Abstract
This study examines trends in heat-related mortality rates in the US population from 1999 to 2023.
The warmest average temperature recorded since 1850 occurred in 2023.1 Recent studies have found exposure to extreme heat to be associated with mortality, with variability by age, sex, and race and ethnicity.2,3 Recent research suggests that heat-related mortality risk is increasing globally,4 but formal analyses of heat-related mortality trends in the US through 2023 are lacking. This study examined trends in heat-related mortality rates in the US population from 1999 to 2023.
Methods
We analyzed all deaths from 1999 to 2023 in which the International Statistical Classification of Diseases and Related Health Problems, 10th Revision code was P81 (environmental hyperthermia of newborn), T67 (effects of heat and light), or X30 (exposure to excessive natural heat) as either the underlying cause or as a contributing cause of death, as recorded in the Multiple Cause of Death file. Data were accessed through the Centers for Disease Control and Prevention’s WONDER platform,5 which combines death counts with population estimates produced by the US Census Bureau to calculate mortality rates. For each year, we extracted age-adjusted mortality rates (AAMRs) per 100 000 person-years for heat-related deaths. The AAMR accounts for differences due to age structures, allowing direct comparisons across time. The approach of analyzing cause-specific mortality rates rather than excess mortality is warranted because the excess mortality methodology is subject to confounding from the COVID-19 pandemic from 2020 to 2023. This study used publicly available, deidentified aggregate data; thus, it was not considered human subjects research.
Joinpoint version 5.2.0 (National Cancer Institute) regression6 was used to analyze AAMRs to assess trends and determine elbow points where the trend began to shift to a new trajectory. Results of joinpoint analyses are reported as average annual percentage change (AAPC) in rates with 95% CIs. Statistical significance was defined as 2-sided P < .05. Data were visualized with R version 4.2.2 (R Foundation for Statistical Computing).
Results
From 1999 to 2023, 21 518 deaths were recorded as heat-related underlying or contributing cause of death, with an AAMR of 0.26 per 100 000 person-years (95% CI, 0.24-0.27) (Table). The number of heat-related deaths increased from 1069 (AAMR = 0.38; 95% CI, 0.36-0.40) in 1999 to 2325 (AAMR = 0.62; 95% CI, 0.60-0.65) in 2023, a 117% increase in the number of heat-related deaths and a 63% increase in the AAMR. The lowest number of heat-related deaths in the study period was 311 in 2004, whereas the highest, 2325, was in 2023.
Table. Number of Heat-Related Deaths and Age-Adjusted Mortality Rates From 1999-2023.
| Year | Heat-related deaths | |
|---|---|---|
| No. | Age-adjusted mortality rate per 100 000 person-years (95% CI) | |
| 1999 | 1069 | 0.38 (0.36-0.40) |
| 2000 | 495 | 0.19 (0.17-0.21) |
| 2001 | 540 | 0.19 (0.18-0.21) |
| 2002 | 629 | 0.22 (0.20-0.23) |
| 2003 | 417 | 0.12 (0.11-0.13) |
| 2004 | 311 | 0.09 (0.07-0.10) |
| 2005 | 706 | 0.22 (0.20-0.23) |
| 2006 | 1009 | 0.34 (0.32-0.36) |
| 2007 | 547 | 0.18 (0.16-0.19) |
| 2008 | 504 | 0.15 (0.14-0.17) |
| 2009 | 558 | 0.15 (0.14-0.17) |
| 2010 | 800 | 0.26 (0.25-0.28) |
| 2011 | 1009 | 0.31 (0.29-0.33) |
| 2012 | 840 | 0.24 (0.22-0.26) |
| 2013 | 652 | 0.20 (0.18-0.22) |
| 2014 | 412 | 0.12 (0.11-0.13) |
| 2015 | 579 | 0.17 (0.15-0.18) |
| 2016 | 824 | 0.22 (0.21-0.24) |
| 2017 | 882 | 0.25 (0.23-0.26) |
| 2018 | 1014 | 0.28 (0.27-0.30) |
| 2019 | 913 | 0.26 (0.24-0.27) |
| 2020 | 1159 | 0.31 (0.29-0.33) |
| 2021 | 1602 | 0.43 (0.40-0.45) |
| 2022 | 1722 | 0.47 (0.45-0.49) |
| 2023a | 2325 | 0.62 (0.60-0.65) |
| Total | 21 518 | 0.26 (0.24-0.27) |
Provisional data.
