A novel ensemble-based statistical approach to estimate daily wildfire-specific PM2.5 in California (2006-2020)

Global landscape fire-sourced ambient benzene and health risks in different future scenarios

Landscape fire (wildfire) has a detrimental impact on air quality and human health. However, the contribution of wildfire emissions to ambient benzene concentrations under future climate scenarios remains unknown. To effectively mitigate the adverse effects of wildfire, it is crucial to accurately capture the spatiotemporal trends of wildfire-related ambient benzene and its health impacts. In this study, the GEOS-Chem model was applied to simulate ambient fire-sourced benzene concentrations for 2015-2019 and 2045-2049 across four scenarios. Overall, global observed ambient benzene concentrations and simulated concentrations showed a correlation coefficient of 0.65, indicating robust model performance (p < 0.05). The wildfire-related ambient benzene concentration during 2015-2019 was 0.015 ± 0.011 (mean ± standard deviation) ppb, and then decreased through 2045-2049. Specifically, under the SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5 scenarios, the ambient benzene concentrations were projected at 0.005 ± 0.003, 0.008 ± 0.006, 0.009 ± 0.007, and 0.009 ± 0.007 ppb, respectively. These variations are closely linked to differences in the burned areas, fire activity, and fuel consumption. Wildfire-related ambient benzene concentrations showed significant seasonal variations with high concentrations in summer and autumn, and lower concentrations in winter. This seasonal pattern is likely associated with temperature fluctuations and varying fire activity. Spatially, tropical regions and the northern parts of the US, Canada, and Russia exhibited high wildfire-related ambient benzene concentrations. Although most regions are unaffected by marked carcinogenic or non-carcinogenic risks from ambient benzene, certain boreal and tropical countries remained at higher risks. To address this, better planning and design of natural and urban landscapes is essential to reduce wildfire occurrences, especially in wildfire hotspots. Additionally, controlling carbon emissions can significantly mitigate wildfire-related impacts on air quality.

AI-driven approaches for air pollution modelling: A comprehensive systematic review

In recent years, air quality levels have become a global issue with the rise of harmful pollutants and their effects on climate change. Urban areas are especially affected by air pollution, resulting in a deterioration of the environment and a surge in health complications. Research has been conducted on different studies that accurately predict future pollution concentration levels utilising different methods. This paper introduces the current physical models for air quality prediction and conducts an extensive systematic literature review on Machine Learning and Deep Learning techniques for predicting pollutants. This work compares different methodologies and techniques by grouping studies that utilise similar approaches together and comparing them. Furthermore, a distinction is made between temporal and spatiotemporal models to understand and highlight how both approaches impact future air pollutant concentration level predictions. The review differs from similar works as it focuses not only on comparing models and approaches but by analysing how the usage of external features, such as meteorological data, traffic information, and land usage, affect pollutant levels and the model's accuracy on air quality forecasting. Performances and limitations are explored for both Machine and Deep Learning approaches, and the work offers a discussion on their comparison and possible future developments in this research space. This review highlights how Deep Learning models tend to be more suitable for forecasting problems due to their feature and spatio-temporal correlation representation abilities, as well as providing different directions for further work, from models utilisation to feature inclusion.

Impact of the 2023 wildfire smoke episodes in Ontario, Canada, on asthma and other health outcomes: an interrupted time-series analysis

BACKGROUND

During the 2023 wildfire season, Ontario, Canada, had unprecedented wildfire smoke, but the health impact on the population is unknown. We aimed to quantify the acute impact of the wildfire smoke on respiratory and cardiovascular outcomes across Ontario.

METHODS

We conducted a quasi-experimental study by leveraging the timing of 2 consecutive wildfire smoke episodes in June 2023. Heavy wildfire smoke blanketed much of Ontario on 2 occasions, in early June and again in late June, causing severely degraded daily air quality. Following the epidemiologic triangulation framework, we collected health data on emergency department visits for 4 outcomes (asthma-related causes, other respiratory causes, ischemic heart disease, and non-cardiorespiratory causes) from Ontario's real-time syndromic surveillance system and the National Ambulatory Care Reporting System. We also employed different epidemiologic methodologies, including interrupted time-series and case-crossover analyses.

