OBJECTIVES:

Exposure to air pollutants increases morbidity and mortality. Traffic contributes about half of the emitted NOx to the atmosphere. Human activities play a key role in explaining exposure variation. Despite the importance of healthy air in indoor spaces, most studies focus on outdoor air. Misclassification of exposure has been recognized as a limitation of epidemiologic studies. Here we estimated the effect of outdoor traffic related air pollution on indoor air.

METHODS:

Simultaneous indoor and outdoor monitoring data of pollutants were recorded during the years 2005-2007 every half hour. We examined variables associated with levels of indoor pollutant NOx(µg/m). Independent variables included outdoor levels of NOx, meteorological and seasonal factors and human related activities. In multivariate analysis, we used linear regression models and a ‘Cosinor’ time series technique, to examine predictors for indoor pollutants levels, adjusted for trend and seasonality. We further conducted a multivariable quantile regression analysis, adjusted for the same independent variables as the former model.

RESULTS:

Indoor NOx level differed between weekdays and weekends (F=4.25, p = .015) and between seasons (F=14.56, p<.001). Outdoor NOx was the main predictor for indoor NOx levels (β=0.67, p<.001). In multiple linear regression analysis adjusted for liner trend (β= 0.134, p<.001), season (winter β=-1.40, p<.001, summer β=0.628, p=0.224, fall β=-0.09, p=0.83, compared to spring), outdoor NOx (β=-0.81, p=.001), temperature (β=-0.75, p=.028) rainfall (β=-0.50, p=.006), relative humidity (β=0.42, p=0.03) wind speed (β=0.41, p=.01) were associated with indoor NOx levels.

CONCLUSIONS:

Our findings contribute to a growing body of literature describing the relationships between indoor and outdoor of traffic related air pollutants. Further understanding the extent of the effect of indoor air on health is essential especially when considering the amount of time people spend indoors.