Projected temperature-related deaths in ten large U.S. metropolitan areas under different climate change scenarios

Abstract

Background There is an established U-shaped association between daily temperature and mortality. Temperature changes projected through the end of century are expected to lead to higher rates of heat-related mortality but also lower rates of cold-related mortality, such that the net change in temperature-related mortality will depend on location. Objectives We quantified the change in heat-, cold-, and temperature-related mortality rates through the end of the century across 10 large US metropolitan areas. Methods We applied location-specific projections of future temperature from over 40 downscaled climate models to exposure-response functions relating daily temperature and mortality in 10 US metropolitan areas to estimate the change in temperature-related mortality rates in 2045–2055 and 2085–2095 compared to 1992–2002, under two greenhouse gas emissions scenarios (RCP 4.5 and 8.5). We further calculated the total number of deaths attributable to temperature in 1997, 2050, and 2090 in each metropolitan area, either assuming constant population or accounting for projected population growth. Results In each of the 10 metropolitan areas, projected future temperatures were associated with lower rates of cold-related deaths and higher rates of heat-related deaths. Under the higher-emission RCP 8.5 scenario, 8 of the 10 metropolitan areas are projected to experience a net increase in annual temperature-related deaths per million people by 2086–2095, ranging from a net increase of 627 (95% empirical confidence intervals [eCI]: 239, 1018) deaths per million in Los Angeles to a net decrease of 59 (95% eCI: −485, 314) deaths per million in Boston. Applying these projected temperature-related mortality rates to projected population size underscores the large public health burden of temperature. Conclusions Increases in the heat-related death rate are projected to outweigh decreases in the cold-related death rate in 8 out of 10 cities studied under a high emissions scenario. Adhering to a lower emission scenario has the potential to substantially reduce future temperature-related mortality.

Impact of climate and population change on temperature-related mortality burden in Bavaria, Germany

“…The association between ambient temperature and mortality outcomes has been studied extensively [1][2][3][4]. There is agreement that there exists a temperature of minimum mortality (MMT) at which the Relative Risk (RR) of temperature-related mortality is one [5][6][7][8]. Exposure-response functions (ERFs) between temperature and mortality are found to be U-, J-or V-shaped deviating from this MMT [6,8] and are location-specific depending upon climatic, geographic and demographic characteristics [5,7,8].…”

“…There is agreement that there exists a temperature of minimum mortality (MMT) at which the Relative Risk (RR) of temperature-related mortality is one [5][6][7][8]. Exposure-response functions (ERFs) between temperature and mortality are found to be U-, J-or V-shaped deviating from this MMT [6,8] and are location-specific depending upon climatic, geographic and demographic characteristics [5,7,8]. Nevertheless, it makes, of course, a difference, if the deaths can be attributed to cold or to heat.…”

“…Several studies also show an increasing impact of climate change on heat-related cause-specific mortality burden, such as cardiovascular and respiratory causes [15,16]. Yet, with increasing temperatures and assuming no adaptation of the future population, cold-related mortality burden will decrease in the future [8,[17][18][19][20]. Thus, heat-related mortality might be outnumbered by the reduction in cold-related mortality, resulting in a decrease of the net temperature-related mortality [8].…”

“…Yet, with increasing temperatures and assuming no adaptation of the future population, cold-related mortality burden will decrease in the future [8,[17][18][19][20]. Thus, heat-related mortality might be outnumbered by the reduction in cold-related mortality, resulting in a decrease of the net temperature-related mortality [8]. Given a certain geographic location, the direction of the net change however, depends on the ERF, the projected temperature, and the population changes of that specific location.…”

“…both heat and cold-related mortality burden in different locations of the world [3,8,18,[20][21][22][23]. Most of the studies so far have incorporated only the climate change scenarios [3,8,18,20,21,23] and only a few have considered the range of possible future population scenarios [22]. Furthermore, only a limited number of these studies have incorporated age-specific exposureresponse curves for future mortality projection [3,20,22,23].…”