RESULTS

After the initial heavy wildfire smoke in early June 2023, daily asthma-related visits increased substantially across Ontario, peaking at a 23.6% increase (95% confidence interval 13.2%-34.9%) at a 1-day lag and lasting up to a lag of 5 days after the start of the smoke episode. The later episode of heavy smoke, despite causing higher exposures, had a reduced effect on asthma-related visits. We did not detect any effect on other outcomes in either episode. These findings were consistent across different methodologies and data sources. Post hoc analysis revealed that asthma-related visits were briefly elevated after the wildfire smoke among children (40% higher), but we observed a more sustained effect among adults (48% higher, lasting 1 week).

INTERPRETATION

The 2023 wildfires substantially increased asthma-related emergency department visits in Ontario, with age and timing of exposure being important factors influencing the impact. As wildfires emerge as one of the fastest-growing environmental risk factors globally, future research should identify and evaluate measures to effectively mitigate the acute health impacts of wildfire smoke.

Impact of Wildfire Smoke PM2.5 on Indoor Air Quality of Public Buildings on a University Campus

With increasing wildfire events impacting many regions worldwide, understanding and mitigating the effects of wildfire smoke on indoor air quality (IAQ) in public buildings are essential for protecting occupant health. This study investigated the impact of wildfire smoke on the IAQ across 24 campus buildings in Alberta, Canada, representing public spaces with varied ventilation systems. Using a network of low-cost sensors to monitor indoor PM2.5, the study identified significant spikes during wildfire smoke events, with 71% of buildings exceeding the Canadian Ambient Air Quality Standards daily limit of 27 μg/m3. The buildings had mechanical ventilation systems with filters with different Minimum Efficiency Reporting Value (MERV) ratings. MERV13 filters were found to be more efficient at capturing PM2.5 particles, resulting in lower indoor/outdoor PM2.5 ratios (0.12 ± 0.07) compared to MERV8 filters (0.28 ± 0.14). Buildings with air change rates (ACH) ranging from 5 to 15 per hour exhibited different infiltration patterns, with higher ACH generally leading to elevated indoor PM2.5 concentrations during wildfire events. This highlights the need to balance ventilation and pollutant infiltration by optimizing ACH rates and filtration efficiency to reduce indoor PM2.5. The trajectory-fire interception method, combined with satellite data, enhanced the identification of wildfire-influenced periods, contributing to a better understanding of smoke infiltration dynamics. These findings underscore that even advanced filtration and ventilation systems alone may not ensure a healthy IAQ during extreme pollution. Real-time pollutant measurements are crucial for effective IAQ management. The findings offer valuable insights for building administrators and policymakers, helping them develop strategies to mitigate the effects of wildfire smoke and to support healthier indoor environments during wildfire seasons.

Wildfire Air Pollution and Rates of Cardiovascular Events and Mortality in Northern California in 2018

BACKGROUND

We examined the association between acute cardiovascular disease (CVD) events and wildfire air pollution in California in 2018.

METHODS

The study included adult (≥18 years) members of Kaiser Permanente Northern California, an integrated health care system. Outcomes included CVD events (hospitalizations for acute myocardial infarction, heart failure, or stroke, and CVD death) and death from any cause. Fine particulate air pollution (particulate matter <2.5 microns in diameter; PM2.5) exposure was assessed in categories (Good <12 μg/m3, Moderate 12-34 μg/m3, High ≥35 μg/m3) and continuously. Poisson time series regression was used to model daily event rates during July 1 to December 31, 2018, using a spline to adjust for long-term time trends. We calculated rate ratios (RR) to estimate the association between wildfire air pollution and daily rate of CVD events and deaths.

RESULTS

Our study included 3.2 million adults with a total follow-up of 587.9 million person-days. High PM2.5 concentrations during the Mendocino Complex wildfire in July to August was associated with an increased rate of CVD events (RR, 1.231 [95% CI, 1.039-1.458]) and death (RR, 1.358 [95% CI, 1.128-1.635]) compared with Good PM2.5 concentrations. In contrast, there was no evidence of increased risk during the Camp wildfire in November (RR for CVD events, 0.966 [95% CI, 0.894-1.044]; RR for all-cause mortality, 0.985 [95% CI, 0.904-1.074] High versus Good PM2.5 concentrations).

