OBJECTIVETo assess the effects of ambient air pollutants on hospital outpatient visits for allergic disease and pollinosis.

METHODSThe monitoring data of daily air pollution (SO(2), NO(2) and PM(10)) in 8 national monitoring and controlling sites of Beijing air quality, airborne pollen in 4 monitoring sites of Beijing, and daily meteorological data, along with the daily numbers of outpatients visits for allergic disease and pollinosis at the Allergy Department of Beijing Shijitan Hospital from April to September in 2004 were collected. Associations between the levels of air pollutants and outpatient visits for allergic disease and pollinosis were estimated by time serial analysis using a generalized addictive model (GAM), considering lag effect and the influence of multipollutants.

RESULTSDuring the study period, the average daily ambient concentrations of SO(2), NO(2) and PM(10) were (20.9 ± 12.9), (58.6 ± 13.6) and (126.8 ± 64.1) µg/m(3), respectively. The mean daily pollen count was (163.8 ± 209.0) grains/1000 mm(2), while the average daily numbers of doctor visits for allergic disease and pollinosis were (16.3 ± 5.3) and (3.5 ± 5.0), respectively. Time serial analysis showed that significant positive associations were found between levels of airborne pollen and doctor visits, with an excess risk (ER) of 2.44% (95%CI: 0.75% - 4.13%)for allergic disease and 6.58% (95%CI: 3.82% - 9.34%) for pollinosis per 100 grains/1000 mm(2) increase in pollen, in single-pollutant models. There were associations between ambient air pollutants (SO(2), NO(2), PM(10)) and doctor visits with lag effects. A 10 µg/m(3) increase in NO(2) was associated with a 3.14% (95%CI: 0.42% - 5.85%) increase in doctor visits for allergic disease at lag 6 d. For pollinosis, the highest ER was observed for 10 µg/m(3) increases in SO(2) at lag 1 d, NO(2) at lag 6 d, PM(10) at lag 3 d. However, this association was not statistically significant (RR (95%CI) were 1.0460 (0.9640 - 1.1280), 1.0325 (0.9633 - 1.1017), 1.0079 (0.9942 - 1.0217), respectively). The associations enhanced slightly in multi-pollutant models with an ER of 2.56% (95%CI: 0.80% - 4.31%) for allergic disease and 6.81% (95%CI: 3.91% - 9.71%) for pollinosis per 100 grains/1000 mm(2) of pollen.

CONCLUSIONSOur results suggest that level of airborne pollen may have a stronger effect than ambient air pollutants on allergic disease and pollinosis.