The burden of stroke mortality attributable to cold and hot ambient temperatures: Epidemiological evidence from China

Few data are available on the attributable burden, such as absolute excess or relative excess, of stroke death due to temperature.

We collected data on daily temperature and stroke mortality from 16 large Chinese cities during 2007-2013. First, we applied a distributed lag non-linear model to estimate the city-/age-/gender-specific temperature-mortality association over lag 0-14days. Then, pooled estimates were calculated using a multivariate meta-analysis. Attributable deaths were calculated for cold and heat, defined as temperatures below and above the minimum-mortality temperature (MMT). Moderate and extreme temperatures were defined using cut-offs at the 2.5th and 97.5th percentiles of temperature.

The city-specific MMT increased from the north to the south, with a median of 24.9(o)C. Overall, 14.5% (95% empirical confidence interval: 11.5-17.0%) of stroke mortality (114, 662 deaths) was attributed to non-optimum temperatures, with the majority being attributable to cold (13.1%, 9.7-15.7%). The proportion of temperature-related death had a decreasing trend by latitude, ranging from 22.7% in Guangzhou to 6.3% in Shenyang. Moderate temperatures accounted for 12.6% (9.1-15.3%) of stroke mortality, whereas extreme temperatures accounted for only 2.0% (1.6-2.2%) of stroke mortality. Estimates of death burden due to both cold and heat were higher among males and the elderly, compared with females and the youth.

The burden of temperature-related stroke mortality increased from the north to the south. Most of this burden was caused by cold temperatures. The stroke burden was higher among males and the elderly. This information has important implications for preventing stroke due to adverse temperatures in vulnerable subpopulations in China.