CONCLUSIONS

There was some evidence of increased rates of CVD events and death during wildfires, but results were inconsistent. With ongoing climate change, large wildfires are a pressing public health concern and future work is needed to understand differences in health outcomes by wildfire.

Wildfire Smoke Exposure and Incident Dementia

Importance

Long-term exposure to total fine particulate matter (PM2.5) is a recognized dementia risk factor, but less is known about wildfire-generated PM2.5, an increasingly common PM2.5 source.

Objective

To assess the association between long-term wildfire and nonwildfire PM2.5 exposure and risk of incident dementia.

Design, Setting, and Participants

This open cohort study was conducted using January 2008 to December 2019 electronic health record (EHR) data among members of Kaiser Permanente Southern California (KPSC), which serves 4.7 million people across 10 California counties. KPSC members aged 60 years or older were eligible for inclusion. Members were excluded if they did not meet eligibility criteria, if they had a dementia diagnosis before cohort entry, or if EHR data lacked address information. Data analysis was conducted from May 2023 to May 2024.

Exposures

Three-year rolling mean wildfire and nonwildfire PM2.5 in member census tracts from January 2006 to December 2019, updated quarterly and estimated via monitoring and remote-sensing data and statistical techniques.

Main Outcome and Measures

The primary outcome was incident dementia, identified using diagnostic codes in the EHR. Odds of dementia diagnoses associated with 3-year mean wildfire and nonwildfire PM2.5 exposure were estimated using a discrete-time approach with pooled logistic regression. Models adjusted for age, sex, race and ethnicity (considered as a social construct rather than as a biological determinant), marital status, smoking status, calendar year, and census tract-level poverty and population density. Stratified models assessed effect measure modification by age, sex, race and ethnicity, and census tract-level poverty.

Results

Among 1.64 million KPSC members aged 60 years or older during the study period, 1 223 107 members were eligible for inclusion in this study. The study population consisted of 644 766 female members (53.0%). In total, 319 521 members identified as Hispanic (26.0%), 601 334 members identified as non-Hispanic White (49.0%), and 80 993 members received a dementia diagnosis during follow-up (6.6%). In adjusted models, a 1-μg/m3 increase in the 3-year mean of wildfire PM2.5 exposure was associated with an 18% increase in the odds of dementia diagnosis (odds ratio [OR], 1.18; 95% CI, 1.03-1.34). In comparison, a 1-μg/m3 increase in nonwildfire PM2.5 exposure was associated with a 1% increase (OR, 1.01; 95% CI, 1.01-1.02). For wildfire PM2.5 exposure, associations were stronger among members less than 75 years old upon cohort entry, members from racially minoritized subgroups, and those living in high-poverty vs low-poverty census tracts.

Conclusions and Relevance

In this cohort study, after adjusting for measured confounders, long-term exposure to wildfire and nonwildfire PM2.5 over a 3-year period was associated with dementia diagnoses. As the climate changes, interventions focused on reducing wildfire PM2.5 exposure may reduce dementia diagnoses and related inequities.

Evaluating Chemical Transport and Machine Learning Models for Wildfire Smoke PM2.5: Implications for Assessment of Health Impacts

Growing wildfire smoke represents a substantial threat to air quality and human health. However, the impact of wildfire smoke on human health remains imprecisely understood due to uncertainties in both the measurement of exposure of population to wildfire smoke and dose-response functions linking exposure to health. Here, we compare daily wildfire smoke-related surface fine particulate matter (PM2.5) concentrations estimated using three approaches, including two chemical transport models (CTMs): GEOS-Chem and the Community Multiscale Air Quality (CMAQ) and one machine learning (ML) model over the contiguous US in 2020, a historically active fire year. In the western US, compared against surface PM2.5 measurements from the US Environmental Protection Agency (EPA) and PurpleAir sensors, we find that CTMs overestimate PM2.5 concentrations during extreme smoke episodes by up to 3-5 fold, while ML estimates are largely consistent with surface measurements. However, in the eastern US, where smoke levels were much lower in 2020, CTMs show modestly better agreement with surface measurements. We develop a calibration framework that integrates CTM- and ML-based approaches to yield estimates of smoke PM2.5 concentrations that outperform individual approach. When combining the estimated smoke PM2.5 concentrations with county-level mortality rates, we find consistent effects of low-level smoke on mortality but large discrepancies in effects of high-level smoke exposure across different methods. Our research highlights the differences across estimation methods for understanding the health impacts of wildfire smoke and demonstrates the importance of bench-marking estimates with available surface measurements.

