极端高温天气与宜昌市居民非意外死亡关系时间分层病例交叉分析

Association of extreme high temperature weather with non-accidental deaths among residents of Yichang city:a time-stratified case-crossover analysis

目的分析极端高温天气与湖北省宜昌市居民非意外死亡的关系,为降低居民非意外死亡的风险提供参考依据。方法收集宜昌市疾病预防控制中心死因系统记录的宜昌市城区2014年1月1日—2022年12月31日46170例非意外死亡者的相关数据以及同期宜昌市城区5个国控环境空气质量自动监测站每日的污染物监测数据和城区气象数据,采用时间分层病例交叉设计通过条件logistic回归模型分析温度不低于研究期间平均温度的P_(92.5)且持续1 d(P_(92.5)–1 d)、温度不低于研究期间平均温度的P_(92.5)且持续2 d(P_(92.5)–2 d)、温度不低于研究期间平均温度的P_(92.5)且持续3 d(P_(92.5)–3 d)、温度不低于研究期间平均温度的P_(95.0)且持续1 d(P_(95.0)–1 d)、温度不低于研究期间平均温度的P_(95.0)且持续2 d(P_(95.0)–2 d)、温度不低于研究期间平均温度的P_(95.0)且持续3 d(P_(95.0)–3 d)、温度不低于研究期间平均温度的P_(97.5)且持续1 d(P_(97.5)–1 d)、温度不低于研究期间平均温度的P_(97.5)且持续2 d(P_(97.5)–2 d)和温度不低于研究期间平均温度的P_(97.5)且持续3 d(P_(97.5)–3 d)共9种极端高温事件对宜昌市居民非意外死亡风险的影响。结果湖北省宜昌市46170例非意外死亡者的病例天数为46170 d,匹配的对照天数为157134 d;宜昌市46170例非意外死亡者中,P_(92.5)–1 d死亡3371例(7.30%),P_(92.5)–2 d死亡3041例(6.59%),P_(92.5)–3 d死亡2597例(5.62%),P_(95.0)–1 d死亡2436例(5.28%),P_(95.0)–2 d死亡2205例(4.78%),P_(95.0)–3 d死亡1960例(4.25%),P_(97.5)–1 d死亡1343例(2.91%),P_(97.5)–2 d死亡1148例(2.49%),P_(97.5)–3 d死亡882例(1.91%);在排除了湿度(RH)、细颗粒物(PM_(2.5))、可吸入颗粒物(PM10)、二氧化硫(SO_(2))、二氧化氮(NO_(2))、臭氧(O3)、一氧化碳(CO)以及死亡者性别、年龄等因素的影响后,时间分层病例交叉分析结果显示,P_(92.5)–1 d、P_(92.5)–2 d、P_(92.5)–3 d、P_(95.0)–1 d、P_(95.0)–2 d、P_(95.0)–3 d、P_(97.5)–1 d、P_(97.5)–2 d和P_(97.5)–3 d等9种极端高温事件均与宜昌市居民非意外死亡风险呈正相关且关联具有滞后性,除P_(92.5)–2 d和P_(92.5)–3 d极端高温事件的作用持续到Lag3 d外,其他7种极端高温事件的作用均持续到Lag1 d,9种极端高温事件均以Lag0 d的效应值为最大,分别为P_(92.5)–1 d(OR=1.09,95%CI=1.04~1.14)、P_(92.5)–2 d(OR=1.08,95%CI=1.03~1.14)、P_(92.5)–3 d(OR=1.11,95%CI=1.05~1.16)、P_(95.0)–1 d(OR=1.15,95%CI=1.09~1.22)、P_(95.0)–2 d(OR=1.17,95%CI=1.10~1.24)、P_(95.0)–3 d(OR=1.17,95%CI=1.10~1.25)、P_(97.5)–1 d(OR=1.23,95%CI=1.15~1.32)、P_(97.5)–2 d(OR=1.27,95%CI=1.18~1.38)和P_(97.5)–3 d(OR=1.31,95%CI=1.19~1.43);分层分析结果显示,极端高温事件对宜昌市居民非意外死亡的影响在性别亚组和年龄亚组中依然存在,男性和女性居民均以P_(97.5)–3 d的Lag0 d效应值为最大,分别为男性(OR=1.36,95%CI=1.21~1.54)和女性(OR=1.24,95%CI=1.08~1.43),高年龄组居民对极端高温事件更为敏感,P_(92.5)–1 d、P_(92.5)–2 d和P_(92.5)–3 d极端高温事件仅对70~79岁和≥80岁年龄组居民非意外死亡有显著影响,P_(92.5)–1 d在Lag0 d的效应值分别为70~79岁(OR=1.17,95%CI=1.06~1.29)和≥80岁(OR=1.19,95%CI=1.10~1.29),P_(92.5)–2 d在Lag0 d的效应值分别为70~79岁(OR=1.15,95%CI=1.05~1.28)和≥80岁(OR=1.22,95%CI=1.12~1.32),P_(92.5)–3 d在Lag0 d的效应值分别为70~79岁(OR=1.15,95%CI=1.04~1.29)和≥80岁(OR=1.23,95%CI=1.13~1.34)。结论极端高温天气可增加宜昌市居民非意外死亡的风险。

