京津两地不同温度水平下大气PM_(10)与居民每日死亡关系的时间序列研究

Time-series analysis of the association between PM_(10) and daily mortality at different ambient temperature level in Beijing and Tianjin

目的探讨北京市和天津市不同温度水平下大气可吸入颗粒物(PM_(10))与居民每日死亡的定量关系。方法收集北京市和天津市2007年1月1日至2009年12月31日的每日大气PM_(10)、每日平均温度、每日平均相对湿度、每日居民不同病因别死亡人数,建立时间序列数据库。将每日温度分成低(<P_(50))、中(P_(50)~P_(75))、高(>P_(75))3个层次,采用广义相加模型,分析两城市大气PM_(10)与居民每日不同病因别死亡的相关性。结果对于总死亡,北京市和天津市高、中温水平下大气PM_(10)的超额危险度(excess risk,ER)明显高于低温水平,且中温水平下效应值最大,两城市的ER分别为0.46%(95%CI:0.18%~0.75%)和0.87%(95%CI:0.34%~1.41%);对于循环系统死亡,北京市和天津市高、中温水平下大气PM_(10)的超额危险度明显高于低温水平,且高温水平下的超额危险度值更大,两城市的ER分别为0.52%(95%CI:0.07%~0.98%)和1.13%(95%CI:0.43%~1.83%);对于呼吸系统死亡,北京和天津市中温水平下大气PM_(10)的超额危险度最大,两城市的ER分别为0.99%(95%CI:0.16%~1.74%)和1.48%(95%CI:0.25%~1.95%)。结论不同温度水平下,北京市和天津市大气PM_(10)污染致居民死亡的急性效应不同,中高温水平下,大气PM_(10)的健康效应值更大。制定人群大气污染健康效应公共政策时,需考虑温度对其影响的修饰作用。

Objective To explore the quantitative association between thoracic particulate matter（PM_（10）） and daily mortality at different ambient temperature in Beijing and Tianjin, China. Methods Surveillance data on daily average PM_（10）, ambient temperature, relative humidity and mortality were obtained from January 1, 2007 to December 31, 2009 in Beijing and Tianjin to establish time-series database. Daily ambient temperature data was divided into three levels,such as low（P_（50））, moderate（P_（50）-P_（75））, high（P_（75）） temperature levels, and generalized additive model was used to explore the associations between the two urban ambient PM_（10） and daily mortality. Results The effects of PM_（10） on total mortality were stronger on high and moderate temperature level days than that on low temperature level days in Beijing and Tianjin, the maximum effects were found on the moderate ambient temperature level days and the excess risks were 0.46%（95%CI： 0.18%-0.75%） and 0.87%（95%CI： 0.34%-1.41%）, respectively. The effects of PM_（10） on cardiovascular system mortality were stronger on high and moderate temperature level days than that on low temperature level days in Beijing and Tianjin, the maximum effects were found on the high ambient temperature level days and the excess risks were 0.52%（95% CI： 0.07%-0.98%） and 1.13%（95% CI： 0.43%-1.83%）,respectively. The maximum effects of PM_（10） on respiratory system mortality were found on the moderate ambient temperature level days in Beijing and Tianjin and the excess risks were 0.99%（95%CI：0.16%-1.74%） and 1.48%（95% CI： 0.25%-1.95%）,respectively. Conclusions At different ambient temperature level, the acute effects of PM_（10） on mortality were different, and these effects were stronger on the high and moderate temperature level days than that on the low temperature level days. This result suggests that the modified effects of the temperature should be considered during the air pollution related public policymaking.