"Although previous studies have shown a potential rise in heat-related mortality, little was known about the extent to which this increase would be balanced by a reduction in cold-related deaths," said Antonio Gasparrini of the London School of Hygiene & Tropical Medicine. "In addition, effects tend to vary across regions, depending on local climate and other characteristics, making global comparisons very difficult."

The surge in deaths during hot weather would outweigh any decrease in deaths in cold weather, with cooler regions such as Northern Europe experiencing either no change or a marginal decrease in deaths, Gasparrini and colleagues found.

The study, published in The Lancet Planetary Health, compared heat- and cold-related mortality in 451 locations around the world. The team used data from 85 million deaths to create the first global model of how mortality rates change with hot or cold weather.

If emissions follow the RCP2.6 scenario, however, peaking soon and then declining substantially, the net increase in mortality rates by 2100 could be minimal, the researchers found – between –0.4% and 0.6%.

"This paper shows how heat-related deaths will escalate in the absence of decisive action to reduce the emissions of carbon dioxide and short-lived climate pollutants such as methane and black carbon," said Andy Haines of the London School of Hygiene & Tropical Medicine. "Such action could also result in major health benefits in the near term by reducing deaths from air pollution."