Long-term exposure to wildfire smoke particulate matter and incident stroke: a US nationwide study.

Long-term exposure to wildfire smoke particulate matter and incident stroke: a US nationwide study - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Eur Heart J . 2026 Jan 27;47(21):2673–2682. doi: 10.1093/eurheartj/ehaf875 Long-term exposure to wildfire smoke particulate matter and incident stroke: a US nationwide study Hua Hao 1 , Ke Xu 2 , Danlu Zhang 3 , Yanling Deng 4 , Tarik Benmarhnia 5 , Qiang Pu 6 , Varun K Pattisapu 7 , Kyle Steenland 8 , Howard H Chang 9 , Alvaro Alonso 10 , Yang Liu 11, ✉ Author information Article notes Copyright and License information 1 Gangarosa Department of Environmental Health, Rollins School of Public Health, Emory University, 1518 Clifton Road NE, Atlanta, GA 30322, USA 2 Gangarosa Department of Environmental Health, Rollins School of Public Health, Emory University, 1518 Clifton Road NE, Atlanta, GA 30322, USA 3 Gangarosa Department of Environmental Health, Rollins School of Public Health, Emory University, 1518 Clifton Road NE, Atlanta, GA 30322, USA 4 Gangarosa Department of Environmental Health, Rollins School of Public Health, Emory University, 1518 Clifton Road NE, Atlanta, GA 30322, USA 5 Scripps Institution of Oceanography Department, University of California San Diego, La Jolla, CA, USA 6 College for Public Health and Social Justice, Saint Louis University, St. Louis, MO, USA 7 Department of Cardiology, Piedmont Heart Institute, Piedmont Hospital, Atlanta, GA, USA 8 Gangarosa Department of Environmental Health, Rollins School of Public Health, Emory University, 1518 Clifton Road NE, Atlanta, GA 30322, USA 9 Department of Biostatistics and Bioinformatics, Rollins School of Public Health, Emory University, Atlanta, GA, USA 10 Department of Epidemiology, Rollins School of Public Health, Emory University, Atlanta, GA, USA 11 Gangarosa Department of Environmental Health, Rollins School of Public Health, Emory University, 1518 Clifton Road NE, Atlanta, GA 30322, USA ✉ Corresponding author. Tel: +404 727 2131, Email: yang.liu@emory.edu Received 2025 Mar 4; Revised 2025 Jul 22; Accepted 2025 Oct 12; Collection date 2026 Jun 1. © The Author(s) 2026. Published by Oxford University Press on behalf of the European Society of Cardiology. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License ( https://creativecommons.org/licenses/by-nc/4.0/ ), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact reprints@oup.com for reprints and translation rights for reprints. All other permissions can be obtained through our RightsLink service via the Permissions link on the article page on our site—for further information please contact journals.permissions@oup.com. PMC Copyright notice PMCID: PMC13225884  PMID: 41587749 See " Wildfire smoke and stroke: emerging evidence on long-term cardiovascular risk. " with doi: 10.1093/eurheartj/ehaf1003. Abstract Background and Aims Wildfire smoke is a growing public health concern, particularly due to its high fine particulate matter (PM 2.5 ) concentrations. While short-term exposure has been linked to adverse cardiovascular outcomes, long-term effects remain understudied. This study examined the association between long-term wildfire smoke PM 2.5 exposure and incident stroke among older US adults. Methods A national open cohort study was conducted among Medicare Fee-for-Service beneficiaries aged ≥65 years from 2007 to 2018. Stroke cases were identified using validated Medicare claims. Wildfire smoke PM 2.5 exposure was estimated at a 1 km 2 resolution using a machine learning model distinguishing smoke-specific PM 2.5 from other ambient sources. Stratified Cox proportional hazards models with generalized estimating equations were used to estimate hazard ratios (HRs) per 1 µg/m 3 increase in wildfire smoke PM 2.5 , adjusting for non-smoke PM 2.5 , meteorological variables, socio-economic status, and healthcare capacity. Results Among ∼25 million beneficiaries, ∼2.9 million incident stroke cases occurred. Long-term wildfire smoke PM 2.5 exposure was significantly associated with increased stroke risk. Stronger associations were observed for longer exposure windows. For instance, per 1 µg/m 3 increase in 3-year average wildfire smoke PM 2.5 was associated with a 1.3% increase in stroke risk (HR: 1.013; 95% CI: 1.006, 1.020). Associations for wildfire smoke PM 2.5 were generally stronger than for non-smoke PM 2.5 per unit exposure. Conclusions This study provides national-scale evidence that chronic exposure to wildfire smoke PM 2.5 increases stroke risk in older adults. These findings underscore the long-term health risks of wildfire smoke and support policies aimed at reducing exposure, especially among vulnerable populations. Keywords: Wildfire, Stroke, Fine particulate matter, Long-term exposure, Elderly Structured Graphical Abstract Structured Graphical Abstract. Open in a new tab This nationwide cohort study of approximately 25 million US Medicare fee-for-service beneficiaries aged ≥65 years (2007–18) examined the association between long-term exposure to wildfire smoke–related fine particulate matter (PM 2.5 ) and incident stroke. Wildfire smoke PM 2.5 exposure, assigned at the ZIP code level, was significantly associated with higher stroke risk compared with non-smoke PM 2.5 , with stronger associations observed over longer exposure windows (1-, 2-, and 3-year averages). Findings suggest increased toxicity of wildfire-derived PM 2.5 and highlight wildfire smoke as a distinct environmental risk factor for stroke in older adults See the editorial comment for this article ‘Wildfire smoke and stroke: emerging evidence on long-term cardiovascular risk’, by S. Zhang and K. Chen, https://doi.org/10.1093/eurheartj/ehaf1003 . Introduction Cardiovascular disease (CVD) stands as one of the primary causes of death in the US and worldwide, 1 and approximately every 1 in 3 American adults have some form of prevalent CVD. 2 Stroke, a major type of CVD, is a leading cause of morbidity and mortality in the USA, with 795 000 people experiencing a new or recurrent stroke each year, resulting in ∼140 000 annual deaths. 3 , 4 The economic burden of stroke in the USA is estimated to be around $40 billion annually. 5 , 6 While stroke incidence among elderly has decreased over recent decades, this decline has begun to plateau compared with the steep decreases seen in the 1990s suggesting a change in the potential combination of risk factors. 7 Conventional risk factors such as hypertension, diabetes, and smoking do not fully explain the variations in stroke risk over time. 8 Emerging evidence suggests that environmental factors, particularly air pollution, may play a critical role in CVD risk. 9 The American Heart Association has identified a causal relationship between long-term exposure to fine particulate matter (PM 2.5 ; particle diameter < 2.5 µm) and cardiovascular morbidity and mortality including stroke. 10 Furthermore, several reviews have documented the association between chronic PM 2.5 exposure and increased stroke incidence, likely mediated by oxidative stress, systemic inflammation, endothelial dysfunction, and thrombotic activation, with animal studies also suggesting its role in blood–brain barrier disruption and neuroinflammation. 11 , 12 While emissions control policies have successfully reduced ambient PM 2.5 levels in the USA over the past few decades, 13 this trend is being reversed in some parts of the USA 14 given the growing frequency, intensity, and duration of wildfires—exacerbated by climate change. 15 Wildfires emit vast quantities of smoke that can travel over long distances, resulting in sharp increases in PM 2.5 concentrations in populated areas that can persist over extended periods, contributing to long-term air quality deterioration. 