Skip to content
What Causes What
SMOKE+ INCREASES

Ambient air pollution → All-cause mortality

Higher Ambient air pollution increases All-cause mortality.

Hao 2023's Medicare cohort of about 73.4 million adults 65 and older found a 4.0 percent higher all-cause death rate per 3.68 µg/m³ of annual PM2.5 (HR 1.040, 1.039 to 1.041). Nobody randomized the air.

Observational · Low · N=73400000

The association disappears when you account for other factors, like socioeconomic status or healthy-user bias.

Mechanism

Particles small enough to reach the alveoli can enter blood. Authors name lung injury, inflammation, and cardiovascular strain. The files do not isolate one path. They assign outdoor concentrations to a postcode or ZIP code and later count a death. Abed Al Ahad, Demsar, Sullivan, and Kulu (2023) linked the Scottish Longitudinal Study: 202,237 adults aged 17 and older, 2002 to 2017, 42,415 deaths. Annual NO2, SO2, PM10, and PM2.5 at 1 km² on the residential postcode. All-cause death rose with PM2.5 (HR 1.035, 1.023 to 1.047 per 1 µg/m³) and with PM10, NO2, and SO2. The fullest model added census education, occupation, and urban-rural place. Hao, Wang, Zhu and colleagues (2023) followed about 73.4 million Medicare beneficiaries aged 65 and older in the contiguous United States, 2000 to 2017, 29 million deaths. Yearly PM2.5 mass and six components at the ZIP code, from two prediction surfaces. Per interquartile range of PM2.5 mass (3.68 µg/m³), all-cause HR 1.040 (1.039 to 1.041). Sulfate and ammonium had the largest component HRs per IQR. Models included ZIP-code income, schooling, and Medicaid eligibility. Chen, Dan, Sun and colleagues (2023) followed 4,708 UK Biobank adults who already had inflammatory bowel disease, mean 12 years, 420 deaths. Land-use regression for NOx, NO2, PM10, and PM2.5. Each IQR of PM2.5 tracked all-cause HR 1.14 (1.02 to 1.28). NOx, NO2, and PM10 moved the same way. Enterotomy and gastrointestinal cancer are different outcomes and stay off this page.

Caveats

The stamp is smoke because nobody randomized the air. These are registers and cohorts. Observation alone cannot close the backdoor, so the stamp stays smoke. More papers of the same design do not settle the claim. Neighborhood confounding sits on the backdoor. Dirtier air tracks poorer ZIP codes, less schooling, more traffic noise, and different smoking. Hao adjusts ZIP-code income and Medicaid. The Scottish file adjusts education, occupation, and urban-rural class. Residual confounding remains. Hao's sample is Medicare at 65 and older, 84 percent White. The Scottish file is 202,237 adults from a census-linked register. Chen's 4,708 already have IBD. That is not a general-population death rate. Hao's curves for black carbon, organic matter, and dust bent at high levels. This page still names the sign from PM2.5 mass: more particles, more death. This is a journal entry, not medical advice and not a reason to move, to buy a filter, or to treat a number.

Effect

qualitative. Hao 2023 Medicare ~73.4M: PM2.5 mass HR 1.040 (1.039-1.041) per IQR 3.68 µg/m³. Scotland SLS 202,237: HR 1.035 (1.023-1.047) per 1 µg/m³. Chen IBD 4,708: HR 1.14 per IQR. Observational. Design does not identify.

Nodes

Lifestyle & environmentAmbient air pollutionBiology & endocrinologyAll-cause mortality

Named confounders

Cite this page

What Causes What. “Ambient air pollution → All-cause mortality.” https://whatcauseswhat.org/edges/e-pm25-mortality (atlas updated 2026-09-02).

https://whatcauseswhat.org/edges/e-pm25-mortality

Papers