Researchers put a number on this summer's heat deaths, and on climate's share

Heat kills quietly. It rarely appears on a death certificate, and the people it kills are overwhelmingly already unwell — which is why heat mortality has historically been argued about rather than counted.

Attribution science has changed that. Researchers from Imperial College London, the Met Office, and the London School of Hygiene and Tropical Medicine estimate that more than 2,700 people died from heat-related causes during the May and June heatwaves in England and Wales, and that climate change added between 3 and 4°C to maximum temperatures — making it responsible for roughly 42% of those deaths [s1].

How you attribute a death to warming

The method does not identify individual victims. It compares two worlds.

Researchers model the heatwave as it occurred, then model the same weather system in a counterfactual climate without human-caused warming. Established relationships between temperature and mortality — built from historical records — convert the temperature difference into a difference in deaths [s1]. The gap between the two is the attributed toll.

It produces a population-level estimate with real uncertainty bounds, not a verdict on any particular person's death. That distinction matters, and it is usually the first thing lost in summary coverage.

Why the deaths are not mostly heat stroke

The intuitive picture of heat death — collapse, heat stroke — is a minority of the total. Most deaths come from cardiovascular disease, respiratory disease, and cerebrovascular disease [s2]. Heat is a stressor on systems already close to failing. Heatwaves also drive increased hospital admissions for cardiovascular, kidney, and respiratory disorders [s2].

This is why heat mortality is systematically undercounted without attribution methods. A death recorded as cardiac arrest during a heatwave is a cardiac arrest on the certificate.

The distribution

Europe recorded more than 10,000 heat deaths in June alone [s1]. But the burden is not going to sit where the reporting infrastructure is.

The Climate Impact Lab projects that 90% of deaths caused by rising temperatures will occur in low- and middle-income countries, with ten times more people dying in poor countries each year than in rich ones by 2050 [s3].

That asymmetry is the story underneath the summer's numbers. The places with the strongest mortality surveillance, the best-resourced health systems, and the most attribution research are not the places that will absorb most of the deaths.