038/jes.2014.83

PMID

25492241

Abstract

Exposure during transport and at non-residential locations is ignored in most epidemiological studies of traffic-related air pollution. We investigated the impact of separately estimating NO2 long-term outdoor exposures at home, work/school, and while commuting on the association between this marker of exposure and potential health outcomes. We used spatially and temporally resolved commuter route data and model-based NO2 estimates of a population sample in Basel, Switzerland, to assign individual NO2-exposure estimates of increasing complexity, namely (1) home outdoor concentration; (2) time-weighted home and work/school concentrations; and (3) time-weighted concentration incorporating home, work/school and commute. On the basis of their covariance structure, we estimated the expectable relative differences in the regression slopes between a quantitative health outcome and our measures of individual NO2 exposure using a standard measurement error model. The traditional use of home outdoor NO2 alone indicated a 12% (95% CI: 11-14%) underestimation of related health effects as compared with integrating both home and work/school outdoor concentrations. Mean contribution of commuting to total weekly exposure was small (3.2%; range 0.1-13.5%). Thus, ignoring commute in the total population may not significantly underestimate health effects as compared with the model combining home and work/school. For individuals commuting between Basel-City and Basel-Country, ignoring commute may produce, however, a significant attenuation bias of 4% (95% CI: 4-5%). Our results illustrate the importance of including work/school locations in assessments of long-term exposures to traffic-related air pollutants such as NO2. Information on individuals' commuting behavior may further improve exposure estimates, especially for subjects having lengthy commutes along major transportation routes.Journal of Exposure Science and Environmental Epidemiology advance online publication, 10 December 2014; doi:10.1038/jes.2014.83.

Language: en