The United States just printed a record-low age-adjusted death rate. It did not print a 2025 life expectancy. Those are different files.
On 2 July 2026, CDC’s National Center for Health Statistics released Vital Statistics Rapid Release no. 44, “Mortality in the United States: Provisional Data, 2025,” by Farida B. Ahmad, Jodi A. Cisewski and Robert N. Anderson (DOI 10.15620/cdc/252455). The file covers 3,094,593 deaths among U.S. residents in calendar 2025 — 99.9% of death records received and processed as of 10 May 2026. The age-adjusted rate is 689.2 deaths per 100,000 U.S. standard population, 4.6% below the final 2024 rate of 722.1. NCHS writes that 689.2 “was the lowest death rate recorded in the United States.”
Data Brief 548 (January 2026) is the other file: final 2024 mortality and the period life table, life expectancy at birth 79.0 years. VSRR 44 does not compute a 2025 life-expectancy figure. Do not read 689.2 as “Americans lived longer again.”
What happened
NVSS collects annual mortality from U.S. death certificates. Final statistics typically land about 11 months after 31 December; VSRR 44 is the early estimate for resident deaths in the United States, January–December 2025. Territorial and foreign residents are excluded. Age was unknown for 40 decedents (less than 0.01%); race and ethnicity for 12,728 (0.4%). Causes are coded to ICD-10 and ranked by counts of underlying cause. Age-adjusted rates are given overall and by sex and race and Hispanic origin; crude rates by age. 2025 denominators are July 1, 2025 estimates based on the 2020 census through Census’s Modified Age and Race Census (MARC) file. Rate comparisons are significant at p < 0.05 on a z test unless the text says otherwise.
The headline rate fell. The raw count did not. Final 2024 registered 3,072,666 resident deaths. Provisional 2025 is 3,094,593 — about 22,000 more deaths in a larger, older population, under a lower age-adjusted rate. Ordinary arithmetic of an aging country, not a contradiction.
Death rates decreased for every age group and for both sexes. Crude rates per 100,000 were lowest for children ages 5–14 (14.0, from 14.5) and highest for people 85 and older (12,787.5, from 13,834.0). Working-age bands also moved: 15–24 from 66.9 to 61.6; 25–34 from 124.5 to 111.1. Ages 75–84 and 85+ still account for most deaths (853,780 and 900,375) even as their crude rates fell.
Age-adjusted rates from the VSRR 44 table (2024 final; 2025 provisional as of 10 May 2026):
| Characteristic | 2024 deaths (age-adjusted rate) | 2025 deaths (age-adjusted rate) |
|---|---|---|
| Total | 3,072,666 (722.1) | 3,094,593 (689.2) |
| Female | 1,473,226 (613.6) | 1,486,896 (582.9) |
| Male | 1,599,440 (845.0) | 1,607,697 (811.1) |
| Hispanic | 260,799 (531.8) | 264,768 (500.7) |
| American Indian and Alaska Native, non-Hispanic | 20,924 (786.1) | 20,918 (803.8) |
| Asian, non-Hispanic | 87,728 (369.0) | 90,681 (366.6) |
| Black, non-Hispanic | 380,945 (884.4) | 382,579 (869.0) |
| More than one race, non-Hispanic | 16,782 (333.9) | 17,082 (187.3) |
| Native Hawaiian or Other Pacific Islander, non-Hispanic | 4,431 (682.7) | 4,625 (746.0) |
| White, non-Hispanic | 2,290,348 (753.3) | 2,301,212 (724.2) |
NCHS’s notes on those cells: the Asian non-Hispanic rate did not change significantly. American Indian and Alaska Native and Native Hawaiian or Other Pacific Islander rates rose. The 2025 low is among multiracial non-Hispanic people (187.3); the high is among Black non-Hispanic people (869.0).
The three leading causes in 2025 were heart disease, cancer and unintentional injuries. Suicide left the top ten. Influenza and pneumonia entered it. Counts VSRR 44 states in the text:
| Underlying cause | 2024 (final) | 2025 (provisional) | Rank note from the report |
|---|---|---|---|
| Heart disease | 683,491 | 694,708 | Leading cause, 2025 |
| Cancer | 619,876 | 622,832 | Second, 2025 |
| Unintentional injuries | — | 184,265 | Third, 2025 (2024 count not in VSRR 44) |
| Influenza and pneumonia | 48,139 | 56,511 | 11th in 2024 → 8th in 2025 |
| Suicide | 48,824 | 48,789 | 10th in 2024 → 11th in 2025 |
Those two counts rose. The all-cause age-adjusted rate still fell. More people can die of the largest causes in a larger, older population while that rate improves. The progress is real. It is not a claim that fewer Americans had heart attacks or tumours.