The temporal trend and disparity in short-term health impacts of fine particulate matter in California (2006-2019)

Most epidemiological studies assume that the relationship between short-term air pollution exposure and health outcomes is constant over time, which ignores potential changes in population composition and particulate matter emission sources. Limited studies have assessed changes in the relationship between fine particulate matter (PM2.5) and adverse health outcomes over time, with mixed results. Additionally, there is a need to identify which subgroups are disproportionately impacted over time by PM2.5-related health consequences. Therefore, we aimed to examine whether temporal trends exist in the relationships between daily PM2.5 exposure and circulatory and respiratory acute care utilization in California from 2006 to 2019. We further assessed whether certain subpopulations are more susceptible to PM2.5 exposure by demographic characteristics and extreme wildfire frequency. Daily PM2.5 concentrations estimated from a stacked ensemble model and daily cause-specific acute care utilization and demographic data from the California Department of Health Care Access and Information. We analyzed this relationship using modified two-stage Bayesian hierarchical models, where we first did not consider temporal trends, then stratified by two periods, and finally flexibly considered non-linear changes over time. Increases in circulatory (0.56 %; 95 % credible interval (CI): 0.17 %, 0.96 %) and respiratory acute care utilization risk (2.61 %; 95%CI: 2.29 %, 2.94 %) were found with every 10 μg/m3 increase in PM2.5 on the same day and previous two days. These risks were found to increase over time, where 0.13 % (95%CI: 0.02 %, 0.22 %) and 1.40 % (95%CI: 1.24 %, 1.54 %) increases were identified for circulatory and respiratory acute care utilizations, respectively, from the first (2006-2012) to second period (2013-2019). Differences by age, sex, race/ethnicity, and extreme wildfire frequency were noted. These findings confirm that air pollution guidelines should consider the dynamic nature of epidemiological dose-response and can provide insight for targeted air pollution control and adaptation policies designed to reduce PM2.5 exposure, particularly for the most susceptible subpopulations.

Multiresolution Analysis of HRRR Meteorological Parameters and GOES-R AOD for Hourly PM2.5 Prediction

High-resolution, comprehensive exposure data are crucial for accurately estimating the human health impact of PM2.5. In recent years, satellite remote sensing data have been increasingly utilized in PM2.5 models to overcome the limited spatial coverage of ground monitoring stations. However, data gaps in satellite-retrieved parameters such as aerosol optical depth (AOD), the sparsity of regulatory air quality monitors for model training, and nonlinear relationships between PM2.5 and meteorological conditions can affect model performance and cause data gaps in most existing PM2.5 models. In this study, spatial gaps in AOD obtained from Geostationary Operational Environmental Satellite-16 are filled using Goddard Earth Observing System Composition Forecasting AOD estimations. Furthermore, to improve model performance, meteorological predictors such as temperature from the High-Resolution Rapid Refresh model are preprocessed using Daubechies wavelet to extract low and high frequency components. The spatially gap-filled AOD, along with meteorological data, are ingested into a machine learning model to predict hourly PM2.5 at a 1 km spatial resolution in California. The model evaluation metrics (OOB (out-of-bag) R2 = 0.86 and RMSE (root-mean-square error) = 9.27 μg/m3 and 10-fold spatial cross-validation R2 = 0.82 and RMSE = 9.82 μg/m3) demonstrate the model's reliability in predicting ambient PM2.5, especially for states like California that experience frequent episodes of wildfires.

Fine Particulate Matter, Its Constituents, and Spontaneous Preterm Birth

Importance

The associations of exposure to fine particulate matter (PM2.5) and its constituents with spontaneous preterm birth (sPTB) remain understudied. Identifying subpopulations at increased risk characterized by socioeconomic status and other environmental factors is critical for targeted interventions.

Objective

To examine associations of PM2.5 and its constituents with sPTB.

Design, Setting, and Participants

This population-based retrospective cohort study was conducted from 2008 to 2018 within a large integrated health care system, Kaiser Permanente Southern California. Singleton live births with recorded residential information of pregnant individuals during pregnancy were included. Data were analyzed from December 2023 to March 2024.

