Diurnal temperature range in relation to death from stroke in China

Stroke is the second leading cause of death in the world. It has multiple risk factors of which some, such as ambient temperature, are less well documented.

We aimed to examine the association between diurnal temperature range (DTR) and stroke mortality, and to test the possible effect modification of this association according to gender, age and educational level.

Daily data on weather and stroke mortality from 16 provincial capital cities in China for the years 2007-2013 were obtained, with a total of 788,783 deaths from stroke. A quasi-Poisson generalized linear regression combined with a distributed lag non-linear model was used to examine the city-specific DTR effect on stroke mortality. The pooled effects of DTR on stroke mortality were then obtained using a meta-analysis, which was based on restricted maximum likelihood estimation.

The DTR impacts were generally limited to a period of eight days, while significant effects during lag 0-8 days were only found in the cities of Beijing, Zhengzhou, Nanjing, Hefei, Chongqing and Changsha. The DTR effects were significantly and negatively associated with latitudes at lag 0-10 days (r_s = - 0.640, P = 0.008). An increase of 1 °C in DTR was associated with pooled estimate of 0.66% (95%CI: 0.28-1.05%), 0.12% (- 0.26% to 0.51%) and 0.67% (0.26-1.07%) increases in stroke mortality at lag 0-10 days during the total, hot and cold days, respectively. The impact of DTR was much higher in southern China than in northern China [1.02% (0.62% to 1.43%) versus 0.10% (-0.27% to 0.47%) ]. For the individual characteristics, only females, the elderly aged ≥ 65 years, and those with lower educational attainment were vulnerable to DTR.

DTR has considerable effects on risk of mortality from stroke in various cities in China, especially among the elderly, females, those with low educational level, and people living in southern China. The results can inform decisions on developing programs to protect vulnerable subpopulations from adverse impacts of DTR.