Modelling climate change impacts on attributable-related deaths and demographic changes in the largest metropolitan area in Portugal: A time-series analysis.

Previous studies have consistently analyzed the impact that extreme temperatures will have on human health. However, there are very few data on temperature-related mortality burden considering future demographic changes in a context of climate change in Portugal. This study aims to quantify the impact of climate change on heat-, cold-, and net change mortality burdens, taking into account the future demographic changes in Lisbon Metropolitan Area, Portugal. We applied a time-series generalized linear model with a quasi-Poisson model via a distributed lag nonlinear model to project temperature-related mortality burden for two climatological scenarios: a present (or reference, 1986-2005) scenario and a future scenario (2046-2065), in this case the Representative Concentration Pathway RCP8.5, which reflects the worst set of expectations (with the most onerous impacts). The results show that the total attributable fraction due to temperature, extreme and moderate cold, is statistically significant in the historical period and the future projected scenarios, while extreme and moderate heat were only significant in the projected future summer period. Net differences were attributed to moderate cold in the future winter months. Projections show a consistent and significant increase in future heat-related mortality burden. The attributable fraction due to heat in the future period, compared to the historical period, ranges from 0 to 1.5% for moderate heat and from 0 to 0.5% for extreme heat. Adaptation should be implemented at the local level, so as to prevent and diminish the effects on citizens and healthcare services, in a context of climate change.