16 These recurrent pollution events raise concerns about the long-term health impacts of wildfire smoke exposure, 17 as episodic spikes in pollution can accumulate over time, causing chronic health conditions. Recent studies suggest that short-term exposure to wildfire smoke PM 2.5 has stronger health effects than PM 2.5 from other sources at the same concentration, likely due to its higher oxidative potential and unique composition. 18 , 19 Similar patterns have been observed for long-term exposure, including increased dementia risk, 20 raising concerns about potential cerebrovascular effects such as stroke. While short-term wildfire smoke exposure is well-documented, growing evidence suggests long-term exposure may have distinct and underestimated health effects. 21 , 22 A recent model estimates that nearly 40% of air quality monitoring sites experience over 30 days of smoke-related pollution annually, 23 underscoring the urgency of studying its long-term health implications as wildfire seasons lengthen and intensify. However, many previous studies have relied on proxy measures such as total PM 2.5 concentrations or classified exposure based on smoky days vs non-smoky days, introducing large exposure errors. 21 , 24 , 25 Addressing these limitations requires high-resolution, wildfire-specific PM 2.5 exposure data to accurately quantify long-term health risks, particularly for stroke. To bridge the above knowledge gaps, we conducted the first national, population-based open cohort study to quantify the associations between long-term wildfire smoke PM 2.5 exposure and stroke risk among US older adults, from 2007 to 2018. Leveraging nationwide Medicare claims data, we utilized high-resolution, smoke-specific PM 2.5 estimates generated by a two-stage machine learning model to ensure precise exposure assessment. Methods Study population We constructed our study cohort using the national Medicare denominator file and the Medicare Chronic Conditions Warehouse (CCW) dataset from the Centers for Medicare and Medicaid Services (CMS). The CCW claims data contains predefined indicators for chronic conditions among Medicare Fee-for-Service (FFS) patients, and it provides the year of the initial event and the corresponding diagnosis code for a specific disease. The denominator file contains each Medicare beneficiary’s enrolment information, which was updated annually. This information includes demographics, the death date (if any), the beneficiary’s ZIP code, and Medicaid insurance enrolment, serving as a proxy indicator for socio-economic status (SES). The study included all Medicare beneficiaries continuously enrolled in both the Medicare FFS programme and Parts A (hospital insurance) and B (medical insurance) in the contiguous USA from 2007 to 2018. We tracked each enrollee’s annual residential zip code changes over time. The year of initial stroke diagnosis was identified per individual and linked to the denominator. A 3-year washout period was applied after the first enrolment, during which no stroke diagnoses occurred. This approach helped ensure that only new stroke cases were captured. By removing prevalent cases of stroke during the initial 3 years of follow-up, we improved our ability to approximate stroke incidence in later life. We determined 3 years to be a reasonable timeframe to reduce the likelihood of including pre-existing stroke cases. This decision is consistent with previous Medicare-based studies on chronic disease outcomes, 26–29 which used clean periods ranging from 1 to 5 years. While longer clean periods may improve confidence in capturing incident cases, they also substantially reduce sample size and statistical power, especially in older adult populations. Since we limited enrolment to beneficiaries aged 65 and older at baseline, applying a 3-year washout period resulted in a study population that was at least 68 years old at the beginning of follow-up. In the final cohort, participants entered the study the year following the washout period and were followed until stroke diagnosis, death, or end of the study period. To avoid immortal time bias, 30 the washout period was excluded from follow-up time, meaning follow-up started as early as 2010. The Medicare dataset has been preserved and processed in a secure computing infrastructure compliant with the Health Insurance Portability and Accountability Act (HIPAA) at the Rollins School of Public Health of Emory University. Outcome classification Stroke cases were identified through an algorithm within the CCW that utilizes data from various Medicare claims, including hospital inpatient, outpatient, and carrier claims (such as physician visits), all coded with valid International Classification of Diseases, Ninth Revision (ICD-9) or Tenth Revision (ICD-10) codes. 26 , 27 The year of the first stroke diagnosis was obtained from the CCW following the beneficiary’s enrolment in the Medicare FFS programme and both Part A and Part B. Detailed ICD codes and the algorithm used for identifying stroke conditions in Medicare CCW can be found in Supplementary data online , Table S1 . Exposure assessment Daily wildfire smoke PM 2.5 and non-smoke PM 2.5 concentrations for the contiguous United States (CONUS) from 2007 to 2018 were estimated at 1 km 2 spatial resolution using a two-stage, twin-model machine learning modelling framework. A comprehensive description of this framework is published elsewhere and summarized here. 23 On a daily basis, 1 km 2 grid cells in the CONUS domain were classified as smoke-impacted or smoke-free based on fire smoke contours from the NOAA Hazard Mapping System 31 , 32 and wildfire-attributable contributions to total PM₂.₅ simulated by the Community Multiscale Air Quality Modeling System (CMAQ). The CMAQ provided a mechanistic estimate of how much PM 2.5 in a given grid cell and day could be attributed to biomass burning emissions, enabling classification of smoke-impacted areas. 33 Two separate random forest models were trained: one using data from smoke-impacted grid cells to predict total PM 2.5 and another using data from smoke-free grid cells to predict background non-smoke PM 2.5 . The smoke-specific PM 2.5 concentration was then derived by subtracting the predicted non-smoke PM 2.5 from the predicted total PM 2.5 within smoke-impacted cells. The modelling framework integrated multiple data sources including satellite Aerosol Optical Depth, meteorological variables, ground-based regulatory monitors, and low-cost sensors (e.g. PurpleAir). The total PM 2.5 model demonstrated strong predictive performance, with a cross-validated R 2 of .9 at monthly and annual scales. The smoke-specific PM 2.5 model also performed well, with a daily cross-validated R 2 of .75 in smoke-impacted areas, as reported in Zhang et al . 