A systematic review and meta-analysis of the association between daily mean temperature and mortality in China

We summarized the evidence on the effects of heat and cold exposures on mortality in China. We included studies published on this topic in both Chinese and English, thereby filling a gap in knowledge using data from a country that consists of one-fifth of the world's population.We conducted a systematic search of peer-reviewed studies on the association between daily mean temperature and mortality published from 2001 up to July 2018. We searched one Chinese database (China National Knowledge infrastructure, http://www.cnki.net) and three English databases (PubMed, Scopus, Web of Science). We converted the effect estimates of heat/cold to rate ratios (RRs) associated with 1° increase/decrease beyond the heat/cold reference temperatures. For studies that provided lag-specific estimates, we used both the maximum and minimum of RR estimates. We calculated summary effect estimates for all-cause and cause-specific mortalities, as well as RRs stratified by sex, age, and socioeconomic status. We also investigated patterns of heat and cold adaptation at different latitudes, and at different reference temperatures.In total, 45 articles were included in this systematic review. For every 1° temperature increase/decrease beyond reference points, the rate of non-accidental mortality increased by 2% (RR, 1.02; 95% confidence interval (95% CI [1.01–1.02]) for heat and 4% (RR, 1.04; 95% CI [1.03–1.04]) for cold, respectively; the rate of cardiovascular mortality increased 3% (RR, 1.03; 95% CI [1.03–1.04]) for heat and 6% (RR, 1.06; 95% CI [1.04–1.07]) for cold; the rate of respiratory mortality increased 2% (RR, 1.02; 95% CI [1.01–1.03]) for heat and 2% (RR, 1.02; 95% CI [1.00–1.04]) for cold; the rate of cerebrovascular mortality increased 2% (RR, 1.02; 95% CI [1.02–1.03]) for heat and 3% (RR, 1.03; 95% CI [1.02–1.04]) for cold. We identified a variation in optimal temperature range related to latitude of the residential area, and differences in people's capability to adapt to heat versus cold.We found consistent evidence of the association between temperature and mortality, as well as evidence of patterns in human adaptation, and we discussed the implications of our findings.