Fine particulate matter, known scientifically as PM2.5, refers to particles smaller than 2.5 micrometers in diameter, roughly one-thirtieth the width of a human hair. Because of their tiny size, these particles can bypass the body’s natural defenses in the nose and throat, penetrate deep into the lungs, and even cross into the bloodstream, where they damage the heart, blood vessels, and other vital organs. For decades, PM2.5 has been recognized as the leading environmental risk factor for cardiopulmonary disease, and regulatory efforts under the Clean Air Act drove steady improvements in US air quality through the early 2000s. But that progress has begun to reverse. In extreme smoke years, wildfire emissions have accounted for more than half of the annual PM2.5 concentration in many parts of the country, and climate change is expected to expand the extent of global fire-prone areas by roughly 29 percent by the late twenty-first century.

The research team, led by investigators at the Harvard T.H. Chan School of Public Health, drew on the State Inpatient Databases, a collection of hospital discharge records that represents more than 95 percent of all inpatient admissions nationwide. They analyzed 57,264,966 hospitalizations for cardiovascular diseases and 32,691,394 for pulmonary diseases among residents of 20 states between 2006 and 2019, including heavily fire-affected states such as California, Oregon, and Washington. By linking each patient’s residential ZIP code to sophisticated spatiotemporal models of ambient pollution, the researchers estimated individual exposure to both wildfire-specific PM2.5 and non-wildfire PM2.5, then examined how two-year average concentrations related to the risk of hospital admission.

A central methodological innovation was the use of a self-controlled study design, an extension of the case-crossover approach traditionally reserved for short-term effects. Instead of comparing different people, the design compares each hospitalized individual with themselves during control years selected both before and after the hospitalization year. This structure inherently adjusts for time-invariant confounders such as genetic susceptibility, smoking history, and educational attainment, factors that are notoriously difficult to measure but can distort conventional epidemiological comparisons. The researchers additionally adjusted for measured time-varying covariates, including calendar year, neighborhood-level socioeconomic indicators, and seasonal temperature statistics, and applied Bonferroni-corrected confidence intervals to guard against false-positive findings across the many disease-outcome combinations tested.

The results were striking. For every 1-microgram-per-cubic-meter increase in two-year average wildfire-specific PM2.5, the risk of hospitalization rose by 13.1 percent for cerebrovascular disease, 12.9 percent for ischemic heart disease, 12.2 percent for hypertension, 11.7 percent for arrhythmia, and 10.0 percent for heart failure. Among pulmonary conditions, asthma showed the strongest association, with a 16.0 percent increase in hospitalization risk per unit of wildfire PM2.5, followed by acute respiratory infections and pneumonia. Non-wildfire PM2.5 was also associated with elevated risks across all examined conditions, but the per-unit effect sizes were consistently smaller, ranging from roughly 5.2 to 8.5 percent. In other words, when the same increment of pollution comes from a wildfire rather than a tailpipe or smokestack, the body appears to pay a substantially higher price.

This per-unit comparison, however, tells only part of the story. Because non-wildfire PM2.5 concentrations were approximately 20 to 27 times higher than wildfire-specific concentrations during the study period, the risk increase per standard deviation of exposure, a measure that better reflects real-world variation, was actually greater for non-wildfire sources. The average annual wildfire-specific PM2.5 concentration across the studied ZIP codes was just 0.29 micrograms per cubic meter, compared with 7.78 micrograms per cubic meter for non-wildfire PM2.5. California recorded the highest levels of both pollutant types. The researchers emphasize that this distinction matters for policy: wildfire smoke is currently more toxic per unit, but traditional sources still dominate the overall pollution burden, meaning both require attention.

