How many people die from heat in the US each year?

It depends on the method. US death certificates recorded about 702 heat-related deaths per year on average during 2004–2018 (CDC). Statistical models that capture heat's indirect toll estimate far more, roughly 5,600 excess deaths a year in one large analysis. Single-year coded counts have ranged from about 300 to over 1,100 depending on the weather.

Why is the modeled estimate so much higher than the coded count?

Because heat often kills by worsening existing conditions such as heart disease or stroke, and those deaths get certified under the other cause rather than heat. Death-certificate counts capture only cases explicitly attributed to heat; models detect the broader rise in mortality during hot periods, so they find several times more heat-related deaths.

What is the difference between an underlying and a contributing cause?

The underlying cause is the condition that started the chain of events leading to death; a contributing cause played a role but was not the primary trigger. In the CDC's 2004–2018 data, 59% of coded heat deaths had heat as the underlying cause and 41% had it as a contributing cause, often alongside cardiovascular disease.

How many emergency department visits are caused by heat?

The CDC's syndromic surveillance recorded 119,605 heat-related illness ED visits in 2023, with 92% in May–September. Rates peaked in July and August and were about 2.7 times higher for men than women. These figures count visits rather than unique patients.

Who is most at risk of dying from heat?

In CDC data, about 70% of heat deaths occurred in men, adults 65 and older had the highest death rate, and American Indian/Alaska Native people had the highest rate by race/ethnicity. Arizona, California, and Texas accounted for 37% of deaths. Some outdoor workers, especially in agriculture and construction, also face elevated risk.

How dangerous is heat for outdoor workers?

An analysis of federal occupational-injury data found 359 heat-related worker deaths over 2000–2010 (about 33 per year). Hispanic workers died at roughly 3.2 times the rate of others, and agriculture and construction accounted for 58% of deaths. Recent BLS reporting shows environmental-heat worker deaths in the mid-30s to mid-40s per year.

Research

Heat-Related Illness in the United States: Key Statistics

How many Americans does heat harm each year? The answer depends on how you count. US death certificates recorded an average of about 702 heat-related deaths per year during 2004–2018 (CDC), but statistical models that capture heat's indirect toll estimate roughly 5,600 excess deaths a year. On top of the deaths, there were more than 119,000 emergency department visits for heat-related illness in 2023 alone. This page compiles the verified figures (deaths coded on certificates, modeled excess deaths, ED visits, and worker fatalities) and explains why the headline numbers differ so much.

By the Sweat Explained Editorial Team · Published 2026-07-13 · Last reviewed 2026-07-13 · Educational information, not medical advice.

Key statistics at a glance

~702/yr
heat-related deaths recorded on US death certificates, 2004–2018 average
CDC MMWR 2020 (NVSS)
59% / 41%
of those deaths had heat as the underlying vs. a contributing cause
CDC MMWR 2020
~5,608/yr
excess deaths attributable to heat estimated by modeling (297 counties, 1997–2006), far above coded counts
Weinberger 2020
119,605
US emergency department visits for heat-related illness recorded in 2023 (92% in May–Sept)
CDC MMWR 2024 (NSSP)
271 vs 104
warm-season HRI ED visits per 100,000 ED visits, males vs females, 2023
CDC MMWR 2024
359
US occupational heat-related deaths identified over 2000–2010 (~33/yr)
Gubernot 2015 (BLS CFOI)

How many people die from heat each year?

There is no single number, because heat deaths are counted in two very different ways. The first way is to tally death certificates that name heat. Using national death records, the CDC found an average of about 702 heat-related deaths per year in the United States from 2004 through 2018: 10,527 deaths over the 15-year period. Of those, 6,220 (59%) had exposure to natural heat listed as the underlying cause, while 4,307 (41%) had heat as a contributing cause alongside conditions such as cardiovascular disease.

The second way is statistical: because heat pushes people with existing illness over the edge, many heat deaths are recorded under other causes. Models that compare deaths during hot periods with expected baseline mortality capture this hidden toll. One widely cited analysis of 297 US counties (covering about 62% of the population, 1997–2006) estimated an average of 5,608 deaths attributable to heat each year, roughly eight times the certificate count for a comparable period. Neither number is 'wrong'; they measure different things.

