062 Traffic-related air pollution assessed using a dispersion model related to allergie and respiratory health in children

Abstract

Introduction The impact of traffic-related air pollution (TAP) on respiratory and allergie health is still debated. This might hâve been the resuit of inaccurate classification of exposure. Our cross-sectional study was intended to assess the role of TAP modelled using a dispersion model in the prevalence of allergie and respiratory outeomes in children. Methods 6,683 children aged 9-ll years recruited between 1999 and 2000 from 108 schools randomly selected in 6 French cities underwent a clinical examination including skin prick test and exercise-induced bronchial reactivity (EIB). The prevalence of asthma, allergie rhinitis (AR) and eczema were assessed by a stan-dardised questionnaire completed by the parents. A validated 3-D dispersion model, which combines data on traffic conditions, topo-graphy, meteorology and background air pollution, was used to assign outdoor annual mean concentrations of benzene, VOC, SO2, PM10, NO2, NOx, and CO at the 108 schools attended by the children. For each air pollutant, a 2-categories-exposure variable was defined with respect to the median value of the modelled concentrations at school. Results Adjusted odds ratios (OR) for EIB indicated significant positive associations with ail modelled air pollutants (OR =1.7 (95% CI: 1.3-2.1) with SO2, 1.5 (1.2-1.8) with benzene, 1.3 (1.1-1.7) with PM10), except NO2. Asthma was significantly related to benzene (1.3 (1.0-1.6)), SO2 (1.3 (1.0-1.6)) and PM10 (1.3 (1.0-1.6)), AR to PM10 (1.2 (1.0-1.4), eczema to NO2 (1.2 (1.1 -1.4)) and sensitisation to pollen to benzene (1.2 (1.0-1.5)) and PM10 (1.4 (1.1-1.7)). Positive associations of asthma with VOC and CO, of AR with benzene and SO2, of eczema with PM10, and of sensitisation to pollen with VOC reached borderline statistical significance. Sensitivity analyses suggested stronger associations between sensitisation to pollen and TAP in girls than in boys. Positive associations were more pro-nounced among long-term resident children (current address for at least 8 years). Conclusion Accurate modelization of TAP shows an augmented prevalence of respiratory and allergie diseases in areas with high concentrations of air pollutants.