Effects of ambient ozone exposures during the spring and summer of 1994 on pulmonary function of schoolchildren

To investigate the effect of natural exposure to ambient ozone over time, a follow-up study of school-aged children was performed in two small towns in southwestern Germany (Freudenstadt and Villingen) between March and October of 1994. Ozone half-hour mean concentrations were measured continuously and pulmonary function was tested in each child on four occasions (April, June, August, and September). To obtain an average short- term ozone effect, we first analyzed the data from the four time points separately and then constructed a model that included all information. During the study period the median (5th to 95th percentile) of all half-hour values of the ozone concentration was 101 μg/m³ or 50.6 ppb (45-179 μg/m³ or 22.5-89.8 ppb) in Freudenstadt and 64 μg/m³ or 32.1 ppb (1 to 140 μg/m³ or 0.5-70.1 ppb) in Villingen. To assess the effects of an individual ozone exposure we related the highest ozone concentration in the respective 24 hours before lung function testing to the results of the subsequent pulmonary function tests. In the lung function test following the highest ozone exposure, the results of our cross-sectional linear regression analysis showed a significant negative correlation (P = 0.0181) between ozone exposure and forced vital capacity (FVC). In the longitudinal linear regression model we observed a negative statistical correlation between ozone exposure and lung function for the subpopulation living in the town with the high ozone levels (Freudenstadt). The association was more pronounced in boys than girls. For the children in Freudenstadt the decrement of FVC was -12.31 ml/10 μg/m³ ozone and the decrease in the forced expiratory volume in 1 second (FEV₁) was -1129 ml/10 μg/m₃ ozone.