23 To link the exposure with the Medicare data, we calculated annual wildfire smoke and non-smoke PM 2.5 concentrations at the ZIP code level by averaging the daily 1 km 2 PM 2.5 concentrations whose centroids fall within the ZIP code boundary and for each year of the study period. Exposure levels were then estimated for each participant based on their residential ZIP code. Moving averages for wildfire smoke and non-smoke PM 2.5 were calculated for past 1-, 2-, and 3-year windows for each follow-up year, representing long-term exposure. Building on previous research 17 to assess long-term wildfire PM 2.5 exposure, we calculated three additional metrics to capture key wildfire characteristics: frequency, duration, and intensity. Specifically, we calculated the number of days with wildfire smoke PM 2.5 > 0 µg/m 3 , >1 µg/m 3 , and >2.5 µg/m 3 using 1-, 2-, and 3-year moving averages to characterize long-term exposure to varying smoke intensities. The >0 µg/m 3 threshold was used to capture any detectable smoke exposure, while >1 µg/m 3 and >2.5 µg/m 3 correspond approximately to the 75th and 95th percentiles, respectively, of the daily wildfire smoke PM 2.5 distribution in our dataset. These empirically derived thresholds allowed us to characterize a range of smoke exposure conditions, from moderate, frequent events to rarer, higher-intensity episodes. Covariates Individual-level data on age, sex, race, Medicaid status, and date of death were obtained from the Medicare denominator file. For each beneficiary, based on their ZIP code of residence and follow-up year, we incorporated information on US geographic regions (i.e. Northeast, Southeast, Midwest, Southwest, and West); ZIP code-level meteorological variables (relative humidity and temperature) from the North American Land Data Assimilation System (NLDAS); ZIP code-level healthcare capacity-related information (total number of beds regularly maintained, total number of active medical doctors per 1000 people, and the number of hospitals) from the American Hospital Association Annual Survey Database; ZIP code-level behavioural risk factors (smoking rate) from the Behavioral Risk Factor Surveillance System (BRFSS); and ZIP code-level SES factors, including population, poverty, education, median household income, per cent Black, per cent White, per cent using auto transportation, and per cent renting obtained from the US Census and the American Community Survey. Alcohol use disorder, drug use disorder, and tobacco use disorder at the individual level were derived from the ‘Other Chronic or Potentially Disabling Conditions’ segment of the CCW. Statistical analysis We applied stratified Cox proportional hazards models with generalized estimating equations (GEE) to account for clustering of individuals within ZIP codes, recognizing that residents in the same area may share similar environmental exposures, socio-economic characteristics, healthcare access, and other contextual factors that could influence stroke risk. By using GEE, we adjusted for within-ZIP code correlation, ensuring more robust standard errors and reliable inference when estimating hazard ratios (HRs) of incident stroke risk among Medicare enrollees exposed to long-term wildfire smoke PM 2.5 . Long-term exposure was measured using nine metrics, which included the moving averages of wildfire smoke PM 2.5 concentrations over the past 1, 2, and 3 years, as well as the number of days with wildfire smoke PM 2.5 exceeding 0 µg/m 3 , 1 µg/m 3 , and 2.5 µg/m 3 over the same time periods. All models were adjusted for non-smoke PM 2.5 , meteorological variables (including annual mean of relative humidity and temperature), SES variables, behavioural risk factors, and healthcare capacity-related covariates. Non-smoke PM 2.5 , temperature, and relative humidity were assigned using the same 1-, 2-, and 3-year moving average windows as wildfire smoke PM 2.5 to ensure temporal consistency. All exposure and covariate variables were linked annually by ZIP code and follow-up year and modelled as time-varying covariates. To account for potential residual temporal and spatial patterns, categorical indicators of the calendar year and geographic regions were also added to the model. To avoid collinearity, only covariates with correlations below .7 were included (see Supplementary data online , Figure S1 ). The final model adjusted variables of non-smoke PM 2.5, relative humidity, temperature, calendar year, geographic region, total number of beds, smoking rate, population, per cent Black, per cent poverty, per cent using auto transportation, and per cent renting. All covariates were included as linear terms unless noted otherwise. Person-year observations with missing values in any covariates were excluded. The overall proportion of missingness was low (4.14%) and therefore unlikely to introduce substantial bias. To allow for flexible baseline hazard functions, models were stratified by sex, race (White, Black, other), Medicaid eligibility, and age at study entry in 1-year increments. Hazard ratios with 95% confidence intervals (CIs) per 1 µg/m 3 increase in smoke or non-smoke PM 2.5 concentrations were reported. We further quantified the stroke burden attributable to smoke PM 2.5 exposure using two indicators: the number of attributable cases and the population attributable fraction (PAF), both of which reflect the population-level impact of exposure. 34 Additionally, penalized spline models were used to test for potential non-linear concentration–response (C-R) relationships between smoke PM 2.5 and stroke, controlling for the same covariates across models. We further evaluated effect modification by entry age, sex, and socio-economic indicators, including Medicaid dual enrolment and ZIP code-level median household income, by fitting separate models within each subgroup. These modifiers were chosen based on prior research indicating their relevance in influencing vulnerability to PM 2.5 -related cardiovascular outcomes. 35 We conducted a number of sensitivity studies to assess the robustness of our findings. First, we repeated the analyses using a washout period of 5 years. Compared with the primary analysis, which used a 3-year washout period, this would enhance the likelihood of capturing incidents rather than prevalent diagnosis. Second, we conducted a non-mover cohort analysis by only including enrollees who did not relocate during the follow-up period to reduce potential measurement errors related to address change. Third, although we used ambient air pollution as the exposure metric in this analysis, which is more closely associated with area-level factors and less influenced by individual-level confounders, we still accounted for potential confounding by including personal-level variables such as alcohol use disorders, drug use disorders, and tobacco use disorders in the model. These variables were excluded from the primary models because they were sourced from the ‘Other Chronic or Potentially Disabling Conditions’ segment of the CCW, which employs a different diagnostic identification approach than the standard annual algorithm used for other covariates. Additionally, these variables are more likely to represent severe clinical diagnoses rather than general behavioural tendencies, which could introduce misclassification. As a result, we used them only in sensitivity analyses to test the robustness of our results. Fourth, we implemented a censoring-by-death weighting (CDW) approach to account for death as a competing risk, using stabilized inverse probability weights derived from logistic regression models of mortality. 