Objective To analyze the correlation between extreme high temperature weather and non-accidental deaths among residents in Yichang city,Hubei province,and to provide a reference for reducing the risk of non-accidental deaths.Methods Data on 46170 non-accidental deaths in urban areas of Yichang from 2014 to 2022 were collected from the death registration system of the Yichang Center for Disease Control and Prevention.Daily air pollution monitoring data from five national automatic air quality monitoring stations,as well as urban meteorological data during the same period,were also collected.A time-stratified case-crossover design with a conditional logistic regression model was used to analyze the effects of nine extreme high temperature events on the risk of non-accidental death among Yichang residents.These events included daily mean ambient air temperatures not lower than the 92.5th/95.0th/97.5th percentile during the study period and lasting 1,2,or 3 days,abbreviated as P_(92.5)/95.0/97.5-1/2/3 d.Results The analysis included 46170 case days and 157134 matched control days.Of the 46170 non-accidental deaths registered during the period,3371(7.30%)occurred during the P_(92.5)-1 d event,followed by 3041(6.59%)during P_(92.5)-2 d,2597(5.62%)during P_(92.5)-3 d,2436(5.28%)during P_(95.0)-1 d,2205(4.78%)during P_(95.0)-2 d,1960(4.25%)during P_(95.0)-3 d,1343(2.91%)during P_(97.5)-1 d,1148(2.49%)during P_(97.5)-2 d,and 882(1.91%)during P_(97.5)-3 d,respectively.After controlling for relative humidity(RH),fine particulate matter≤2.5μm in aerodynamic diameter(PM_(2.5)),inhalable particulate matter≤10μm in aerodynamic diameter(PM10),sulfur dioxide(SO_(2)),nitrogen dioxide(NO_(2)),ozone(O3),carbon monoxide(CO),and the gender and age of the deceased,the time-stratified case-crossover analysis showed that all nine extreme high temperature events were positively correlated with increased risk of non-accidental deaths among Yichang residents,with a lag effect Except for the P_(92.5)-2 d and P_(92.5)-3 d events,which lasted until lag day 3,the effects of the other seven extreme high temperature events lasted until lag day 1.The maximum effect values for all nine extreme high temperature events were at lag day 0,with odds ratios(OR)and their 95%confidence intervals(CI)of 1.09(1.04–1.14)for P_(92.5)-1 d,1.08(1.03–1.14)for P_(92.5)-2 d,1.11(1.05–1.16)for P_(92.5)-3 d,1.15(1.09–1.22)for P_(95.0)-1 d,1.17,(1.10–1.24)for P_(95.0)-2 d,1.17(1.10–1.25)for P_(95.0)-3 d,1.23(1.15–1.32)for P_(97.5)-1 d,1.27(1.18–1.38)for P_(97.5)-2 d,and 1.31(1.19–1.43)for P_(97.5)-3 d.Stratified analysis showed that the effects of extreme high temperature events on non-accidental deaths persisted in different sex and age subgroups.For both male and female subgroups,the effect of P_(97.5)-3 d was strongest at lag day 0,with ORs(95%CI)of 1.36(1.21–1.54)for males and 1.24(1.08–1.43)for females.Older residents were more sensitive to extreme high temperature events.The P_(92.5)-1 d,P_(92.5)-2 d,and P_(92.5)-3 d events had significant effects on non-accidental deaths only in residents aged 70–79 and≥80 years,with the ORs(95%CI)of 1.17(1.06–1.29)and 1.19(1.10–1.29)for P_(92.5)-1 d at lag day 0 for residents aged 70–79 and≥80 years,followed by 1.15(1.05–1.28)and 1.19(1.10–1.29)for P_(92.5)-2 d,1.15(1.04–1.29)and 1.23(1.13–1.34)for P_(92.5)-3 d.Conclusion Extremely high temperatures may increase the risk of non-accidental deaths among Yichang residents.