Every year heat kills tens of thousands of people. Their breathing grows shallow, their heart rates flutter, their muscles spasm, and then they die. Heat killed over 100,000 people in 2018, when high temperatures broiled the European Union.

A new study suggests that climate change was responsible for many of those deaths. The paper, published Monday in Nature Climate Change, scrutinized summertime deaths in 43 countries between 1991 and 2018 — the largest collection of heat mortality data ever assembled.

The researchers estimate that higher temperatures driven by greenhouse gas emissions caused more than half the heat-related deaths in several countries, including Thailand, Peru and the Philippines. On average, climate change was at fault for 37 percent of heat-related deaths. The world has only warmed around 2 degrees Fahrenheit so far, but that’s already enough to kill roughly 100,000 people every year, if you apply this paper’s estimate to the entire world.

Reader support helps sustain our work. Donate today to keep our climate news free. All donations DOUBLED!

There are, however, some pretty big holes in the data for anyone trying to do that kind of extrapolation. There’s simply no data on heat-related deaths from huge swaths of the world, including major population centers in equatorial Africa, and India. “The main point of this paper is that the map is mainly empty!” wrote Friedi Otto, a climate scientist at the University of Oxford who was not involved in the study, in an email. And Otto’s own research suggests that that heat is particularly deadly in these places that don’t have sophisticated systems to record the cause of deaths.

Grist thanks its sponsors. Become one.

In other words, reality could be worse than the estimates. “The countries where we do not have the necessary health data are often among the poorest and most susceptible to climate change, and, concerningly, are also the projected major hotspots of future population growth, ” according to Dann Mitchell, a climate scientist at the University of Bristol, in a piece accompanying the study.

To make their estimates, the authors analyzed 29,936,896 death records from around the world, zeroing in on the heat-related deaths. Then they did some fancy modelling to determine how much cooler it would have been if our pollution hadn’t wrapped round the planet like an overheavy comforter. And they finished it off with a little more math to estimate the number of people who would not have succumbed in that alternative (read: one without climate change) world.

These kinds of studies “are compelling for motivating the policy process because you can show there has been some number of deaths from climate change,” said Kristie Ebi, a professor of Global Health at the University of Washington.

Still, models that gin up these kinds of alternative worlds are never foolproof. Who knows what else would be different about a non-climate-changed world besides the temperatures? “There has been some acclimatization,” Ebi said. “In high-income countries there have been declining heat related deaths because of better air conditioning and better health care.”

Grist thanks its sponsors. Become one.

But if you want to make the best estimate of how many heat deaths were caused by climate change, you’d use the methods in this study, Otto said. “So the numbers here are a conservative lower bound estimate of the true heat deaths due to climate change,” she said. “Heat kills, and climate change is an absolute game changer when it comes to heat, and we do not talk about this enough.”