Exposures

Daily total PM2.5 concentrations and monthly data on 5 PM2.5 constituents (sulfate, nitrate, ammonium, organic matter, and black carbon) in California were assessed, and mean exposures to these pollutants during pregnancy and by trimester were calculated. Exposures to total green space, trees, low-lying vegetation, and grass were estimated using street view images. Wildfire-related exposure was measured by the mean concentration of wildfire-specific PM2.5 during pregnancy. Additionally, the mean exposure to daily maximum temperature during pregnancy was calculated.

Main Outcomes and Measures

The primary outcome was sPTB identified through a natural language processing algorithm. Discrete-time survival models were used to estimate associations of total PM2.5 concentration and its 5 constituents with sPTB. Interaction terms were used to examine the effect modification by race and ethnicity, educational attainment, household income, and exposures to green space, wildfire smoke, and temperature.

Results

Among 409 037 births (mean [SD] age of mothers at delivery, 30.3 [5.8] years), there were positive associations of PM2.5, black carbon, nitrate, and sulfate with sPTB. Adjusted odds ratios (aORs) per IQR increase were 1.15 (95% CI, 1.12-1.18; P < .001) for PM2.5 (IQR, 2.76 μg/m3), 1.15 (95% CI, 1.11-1.20; P < .001) for black carbon (IQR, 1.05 μg/m3), 1.09 (95% CI, 1.06-1.13; P < .001) for nitrate (IQR, 0.93 μg/m3), and 1.06 (95% CI, 1.03-1.09; P < .001) for sulfate (IQR, 0.40 μg/m3) over the entire pregnancy. The second trimester was the most susceptible window; for example, aORs for total PM2.5 concentration were 1.07 (95% CI, 1.05-1.09; P < .001) in the first, 1.10 (95% CI, 1.08-1.12; P < .001) in the second, and 1.09 (95% CI, 1.07-1.11; P < .001) in the third trimester. Significantly higher aORs were observed among individuals with lower educational attainment (eg, less than college: aOR, 1.16; 95% CI, 1.12-1.21 vs college [≥4 years]: aOR, 1.10; 95% CI, 1.06-1.14; P = .03) or income (<50th percentile: aOR, 1.17; 95% CI, 1.14-1.21 vs ≥50th percentile: aOR, 1.12; 95% CI, 1.09-1.16; P = .02) or who were exposed to limited green space (<50th percentile: aOR, 1.19; 95% CI, 1.15-1.23 vs ≥50th percentile: aOR, 1.12; 95% CI, 1.09-1.15; P = .003), more wildfire smoke (≥50th percentile: aOR, 1.19; 95% CI, 1.16-1.23 vs <50th percentile: aOR, 1.13; 95% CI, 1.09-1.16; P = .009), or extreme heat (aOR, 1.51; 95% CI, 1.42-1.59 vs mild temperature: aOR, 1.11; 95% CI, 1.09-1.14; P < .001).

Conclusions and Relevance

In this study, exposures to PM2.5 and specific PM2.5 constituents during pregnancy were associated with increased odds of sPTB. Socioeconomic status and other environmental exposures modified this association.

Effects of fine particulate matter from wildfire and non-wildfire sources on emergency-department visits in people who were housed and unhoused in San Diego County (CA, USA) during 2012-20: a time-stratified case-crossover study

BACKGROUND

Being unhoused can increase vulnerability to adverse health effects due to air pollution. We aimed to quantify changes in emergency-department visits during and after exposure to wildfire-specific and non-wildfire particulate matter 2·5 μm or less in diameter (PM2·5) in San Diego County (CA, USA) in people who were both unhoused and housed.

METHODS

For this time-stratified case-crossover study, we used data on exposure to wildfire-specific PM2·5 in California and individual-level data for people admitted to the emergency departments of two hospitals (UC San Diego Health emergency departments at La Jolla and Hillcrest, San Diego) in San Diego County between July 1, 2012, and Dec 31, 2020. People with a postcode outside of San Diego County were excluded. Demographic information was age group, race or ethnicity, and transport to the emergency department. Wildfire-specific PM2·5 concentration at the postcode level was previously estimated using an ensemble model that combined multiple machine-learning algorithms and explanatory variables obtained via data on 24-h mean PM2·5 concentrations from the US Environmental Protection Agency Air Quality System. Conditional logistic regression models were applied, adjusting for specific humidity, wind velocity, and maximum temperature extracted from the US Gridded Surface Meteorological Dataset. Housing status was established by registration staff or triage nurses on arrival at the emergency department. For people who were unhoused, exposure was defined based on the weighted mean PM2·5 concentration at the city level proportional to the number of people who were unhoused in each specific city across urban centres in San Diego County. For people who were housed, we used residence postcode to measure exposure. We assessed the association between PM2·5 from wildfire and non-wildfire sources and emergency-department visits in people who were housed and unhoused.