36 , 37 Last, to account for potential state-level heterogeneity in baseline stroke risk and unmeasured contextual factors, we further stratified the models by US state. The Rollins High Performance Computing Cluster at Emory University was used for all computations, and all statistical analyses were implemented in R (version 4.0.2). Results Study population characteristics Table 1 presents the demographic statistics and exposure distributions of our stroke study cohort. We have included approximately 25 million Medicare beneficiaries in each cohort, with a mean follow-up period of 5 years. The mean age at entry of the Medicare programme was around 75 years, females took a higher proportion (57.36%), White race was the majority (87.90%), and ∼12% of the cohort populations were eligible for the Medicaid programme. We observed ∼2.9 million stroke events (12.06%) during the follow-up periods. Supplementary data online , Table S2 presents additional ZIP code level demographic information. Supplementary data online , Figure S2A illustrates the ZIP code-level incidence of stroke events per 100 000 Medicare enrollees across the contiguous USA, with high-rate clusters observed in the Southeast. Supplementary data online , Figure S2B shows the temporal trend of stroke incidence among Medicare enrollees from 2010 to 2018, revealing a slow gradual decline over time. Table 1. Demographic characteristics and exposure summary of the stroke study cohort after a 3-year clean period Characteristics Stroke cohort (% or IQR) Number of events 2 981 141 (12.06%) Number of total populations 24 716 855 Total person-years 132 754 940 Median follow-up years 5 Mean age at entry 75.05 Gender Male 10 540 489 (42.64%) Female 14 176 366 (57.36%) Race White 21 725 231 (87.90%) Black 1 654 369 (6.69%) Other 1 337 255 (5.41%) Medicaid eligibility Eligible 2 946 268 (11.92%) Ineligible 21 770 587 (88.08%) Air pollution concentrations (µg/m 3 ) Mean (IQR) Wildfire smoke PM 2.5 (WSP) Moving average of past 1 year .53 (.59) Moving average of past 2 years .50 (.47) Moving average of past 3 years .51 (.48) Days with WSP > 1 µg/m 3 of past 1 year 83 (65) Average days with WSP > 1 µg/m 3 of past 2 years 82 (65) Average days with WSP > 1 µg/m 3 of past 3 years 83 (66) Non-smoke PM 2.5 Moving average of past 1 year 8.55 (3.42) Moving average of past 2 years 8.56 (3.43) Moving average of past 3 years 8.61 (3.55) Open in a new tab Air pollution levels The mean concentrations for wildfire smoke PM 2.5 for the past 1-, 2-, and 3-year exposure metrics were .53 μg/m 3 , .50 μg/m 3 , and .51 μg/m 3 , respectively. The corresponding mean non-smoke PM 2.5 concentrations were ∼8.55 μg/m 3 , 8.56 μg/m 3 , and 8.61 μg/m 3 . The average number of days with wildfire smoke PM 2.5 > 1 μg/m 3 over the past 1-, 2-, and 3-year were 83, 82, and 83 days ( Table 1 ). Additional details on exposure distribution by each metric are provided in Supplementary data online , Table S3 . Figure 1 presents the geographic distribution of wildfire smoke and non-smoke PM 2.5 across the contiguous USA, averaged from 2010 to 2018. Wildfire smoke PM 2.5 concentrations were highest in the West (mean: .74 μg/m 3 ) and Southeast (mean: .80 μg/m 3 ), while non-smoke PM 2.5 concentrations peaked in the Midwest (mean: 9.11 μg/m 3 ) and west (mean: 8.13 μg/m 3 ). Supplementary data online , Figure S3 shows the temporal distributions of wildfire smoke and non-smoke PM 2.5 during the study period. Smoke PM 2.5 peaked sharply in 2017, showing significant fluctuations throughout the follow-up, while non-smoke PM 2.5 displayed a relatively gradual decreasing trend over the study period. Supplementary data online , Table S4 shows the correlation coefficients between wildfire smoke PM 2.5 and non-smoke PM 2.5 across the three continuous exposure metrics, indicating low correlations between them. Figure 1. Open in a new tab ( A ) Mean annual wildfire smoke PM 2.5 concentrations (µg/m 3 ) across the contiguous USA during the study period. ( B ) Mean annual non-smoke PM 2.5 concentrations (µg/m 3 ) across the contiguous USA during the study period Health effects estimates Figure 2 and Supplementary data online , Table S5 present the associations between long-term exposure to wildfire smoke and non-smoke PM 2.5 and stroke incidence among older adults in the USA. Both wildfire smoke and non-smoke PM 2.5 were associated with elevated stroke risk. For the past 1-year exposure window, the HR per 1 µg/m 3 increase in wildfire smoke PM 2.5 was lower than for non-smoke PM 2.5 . However, at longer exposure windows, HRs for wildfire smoke PM 2.5 increased, while HRs for non-smoke PM 2.5 remained relatively stable. Specifically, each 1 µg/m 3 increase in 3-year average wildfire smoke PM 2.5 was associated with a 1.3% increase in stroke risk (HR: 1.013; 95% CI: 1.006–1.020), compared with a .7% increase for non-smoke PM 2.5 (HR: 1.007; 95% CI: 1.006–1.008), suggesting potentially greater toxicity of smoke-derived particles. Additionally, long-term exposure—measured as 1-, 2-, and 3-year moving averages—of days with wildfire smoke PM 2.5 > 1 µg/m 3 and > 2.5 µg/m 3 was associated with a higher risk of stroke (see Supplementary data online , Figure S4 ). Assuming a baseline annual stroke incidence rate of 2211 per 100 000 Medicare beneficiaries, 26 this corresponds to an estimated 17 226 (95% CI: 7,830, 26,100) additional stroke cases per year and a PAF of 1.30% (95% CI: .60%, 1.96%) among US older adults. Figure 2. Open in a new tab Hazard ratios of stroke associated with per 1 µg/m 3 increase in mean concentration of wildfire smoke and non-smoke PM 2.5 across different exposure windows (past 1, 2, and 3 years). The estimated hazard ratios were obtained from Cox proportional hazard models, and error bars stand for the 95% confidence intervals. The corresponding hazard ratio values can be found in Supplementary data online , Table S5 Concentration–response relationships Figure 3 illustrates the estimated C-R relationship for wildfire smoke and non-smoke PM 2.5 and stroke and three different exposure metrics (moving average past 1, 2, and 3 years). Linear associations with no apparent threshold were observed for all three wildfire smoke PM 2.5 exposure windows. For non-smoke PM 2.5 , the relationships are essentially near linear until high and rarely occurring concentrations. For the C-R curves based on smoke day metrics using the >2.5 µg/m 3 cut point, we observed a positive and approximately linear association at commonly occurring concentrations among all exposure windows (see Supplementary data online , Figure S5 ). Figure 3. Open in a new tab The concentration–response curves for each exposure metric (moving average of past 1, 2, and 3 years) of wildfire smoke and non-smoke PM 2.5 and stroke. The concentration–response curves are shown for exposure windows concentration range between .025th and 99.975th percentiles of the pollutants, i.e. with .05% poorly constrained extreme values excluded Stratification analysis We examined effect modification by individual and community-level characteristics and observed variation in the associations between wildfire smoke PM 2.5 and stroke incidence risk. Among all subgroups, Medicaid-eligible individuals—a proxy for lower personal SES—consistently showed stronger associations across all three exposure windows (moving average of past 1, 2, and 3 years). In contrast, other modifiers showed less consistent patterns across time windows (see Supplementary data online , Table S6 ). Sensitivity analysis Relationships between long-term exposure to wildfire smoke PM 2.5 and CVD outcomes were consistent in multiple sets of sensitivity assessments. First, a more rigorous ‘cleaner period’ that excluded anyone diagnosed with stroke in their first 5 years of follow-up showed similar results to the primary analysis (see Supplementary data online , Table S7 ). Second, after we restricted the analysis to participants who did not move throughout the follow-up period to adjust for address change bias, the results agreed with our main findings, suggesting statistically precise and positive associations between exposure to wildfire smoke PM 2.5 and stroke incidence (see Supplementary data online , Table S8 ). Third, our model yielded similar results even after incorporating personal-level covariates (see Supplementary data online , Table S9 ). Fourth, when accounting for death as a competing risk using CDWs, the results remained consistent with those of the main analysis (see Supplementary data online , Table S10 ). Last, stratifying the model by US state yielded results consistent with the primary analysis (see Supplementary data online , Table S11 ). Discussion Our study revealed significant associations between long-term exposure to wildfire smoke PM 2.5 and an increased risk of stroke incidence ( Structured Graphical Abstract ). This is the first study focusing on long-term effects of wildfire smoke PM 2.5 on stroke incidence. Our findings aligned with recently published systematic reviews by Reid et al . 