I analysed the impact of summer heat on human health by using a daily mortality database with nearly 60 million counts of death in 147 European regions representing over 400 million people. I first modelled region-specific temperature-mortality associations through a distributed lag nonlinear model, and then I performed a multivariate random-effects meta-analysis to estimate the average temperature-mortality association, both across all the regions in the whole domain, and across all the regions in a given country. I computed the trends in heat-attributable vulnerability and mortality, and characterized their relationship with summer temperature changes and macroeconomic growth.
I observed a generalized decrease in vulnerability in most of Europe, mainly in the Mediterranean countries, but with a few exceptions in eastern and central Europe. Results indicate that, on average, Europe is reducing human vulnerability to heat, but as a result of temperature changes, I did not observe a parallel decrease in overall heat-attributable mortality. I showed for the first time that the trends in summer temperature and relative risk are strongly and significantly associated with the trends of the attributable fraction. I also found that macroeconomic growth is associated with increasing trends in both heat-attributable vulnerability and mortality, and that this relationship is strong and mostly linear for all temperatures during at least the warmest 60-75 days of the year.