用不同指标度量隔日气温变异对居民寿命损失年影响的比较研究
Comparison of indicators in predicting impact of temperature variability between neighboring days on years of life lost
-
摘要:
背景 大量研究表明气温是影响人群健康的重要因素,而气温变异,尤其隔日气温变异对人群健康影响的研究较少。
目的 比较隔日温差(
TCN ,隔日平均气温之差)、气温变异(TV ,隔日最高气温和最低气温的标准差)以及本研究新提出的根据隔日气温变异的方向和效应大小计算得到的隔日温度总变异(TTV )这三个隔日气温变异指标与居民寿命损失年(YLL )的暴露-反应关系,探索能更好反映隔日气温变异对居民死亡影响的指标。方法 收集2013-2017年广东省40个区(县)气象数据以及死亡登记资料。采用分布滞后非线性模型(DLNM)和多变量meta分析的两阶段分析方法,分别拟合日夜温差和夜日温差与
YLL 率(每10万人口YLL 值)的暴露-反应关系,提取日夜温差和夜日温差的归因YLL 率作为各自权重计算TTV 。计算Pearson相关系数,分析三个隔日气象变异指标间的相关性。采用DLNM和多变量meta分析两阶段分析方法,分别分析TCN 、TV 和TTV 与居民YLL 率的暴露-反应关系,比较不同隔日气温变异指标对人群死亡影响的差异。结果 研究期间内广东省40个区(县)日均
YLL 率为22.3/10万。经计算,TCN 平均值为(0.0±1.8)℃,TV 为4.6±1.5,TTV 平均值为(8.1±2.7)℃,三个指标均趋近正态分布。TCN 与TV 和TTV 相关性较弱(r =0.097 9,r =0.088 0),而TV 与TTV 相关性较强(r =0.889 1)。在控制平均气温的滞后效应后,TCN 与YLL 率的暴露-反应关系无统计学意义,而TV 和TTV 与YLL 率的暴露-反应关系有统计学意义。TV -YLL 和TTV -YLL 的暴露-反应关系曲线相似,均呈类似"U"型关系,过低或过高的TV 和TTV 均会增加人群的YLL 率。极端低(第5百分位数)的TV (TV =2.2)和TTV (TTV =2.8℃)的归因YLL 率及其95%CI 依次为1.0/10万(0.1/10万~1.9/10万)和2.1/10万(0.2/10万~4.0/10万),极端高(第95百分位数)的TV (TV =7.2)和TTV (TTV =12.1℃)的归因YLL 率效应值及其95%CI 依次为3.1/10万(1.2/10万~5.1/10万)和4.1/10万(2.3/10万~5.8/10万),在极端低和极端高节点上,TTV 的YLL 率效应值均大于TV ,而在中等低和中等高节点上,两个指标的效应相近。结论 TCN 、TV 、TTV 与YLL 的暴露-反应关系存在差异,其中TTV 综合考虑了气温变异的程度、方向以及健康效应,更加全面地反映了短时气温变异对人群健康的影响。Abstract:Background Numerous epidemiological studies have demonstrated a significant association between ambient temperature and population health, but evidence is limited for the health impact of temperature variability between neighboring days.
Objective The study compares exposure-response associations of years of life lost (
YLL ) with different indicators of temperature variability between neighboring days, including temperature change between neighboring days (TCN , difference of mean temperature between neighboring days), temperature variability (TV , the standard deviation of maximum and minimum temperatures between neighboring days), and total temperature variability between neighboring days (TTV ) according to the directions and effects of temperature variability between neighboring days which was developed in the present study. The study aims to explore which measure can better assess the impact of temperature variability between neighboring days on mortality.Methods Death registration data and meteorological data during 2013-2017 were collected from 40 districts/counties in Guangdong, China. The exposure-response association of diurnal temperature range with
YLL rate (YLL per 100 000 population) and the association of nocturnal temperature range withYLL rate were investigated using a two-stage approach including distributed lag non-linear model (DLNM) and multivariable meta-analysis. ThenTTV was weighted by attributableYLL rate of diurnal temperature range and nocturnal temperature range. The correlations of the three indicators of temperature variability between neighboring days were examined by Pearson correlation analysis. The exposure-response associations ofYLL rate withTCN ,TV , andTTV were evaluated by DLNM model and multivariable meta-analysis. The effects of different indicators of temperature variability between neighboring days on mortality were compared.Results The daily average
YLL rate of the 40 study locations in Guangdong Province was 22.3 per 105 inhabitants during the study period. The means ofTCN ,TV , andTTV were (0.0±1.8)℃, 4.6±1.5, and (8.1±2.7)℃, respectively. These indicators all approximated a normal distribution.TCN had a weak correlation withTV andTTV (r =0.097 9,r =0.088 0), andTV had a strong correlation withTTV (r =0.889 1). The exposure-response association betweenTV andYLL rate was insignificant after controlling the overall lag effect of temperature, butTV andTTV were statistically associated withYLL rate. Both theTV -YLL and theTTV -YLL exposure-response relationships were U-shaped, suggesting that both ends ofTV andTTV increasedYLL rate in the study population. The attributableYLL rates of extremely lowTV (P 5,TV =2.2) and extremely highTV (P 95,TV =7.2) were 1.0/105 (95%CI :0.1/105-1.9/105) and 3.1/105 (95%CI :1.2/105-5.1/105) respectively, and both were lower than the attributableYLL rates of extremely lowTTV (P 5,TTV =2.1℃) (2.1/105, 95%CI :0.2/105-4.0/105) and extremely highTTV (P 95,TTV =12.1℃) (4.1/105, 95%CI :2.3/105-5.8/105). The effects of moderately low or highTV andTTV were similar.Conclusion Associations of
YLL withTCN ,TV , andTTV are inconsistent.TTV takes both degree and direction of temperature variability into account, and is a better predictor of the impact of temperature variability between neighboring days on human health.