2001-2011年武汉市寒潮天气对江岸区居民死亡的急性影响

Acute impact of cold spells on mortality during 2001-2011 in Jiang＇an district of Wuhan, China

目的：分析2001—2011年武汉市寒潮天气对江岸区居民死亡的急性影响。方法收集武汉市江岸区2001至2010年每年12月和2002至2011年每年1—3月的居民死亡监测资料，根据国际疾病分类编码将死亡居民以不同死因进行分类，提取其中非意外死亡、心血管疾病死亡、呼吸系统死亡例数，并收集研究时期内国家气象中心的气象因素以及武汉市环境监测中心的大气污染物数据，将寒潮定义为日均气温不高于研究时间内日均气温的P5（2.58℃），且持续不少于5 d。应用分布滞后非线性模型（DLNM）评价寒潮天气对居民死亡的急性影响。结果研究期间，非意外死亡人数共17119例，其中心血管疾病死亡7968例（占46.5%），65岁以上老年人9403例（占75.5%）。研究期内共发生13次寒潮事件，寒潮持续时间合计111 d。非寒潮日、2008年寒潮和其他年份寒潮期间的日均气温分别为（8.2±4.5）、（-0.7±1.4）和（0.8±1.2）℃；日均死亡例数分别为（14.0±4.2）、（18.2±4.5）、（14.9±4.9）例。DLNM模型分析结果显示，寒潮会增加全人群非意外死亡风险，与非寒潮日相比，2008年寒潮累积滞后0~27 d的死亡风险较高[RR=1.56（95%CI：1.36~1.79）]，且高于其他年份寒潮日（RR=1.23，95%CI：1.08~1.41）。2008年寒潮期间，心血管疾病死亡和≥65岁人群发生死亡风险的RR （95%CI）值分别为1.96（1.62~2.37）、1.67（1.43~1.95），男性和女性发生死亡风险的RR（95%CI）值分别为1.60（1.33~1.92）、1.50（1.23~1.84）。结论寒潮天气增加了武汉市居民非意外死亡风险，且心血管疾病患者和老年人群可能对寒潮的影响更为敏感。

Objective To evaluate the association between cold spells and nonaccidental mortality from 2001 to 2011 in the Jiang＇an District of Wuhan, China. Methods We collected mortality data for December 2001 to 2010 and January to March 2002 to 2011 in the study area. According to the International Classification of Diseases, we stratified the mortality data into three cause-specific categories：nonaccidental mortality, cardiovascular mortality, and respiratory mortality. We also obtained meteorological data （from the China Meteorological Administration） and ambient pollution data （from the Wuhan Environmental Monitoring Center） during the same periods. In the present study, a cold spell was defined as 5 or more consecutive days with daily average temperatures below the 5th percentile of daily mean temperatures （2.58℃） from January to December in 2001-2011. A distributed lag nonlinear model （DLNM） was applied to assess the acute effect of cold spells on daily nonaccidental mortality. Results During the study period, the total number of nonaccidental deaths was 17 119, including 9 403 （75.5%） among individuals aged over 65 years; 7 968 （46.5%） people died of cardiovascular disease. According to this definition, there were a total of 13 cold spell events and 111 days of duration in Wuhan during the study period. Study days were divided into three periods： non-cold spell days, 2008 cold spell days, and cold spell days in other years. Average daily mean temperatures of the above three periods were（8.2±4.5）,（-0.7±1.4）, and（0.8±1.2）℃,＆amp;nbsp;respectively, corresponding to average daily deaths of 14.0 ± 4.2, 18.2 ± 4.5, and 14.9 ± 4.9 for nonaccidental mortality. After adjusting for long-term trends, seasonal trends, weekdays, holidays, and relative humidity, analysis by the DLNM revealed that cold spells were associated with increased mortality risk, with a cumulative relative risk （RR） of 1.56 （95%CI：1.36-1.79） at lag 0-27 days in 2008, higher than that in other years with 1.23 （95%CI：1.08-1.41）. Cold spells were not significantly associated with respiratory mortality and people under 65 years of age;however, during the 2008 cold spell RR increased to 1.96 （95%CI：1.62-2.37） and 1.67 （95%CI： 1.43-1.95） for cardiovascular mortality and older adults （≥65 years old）, respectively;both males and females had high mortality risk, with RRs of 1.60 （95%CI：1.33-1.92） and 1.50 （95%CI： 1.23-1.84）, respectively. The association between cold spells and mortality remained nearly unchanged with and without adjustment for ambient pollutants （PM10, SO2, and NO2） in the DLNMs. Conclusion In Wuhan, both the 2008 cold spell and cold spells in other years were significantly associated with increased nonaccidental mortality. People with cardiovascular disease and elderly adults may be more susceptible to the impact of cold spells on mortality.