38 and Chen et al . 39 that exposure to wildfire smoke is associated with acute cardiovascular and cerebrovascular outcomes. However, most prior research on stroke morbidity linked to wildfire smoke has focused on transient effects, such as emergency department visits and hospitalizations under acute exposure. Our study, in contrast, extends this knowledge by illuminating the potent impact of long-term wildfire smoke PM 2.5 on stroke. While the increase in stroke risk per unit of wildfire smoke PM 2.5 was relatively small at the individual level, the potential population-level burden is substantial given the widespread exposure among older adults. We estimate that a long-term increase of just 1 µg/m 3 in wildfire smoke PM 2.5 could result in approximately 17 226 additional stroke cases annually in the US elderly population. These findings underscore the public health significance of chronic wildfire smoke exposure and the urgency of effective mitigation strategies. Interestingly, we observed stronger wildfire smoke PM 2.5 effects on stroke than non-smoke PM 2.5 per 1 μg/m 3 change in exposure, suggesting that wildfire smoke PM 2.5 impacts cerebrovascular health more than PM 2.5 from other sources, with the most marked difference seen in the moving average of past 3-year results. This aligns with previous evidence showing similar patterns for other outcomes. 20 While our study contributes to the limited research comparing wildfire smoke and ambient PM 2.5 , direct comparisons of toxicity remain challenging due to differences in exposure distributions and chemical composition. Recent in vivo animal studies indicated that particles in wildfire smoke are more harmful to the lungs than equal dosages from ambient air pollution sources. 40 This observed toxicity has manifested in increased neutrophil count and protein concentration within the pulmonary lavage fluid. 38 Previous clinical studies have demonstrated that neutrophils can exacerbate cardiac damage through the release of granular components, reactive oxygen species, and pro-inflammatory mediators. 41 Wildfire smoke PM 2.5 possesses a different chemical composition, characterized by a higher proportion of polar organic compounds, 42 containing a substantial organic carbon component (at least 50%) and a notable element carbon content (5%–20%). 39 This composition renders it more susceptible to oxidation when compared with ambient PM 2.5 . Moreover, the polar organic compounds in wildfire smoke have the propensity to produce additional free radicals, which raises the risk of increased inflammation and oxidative stress and further triggers CVD onsets. 43 Additionally, variations in the duration and intensity of particulate matter exposure may underlie disparities in the magnitude of the inflammatory response. 10 Unlike PM 2.5 generated from urban resources, wildfire smoke PM 2.5 can exhibit dramatic fluctuations in concentration levels, contingent on the durations and intensities of wildfire events. 44 Finally, the physical and mental stress produced by the fear of a wildfire, evacuation, possible unemployment, and lack of healthcare access after the disaster might increase CVD incidence. Several prior studies have established links between CVD and natural disasters, such as hurricanes, typhoons, and earthquakes, 45 and those effects can sometimes last for up to 6 years after the disaster. 46 Our research also found that long-term exposure to non-smoke PM 2.5 is associated with an increased risk of stroke incidence. Although previous meta-analyses 11 , 12 concluded that long-term ambient PM 2.5 exposure is linked to incident stroke, findings from US studies have been inconsistent. Alexeeff et al .’s meta-analysis, 12 which pooled 14 studies, included five from the USA. Among these, Lipsett et al . 47 conducted a prospective cohort study of over 100 000 California teachers, with a median follow-up of 8 years, reporting a marginally significant association between long-term PM 2.5 exposure and stroke [HR: 1.14 (95% CI: .99, 1.32)]. Similarly, Hart et al . 48 followed 114 537 women in the Nurses’ Health Study and also found a marginal association [HR: 1.18 (95% CI: .96, 1.45)]. Conversely, Puett et al .’s study 49 of 17 545 men with higher SES living in the midwestern and northeastern USA reported no association [HR: .69 (95% CI: .39, 1.21)]. Another prospective cohort study of 6575 Ohio residents who underwent coronary angiography also found no association [HR: 1.34 (95% CI: .23, 7.93)]. 50 However, Miller et al .’s study, 51 which followed 65 893 postmenopausal women without previous CVD for a median of 6 years, reported a significant association between long-term PM 2.5 exposure and stroke risk [HR: 1.28 (95% CI: 1.02, 1.61)]. In our study, we observed that the adverse effects of wildfire smoke PM 2.5 on stroke risk became stronger with longer exposure windows and higher smoke-day thresholds—reflecting both increasing cumulative concentrations and more frequent or intense smoke days over time. This pattern was not observed with non-smoke PM 2.5 , which showed relatively stable associations across different exposure durations. The trend for wildfire smoke PM 2.5 may be explained by a combination of cumulative exposure burden and intensity-driven toxicity. 52 Longer durations of wildfire smoke exposure may promote systemic inflammation, endothelial dysfunction, and neurovascular stress—key mechanisms in stroke development. 38 , 53 In addition, repeated exposure to days with higher smoke concentrations may lead to disproportionate biological effects, as more intense wildfire PM 2.5 has been shown to contain elevated levels of reactive and toxic constituents. 39 , 42 Epidemiological evidence on the chronic effects of smoke PM 2.5 on cardiovascular and cerebrovascular outcomes has been limited. One recent study showed that long-term smoke PM 2.5 exposure is linked to cardiovascular mortality, 54 but no research has examined the effects of long-term cumulative exposure to wildfire smoke. For short-term effects, Requia et al . 55 conducted a national study in Brazil with more than 2 million cardiovascular hospital admissions but found no specific patterns with 1–5 day moving averages. Future research is warranted to reproduce our findings on cumulative effects in different populations and settings. While several studies have explored the cumulative effects of long-term exposure to ambient PM 2.5 on cardiovascular outcomes, their conclusions remain inconclusive. For example, a national study in the USA suggested that cardiovascular mortality risk increases with longer exposure windows. 56 However, another national study from South Korea reported no cumulative effect of ambient PM 2.5 exposure on cardiovascular cause-specific deaths. 57 Additionally, a Danish cohort study with more than 15 years of follow-up reported that shorter ambient PM 2.5 exposure windows (5 years and 10 years) showed similar results for CVD-related deaths compared with a 15-year exposure window. 58 Our findings on non-smoke PM 2.5 align with the South Korean study and the Danish study. The assessment of the dose–response relationship is crucial for informing public health policies. We observed linear relationships with no threshold of smoke PM 2.5 exposure windows for stroke, suggesting that there is no safe dose of wildfire smoke. A previous multi-country study reported approximately threshold type C-R curves for stroke incidence and long-term ambient PM 2.5 , 59 but an Indian national study reported a linear relationship. 