Acute effects of summer air pollution on respiratory function in primary school children in southern England.

BACKGROUND: There is growing concern about health effects of air pollution in the UK. Studies in the USA have reported adverse effects on lung function among children but no comparable studies have been published in the UK. This study investigates the relationship between daily changes in ambient air pollution and short term variations in lung function in a panel of school children. METHODS: One hundred and fifty four children aged 7-11 attending a primary school adjacent to a major motorway in Surrey, south-east England, were studied. Bellows spirometry was performed daily on 31 schooldays between 6 June and 21 July 1994. Levels of ozone, nitrogen dioxide, and particulates of less than 10 microns in diameter (PM10) were measured continuously at the school and the pollen count was measured six miles away. Relationships between daily changes in forced expiratory volume in 0.75 seconds (FEV0.75), forced vital capacity (FVC), the FEV0.75/FVC ratio and pollutants were analysed using separate autoregressive models for each child. A weighted average of the resulting slopes was then calculated. RESULTS: There was a significant inverse relationship between daily mean PM10 levels lagged one day and FVC, with a reduction in lung function of 1% (95% CI 0.3% to 2%) across the whole range of PM10 levels (20-150 micrograms/m3). The effect on FEV0.75 was similar (-0.5%) but was not significant when weighted by 1/SE2 (95% CI -1.2% to 0.2%). There was no effect of PM10 levels on the FEV0.75/FVC ratio. No significant association was seen between FEV0.75, FVC, or the FEV0.75/FVC ratio and either ozone or nitrogen dioxide levels. There was no evidence that wheezy children were more affected than healthy children. Pollen levels on the previous day had no effect on lung function and did not change the air pollution results. CONCLUSIONS: There is a very small, but statistically significant, adverse effect of airborne respirable particulate matter, measured as PM10, on lung function in this study group. There is no evidence for an inverse association of lung function with levels of ozone or NO2 measured on the previous day.