FINDINGS

There were 587 562 emergency-department visits at the two hospitals, 76 407 (13·0%) of which were by people who were unhoused. People who were housed had a higher exposure to overall PM2·5 (24-h mean over the study period of 9·904 mg/m3, SD 3·445) and non-wildfire PM2·5 (9·663, 2·977) than people who were unhoused (9·863, 3·221; 9·557, 2·599). However, people who were unhoused had a higher exposure to wildfire-specific PM2·5 (0·305, 1·797) than people who were housed (0·240, 1·690). Overall PM2·5 exposure was associated with increased odds of emergency-department visits for both people who were housed (odds ratio 1·003, 95% CI 1·001-1·004 per 1 μg/m3 PM2·5 for 0-3 days after exposure) and people who were unhoused (1·004, 1·000-1·008 for 0-3 days after exposure). We found that non-wildfire PM2·5 was associated with emergency-department visits among people who were housed (1·003, 1·002-1·005 for 0-3 days after exposure) and wildfire-specific PM2·5 was associated with emergency-department visits in people who were unhoused (1·006, 1·001-1·011 for 0-3 days after exposure).

INTERPRETATION

People who were unhoused in San Diego County were more likely to visit emergency departments after exposure to increased wildfire-specific PM2·5. As the intensity and frequency of wildfires increase, understanding risk factors for vulnerable populations, such as people who are unhoused, is crucial to develop effective adaptation strategies.

FUNDING

US National Institutes of Health, National Institute on Aging.

Impact of the 2022 New Mexico, US wildfires on air quality and health

The 2022 wildfires in New Mexico, United States, were unparalleled compared to past wildfires in the state in both their scale and intensity, resulting in poor air quality and a catastrophic loss of habitat and livelihood. Among all wildfires in New Mexico in 2022, six wildfires were selected for our study based on the size of the burn area and their proximity to populated areas. These fires accounted for approximately 90 % of the total burn area in New Mexico in 2022. We used a regional chemical transport model and data-fusion technique to quantify the contribution of these six wildfires (April 6 to August 22) on particulate matter (PM2.5: diameter ≤ 2.5 μm) and ozone (O3) concentrations, as well as the associated health impacts from short-term exposure. We estimated that these six wildfires emitted 152 thousand tons of PM2.5 and 287 thousand tons of volatile organic compounds to the atmosphere. We estimated that the average daily wildfire smoke PM2.5 across New Mexico was 0.3 μg/m3, though 1 h maximum exceeded 120 μg/m3 near Santa Fe. Average wildfire smoke maximum daily average 8-h O3 (MDA8-O3) contribution was 0.2 ppb during the study period over New Mexico. However, over the state 1 h maximum smoke O3 exceeded 60 ppb in some locations near Santa Fe. Estimated all-cause excess mortality attributable to short term exposure to wildfire PM2.5 and MDA8-O3 from these six wildfires were 18 (95 % Confidence Interval (CI), 15-21) and 4 (95 % CI: 3-6) deaths. Additionally, we estimate that wildfire PM2.5 was responsible for 171 (95 %: 124-217) excess cases of asthma emergency department visits. Our findings underscore the impact of wildfires on air quality and human health risks, which are anticipated to intensify with global warming, even as local anthropogenic emissions decline.

