The Concentration-response Function For Air Pollutant And Outpatient Visits Under Different Exposure Levels

Objective:The aim of this study was to investigate the impact of air pollution on the outpatient visits and the change of health risk when air quality improved. Provide basic accurate data for the accountability research about health benefits. Method:Outpatient visits data was from a major hospital in Taiyuan; Air pollution data was from Jianhe Environment Monitoring Station; Weather data was from Shanxi Meteorological Service. The research period was from 2001 to 2010. According to the local environmental protection policy and the change of air pollutants concentration, we divided the research period into 2001-2002 and 2003-2010 two parts. Compare the health risk of 2003-2010 with it of 2001-2002. Concentration-response functions(CRFs) were estimated using generalized additive modeling(GAM) with natural log link function while multiple structural change test was used to determine the changing point in CRFs. Constrained piecewise linear function was used to estimate the risks for different segments of CRFs. Results:Concentration of the air pollutants decreased from 2001 to 2010. Concentration of SO2 reduced obviously after 2003. PM10 and CO reduced slowly. Compared with other air pollutants, NO2 has a small drop. When the concentration of SO2 reduced, the health risk of paediatrics, respiratory and endocrine decreased. At the same time, the impact of other air pollutants increased. In heating period, the impact of CO is obvious. When the air quality improved, the shape of CRFs and the relative risk became different. Under the same concentration, the relative risk decreased, especially for CO. During the research period, the relative risk decreased obviously with the concentration of air pollutants increased. The shape of CRFs and relative risk are different in different departments. The relative risk of paediatrics is bigger than that of adults. The relative risk of different diseases is different, too. The health risk of SO2 and CO is obvious in cardiovascular disease, and the health risk of PM10 is obvious in respiratory and endocrine diseases. Conclusions:The relative risk of air pollutants is different in different regions, exposure levels and groups of people. When access the health benefits of environmental policy, we should consider above elements. Choosing the suitable relative risk can obtain accurate results.