Why would wildfire-derived particles be more dangerous? The answer likely lies in their unique chemical and physical composition. Wildfire-specific PM2.5 contains a higher proportion of submicron and ultrafine particles, elevated levels of elemental and organic carbon, and greater concentrations of polar organic compounds, including oxidative and pro-inflammatory substances such as polycyclic aromatic hydrocarbons. Because these particles are smaller, they penetrate more deeply into the respiratory tract and bloodstream. Toxicological experiments support the epidemiological findings: mice exposed to wildfire-derived particulate matter at doses equivalent to ambient levels developed stronger pulmonary inflammation and histological lung damage than mice exposed to comparable doses of PM from other sources, and exposure to particles from smoldering peat fires has been shown to trigger lung inflammation, reduced cardiac function, and post-ischemic myocardial infarction in animal models.

Perhaps the most consequential finding involved exposure duration. When the researchers tested alternative exposure windows using three- and four-year moving averages, the association between wildfire-specific PM2.5 and cardiopulmonary hospitalizations grew stronger with longer exposure, while the association with non-wildfire PM2.5 remained relatively stable. This pattern suggests that wildfire smoke exerts cumulative toxic effects that build up in the body over time, a mechanism consistent with prior evidence that chronic exposure to air pollutants poses substantially larger risks than acute exposures lasting only days. It also helps explain why earlier studies, which typically examined short-term wildfire episodes within a week of exposure, produced inconsistent and often null results for cardiovascular outcomes. One prior study found that cardiovascular effects of wildfire-related PM2.5 emerged on lag day 10, considerably later than the pulmonary effects, which appeared immediately.

The burden of this pollution is not distributed equally. Hispanic individuals and those from other racial and ethnic minority groups experienced greater cardiopulmonary hospitalization risks from both pollutant types, as did residents of metropolitan areas, communities with lower educational attainment, and communities with higher deprivation levels as measured by the Townsend Deprivation Index, which combines unemployment, household overcrowding, and lack of car and home ownership. Individuals with co-existing obesity or diabetes also faced heightened susceptibility, with obesity generally associated with stronger effects, possibly because obesity involves reduced cardiopulmonary function, chronic systemic inflammation, and elevated oxidative stress. The researchers attribute the racial and socioeconomic patterns to structural inequities, including poorer healthcare access, higher baseline pollution burdens, and fewer resources to mitigate smoke exposure. Metropolitan vulnerability may reflect interactions with other urban pollutants such as nitrogen dioxide and ozone, as well as social stress pathways that promote chronic inflammation.

The implications for air quality policy are profound. Current regulatory frameworks largely exempt wildfire smoke from determining attainment of the National Ambient Air Quality Standards, and traditional wildfire management approaches such as controlled burns have historically prioritized property protection over minimizing population smoke exposure. Under climate change scenarios, annual wildfire-specific PM2.5 concentrations in some US regions could reach 10 micrograms per cubic meter, levels that would rival or exceed today’s non-wildfire pollution. The study’s authors argue that wildfire management deserves far greater emphasis alongside conventional air quality control, and that the persistent, cumulative nature of smoke’s health effects demands long-term protective strategies rather than episodic responses. With wildfire now the fastest-growing source of ambient air pollution in the United States, the era of treating smoke as an unavoidable natural hazard, rather than a modifiable public health threat, may be coming to an end.

Subject of Research: Long-term cardiopulmonary hospitalization risks of wildfire-specific versus non-wildfire fine particulate matter exposure

Article Title: Cardiopulmonary hospitalization risks from wildfire-specific and non-wildfire PM2.5 in 20 US states

Article References: Zhang, M., Castro, E., Qiu, M., Danesh Yazdi, M., Li, B., Wright, R. J., Schwartz, J. D., O. Wright, R., & Wei, Y. (2026). Cardiopulmonary hospitalization risks from wildfire-specific and non-wildfire PM2.5 in 20 US states. Nature Communications, 17(1), Article 9349. https://doi.org/10.1038/s41467-026-76974-7

Image Credits: AI Generated

DOI: 10.1038/s41467-026-76974-7

Keywords: wildfire smoke, PM2.5, air pollution, cardiovascular disease, pulmonary disease, hospitalization risk, asthma, heart failure, climate change, environmental health, epidemiology, health disparities