Skip to content

Environmental Health Perspectives

Facebook Page EHP Twitter Feed Open Access icon  

Science Selection November 2015 | Volume 123 | Issue 11

Email this to someoneShare on FacebookTweet about this on TwitterShare on LinkedInShare on Google+Share on StumbleUpon
Environ Health Perspect; DOI:10.1289/ehp.123-A287

Time and Temperature: Changes in Heat-Related Mortality over 27 Years

Wendee Nicole was awarded the inaugural Mongabay Prize for Environmental Reporting in 2013. She writes for Discover, Scientific American, National Wildlife, and other magazines.

About This Article open

Citation: Nicole W. 2015. Time and temperature: changes in heat-related mortality over 27 years. Environ Health Perspect 123:A287;

Published: 1 November 2015

PDF icon PDF Version (557 KB)

Related EHP Article

Temporal Variation in Heat–Mortality Associations: A Multicountry Study

Antonio Gasparrini, Yuming Guo, Masahiro Hashizume, Patrick L. Kinney, Elisaveta P. Petkova, Eric Lavigne, Antonella Zanobetti, Joel D. Schwartz, Aurelio Tobias, Michela Leone, Shilu Tong, Yasushi Honda, Ho Kim, and Ben G. Armstrong

Predictions of how climate change will affect public health1 depend on accurate projections of future risk. Several recent studies suggest the risk of mortality associated with extreme heat has declined with time.2 But methodological and data differences between these studies have made it difficult to fully elucidate what’s now known to be a complex relationship between temperature and health. A study reported in this issue of EHP expands this line of research by applying consistent methodology across a broader geographic scope, including countries with different climates and demographic profiles.3

The authors analyzed data from multiple cities and nations using two-stage time-series modeling, including a meta-analysis. The data set included more than 20 million deaths from nonaccidental causes in seven countries: Australia, Canada, Japan, South Korea, Spain, the United Kingdom, and the United States. The team used mean daily temperature as the exposure index, with data back to 1985 in some locations. They restricted the analysis to the four warmest months of the year for each location.

People on a hot city street.The estimated risk of heat-related mortality fell over time in more than half the countries studied, possibly a result of factors such as better housing and healthcare, increased use of air conditioning, and public education.

© Spencer Platt/Getty Images

Results showed that the estimated relative risk of heat-related mortality significantly declined with time in the United States, Japan, and Spain, but the authors found no evidence of a decline in relative risk in the United Kingdom. Canada had a nonsignificant decrease in heat-related mortality, and low statistical power for South Korea and Australia made the results difficult to analyze for those countries.3

“Little is known about which individual or population-level factors modify the association between heat and mortality,” says lead author Antonio Gasparrini, senior lecturer in biostatistics and epidemiology at the London School of Hygiene & Tropical Medicine. Reasons for the lowered mortality were not analyzed in this work. However, the authors suggest the reduction could be due to improvements in infrastructure, such as better housing and healthcare, increased use of air conditioning, and education about the risks of extreme heat.3

Air conditioning prevalence has increased over time in four of the countries where Gasparrini and colleagues reported a decrease in heat-related mortality (Canada, Japan, Spain, and the United States), but has stayed the same in one of the countries where there was no change in mortality over time (the United Kingdom).3 Studies conducted elsewhere have found evidence of reduced mortality associated with higher use of air conditioning across time or geographical regions,4,5 although others have found only a weak association.6

“The study design is appropriate and technical analysis quite elegant,” says Joacim Rocklöv, an associate professor of epidemiology at Umeå University in Sweden, who was not involved with the study. “However, the aim of the study is temperature and not climatic extreme heat waves, which can be harder to study in shorter time periods but which can show different trends in effects not captured here.” Furthermore, Rocklöv says, pooling at the country level can mask considerable within-country differences.7

It is unclear why the authors found no decline in heat-related mortality in the United Kingdom after the severe European heat wave of 2003, whereas they did in Spain. Gasparrini suggests this could be due to lack of more current data from the United Kingdom. He explains that Spain’s heat adaptation plans after the 2003 heat wave may have provided benefits that were measurable in the Spanish data, which extended to 2010. In contrast, U.K. data covered only the years 1993–2006, with only three post-2003 years.3

Rocklöv also points out that the study includes only high-income countries where factors such as urbanization and housing standards differ from locations where billions of other people live.

One limitation of several heat-related mortality studies has been that the heat–mortality relationship is assumed to change in a linear fashion over time. “The authors emphasize how their approach flexibly models the association between temperature and mortality, allowing for a nonlinear relationship as well as delayed effects,” says Jennifer Bobb, assistant scientific investigator at Group Health Research Institute, who was not involved with the current study. “And this is indeed a strength of the analysis.”


1. IPCC. Climate Change 2013: The Physical Science Basis. Working Group I Contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (Stocker TF, et al., eds.). Cambridge, United Kingdom and New York, NY:Cambridge University Press Available:​t/ar5/wg1/WG1AR5_Frontmatter_FINAL.pdf [accessed 22 September 2015].

2. Hondula DM, et al. Rising temperatures, human health, and the role of adaptation. Curr Clim Change Rep 1(3):144–154 (2015); doi: 10.1007/s40641-015-0016-4.

3. Gasparrini A, et al. Temporal variation in heat–mortality associations: a multicountry study. Environ Health Perspect 123(11):1200–1207 (2015); doi: 10.1289/ehp.1409070.

4. Anderson GB, Bell ML. Weather-related mortality: how heat, cold, and heat waves affect mortality in the United States. Epidemiology 20(2):205–213 (2009); doi: 10.1097/EDE.0b013e318190ee08.

5. Ostro B, et al. The effects of temperature and use of air conditioning on hospitalizations. Am J Epidemiol 172(9):1053–1061 (2010); doi: 10.1093/aje/kwq231.

6. Bobb JF, et al. Heat-related mortality and adaptation to heat in the United States. Environ Health Perspect 122(8):811–816 (2014); doi: 10.1289/ehp.1307392.

7. Rocklöv J, et al. Mortality related to temperature and persistent extreme temperatures: a study of cause-specific and age-stratified mortality. Occup Environ Med 68(7):531–536 (2011); doi: 10.1136/oem.2010.058818.

WP-Backgrounds Lite by InoPlugs Web Design and Juwelier Schönmann 1010 Wien