Year to year, the coded count also swings with the weather. Records of deaths from excessive natural heat have ranged from a low near 300 in a mild year to more than 1,100 in a severe heat year, so any single-year figure should be read in the context of that variability.

Coded deaths vs. modeled excess deaths

The gap between the two counting methods is large. Death-certificate coding captures cases explicitly attributed to heat; modeling captures the broader increase in mortality when temperatures spike, regardless of the cause written on the certificate.

US heat deaths per year by counting method (approximate annual averages)

US heat deaths per year by counting method (approximate annual averages)
Group	Value
Coded on death certificates	702 deaths/yr (CDC / NVSS, 2004–2018 average)
Modeled excess deaths	5608 deaths/yr (297 counties, 1997–2006 average)

Source: CDC MMWR 2020 (coded, 2004–2018); Weinberger et al. 2020 (modeled excess, 1997–2006). Chart is an original rendering of the cited data.

Who is most affected

Heat-related deaths are not spread evenly. The CDC analysis of 2004–2018 death records shows clear patterns by sex, age, race/ethnicity, and geography.

Distribution of US heat-related deaths, 2004–2018 (CDC MMWR 2020)
Group	Figure	What it means
Males	70% of deaths	Roughly two-thirds to three-quarters of heat deaths occur in men
Aged 65 and older	4,019 deaths (39%)	Highest death rate of any age group (0.7 per 100,000)
American Indian / Alaska Native	0.6 per 100,000	Highest age-adjusted rate of any race/ethnicity group
Arizona, California, Texas	3,852 deaths (37%)	Three states accounted for over a third of all heat deaths

Non-Hispanic White decedents were the largest count (6,602) but not the highest rate. Rates are per 100,000 population.

Emergency department visits

Deaths are only the most severe outcome. Far more people seek emergency care. Using the CDC's National Syndromic Surveillance Program, 119,605 heat-related illness (HRI) emergency department visits were recorded in 2023, with 92% occurring in the warm-season months of May through September.

Rates rose sharply in the summer heat and fell hardest on men. In July and August 2023 the HRI visit rate averaged 303 per 100,000 ED visits, versus 97 in the other warm-season months. Across the warm season, men had a rate of 271 per 100,000 ED visits compared with 104 for women, about 2.7 times higher. The 2023 overall warm-season rate (180 per 100,000) ran above the 2018–2022 average (151), and several southern and western regions saw rates well above the rest of the country.

Because syndromic surveillance counts visits rather than individuals, and coding practices vary, these figures track the pattern and burden of heat illness rather than an exact case count.

Heat deaths among workers

Outdoor and hot indoor jobs carry added risk. An analysis of the Bureau of Labor Statistics Census of Fatal Occupational Injuries identified 359 occupational heat-related deaths in the United States from 2000 through 2010: an average of about 33 per year, or 0.22 deaths per million workers.

The burden is concentrated. Hispanic workers died at 0.54 per million (about 3.2 times the rate of non-Hispanic workers), and the agriculture (3.06 per million) and construction (1.13 per million) industries together accounted for 207 (58%) of the deaths. The great majority occurred in summer, with 86% in June through August. In more recent BLS reporting, work-related deaths from environmental heat exposure have continued to run in the mid-30s to mid-40s per year.

How to read these numbers

The single most important thing to understand is why the death figures disagree. A count of about 702 per year is what appears when someone writes heat on a death certificate. An estimate of about 5,600 per year is what statistical models find when they measure how much mortality rises during hot weather, including deaths recorded under heart disease, stroke, and other causes that heat helped trigger. Both are legitimate; the modeled figure is generally considered a fuller picture of heat's true burden, while the coded figure is a precise, if conservative, tally.

For everyday interpretation: heat kills hundreds of Americans a year by the strictest count and likely thousands by broader measures, sends well over a hundred thousand to emergency departments, and disproportionately harms older adults, men, some outdoor workers, and communities in the hottest regions. Extreme heat is a medical emergency. Anyone with confusion, fainting, a very high temperature, or hot dry skin needs urgent care.

Methodology and limitations

This page compiles US government surveillance and peer-reviewed sources: CDC MMWR reports drawn from the National Vital Statistics System (heat-related deaths, 2004–2018) and the National Syndromic Surveillance Program (ED visits, 2023); a peer-reviewed modeling study of excess deaths across 297 counties (Weinberger et al., 1997–2006); and an analysis of the BLS Census of Fatal Occupational Injuries (Gubernot et al., 2000–2010). Each figure was traced to its source and confirmed.

