Burnett Richard T, Stieb Dave, Brook Jeffrey R, Cakmak Sabit, Dales Robert, Raizenne Mark, Vincent Renaud, Dann Tom
Biostatistics and Epidemiology Division, Safe Environments Directorate, Healthy Environments and Consumer Safety Branch, Health Canada, Ottawa, Canada.
Arch Environ Health. 2004 May;59(5):228-36. doi: 10.3200/AEOH.59.5.228-236.
The association between daily variations in ambient concentrations of nitrogen dioxide (NO2) and mortality was examined in 12 of Canada's largest cities, using a 19-yr time-series analysis (from 1981-1999). The authors employed parametric statistical methods that are not subject to the recently discovered convergence and error estimation problems of generalized additive models. An increase in the 3-d moving average of NO2 concentrations equivalent to the population-weighted study mean of 22.4 ppb was associated with a 2.25% (t = 4.45) increase in the daily nonaccidental mortality rate and was insensitive to adjustment for ozone, sulfur dioxide, carbon monoxide, coefficient of haze, size-fractionated particulate mass, and the sulfate ion measured on an every-6th-day sampling schedule. The 3-d moving average of NO2 was sensitive to adjustment for fine particulate matter measured daily during the 1998-2000 time period.
利用19年时间序列分析（1981年至1999年），在加拿大12个最大城市中研究了二氧化氮（NO₂）环境浓度的每日变化与死亡率之间的关联。作者采用了参数统计方法，这些方法不存在广义相加模型最近发现的收敛和误差估计问题。与人口加权研究平均值22.4 ppb相当的NO₂浓度3天移动平均值的增加，与每日非意外死亡率增加2.25%（t = 4.45）相关，并且对臭氧、二氧化硫、一氧化碳、雾霾系数、粒径分级颗粒物质量以及按每六天采样计划测量的硫酸根离子进行调整不敏感。在1998 - 2000年期间，NO₂的3天移动平均值对每日测量的细颗粒物调整敏感。
