Heat gets the headlines. Across 58 studies, cold is the deadlier extreme.
A meta-analysis spanning two decades of research found extreme cold carried a higher relative risk of death than extreme heat or heatwaves — even as warming shifts public attention toward summer temperatures.
| Group | Value (value) |
|---|---|
| Extreme cold | 1.48 (1.22 to 1.79) |
| Heatwaves | 1.28 (1.16 to 1.42) |
| Extreme heat | 1.15 (1.1 to 1.19) |
Public attention to temperature and health has increasingly centered on summer heatwaves, but a meta-analysis published this month in the International Journal of Hygiene and Environmental Health finds that, measured by relative mortality risk, extreme cold remains the more hazardous of the two extremes [s1].
What the analysis covered
Researchers searched the literature published between 2000 and 2024 for observational studies quantifying how extreme temperature events affect urgent care outcomes — hospitalization, emergency department visits, length of hospital stay, and medical costs [s1]. From an initial pool of 3,411 screened papers, 58 studies met the review's inclusion criteria and were pooled in a meta-analysis [s1].
The core numbers
The pooled relative risk for all-cause mortality was 1.478 (95% CI: 1.219–1.792) for extreme cold, compared with 1.145 (95% CI: 1.103–1.189) for extreme heat and 1.280 (95% CI: 1.155–1.418) for heatwaves specifically [s1]. In plain terms: across the pooled studies, a period of extreme cold was associated with roughly a 48% increase in the relative risk of death, versus about 15% for extreme heat generally and 28% for heatwave conditions specifically [s1].
Heat's signature showed up more clearly in downstream health-system use rather than in the mortality figures. The pooled relative risk for hospitalization during heatwaves was 1.060 (95% CI: 1.018–1.104), and for emergency department visits during heatwaves it was 1.047 (95% CI: 1.022–1.072) [s1]. The review found a limited but notable set of studies showing that both extreme cold and extreme heat increased the risk of hospital stays longer than 24 hours, and estimated that the associated economic losses could reach €25.5 billion [s1].
Why cold outranks heat in this data
The review's authors conclude plainly that extreme cold is the most hazardous temperature event studied, with mortality the most sensitive indicator of that hazard [s1]. That finding runs against the more heat-focused framing common in recent climate-health coverage, though it is consistent with a body of prior epidemiological research showing that cold-related mortality — driven substantially by cardiovascular strain, respiratory illness, and indirect causes like falls and carbon monoxide poisoning from heating — has historically outpaced heat-related mortality in absolute terms across many temperate and cold-climate countries.
What this doesn't settle
A pooled relative risk across 58 heterogeneous studies collapses a lot of local variation — climate, housing quality, healthcare access, and population age structure all shape how dangerous a given cold snap or heatwave actually is in a specific place, and the review's aggregate numbers don't capture that variation. The review's authors also note that fewer studies examined length-of-stay outcomes using measures beyond extreme heat, meaning the cold-specific picture on that particular outcome is less complete than the mortality data [s1]. And a lower relative risk for heat does not mean heat is a smaller problem in absolute terms — heatwave-associated increases in emergency department and hospital use, even at a smaller relative risk than cold-related mortality, translate into large numbers of additional cases as extreme heat events become more frequent. The review's contribution is not to declare heat a non-issue, but to put a number on a comparison that public discussion often treats as settled in heat's direction by assumption rather than data [s1].
Sources
- A systematic review and meta-analysis comparing the health burden of extreme heat and cold on mortality and urgent care utilization — International Journal of Hygiene and Environmental Health, 2026-04-07
Sources
- A systematic review and meta-analysis comparing the health burden of extreme heat and cold on mortality and urgent care utilization — International Journal of Hygiene and Environmental Health , April 7, 2026
More on
US heat deaths tripled among the least-educated adults, and barely moved at the top
A 14-year national analysis finds temperature-related death rates rising fastest in the group that started highest — and the same pattern in cold deaths, which fell for everyone else.
Researchers put a number on this summer's heat deaths, and on climate's share
More than 2,700 people died in England and Wales during the May and June heatwaves. An attribution study assigns roughly 42% of those deaths to warming.
In the county that counts heat deaths best, a statistical model found 57% more of them
Maricopa County runs some of the most detailed heat mortality surveillance in the US. A distributed lag model applied to the same five years put the toll far higher — and the gap is the method, not a scandal.
An 800-times-more-likely heat wave hit the Southwest in March, before anyone was ready
A rapid attribution study finds human-caused warming made this month's record Southwest heat wave far more likely, and doctors warn the early timing caught people unacclimatized.