Time-series study on the impacts of temperature changes between neighboring days on mortality in 21 regions of China

Objective To investigate the impact of temperature changes between neighboring days (TCN) on mortality.

Methods Data of daily meteorology, air pollution and cause-of-death were collected for 21 regions of China in 2014—2018. The distributed lag non-linear model(DLNM) and the multivariate meta-analysis were used to investigate the impact of TCN on total daily deaths in different seasons.

Results The study showed a significant impact of TCN on total daily deaths in different seasons, with different thresholds. During the cold season, P₉₅ TCN (warming) had a 14-day cumulative relative risk (CRR) of 0.868 (95% CI: 0.794-0.948), while P₅ TCN (cooling) had no significant impact. During the warm season, P₉₅ TCN (warming) had a 7-day CRR of 1.078 (95% CI: 1.009-1.152), while P₅ TCN (cooling) had a 7-day CRR of 0.929 (95% CI: 0.889-0.971). During the cold season, the population with respiratory diseases was more likely to be affected by temperature changes, and during the warm season, the population with circulatory diseases, women, and the population aged ≥65 years were more sensitive to temperature changes. The analysis of the north and south regions showed that the population in northern cities was more sensitive to the impact of P₉₅ TCN.

Conclusion Extreme TCN in different seasons is associated with the risk of death among populations; warming in the cold season can reduce the risk of death, while in the warm season, warming can increase the risk of death and cooling can reduce the risk of death. Health promotion strategies should take into account the impact of temperature changes between neighboring days on mortality.