Achebak H, Devolder D, Ballester J (2018) Heat-related mortality trends under recent climate warming in Spain: A 36-year observational study. PLoS Med 15(7): e1002617.

Download press release.



Anthropogenic greenhouse gas emissions have increased summer temperatures in Spain by nearly one degree Celsius on average between 1980 and 2015. However, little is known about the extent to which the association between heat and human mortality has been modified. We here investigate whether the observed warming has been associated with an upward trend in excess mortality attributable to heat or, on the contrary, a decrease in the vulnerability to heat has contributed to a reduction of the mortality burden.

Methods and findings

We analysed a dataset from 47 major cities in Spain for the summer months between 1980 and 2015, which included daily temperatures and 554,491 deaths from circulatory and respiratory causes, by sex. We applied standard quasi-Poisson regression models, controlling for seasonality and long-term trends, and estimated the temporal variation in heat-related mortality with time-varying distributed lag nonlinear models (DLNMs). Results pointed to a reduction in the relative risks of cause-specific and cause-sex mortality across the whole range of summer temperatures. These reductions in turn explained the observed downward trends in heat-attributable deaths, with the only exceptions of respiratory diseases for women and both sexes together. The heat-attributable deaths were consistently higher in women than in men for both circulatory and respiratory causes. The main limitation of our study is that we were not able to account for air pollution in the models because of data unavailability.


Despite the summer warming observed in Spain between 1980 and 2015, the decline in the vulnerability of the population has contributed to a general downward trend in overall heat-attributable mortality. This reduction occurred in parallel with a decline in the vulnerability difference between men and women for circulatory and cardiorespiratory mortality. Despite these advances, the risk of death remained high for respiratory diseases, and particularly in women.

Author summary

Why was this study done?

  • The Euro-Mediterranean region arises as a major climatic hot spot because of global warming.
  • Warmer temperatures should in principle contribute to an increase in the number of deaths because of heat.
  • We do not know yet if and to what extent societal adaptation and/or socioeconomic development is modifying this expected increase.

What did the researchers do and find?

  • We analysed daily mortality records from 47 major cities in Spain.
  • There has been a general and sustained decline in the vulnerability of the population since 1980.
  • Despite the observed warming, the decline of the vulnerability has generally contributed to a progressive reduction in the number of deaths attributed to heat since 1980.

What do these findings mean?

  • It is generally believed that climate change will cause an increase in heat-related mortality.
  • Societal adaptation and/or socioeconomic development contributed, up to now, to a general decline in heat-related mortality.
  • It is still uncertain if this decline in heat-related mortality will also occur at higher future levels of climate warming.