沙尘天气颗粒物对呼吸与心血管系统疾病日入院人数的影响

Association of Particulate Matter Derived from Dust Events with Daily Respiratory and Cardiovascular Hospitalization

[目的]研究沙尘天气及沙尘颗粒物（PM2.5、PM10）与呼吸、心血管系统疾病日入院人数的关系。[方法]采用时间序列的半参数广义相加模型（Semi-parametric generalized additive model，GAM），在控制了长期趋势、星期几效应（day of the week，DOW）以及气象因子等混杂因素的影响后，分析2004年春季沙尘PM2.5和PM10及其大气污染物（SO2和NO2）与呼吸、心血管系统疾病日入院人数的关系，并将沙尘PM2.5和PM10按浓度分类探讨不同沙尘天气对人体健康的影响。[结果]①沙尘PM2.5和PM10在单污染模型中对居民呼吸和心血管系统疾病日入院人数的影响在滞后第2天和第3天有统计学意义。多污染模型分析表明，引入SO2或（和）NO2后，PM10对呼吸和心血管系统疾病日入院人数的影响有统计学意义，PM2.5对心血管系统疾病日入院人数的影响有统计学意义。在模型中引入其他污染物后，SO2和NO2对呼吸和心血管系统日入院人数的影响均无统计学意义。②颗粒物浓度分类模型分析表明，沙尘天气PM2.5只有达到沙尘暴颗粒物水平（〉167．5μg／m^3）对男性呼吸和心血管系统疾病日入院人数的影响才有意义，而对女性的影响PM2.5在扬沙天气水平（95．4～167．5μg/m^3）就有意义。沙尘天气PM10对男、女性呼吸和心血管系统健康的影响在扬沙天气水平（150～250μg／m^3）就有意义，达到沙尘暴水平时，其影响达到最大。[结论]①在几乎没有工业污染的沙尘发生源区或近源区，其沙尘PM2.5、PM10只有达到较高浓度（扬沙天气水平）时，才会对居民呼吸和心血管系统疾病日入院人数有影响，且为滞后效应，并呈一定剂量效应关系，即：沙尘暴天〉扬沙天〉沙尘轻度污染天、清洁天；②本研究根据大气PM10和PM2.5水平划分沙尘天气类型．比以能见度划分更为科学且更易操作．同时也为我国制定PM10和PM2.5大气标准提出了科学依据。

[ Objective ] To explore the association of particulate matter （ PM2.5PM10 ） derived from spring-dust events with daily respiratory and cardiovascular hospitalization. [ Methods ] Using a semi-parametric generalized additive model and controlling for long-term temporal trends, day of the week and meteorological factors, counts of hospitalization for respiratory and cardiovascular disease were analyzed for PM2.5, PM10 and gaseous pollutants （ SO2 and NO2 ） in a Poisson regression. Then, Categorical model of particulate matter were processed to explore the health effect of particulate matter at different dust events levels. [ Results ] 1 ）PM10 and PM2.5 derived from the dust events with lag of 2 days and 3 days were significantly associated with respiratory and cardiovascular hospitalization in simple-pollutant model. The multi-pollutants model showed that after adjusted the effect of SO2 and/or NO2, PM10 was significantly associated with respiratory and cardiovascular disease hospitalization and PM2.5 was significantly associated with cardiovascular diseases, while the effects of PM2.5 on respiratory diseases became weaker. There were non-significantly association between SO2 and NO2 and respiratory and cardiovascular disease in multi-pollutants model. 2 ）PM2.5 at the dust storm levels（ 〉167.5 μg/m^3 ）had a significant association with the male respiratory and cardiovascular disease hospitalization and PM2.5 at the blowing dust levels （ 95.4-167.5 μg/m^3 ） had a significant association with the females. PM10 at the blowing dust levels （ 150-250 μg/m^3 ）had the significant associations with the respiratory and cardiovascular hospitalization in males and females. In addition, stronger health damage effects of particulate matters at the dust storm level were found. [ Conclusion ] 1 ） Only at higher concentrations（ blowing dust levels ）, the particulate matters derived from the source region and nearby of dust events, where there were not almost any industrial pollutant, had the increased lag-effects on respiratory and cardiovascular hospitalization in a dose-related manner： dust storm 〉 blowing dust 〉 low-dust pollution days and normal clear days. 2 ）In this paper, the categories of dust events were divided according to the levels of airborne PM10 and PM2.5. This setting was more scientific and easily operated; meanwhile, it provided new scientific evidence to set down the environmental quality standard of PM10 and PM2.5 in the atmosphere.