60 These disparities underscore the need for future studies to validate our findings. Our study has several strengths. First, this is the first large-scale national cohort study to explore the associations between long-term wildfire smoke PM 2.5 exposure and stroke incidence in the contiguous USA. About 25 million enrollees with millions of incidence cases were included in our analysis and provided ample statistical power. Second, the high-resolution smoke PM 2.5 exposure data enabled us to evaluate the effect of wildfire smoke PM 2.5 on stroke more precisely compared with most previous research that rely on highly uncertain proxies, such as ambient total PM 2.5 concentrations during fire seasons. Third, the CCW database collected data from various healthcare settings like inpatient and outpatient settings. Unlike previous studies that used hospital data, our study cohorts, coupled with a 3-year clean period in data collection, more accurately represent the incidence (rather than prevalence) occurrence of stroke. Our study also has a few limitations. First, despite excellent model performance, our predicted smoke PM 2.5 concentrations may still include measurement errors. Since the estimated smoke PM 2.5 concentration error is presumably independent of the disease outcome status, any biases are expected to be towards the null. Second, there may be unmeasured confounding due to the lack of control for individual-level stroke risk factors. However, since we used ambient air pollution as the exposure metric rather than personal exposure, it is more closely associated with area-level factors, which we have already controlled for, reducing the potential for confounding. 61 Furthermore, sensitivity analyses showed that the model results remained consistent even after incorporating personal-level covariates. Third, while our findings suggest a stronger association per unit increase in wildfire smoke PM 2.5 compared with non-smoke PM 2.5 , we acknowledge that these two pollutant types differ in their distribution patterns and chemical composition. Therefore, direct comparisons should be interpreted with caution, and additional toxicological and epidemiological research is warranted to better characterize the relative toxicity of different PM 2.5 sources. Finally, our results only account for the Medicare FFS enrollees. We potentially missed outcome cases in our data because we do not include the Medicare-HMO (Medicare managed care plan). Therefore, our results cannot be generalized to the entire Medicare population. In conclusion, our study provides strong evidence that chronic exposure to PM 2.5 from wildfire smoke increases the risk of developing stroke. Our findings indicate that regulations leading to effective wildfire management and prevention can have important health benefits for the elderly American population. Supplementary Material ehaf875_Supplementary_Data ehaf875_supplementary_data.docx (965.3KB, docx) Contributor Information Hua Hao, Gangarosa Department of Environmental Health, Rollins School of Public Health, Emory University, 1518 Clifton Road NE, Atlanta, GA 30322, USA. Ke Xu, Gangarosa Department of Environmental Health, Rollins School of Public Health, Emory University, 1518 Clifton Road NE, Atlanta, GA 30322, USA. Danlu Zhang, Gangarosa Department of Environmental Health, Rollins School of Public Health, Emory University, 1518 Clifton Road NE, Atlanta, GA 30322, USA. Yanling Deng, Gangarosa Department of Environmental Health, Rollins School of Public Health, Emory University, 1518 Clifton Road NE, Atlanta, GA 30322, USA. Tarik Benmarhnia, Scripps Institution of Oceanography Department, University of California San Diego, La Jolla, CA, USA. Qiang Pu, College for Public Health and Social Justice, Saint Louis University, St. Louis, MO, USA. Varun K Pattisapu, Department of Cardiology, Piedmont Heart Institute, Piedmont Hospital, Atlanta, GA, USA. Kyle Steenland, Gangarosa Department of Environmental Health, Rollins School of Public Health, Emory University, 1518 Clifton Road NE, Atlanta, GA 30322, USA. Howard H Chang, Department of Biostatistics and Bioinformatics, Rollins School of Public Health, Emory University, Atlanta, GA, USA. Alvaro Alonso, Department of Epidemiology, Rollins School of Public Health, Emory University, Atlanta, GA, USA. Yang Liu, Gangarosa Department of Environmental Health, Rollins School of Public Health, Emory University, 1518 Clifton Road NE, Atlanta, GA 30322, USA. Supplementary data Supplementary data are available at European Heart Journal online. Declarations Disclosure of Interest All authors declare no disclosure of interest for this contribution. Data Availability The Medicare dataset used in this study is maintained in a secure computing infrastructure compliant with HIPAA at the Rollins School of Public Health, Emory University. Due to data privacy regulations, individual-level Medicare data cannot be publicly shared but may be available upon reasonable request to the Center for Medicare & Medicaid Services (CMS). Funding This study was supported by the National Institutes of Health (R01ES034175). The funding agency had no role in the study design, data collection, analysis, interpretation, or manuscript preparation. Ethical Approval This study was conducted using de-identified Medicare claims data and was approved by the relevant institutional review board. Pre-registered Clinical Trial Number None supplied. References 1. Heron  M. Deaths: leading causes for 2016. Natl Vital Stat Rep  2018;67:1–77. [ PubMed ] [ Google Scholar ] 2. Foy  AJ, Mandrola  JM. Heavy heart: the economic burden of heart disease in the United States now and in the future. Prim Care  2018;45:17–24. 10.1016/j.pop.2017.11.002 [ DOI ] [ PubMed ] [ Google Scholar ] 3. Fisher  JA, Puett  RC, Laden  F, Wellenius  GA, Sapkota  A, Liao  D, et al.  Case-crossover analysis of short-term particulate matter exposures and stroke in the health professionals follow-up study. Environ Int  2019;124:153–60. 10.1016/j.envint.2018.12.044 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 4. Benjamin  EJ, Blaha  MJ, Chiuve  SE, Cushman  M, Das  SR, Deo  R, et al.  American heart association statistics committee and stroke statistics subcommittee. Heart disease and stroke statistics-2017 Update: a report from the American Heart Association. Circulation  2017;135:e146–603. 10.1161/CIR.0000000000000485 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 5. Khan  SU, Khan  MZ, Khan  MU, Khan  MS, Mamas  MA, Rashid  M, et al.  Clinical and economic burden of stroke among young, midlife, and older adults in the United States, 2002–2017. Mayo Clin Proc Innov Qual Outcomes  2021;5:431–41. 10.1016/j.mayocpiqo.2021.01.015 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 6. Skolarus  LE, Freedman  VA, Feng  C, Wing  JJ, Burke  JF. Care received by elderly US stroke survivors may be underestimated. Stroke  2016;47:2090–5. 10.1161/STROKEAHA.116.012704 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 7. Leasure A . U.S. Stroke Rate Declining in Adults 75 and Older, Yet Rising in Adults 49 and Younger , 2022. https://newsroom.heart.org/news/u-s-stroke-rate-declining-in-adults-75-and-older-yet-rising-in-adults-49-and-younger (17 November 2025, date last accessed). 8. Kulick  ER, Kaufman  JD, Sack  C. Ambient air pollution and stroke: an updated review. Stroke  2023;54:882–93. 10.1161/STROKEAHA.122.035498 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 9. Montone  RA, Camilli  M, Calvieri  C, Magnani  G, Bonanni  A, Bhatt  DL, et al.  