Limitations: coded death counts undercount heat's true toll because many heat-influenced deaths are certified under other causes, which is why modeled excess-death estimates are several times higher; the two should not be added together. Modeled estimates carry their own statistical uncertainty (the 5,608 figure has a confidence interval of roughly 4,700–6,300) and cover a subset of counties and an earlier decade. Syndromic ED data count visits, not unique patients, and depend on local coding. Occupational figures rely on cases identified in administrative records and likely undercount. Year-to-year totals vary with weather. Nothing here is medical advice.

Frequently asked questions

How many people die from heat in the US each year?: It depends on the method. US death certificates recorded about 702 heat-related deaths per year on average during 2004–2018 (CDC). Statistical models that capture heat's indirect toll estimate far more, roughly 5,600 excess deaths a year in one large analysis. Single-year coded counts have ranged from about 300 to over 1,100 depending on the weather.
Why is the modeled estimate so much higher than the coded count?: Because heat often kills by worsening existing conditions such as heart disease or stroke, and those deaths get certified under the other cause rather than heat. Death-certificate counts capture only cases explicitly attributed to heat; models detect the broader rise in mortality during hot periods, so they find several times more heat-related deaths.
What is the difference between an underlying and a contributing cause?: The underlying cause is the condition that started the chain of events leading to death; a contributing cause played a role but was not the primary trigger. In the CDC's 2004–2018 data, 59% of coded heat deaths had heat as the underlying cause and 41% had it as a contributing cause, often alongside cardiovascular disease.
How many emergency department visits are caused by heat?: The CDC's syndromic surveillance recorded 119,605 heat-related illness ED visits in 2023, with 92% in May–September. Rates peaked in July and August and were about 2.7 times higher for men than women. These figures count visits rather than unique patients.
Who is most at risk of dying from heat?: In CDC data, about 70% of heat deaths occurred in men, adults 65 and older had the highest death rate, and American Indian/Alaska Native people had the highest rate by race/ethnicity. Arizona, California, and Texas accounted for 37% of deaths. Some outdoor workers, especially in agriculture and construction, also face elevated risk.
How dangerous is heat for outdoor workers?: An analysis of federal occupational-injury data found 359 heat-related worker deaths over 2000–2010 (about 33 per year). Hispanic workers died at roughly 3.2 times the rate of others, and agriculture and construction accounted for 58% of deaths. Recent BLS reporting shows environmental-heat worker deaths in the mid-30s to mid-40s per year.

Sources

Primary peer-reviewed studies and official sources first, then reviews and institutional framing (secondary).

Vaidyanathan A, Malilay J, Schramm P, Saha S. Heat-Related Deaths: United States, 2004–2018. MMWR Morb Mortal Wkly Rep. 2020;69(24):729–734. National Vital Statistics System. Full text (CDC MMWR)
Vaidyanathan A, Gates A, Brown C, et al. Heat-Related Emergency Department Visits: United States, May–September 2023. MMWR Morb Mortal Wkly Rep. 2024;73(15):324–329. National Syndromic Surveillance Program. Full text (CDC MMWR)
Weinberger KR, Harris D, Spangler KR, Zanobetti A, Wellenius GA. Estimating the number of excess deaths attributable to heat in 297 United States counties. Environ Epidemiol. 2020;4(3):e096. Estimated 5,608 heat-attributable deaths/yr, 1997–2006. Full text
Gubernot DM, Anderson GB, Hunting KL. Characterizing occupational heat-related mortality in the United States, 2000–2010: an analysis using the Census of Fatal Occupational Injuries database. Am J Ind Med. 2015;58(2):203–211. Full text
US Bureau of Labor Statistics. Work-related deaths due to environmental heat exposure (Census of Fatal Occupational Injuries). The Economics Daily, BLS: e.g., 36 deaths in 2021. (secondary) BLS

How to cite this page

Sweat Explained. Heat-Related Illness in the United States: Key Statistics. Published 2026-07-13; last reviewed 2026-07-13. Available at: https://sweatexplained.com/research/heat-related-illness-statistics-us

Please cite the original studies for the underlying figures. Journalists are welcome to link to this page; the charts are original renderings of the cited data.