Results of the joinpoint trend analysis showed that during the entire period, the AAMR increased by 3.6% per year (AAPC = 3.6%; 95% CI, 0.1%-7.2%; P = .04) from 1999 to 2023 (Figure). The number of heat-related deaths and AAMR showed year-to-year variability, with spikes in 2006 and 2011, before showing steady increases after 2016. Joinpoint results showed a nonsignificant decrease of 1.4% per year from 1999 to 2016 (AAPC = −1.4%; 95% CI, −4.7% to 2.1%; P = .42), followed by a significant increase of 16.8% per year in the AAMR from 2016 to 2023 (AAPC = 16.8%; 95% CI, 6.4%-28.2%; P = .002).
Figure. Age-Adjusted Mortality Rates With Joinpoint Model Estimates for Heat-Related Deaths as Either Underlying or Contributing Cause From 1999-2023.
Data are from the Centers for Disease Control and Prevention’s WONDER database. Heat-related mortality rates are plotted as dots, with a joinpoint model plotted as a line and a single joinpoint in 2016. Overall, heat-related mortality rates increased from 1999 to 2023 (AAPC = 3.6%; 95% CI, 0.1%-7.2%; P = .04). However, there was a nonsignificant decreasing trend from 1999 to 2016 (AAPC = −1.4%; 95% CI, −4.7% to 2.1%; P = .42) and an increasing trend from 2016 to 2023 (AAPC = 16.8%; 95% CI, 6.4%-28.2%; P = .002).
Discussion
This study found that heat-related mortality rates in the US increased between 1999 and 2023, especially during the last 7 years. Although a study using data through 2018 found a downward trend in heat-related mortality in the US,2 this study is the first to our knowledge to demonstrate a reversal of this trend from 2016 to 2023. These results align with site-specific data analyzed in a global study that suggest increases in heat-related mortality.4 As temperatures continue to rise because of climate change,4 the recent increasing trend is likely to continue. Local authorities in high-risk areas should consider investing in the expansion of access to hydration centers and public cooling centers or other buildings with air conditioning. Study limitations include the potential for misclassification of causes of death, leading to possible underestimation of heat-related mortality rates; potential bias from increasing awareness over time; and lack of data for vulnerable subgroups.
Section Editors: Kristin Walter, MD, and Jody W. Zylke, MD, Deputy Editors; Karen Lasser, MD, MPH, Senior Editor.
Data Sharing Statement
References
- 1.National Oceanic and Atmospheric Administration . 2023. Was the world’s warmest year on record, by far. Published January 12, 2024. Accessed June 4, 2024. https://www.noaa.gov/news/2023-was-worlds-warmest-year-on-record-by-far
- 2.Sheridan SC, Dixon PG, Kalkstein AJ, Allen MJ. Recent trends in heat-related mortality in the United States: an update through 2018. Weather Clim Soc. 2021;13(1):95-106. doi: 10.1175/WCAS-D-20-0083.1 [DOI] [Google Scholar]
- 3.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. doi: 10.15585/mmwr.mm6924a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Lüthi S, Fairless C, Fischer EM, et al. Rapid increase in the risk of heat-related mortality. Nat Commun. 2023;14(1):4894. doi: 10.1038/s41467-023-40599-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Centers for Disease Control and Prevention . Wide-ranging Online Data for Epidemiologic Research (WONDER). Accessed June 1, 2024. https://wonder.cdc.gov
- 6.Kim HJ, Fay MP, Feuer EJ, Midthune DN. Permutation tests for joinpoint regression with applications to cancer rates. Stat Med. 2000;19(3):335-351. doi: [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Sharing Statement