Exposome in ischaemic heart disease: beyond traditional risk factors. Eur Heart J  2024;45:419–38. 10.1093/eurheartj/ehae001 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 10. Brook  RD, Rajagopalan  S, Pope  CA, Brook  JR, Bhatnagar  A, Diez-Roux  AV, et al.  Particulate matter air pollution and cardiovascular disease: an update to the scientific statement from the American Heart Association. Circulation  2010;121:2331–78. 10.1161/CIR.0b013e3181dbece1 [ DOI ] [ PubMed ] [ Google Scholar ] 11. Yuan  S, Wang  J, Jiang  Q, He  Z, Huang  Y, Li  Z, et al.  Long-term exposure to PM(2.5) and stroke: a systematic review and meta-analysis of cohort studies. Environ Res  2019;177:108587. 10.1016/j.envres.2019.108587 [ DOI ] [ PubMed ] [ Google Scholar ] 12. Alexeeff  SE, Liao  NS, Liu  X, Van Den Eeden  SK, Sidney  S. Long-term PM(2.5) exposure and risks of ischemic heart disease and stroke events: review and meta-analysis. J Am Heart Assoc  2021;10:e016890. 10.1161/JAHA.120.016890 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 13. McClure  CD, Jaffe  DA. US particulate matter air quality improves except in wildfire-prone areas. Proc Natl Acad Sci U S A  2018;115:7901–6. 10.1073/pnas.1804353115 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 14. Burke  M, Childs  ML, de la Cuesta  B, Qiu  M, Li  J, Gould  CF, et al.  The contribution of wildfire to PM(2.5) trends in the USA. Nature  2023;622:761–6. 10.1038/s41586-023-06522-6 [ DOI ] [ PubMed ] [ Google Scholar ] 15. Wildfires . World Health Organ 2021 . https://www.who.int/health-topics/wildfires (29 October 2021, date last accessed). 16. Baars  H, Ansmann  A, Ohneiser  K, Haarig  M, Engelmann  R, Althausen  D, et al.  The unprecedented 2017–2018 stratospheric smoke event: decay phase and aerosol properties observed with the EARLINET. Atmos Chem Phys  2019;19:15183–98. 10.5194/acp-19-15183-2019 [ DOI ] [ Google Scholar ] 17. Casey  JA, Kioumourtzoglou  MA, Padula  A, Gonzalez  DJX, Elser  H, Aguilera  R, et al.  Measuring long-term exposure to wildfire PM(2.5) in California: time-varying inequities in environmental burden. Proc Natl Acad Sci U S A  2024;121:e2306729121. 10.1073/pnas.2306729121 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Aguilera  R, Corringham  T, Gershunov  A, Benmarhnia  T. Wildfire smoke impacts respiratory health more than fine particles from other sources: observational evidence from Southern California. Nat Commun  2021;12:1493. 10.1038/s41467-021-21708-0 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 19. Connolly  R, Marlier  ME, Garcia-Gonzales  DA, Wilkins  J, Su  J, Bekker  C, et al.  Mortality attributable to PM(2.5) from wildland fires in California from 2008 to 2018. Sci Adv  2024;10:eadl1252. 10.1126/sciadv.adl1252 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 20. Elser  H, Frankland  TB, Chen  C, Tartof  SY, Mayeda  ER, Lee  GS, et al.  Wildfire smoke exposure and incident dementia. JAMA Neurol  2025;82:40–8. 10.1001/jamaneurol.2024.4058 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Rappold  AG, Stone  SL, Cascio  WE, Neas  LM, Kilaru  VJ, Carraway  MS, et al.  Peat bog wildfire smoke exposure in rural North Carolina is associated with cardiopulmonary emergency department visits assessed through syndromic surveillance. Environ Health Perspect  2011;119:1415–20. 10.1289/ehp.1003206 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Haikerwal  A, Akram  M, Del Monaco  A, Smith  K, Sim  MR, Meyer  M, et al.  Impact of fine particulate matter (PM2.5) exposure during wildfires on cardiovascular health outcomes. J Am Heart Assoc  2015;4:e001653. 10.1161/JAHA.114.001653 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 23. Zhang  D, Wang  W, Xi  Y, Bi  J, Hang  Y, Zhu  Q, et al.  Wildland fires worsened population exposure to PM(2.5) pollution in the contiguous United States. Environ Sci Technol  2023;57:19990–8. 10.1021/acs.est.3c05143 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Lee  TS, Falter  K, Meyer  P, Mott  J, Gwynn  C. Risk factors associated with clinic visits during the 1999 forest fires near the Hoopa Valley Indian Reservation, California, USA. Int J Environ Health Res  2009;19:315–27. 10.1080/09603120802712750 [ DOI ] [ PubMed ] [ Google Scholar ] 25. Stowell  JD, Geng  G, Saikawa  E, Chang  HH, Fu  J, Yang  C-E, et al.  Associations of wildfire smoke PM2.5 exposure with cardiorespiratory events in Colorado 2011–2014. Environ Int  2019;133:105151. 10.1016/j.envint.2019.105151 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 26. Jin  T, Di  Q, Requia  WJ, Danesh Yazdi  M, Castro  E, Ma  T, et al.  Associations between long-term air pollution exposure and the incidence of cardiovascular diseases among American older adults. Environ Int  2022;170:107594. 10.1016/j.envint.2022.107594 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 27. Ma  T, Yazdi  MD, Schwartz  J, Requia  WJ, Di  Q, Wei  Y, et al.  Long-term air pollution exposure and incident stroke in American older adults: a national cohort study. Glob Epidemiol  2022;4:100073. 10.1016/j.gloepi.2022.100073 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 28. Shi  L, Zhu  Q, Wang  Y, Hao  H, Zhang  H, Schwartz  J, et al.  Incident dementia and long-term exposure to constituents of fine particle air pollution: a national cohort study in the United States. Proc Natl Acad Sci U S A  2023;120:e2211282119. 10.1073/pnas.2211282119 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 29. Qiu  X, Shi  L, Kubzansky  LD, Wei  Y, Castro  E, Li  H, et al.  Association of long-term exposure to air pollution with late-life depression in older adults in the US. JAMA Netw Open  2023;6:e2253668. 10.1001/jamanetworkopen.2022.53668 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 30. Yadav  K, Lewis  RJ. Immortal time bias in observational studies. JAMA  2021;325:686–7. 10.1001/jama.2020.9151 [ DOI ] [ PubMed ] [ Google Scholar ] 31. Brey  SJ, Ruminski  M, Atwood  SA, Fischer  EV. Connecting smoke plumes to sources using Hazard Mapping System (HMS) smoke and fire location data over North America. Atmos Chem Phys  2018;18:1745–61. 10.5194/acp-18-1745-2018 [ DOI ] [ Google Scholar ] 32. Hu  X, Yu  C, Tian  D, Ruminski  M, Robertson  K, Waller  LA, et al.  Comparison of the Hazard Mapping System (HMS) fire product to ground-based fire records in Georgia, USA. J Geophys Res  2016;121:2015JD024448. 10.1002/2015JD024448 [ DOI ] [ Google Scholar ] 33. Yang  C-E, Fu  JS, Liu  Y, Dong  X, Liu  Y. Projections of future wildfires impacts on air pollutants and air toxics in a changing climate over the western United States. Environ Pollut  2022;304:119213. 10.1016/j.envpol.2022.119213 [ DOI ] [ PubMed ] [ Google Scholar ] 34. Mansournia  MA, Altman  DG. Population attributable fraction. BMJ  2018;360:k757. 10.1136/bmj.k757 [ DOI ] [ PubMed ] [ Google Scholar ] 35. Tibuakuu  M, Michos  ED, Navas-Acien  A, Jones  MR. Air pollution and cardiovascular disease: a focus on vulnerable populations worldwide. Curr Epidemiol Rep  2018;5:370–8. 10.1007/s40471-018-0166-8 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 36. Mansournia  MA, Nazemipour  M, Etminan  M. A practical guide to handling competing events in etiologic time-to-event studies. Glob Epidemiol  2022;4:100080. 10.1016/j.gloepi.2022.100080 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 37. Huang  S-P, Su  C-C, Lin  C-Y, Nethery  R, Josey  K, Bates  B, et al. Exposure-response to high PM (2.5) levels for cardiovascular events in high-risk older adults in Taiwan. medRxiv. 10.1101/2024.05.08.24306967, 9 May 2024, preprint: not peer reviewed. [ DOI ] [ PMC free article ] [ PubMed ] 38. Reid  CE, Brauer  M, Johnston  FH, Jerrett  M, Balmes  JR, Elliott  CT. Critical review of health impacts of wildfire smoke exposure. Environ Health Perspect  2016;124:1334–43. 