Long-term Exposure to Ambient Temperature and Mortality Risk in China: A Nationwide Study Using the Difference-in-differences Design.
Jianxiong Hu,Maigeng Zhou,Mingfang Qin,Shilu Tong,Zhulin Hou,Yanjun Xu,Chunliang Zhou,Yize Xiao,Min Yu,Biao Huang,Xiaojun Xu,Lifeng Lin,Tao Liu,Jianpeng Xiao,Weiwei Gong,Ruying Hu,Junhua Li,Donghui Jin,Qinglong Zhao,Peng Yin,Yiqing Xu,Weilin Zeng,Xing Li,Guanhao He,Cunrui Huang,Wenjun Ma
DOI: https://doi.org/10.1016/j.envpol.2021.118392
IF: 8.9
2021-01-01
Environmental Pollution
Abstract:The short-term effects of ambient temperature on mortality have been widely investigated. However, the epidemiological evidence on the long-term effects of temperature on mortality is rare. In present study, we conducted a nationwide quasi-experimental design, which based on a variant of difference-in-differences (DID) approach, to examine the association between long-term exposure to ambient temperature and mortality risk in China, and to analyze the effect modification of population characteristics and socioeconomic status. Data on mortality were collected from 364 communities across China during 2006 2017, and environmental data were obtained for the same period. We estimated a 2.93 % (95 % CI: 2.68 %, 3.18 %) increase in mortality risk per 1 degrees C decreases in annual temperature, the greater effects were observed on respiratory diseases (5.16 %, 95 % CI: 4.53 %, 5.79 %) than cardiovascular diseases (3.43 %, 95 % CI: 3.06 %, 3.80 %), and on younger people (4.21 %, 95 % CI: 3.73 %, 4.68 %) than the elderly (2.36 %, 95 % CI: 2.06 %, 2.65 %). In seasonal analysis, per 1 degrees C decreases in average temperature was associated with 1.55 % (95 % CI: 1.23 %, 1.87 %), -0.53 % (95 % CI: -0.89 %, -0.16 %), 2.88 % (95 % CI: 2.45 %, 3.31 %) and 4.21 % (95 % CI: 3.98 %, 4.43 %) mortality change in spring, summer, autumn and winter, respectively. The effects of long-term temperature on total mortality were more pronounced among the communities with low urbanization, low education attainment, and low GDP per capita. In total, the decrease of average temperature in summer decreased mortality risk, while increased mortality risk in other seasons, and the associations were modified by demographic characteristics and socioeconomic status. Our findings suggest that populations with disadvantaged characteristics and socioeconomic status are vulnerable to long-term exposure of temperature, and targeted policies should be formulated to strengthen the response to the health threats of temperature exposure.