Innovative approaches for accurate ozone prediction and health risk analysis in South Korea: The combined effectiveness of deep learning and AirQ

Short-term exposure to ground-level ozone (O3) poses significant health risks, particularly respiratory and cardiovascular diseases, and mortality. This study addresses the pressing need for accurate O3 forecasting to mitigate these risks, focusing on South Korea. We introduce Deep Bias Correction (Deep-BC), a novel framework leveraging Convolutional Neural Networks (CNNs), to refine hourly O3 forecasts from the Community Multiscale Air Quality (CMAQ) model. Our approach involves training Deep-BC using data from 2016 to 2019, including CMAQ's 72-hour O3 forecasts, 31 meteorological variables from the Weather Research and Forecasting (WRF) model, and previous days' station measurements of 6 air pollutants. Deep-BC significantly outperforms CMAQ in 2021, reducing biases in O3 forecasts. Furthermore, we utilize Deep-BC's daily maximum 8-hour average O3 (MDA8 O3) forecasts as input for the AirQ+ model to assess O3's potential impact on mortality across seven major provinces of South Korea: Seoul, Busan, Daegu, Incheon, Daejeon, Ulsan, and Sejong. Short-term O3 exposure is associated with 0.40 % to 0.48 % of natural cause and respiratory deaths and 0.67 % to 0.81 % of cardiovascular deaths. Gender-specific analysis reveals higher mortality rates among men, particularly from respiratory causes. Our findings underscore the critical need for region-specific interventions to address air pollution's detrimental effects on public health in South Korea. By providing improved O3 predictions and quantifying its impact on mortality, this research offers valuable insights for formulating targeted strategies to mitigate air pollution's adverse effects. Moreover, we highlight the urgency of proactive measures in health policies, emphasizing the significance of accurate forecasting and effective interventions to safeguard public health from the deleterious effects of air pollution.

Long-term exposure to wildland fire smoke PM2.5 and mortality in the contiguous United States

Despite the substantial evidence on the health effects of short-term exposure to ambient fine particles (PM2.5), including increasing studies focusing on those from wildland fire smoke, the impacts of long-term wildland fire smoke PM2.5 exposure remain unclear. We investigated the association between long-term exposure to wildland fire smoke PM2.5 and nonaccidental mortality and mortality from a wide range of specific causes in all 3,108 counties in the contiguous United States, 2007 to 2020. Controlling for nonsmoke PM2.5, air temperature, and unmeasured spatial and temporal confounders, we found a nonlinear association between 12-mo moving average concentration of smoke PM2.5 and monthly nonaccidental mortality rate. Relative to a month with the long-term smoke PM2.5 exposure below 0.1 μg/m3, nonaccidental mortality increased by 0.16 to 0.63 and 2.11 deaths per 100,000 people per month when the 12-mo moving average of PM2.5 concentration was of 0.1 to 5 and 5+ μg/m3, respectively. Cardiovascular, ischemic heart disease, digestive, endocrine, diabetes, mental, and chronic kidney disease mortality were all found to be associated with long-term wildland fire smoke PM2.5 exposure. Smoke PM2.5 contributed to approximately 11,415 nonaccidental deaths/y (95% CI: 6,754, 16,075) in the contiguous United States. Higher smoke PM2.5-related increases in mortality rates were found for people aged 65 and above. Positive interaction effects with extreme heat were also observed. Our study identified the detrimental effects of long-term exposure to wildland fire smoke PM2.5 on a wide range of mortality outcomes, underscoring the need for public health actions and communications that span the health risks of both short- and long-term exposure.

Understanding the variability of ground-level ozone and fine particulate matter over the Tibetan plateau with data-driven approach

The Tibetan Plateau, known as the "Third Pole", is susceptible to ground-level ozone (O3) and fine particulate matter (PM2.5) pollution due to its unique high-altitude environment. This study constructed random forest regression models using multi-source data from ground measurements and meteorological satellites to predict variations in ground-level O3 and PM2.5 concentrations and their influencing factors across seven major cities in the Tibetan Plateau over two-year periods. The models successfully reproduced O3 and PM2.5 levels with satisfactory R-squared values of 0.71 and 0.73, respectively. Results reveal combustion-related carbon monoxide (CO) and nitrogen dioxide (NO2) as the most substantial influences on O3 and PM2.5 concentrations. Solar radiation, geographical factors, and meteorological variables also played crucial roles in driving pollutant variations. Conversely, transport-related and human activity factors exhibited relatively lower significance. High O3 and PM2.5 pollution occurred during pre-monsoon and post-monsoon/winter seasons, driven by solar radiation and emissions, respectively. While CO consistently contributed across cities and seasons, key influencing factors varied locally. This study unveils the key driving forces governing air pollutant variations across the Tibetan Plateau, shedding light on complex atmospheric processes in this unique high-altitude region.

