Abstract
Panel studies have found 1-3% increases in reports of acute respiratory symptoms associated with each 10 microg x m(-3) increase in ambient concentrations of particles with a 50% cut-off aerodynamic diameter of 10 microm (PM10). Statistically significant decrements in population mean peak expiratory flow (PEF) have also been observed but their magnitude is only about 0.1% for the same exposure. This study evaluated whether this difference is due to the different measurement scales used for the expression of the effects of air pollution on PEF and respiratory symptoms. The association between the prevalence of large decrements in PEF (more than 10% or 20% below the median) and PM10 concentrations was analysed using data from five recent panel studies. The effect estimates were compared to the effect estimates from original studies based on the population mean PEF. The decrement in the population mean PEF for an increase of 10 microg x m(-3) of the same-day PM10 concentration was 0.07%, averaged over all panels. A significant relative increase of 2.7% (95% confidence interval 1.6-3.8) in the prevalence of PEF decrements greater than 10% was associated with the same exposure. In conclusion, defining the peak expiratory flow response by the prevalence of large decrements provides effect estimates of a comparable magnitude to effect estimates for the prevalence of reports of acute lower respiratory symptoms.