Heat wave characteristics, mortality and effect modification by temperature zones: a time-series study in 130 counties of China

Background: The substantial disease burden attributed to heat waves, and their increasing frequency and intensity due to climate change, highlight the importance of understanding the health consequences of heat waves. We explore the mortality risk due to heat wave characteristics, including the timing in the seasons, the day of the heat wave, the intensity and the duration, and the modifying effect of temperature zones. Methods: Heat waves were defined as >= 2 days with a temperature >= 99th percentile for the county from 1 May through 30 September. Heat waves were characterized by their intensity, duration, timing in the season, and day of the heat wave. Within each county, we estimated the total non-accidental death and cardiovascular disease mortality during each heat wave compared with non-heat wave days by controlling for potential confounders in summer. We combined individual heat wave effect estimates using a random-effects model to calculate overall effects at the temperature zone and national levels. Results: The average daily total number of non-accidental deaths was nine in the warm season (across all the counties). Approximately half of the daily total number of nonaccidental deaths were cardiovascular-related deaths (approximately four persons per day). The average and maximum temperatures across the study area were 23.1 degrees C (range: -1.2-35.9 degrees C) and 28.3 degrees C (range: 5.4-42.8 degrees C), respectively. The average relative humidity during the study was 68.9% (range: 8.0-100.0%). Heat waves increase the risk of total non-accidental death by 15.7% [95% confidence interval (CI): 12.5, 18.9] compared with non-heat wave periods, and the risk of cardiovascular-related death increases by 22.0% (95% CI: 16.9, 27.4). The risk of non-accidental death during the first heat wave of the season increases by 16.3% (95% CI: 12.6, 20.2), the risk during the second heat wave increases by 6.3% (95% CI: 2.8, 9.9) and during subsequent heat waves increases by - 2.1% (95% CI: -4.6, 0.4). The first day and the second to third days of heat waves increase the risk of total non-accidental death by 11.7% (95% CI: 7.6, 15.9) and 17.0% (95% CI: 13.1, 21.0), respectively. Effects of heat waves on mortality lasted more than 4 days (6.3%, 95% CI: 2.4, 10.5) and are non-significantly different from the first day of heat waves. We found non-significant differences of the heat wave-associated mortality risks across mid-, warm and subtropical temperature zones. Conclusions: In China, the effect of heat waves on mortality is acute, and varies by certain characteristics of heat waves. Given these results, national heat wave early warning systems should be developed, as well as precautions and protection warranted according to characteristics of heat waves.