中文摘要：

目的分析大气中主要气态污染物二氧化氮（NO2）急性暴露对北京市居民每日呼吸系统疾病死亡人数的影响。方法收集北京市2006--2009年居民每日呼吸系统疾病死亡资料、大气主要污染物（so2、NO2、PM10）日均浓度资料和气象因素（日均气温、日均相对湿度）资料。采用时间序列的广义相加模型，通过平滑样条函数控制时间趋势、“星期几效应”和气象因素等混杂因素，拟合大气NO2浓度与居民每日呼吸系统疾病死亡人数之间的暴露-反应关系模型。对确定的最优拟合模型进行敏感度检验以检测其稳定性，并进一步建立双污染物模型，以检验NO2对居民每日呼吸系统疾病死亡人数的效应是否受到其他大气污染物的影响。结果NO2浓度与北京市居民每日呼吸系统疾病死亡人数的关联元统计学意义（P〉0．05），但与居民每日肺炎死亡人数之间有统计学意义（P〈0．05）。建立的最优拟合模型稳定。日均NO2浓度每升高10μg/m^3，居民每日肺炎死亡人数增加1．46％（95％CI：0．04％-2．90％）。在模型中加入PM10后，NO2对居民每日肺炎死亡的效应有增加，居民每日肺炎死亡人数增幅增加至1．88％（95％CI：0．12％-3．68％）。结论北京市大气中NO2浓度升高可能导致居民每日肺炎死亡人数增加。

英文摘要：

Objective To explore the association between ambient nitrogen dioxide （NO2） level and population mortality of respiratory disease in Beijing. Methods The data of daily respiratory disease （respiratory disease and pneumonia ） mortality of Beijing in 2006-2009 was provided by China CDC. The data of daily air pollutants （SO2, NO2 and PM10） concentration and meteorology at the same time periods were collected from the local departments of Beijing. After adjustment of the confounding of long-term trend, weather conditions and weekdays effect, generalized additive models （GAM） were used to explore the relationship between ambient level of NO2 and respiratory disease mortality. Sensitivity analysis was done to test the stability of the best fitted model. By setting up a double-pollutants model, it was tested that weather the effect of NO2 on daily respiratory disease mortality was affected by other pollutants. Results Ambient NO2 concentration was significantly associated with daily pneumonia mortality in Beijing （P〈0.05）. There was no statistical significant association between NO2 level and daily respiratory disease mortality. The best fitted GAM model was stable. An increase of 10 μg/m^3 in NO2 was associated with 1.46%（ 0.04%- 2.90%） increase in daily pneumonia mortality. Slight increase of the effect of NO2 on daily pneumonia mortality was observed when PM10 was added into the model. Conclusion The level of ambient NO2 can increase the daily pneumonia mortality of residents in Beijing.