The racial picture is not a single arrow. Rates fell for Hispanic, Black non-Hispanic and White non-Hispanic people. They did not fall everywhere. NCHS is explicit that the vintage-2025 denominator is not the vintage-2024 one: 2024 used a blended base (2020 census, 2020 Demographic Analysis, vintage-2020 estimates); 2025 uses the enumerated 2020 census and the MARC file. Changes in denominators “may disproportionately affect race groups with small populations, including American Indian and Alaska Native, Native Hawaiian or Other Pacific Islander, and more than one race.” The multiracial drop from 333.9 to 187.3 is the cell to handle with caution; those comparisons “may change with final mortality or population estimates, or both.”
Three more limits from the Discussion: the data are provisional — previously published 2024 provisional counts ran slightly light of the final file, and causes “such as unintentional injury deaths” lag, so “the final death count will likely be higher than currently reported.” Certificate timeliness varies by jurisdiction. Race and Hispanic origin on certificates can be misclassified, especially for American Indian and Alaska Native and Asian people and for Hispanic ethnicity.
Why it matters
A record-low death rate is a real public-health signal. It is also easy to oversell.
A different mortality print landed the same week: Burkina Faso’s malaria vaccine rollout cut cases by a third in the 2025 national register — not a U.S. death-rate cause, but the other public-health number currently on this desk.
The 2024 decline did not reverse in 2025. The drop reached every age band, including infants (554.5 to 532.5) and the 5–14 group that was statistically unchanged in the 2024 final print. Males remain well above females. Black non-Hispanic people remain the highest listed group (869.0) against White non-Hispanic 724.2 and Hispanic 500.7. That gap did not close because the national line moved.
VSRR 44 does not attribute the 4.6% decline to overdoses, COVID-19, or any other single cause. It names the top three, notes that heart-disease and cancer counts rose, that suicide left the ten, and that influenza and pneumonia rose into eighth (56,511, up from 48,139). A cause moving up the list while the all-cause rate hits a recorded low is ordinary coexistence in a mortality file, not a paradox.
For a Good Signal reader the progress is the print: 99.9% of 2025 certificates, a rate NCHS will call the lowest recorded, and a comparison year that is final 2024. The honest next sentence is that injury deaths will likely rise in the final file, that two of the three leading causes killed more people in raw numbers than in 2024, and that NCHS has not published a 2025 life table. 689.2 is the death-rate number. 79.0 remains the last official life-expectancy number, and it is a 2024 number.
What to watch next
- Final 2025 mortality. Provisional 2024 ran light of the final count. Unintentional injuries are the lag to watch; 184,265 here is an early number, not a closed one.
- A 2025 life table, if NCHS publishes one. Until it does, there is no 2025 life-expectancy figure to quote. Do not back-solve one from 689.2.
- Influenza and pneumonia. Eighth at 56,511, up from 11th at 48,139. Whether that ranking survives the final file is the next cause question this report actually poses.
- AIAN, NHPI and multiracial rates after a consistent denominator. Increases (786.1 → 803.8; 682.7 → 746.0) and the multiracial drop (333.9 → 187.3) carry a vintage warning. Final mortality or a later population estimate can move those cells.
- The Black–White gap. 869.0 versus 724.2 in 2025. A falling national rate that leaves the highest group highest is still a gap.
Sources
- Ahmad FB, Cisewski JA, Anderson RN. Mortality in the United States: Provisional data, 2025. Vital Statistics Rapid Release. 2026 Jul;(44):1–6. DOI: 10.15620/cdc/252455. PDF: https://www.cdc.gov/nchs/data/vsrr/vsrr044.pdf
- CDC Stacks record, published 2 July 2026 — https://stacks.cdc.gov/view/cdc/252455
- NCHS blog, “Provisional Data: U.S. Death Rate Fell to a Record Low in 2025,” 2 July 2026 — https://www.cdc.gov/nchs/blog/posts/2026/07/provisional-data-u-s-death-rate-fell-to-a-record-low-in-2025.html
- Xu JQ, Murphy SL, Kochanek KD, Arias E. Mortality in the United States, 2024. NCHS Data Brief no. 548, January 2026 (final 2024 mortality and life expectancy 79.0; not a 2025 life table) — https://www.cdc.gov/nchs/products/databriefs/db548.htm