10.1289/ehp.1409277 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 39. Chen  H, Samet  JM, Bromberg  PA, Tong  H. Cardiovascular health impacts of wildfire smoke exposure. Part Fibre Toxicol  2021;18:2. 10.1186/s12989-020-00394-8 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 40. Wegesser  TC, Pinkerton  KE, Last  JA. California wildfires of 2008: coarse and fine particulate matter toxicity. Environ Health Perspect  2009;117:893–7. 10.1289/ehp.0800166 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 41. Ma  Y. Role of neutrophils in cardiac injury and repair following myocardial infarction. Cells  2021;10:1676. 10.3390/cells10071676 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 42. Verma  V, Polidori  A, Schauer  JJ, Shafer  MM, Cassee  FR, Sioutas  C. Physicochemical and toxicological profiles of particulate matter in Los Angeles during the October 2007 southern California wildfires. Environ Sci Technol  2009;43:954–60. 10.1021/es8021667 [ DOI ] [ PubMed ] [ Google Scholar ] 43. Steven  S, Frenis  K, Oelze  M, Kalinovic  S, Kuntic  M, Bayo Jimenez  MT, et al.  Vascular inflammation and oxidative stress: major triggers for cardiovascular disease. Oxid Med Cell Longev  2019;2019:7092151. 10.1155/2019/7092151 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 44. Johnston  F, Hanigan  I, Henderson  S, Morgan  G, Bowman  D. Extreme air pollution events from bushfires and dust storms and their association with mortality in Sydney, Australia 1994–2007. Environ Res  2011;111:811–6. 10.1016/j.envres.2011.05.007 [ DOI ] [ PubMed ] [ Google Scholar ] 45. Babaie  J, Pashaei Asl  Y, Naghipour  B, Faridaalaee  G. Cardiovascular diseases in natural disasters; a systematic review. Arch Acad Emerg Med  2021;9:e36. 10.22037/aaem.v9i1.1208 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 46. Peters  MN, Moscona  JC, Katz  MJ, Deandrade  KB, Quevedo  HC, Tiwari  S, et al.  Natural disasters and myocardial infarction: the six years after Hurricane Katrina. Mayo Clin Proc  2014;89:472–7. 10.1016/j.mayocp.2013.12.013 [ DOI ] [ PubMed ] [ Google Scholar ] 47. Lipsett  MJ, Ostro  BD, Reynolds  P, Goldberg  D, Hertz  A, Jerrett  M, et al.  Long-term exposure to air pollution and cardiorespiratory disease in the California teachers study cohort. Am J Respir Crit Care Med  2011;184:828–35. 10.1164/rccm.201012-2082OC [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 48. Hart  JE, Puett  RC, Rexrode  KM, Albert  CM, Laden  F. Effect modification of long-term air pollution exposures and the risk of incident cardiovascular disease in US Women. J Am Heart Assoc  2015;4:e002301. 10.1161/JAHA.115.002301 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 49. Puett  RC, Hart  JE, Suh  H, Mittleman  M, Laden  F. Particulate matter exposures, mortality, and cardiovascular disease in the health professionals follow-up study. Environ Health Perspect  2011;119:1130–5. 10.1289/ehp.1002921 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 50. Hartiala  J, Breton  CV, Tang  WH, Lurmann  F, Hazen  SL, Gilliland  FD, et al.  Ambient air pollution is associated with the severity of coronary atherosclerosis and incident myocardial infarction in patients undergoing elective cardiac evaluation. J Am Heart Assoc  2016;5:e003947. 10.1161/JAHA.116.003947 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 51. Miller  KA, Siscovick  DS, Sheppard  L, Shepherd  K, Sullivan  JH, Anderson  GL, et al.  Long-term exposure to air pollution and incidence of cardiovascular events in women. N Engl J Med  2007;356:447–58. 10.1056/NEJMoa054409 [ DOI ] [ PubMed ] [ Google Scholar ] 52. Navarro  KM, Kleinman  MT, Mackay  CE, Reinhardt  TE, Balmes  JR, Broyles  GA, et al.  Wildland firefighter smoke exposure and risk of lung cancer and cardiovascular disease mortality. Environ Res  2019;173:462–8. 10.1016/j.envres.2019.03.060 [ DOI ] [ PubMed ] [ Google Scholar ] 53. Tuttolomondo  A, Daidone  M, Pinto  A. Endothelial dysfunction and inflammation in ischemic stroke pathogenesis. Curr Pharm Des  2020;26:4209–19. 10.2174/1381612826666200417154126 [ DOI ] [ PubMed ] [ Google Scholar ] 54. Ma  Y, Zang  E, Liu  Y, Wei  J, Lu  Y, Krumholz  HM, et al.  Long-term exposure to wildland fire smoke PM(2.5) and mortality in the contiguous United States. Proc Natl Acad Sci U S A  2024;121:e2403960121. 10.1073/pnas.2403960121 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 55. Requia  WJ, Amini  H, Mukherjee  R, Gold  DR, Schwartz  JD. Health impacts of wildfire-related air pollution in Brazil: a nationwide study of more than 2 million hospital admissions between 2008 and 2018. Nat Commun  2021;12:6555. 10.1038/s41467-021-26822-7 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 56. Pun  VC, Kazemiparkouhi  F, Manjourides  J, Suh  HH. Long-term PM2.5 exposure and respiratory, cancer, and cardiovascular mortality in older US adults. Am J Epidemiol  2017;186:961–9. 10.1093/aje/kwx166 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 57. Oh  J, Choi  S, Han  C, Lee  D-W, Ha  E, Kim  S, et al.  Association of long-term exposure to PM2.5 and survival following ischemic heart disease. Environ Res  2023;216:114440. 10.1016/j.envres.2022.114440 [ DOI ] [ PubMed ] [ Google Scholar ] 58. Hvidtfeldt  UA, Sørensen  M, Geels  C, Ketzel  M, Khan  J, Tjønneland  A, et al.  Long-term residential exposure to PM2.5, PM10, black carbon, NO2, and ozone and mortality in a Danish cohort. Environ Int  2019;123:265–72. 10.1016/j.envint.2018.12.010 [ DOI ] [ PubMed ] [ Google Scholar ] 59. Lin  H, Guo  Y, Di  Q, Zheng  Y, Kowal  P, Xiao  J, et al.  Ambient PM(2.5) and stroke: effect modifiers and population attributable risk in six low- and middle-income countries. Stroke  2017;48:1191–7. 10.1161/STROKEAHA.116.015739 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 60. Limaye  VS, Schopp  W, Amann  M. Applying integrated exposure-response functions to PM(2.5) pollution in India. Int J Environ Res Public Health  2018;16:60. 10.3390/ijerph16010060 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 61. Weisskopf  MG, Webster  TF. Trade-offs of personal versus more proxy exposure measures in environmental epidemiology. Epidemiology  2017;28:635–43. 10.1097/EDE.0000000000000686 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials ehaf875_Supplementary_Data ehaf875_supplementary_data.docx (965.3KB, docx) Data Availability Statement The Medicare dataset used in this study is maintained in a secure computing infrastructure compliant with HIPAA at the Rollins School of Public Health, Emory University. Due to data privacy regulations, individual-level Medicare data cannot be publicly shared but may be available upon reasonable request to the Center for Medicare & Medicaid Services (CMS). Articles from European Heart Journal are provided here courtesy of Oxford University Press ACTIONS View on publisher site PDF (655.0 KB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• Cardiopulmonary hospitalization risks from wildfire-specific and non-wildfire PM&lt;sub&gt;2.5&lt;/sub&gt; in 20 US states#1052844
• Evaluating differential air quality impacts of prescribed fire and wildfire in the United States#1070419
• Comparing effects of gaseous vs. particulate components of traffic-related air pollution on Alzheimer’s disease biomarkers in a genetically susceptible rat model#996365
• Wildfire pollution exposure during childhood adversely affects cognitive and neural development#369742
• Ambient PM2.5 Concentration and the Cardiovascular Response to Exercise Training: A Systematic Review and Meta-Analysis Across Global Pollution Gradients#1029370