Heat Wave Deaths: How Scientists Count Lives Lost Worldwide

Extreme heat is causing hundreds of thousands of deaths worldwide each year, and the annual number of heat-related deaths is increasing.

On average, an estimated 210,000 more people died from heat-related causes each year between 2012 and 2021 than between 1990 and 1999¹ (see ‘Heat’s toll’). From 1990 to 2019, heatwaves were linked to around 75,000 deaths per hot season in Asia². In Europe, nearly 14,000 people are estimated to have died during a record-breaking heatwave in June this year³. This week, dangerous temperatures have once again affected parts of Europe and East Asia.

How do scientists determine how many people die from extreme heat? Producing an accurate estimate of heat-related mortality is challenging. High temperatures can trigger heart attacks and worsen many serious medical conditions, yet death certificates rarely identify heat as the direct cause of death.

Instead of relying solely on death certificates, researchers use statistical methods to estimate the number of heat-related deaths among total deaths. These approaches cannot produce a perfectly precise death count, but scientists say that estimating the mortality impact of heatwaves is essential for developing effective public-health measures. “Nobody needs to die in a heatwave,” says Kristie Ebi, a climate epidemiologist at the University of Washington in Seattle. “If you don’t know how many people died in a heatwave, you’re not going to take appropriate preventive action.”

How are heat-related deaths calculated?

Most public-health agencies estimate heat-related mortality using the ‘excess deaths’ method. This approach compares the total number of deaths recorded during a heatwave with the expected number of deaths during a baseline period without extreme temperatures. The difference between the two figures is treated as the estimated number of deaths associated with heat.

The excess-deaths method is relatively simple and fast, says Antonio Gasparrini, a biostatistician and epidemiologist at the London School of Hygiene and Tropical Medicine. However, it cannot always distinguish deaths caused by a heatwave from those resulting from other factors. It also can be used only after a heatwave has ended, which limits its value for predicting the death toll of an approaching extreme-heat event.

A more advanced approach uses an exposure–response function. This statistical tool describes the relationship between temperature and the number of deaths recorded at different temperatures. Researchers incorporate these functions into models that estimate heat-related mortality for a specific temperature increase. For example, a model might predict that a 5 °C rise in temperature would increase a city’s weekly mortality rate by 3%.

Unlike the excess-deaths method, exposure–response models can be used in real time during a heatwave and to assess future climate scenarios. Their accuracy depends on reliable temperature forecasts, however. Other changes in mortality that are unrelated to heat can also affect the reliability of model predictions.

Do different methods produce similar heat-death estimates?

Last week, the UK Health Security Agency (HSA) announced that an estimated 2,877 people died from heat-related causes in England during heatwaves in May and June. The agency used the excess-deaths method, comparing the average number of deaths on heatwave days with the average number recorded on days without extreme heat.

Gasparrini and his colleagues also studied the May and June heatwaves, but they used an exposure–response model rather than the excess-deaths method. Nevertheless, their estimate of 2,736 heat-related deaths in England and Wales was close to the HSA’s figure. The researchers also concluded that human-caused climate change contributed to almost half of those deaths. The close agreement between the two approaches strengthens confidence in the exposure–response method, says Gasparrini.