Effects of Inhalable Particles on Respiratory Health of Children

The field data collection of an air pollution epidemiologic study was carried out from 1993 to 1996 in four Chinese cities of Lanzhou, Chongqing, Wuhan, and Guangzhou. In each city, an urban district and a suburban district were selected. Ambient concentrations of total suspended particles ( TSP ) , size -fractionated particulate matter including PM 2.5 , PM 2.5 -10 , and PM 10 , sulfur dioxide ( SO 2 ) , and oxides of nitrogen ( NO x ) were measured in these districts. The results indicate the presence of wide inter -city and intra -city gradients in long -term ambient levels of these measured pollutants. Across the eight districts, the 1993 -1996 4 -year means of TSP, SO 2 , and NO x ranged from 198 to 659 g / m 3 , from 14.6 to 331 g / m 3 , and from 31.5 to 239 g / m 3 , respectively, and the 1995 -1996 2 -year means of PM 2.5 , PM 2.5 -10 , and PM 10 ranged from 51.5 to 142 g / m 3 , from 29.2 to 107 g / m 3 , and from 80.7 to 232 g / m 3 , respectively. These pollution ranges substantially extended the upper end of the pollution ranges of previous air pollution epidemiologic studies conducted in North America and Europe. In each district, significant correlations among the measured pollutants were observed for daily concentrations. However, the gradient patterns in longterm means of different pollutants were different across the eight districts. ( e.g., PM 2.5 -10 and TSP were highest in the Lanzhou urban district, PM 2.5 and PM 10 were highest in the Guangzhou urban district, SO 2 was highest in the Chongqing urban district, and NO x was highest in the Guangzhou urban district ) . In general, seasonal variations were present in the ambient concentrations with high levels often occurring in winter months. The eight districts may be classified into four district clusters based on integrated levels of all measured pollutants. These features of the ambient air pollution have important implications for epidemiological studies and may provide unique opportunities to study exposure -effects relationships in the four Chinese cities.

Section: Discussionsupporting

confidence: 69%

Section: Discussionmentioning

confidence: 99%

Long-term ambient air pollution levels in four Chinese cities: inter-city and intra-city concentration gradients for epidemiological studies

Qian

Zhang

Wei

et al. 2001

“…Because community air pollutants tend to be widely distributed, exposure subgroups within a region were difficult to distinguish. As a result, four studies distinguished exposure groups by areas with different measured pollutant levels (Ware et al, 1986;Dockery et al, 1989;Pope, 1991;Horstman et al, 1997 ). In four other studies, exposure groups were aggregated from populations of different areas with comparable pollutant levels (Schenker et al, 1983(Schenker et al, ,1986Dockery et al, 1996;Raizenne et al, 1996 ).…”

Section: Study Characterizationmentioning

confidence: 99%

Residence location as a measure of environmental exposure: a review of air pollution epidemiology studies

Huang

Batterman

2000

160

Residence location has long been used to indicate environmental exposure in many epidemiological studies. This indicator is easy to establish, requires little exposure or monitoring data, and is potentially applicable to many types of investigations. The validity, accuracy and utility of residence location as an exposure indicator, however, is challenged by current concerns regarding multiple exposure pathways, persistent and toxic contaminants, and cumulative exposures from non -point, mobile and point sources. This paper reviews 45 epidemiological studies that use residence location to identify study populations and estimate air pollution exposures. Thirteen ( 29% ) of the studies determined environmental exposures based on``proximity'' measures, usually the distance from a subject's residence to a pollutant source. Other studies used``zones'' presumed to have equal pollutant levels. Several studies combined zone and proximity approaches. Exposures were quantified using monitoring data in 27 ( 60% ) studies and dispersion modeling in two ( 4% ) studies. Sixteen ( 36% ) studies did not use any environmental data to quantify exposure. A total of 31 ( 69% ) of the studies reported significant associations between health endpoints and the pollutant exposures represented by residence location. In general, comprehensive and systematic approaches to identify and estimate population exposures were not used, and the exposure estimates were therefore deemed likely to have great uncertainty. Unless exposure levels among groups are verified, it cannot be determined whether nonsignificant associations between exposures and health endpoints indicate a lack of measurable health effects, or are merely a result of exposure misclassification. Site -specific and quantitative exposure assessments are needed to better quantify and confirm exposures within such studies, as well as to permit interpretations and comparisons across studies. Journal of Exposure Analysis and Environmental Epidemiology (2000) 10, 66 ± 85.

“…The Two -Stage Method To avoid the problems in the non hierarchical method, we can aggregate individual health data to the same level as the pollution measures (Dockery et al, 1989 ). The first stage is therefore to calculate this aggregated measure.…”

Section: Three Approaches To Analysis In Hierarchical Studiesmentioning

confidence: 99%

“…Another method uses distance from a point source as the``dose'' (e.g., Bhopal et al, 1998;Dolk et al, 2000 ). A third uses the ambient air pollution within an area of residence as the estimate of exposure (e.g., Dockery et al, 1989;Zemp et al, 1999) . This third method requires a number of areas with differing levels of pollution, in order to have individuals at different exposures, and is increasingly used in the context of large collaborative projects.…”

Section: Introductionmentioning

confidence: 99%

Methodological approaches to the analysis of hierarchical studies of air pollution and respiratory health — examples from the CESAR study

Pattenden

Armstrong

Houthuijs

et al. 2000

Objectives: Many studies of air pollution and health are carried out over several geographical areas, and sometimes over several countries. This paper explores three approaches to analysis in such studies: a non hierarchical model, a two -stage analysis, and multilevel modelling. Illustrations are given using a preliminary subset of data from the CESAR study. Design: The Central European Study on Air pollution and Respiratory Health ( CESAR ) was conducted in 25 areas within six Central European countries, enrolling 20,271 schoolchildren. Pollution averages were calculated for each area. Associations between pollution and health outcomes were estimated under different models. Main results: A regression analysis of log FVC ( forced vital capacity ) on PM10, ignoring the geographical hierarchy, estimated a significant mean drop in FVC ( adjusted for confounders ) of 2.2% ( 95% CI 0.5% to 1.3% ) , p = 0.007, from the area with the lowest PM10 to that with the highest. A multilevel model ( mlm ) , using data for all children, but with random effects at area and country level, estimated a drop of 2.8% ( À 0.6% to 6.1% ) , p = 0.110. A two -stage analysis ( mean log FVC, adjusted for confounders, was estimated for each area using regression, and these means then regressed on PM10 ) estimated a drop of 2.6% ( À 0.5% to 5.5% ) , p = 0.101. Simulation exercises showed the non hierarchical method to be very inadequate in the context of the CESAR study, with only half of all 95% confidence intervals for the estimated PM10 slope containing the true value ( i.e., that used to create the simulated data ) . The two -stage and multilevel modelling methods gave results which were substantially better, though both underperformed slightly. All three methods appeared to give unbiased slope estimates. Conclusions: Acknowledgement of hierarchical structures is essential in statistical inference Ð standard errors can be substantially incorrect when they are ignored. Multilevel, random -effects models correctly address hierarchical structures, though having few units at higher levels can cause problems in convergence, especially where complex modelling is required. Two -stage analyses, acknowledging hierarchy, provide simple alternatives to random -effects models. Journal of Exposure Analysis and Environmental Epidemiology ( 2000 ) 10, 420 ± 426.