A method to derive the relationship between the annual and short-term air quality limits--analysis using the WHO Air...

The World Health Organization (WHO) Air Quality Guidelines (AQG) were launched in 2006, but gaps remain in evidence on health impacts and relationships between short-term and annual AQG needed for health protection. We tested whether relationships between WHO short-term and annual AQG for particulates (PM10 and PM2.5) and nitrogen dioxide (NO2) are concordant worldwide and derived the annual limits for sulfur dioxide (SO2) and ozone (O3) based on the short-term AQG. We obtained air pollutant data over seven years (2004-2010) in seven cities from Asia-Pacific, North America and Europe. Based on probability distribution concept using maximum as the short-term limit and arithmetic mean as the annual limit, we developed a new method to derive limit value one from another in each paired limits for each pollutant with capability to account for allowable exceedances. We averaged the limit derived each year for each city, then used meta-analysis to pool the limit values in all cities. Pooled mean short-term limit for NO2 (140.5μg/m(3) [130.6-150.4]) was significantly lower than the WHO AQG of 200μg/m(3) while for PM10 (46.4μg/m(3) [95CI:42.1-50.7]) and PM2.5 (28.6μg/m(3) [24.5-32.6]) were not significantly different from the WHO AQG of 50 and 25μg/m(3) respectively. Pooled mean annual limits for SO2 and O3 were 4.6μg/m(3) [3.7-5.5] and 27.0μg/m(3) [21.7-32.2] respectively. Results were robust in various sensitivity analyses. The distribution relationships between the current WHO short-term and annual AQG are supported by empirical data from seven cities for PM10 and PM2.5, but not for NO2. The short-term AQG for NO2 should be lowered for concordance with the selected annual AQG for health protection.