IPCC Fourth Assessment Report: Climate Change 2007
Climate Change 2007: Working Group II: Impacts, Adaptation and Vulnerability Heat- and cold-related mortality

Evidence of the relationship between high ambient temperature and mortality has strengthened since the TAR, with increasing emphasis on the health impacts of heatwaves. Table 8.3 summarises projections of the impact of climate change on heat- and cold-related mortality. There is a lack of information on the effects of thermal stress on mortality outside the industrialised countries.

Table 8.3. Projected impacts of climate change on heat- and cold-related mortality.

Area Health effect Model Climate scenario, time slices Temperature increase and baseline Population projections and other assumptions Main results Reference 
UK Heat- and cold-related mortality Empirical-statistical model derived from observed mortality UKCIP scenarios 2020s, 2050s, 2080s 0.57 to 1.38°C in 2020s; 0.89 to 2.44°C in 2050s; 1.13 to 3.47°C in 2080s compared with 1961-1990 baseline Population held constant at 1996. No acclimatisation assumed.  Annual heat-related deaths increase from 798 in 1990s to 2,793 in 2050s and 3,519 in the 2080s under the medium-high scenario. Annual cold-related deaths decrease from 80,313 in 1990s to 60,021 in 2050s and 51,243 in 2080s under the medium-high scenario. Donaldson et al., 2001 
Germany, Baden-Wuertemberg Heat- and cold-related mortality Thermo-physiological model combined with conceptual model for adaptation ECHAM4-OPYC3 driven by SRES A1B emissions scenario. 2001-2055 compared with 1951-2001   Population growth and aging and short-term adaptation and acclimatisation.  About a 20% increase in heat-related mortality. Increase not likely to be compensated by reductions in cold-related mortality. Koppe, 2005 
Lisbon, Portugal  Heat-related mortality Empirical-statistical model derived from observed summer mortality PROMES and HadRM2 2020s, 2050s, 2080s 1.4 to 1.8°C in 2020s; 2.8 to 3.5°C in 2050s; 5.6 to 7.1°C in 2080s, compared with 1968-1998 baseline SRES population scenarios. Assumes some acclimatisation.  Increase in heat-related mortality from baseline of 5.4 to 6 deaths/100,000 to 5.8 to 15.1 deaths/100,000 by the 2020s, 7.3 to 35.9 deaths/100,000 by the 2050s, 19.5 to 248.4 deaths/100,000 by the 2080s Dessai, 2003 
Four cities in California, USA (Los Angeles, Sacramento, Fresno, Shasta Dam) Annual number of heatwave days, length of heatwave season, and heat-related mortality Empirical-statistical model derived from observed summer mortality PCM and HadCM3 driven by SRES B1 and A1FI emissions scenarios 2030s, 2080s 1.35 to 2.0°C in 2030s; 2.3 to 5.8°C in 2080s compared with 1961-1990 baseline SRES population scenarios. Assumes some adaptation. Increase in annual number of days classified as heatwave conditions. By 2080s, in Los Angeles, number of heatwave days increases 4-fold under B1 and 6 to 8-fold under A1FI. Annual number of heat-related deaths in Los Angeles increases from about 165 in the 1990s to 319 to 1,182 under different scenarios. Hayhoe, 2004 
Australian capital cities (Adelaide, Brisbane, Canberra, Darwin, Hobart, Melbourne, Perth, Sydney) Heat-related mortality in people older than 65 years Empirical-statistical model, derived from observed daily mortality CSIROMk2, ECHAM4, and HADCM2 driven by SRES A2 and B2 emissions scenarios and a stabilisation scenario at 450 ppm 2100 0.8 to 5.5°C increase in annual maximum temperature in the capital cities, compared with 1961-1990 baseline Population growth and population aging. No acclimatisation. Increase in temperature-attributable death rates from 82/100,000 across all cities under the current climate to 246/100,000 in 2100; death rates decreased with implementation of policies to mitigate GHG.  McMichael et al., 2003b 

Reductions in cold-related deaths due to climate change are projected to be greater than increases in heat-related deaths in the UK (Donaldson et al., 2001). However, projections of cold-related deaths, and the potential for decreasing their numbers due to warmer winters, can be overestimated unless they take into account the effects of influenza and season (Armstrong et al., 2004).

Heat-related morbidity and mortality is projected to increase. Heat exposures vary widely, and current studies do not quantify the years of life lost due to high temperatures. Estimates of the burden of heat-related mortality attributable to climate change are reduced, but not eliminated, when assumptions about acclimatisation and adaptation are included in models. On the other hand, increasing numbers of older adults in the population will increase the proportion of the population at risk because a decreased ability to thermo-regulate is a normal part of the aging process. Overall, the health burden could be relatively small for moderate heatwaves in temperate countries, because deaths occur primarily in susceptible persons. Additional research is needed to understand how the balance of heat-related and cold-related mortality could change under different socio-economic scenarios and climate projections.