Objective: To estimate the effect of daily diurnal temperature range (DTR) on mortality in different areas in China. Methods: A time series study using the data collected from 66 areas in China was conducted, and Meta-analysis was used to analyze the estimates of associations between DTR and daily mortality. Modifying effects of extremely low and high DTR-mortality relationship by season and socioeconomic status (SES) were also evaluated respectively. Cumulative excess risk (CER) was used as an index to evaluate the effects. Results: The information about 1 260 913 registered deaths were collected between 1 January 2006 and 31 December 2011, we found the relationship between extreme DTR and mortality was non-linear in all regions and the exposure-response curve was J-shaped. In central and south areas of China, the result indicated the obvious acute effect of extremely high DTR, and the mortality effect in central area (CER=5.1%, 95% CI: 2.4%-7.9%) was significant higher than that in south area (CER=4.5%, 95% CI: 1.7%-7.3%). Regarding to the modification of seasons, the cumulative mortality effect of DTR in cold season (CER=5.8%, 95%CI: 2.5%-9.2%) was higher than that in hot season (CER=3.1%, 95%CI: 1.1%-5.1%). Generally, deaths among the elderly (≥75 years) were associated more strongly with extremely high DTR. Conclusions: The mortality effects of extremely DTR in different areas and seasons showed different characteristics, that in central area and in cold season it was significantly stronger. After modified by season and SES, DTRs were the greatest threat to vulnerable population, especially to the elderly (≥75 years). Therefore, more attention should be paid to vulnerable groups and protection measures should be taken according to the local and seasonal conditions.
目的： 了解中国不同城市日温差对居民死亡风险的影响。 方法： 用时间序列方法分析中国66个县／区日温差与居民日死亡数的关系，利用Meta分析方法整合不同县／区的结果。在调整季节、年龄、性别和死亡地影响后，以累计超额危险度（CER）为指标，分析极端日温差对死亡的影响。 结果： 研究共收集1 260 913例死亡信息，北部、中部和南部县／区的日均死亡人数分别为6.1、9.7和10.7人，日温差均值为11.1、8.0及8.2 ℃。在全国范围内，日温差与居民死亡存在非线性关系，其暴露-反应曲线呈J形，即极高日温差存在显著死亡风险（最大CER＝3.6%，95%CI：1.9%～5.3%），而极低日温差的死亡效应差异无统计学意义；且极高日温差对人群的死亡风险有明显的时空异质性，对中部县／区人群的影响（CER＝5.1%，95%CI：2.4%～7.9%）大于南部（CER＝4.5%，95%CI：1.7%～7.3%），在冬春季的效应高于夏秋季；老年人（≥75岁）为极高日温差的脆弱人群。 结论： 极端日温差对人群的死亡风险存在时空异质性，不同县／区的极高日温差均对居民死亡风险造成影响。对中部的影响高于南部和北部，冬春季的影响强于夏秋季，对老年人等脆弱人群影响最大，应因时因地制定政策保护脆弱人群。.
Keywords: Death; Diurnal temperature range; Temperature; Time series.