Long-term exposure to wildland fire smoke PM2.5 and mortality in the contiguous United States

Despite the substantial evidence on the health effects of short-term exposure to ambient fine particles (PM2.5), including increasing studies focusing on those from wildland fire smoke, the impacts of long-term wildland fire smoke PM2.5 exposure remain unclear. We investigated the association between long-term exposure to wildland fire smoke PM2.5 and non-accidental mortality and mortality from a wide range of specific causes in all 3,108 counties in the contiguous U.S., 2007-2020. Controlling for non-smoke PM2.5, air temperature, and unmeasured spatial and temporal confounders, we found a non-linear association between 12-month moving average concentration of smoke PM2.5 and monthly non-accidental mortality rate. Relative to a month with the long-term smoke PM2.5 exposure below 0.1 μg/m3, non-accidental mortality increased by 0.16-0.63 and 2.11 deaths per 100,000 people per month when the 12-month moving average of PM2.5 concentration was of 0.1-5 and 5+ μg/m3, respectively. Cardiovascular, ischemic heart disease, digestive, endocrine, diabetes, mental, and chronic kidney disease mortality were all found to be associated with long-term wildland fire smoke PM2.5 exposure. Smoke PM2.5 contributed to approximately 11,415 non-accidental deaths/year (95% CI: 6,754, 16,075) in the contiguous U.S. Higher smoke PM2.5-related increases in mortality rates were found for people aged 65 above. Positive interaction effects with extreme heat (monthly number of days with daily mean air temperature higher than the county's 90th percentile warm season air temperature) were also observed. Our study identified the detrimental effects of long-term exposure to wildland fire smoke PM2.5 on a wide range of mortality outcomes, underscoring the need for public health actions and communications that span the health risks of both short- and long-term exposure.

Mortality attributable to PM2.5 from wildland fires in California from 2008 to 2018

In California, wildfire risk and severity have grown substantially in the last several decades. Research has characterized extensive adverse health impacts from exposure to wildfire-attributable fine particulate matter (PM2.5), but few studies have quantified long-term outcomes, and none have used a wildfire-specific chronic dose-response mortality coefficient. Here, we quantified the mortality burden for PM2.5 exposure from California fires from 2008 to 2018 using Community Multiscale Air Quality modeling system wildland fire PM2.5 estimates. We used a concentration-response function for PM2.5, applying ZIP code-level mortality data and an estimated wildfire-specific dose-response coefficient accounting for the likely toxicity of wildfire smoke. We estimate a total of 52,480 to 55,710 premature deaths are attributable to wildland fire PM2.5 over the 11-year period with respect to two exposure scenarios, equating to an economic impact of $432 to $456 billion. These findings extend evidence on climate-related health impacts, suggesting that wildfires account for a greater mortality and economic burden than indicated by earlier studies.

Assessing the 2023 Canadian wildfire smoke impact in Northeastern US: Air quality, exposure and environmental justice

The Canadian wildfires in June 2023 significantly impacted the northeastern United States, particularly in terms of worsened air pollution and environmental justice concerns. While advancements have been made in low-cost sensor deployments and satellite observations of atmospheric composition, integrating dynamic human mobility with wildfire PM2.5 exposure to fully understand the environmental justice implications remains underinvestigated. This study aims to enhance the accuracy of estimating ground-level fine particulate matter (PM2.5) concentrations by fusing chemical transport model outputs with empirical observations, estimating exposures using human mobility data, and evaluating the impact of environmental justice. Employing a novel data fusion technique, the study combines the Weather Research and Forecasting model with Chemistry (WRF-Chem) outputs and surface PM2.5 measurements, providing a more accurate estimation of PM2.5 distribution. The study addresses the gap in traditional exposure assessments by incorporating human mobility data and further investigates the spatial correlation of PM2.5 levels with various environmental and demographic factors from the US Environmental Protection Agency (EPA) Environmental Justice Screening and Mapping Tool (EJScreen). Results reveal that despite reduced mobility during high PM2.5 levels from wildfire smoke, exposure for both residents and individuals on the move remains high. Regions already burdened with high environmental pollution levels face amplified PM2.5 effects from wildfire smoke. Furthermore, we observed mixed correlations between PM2.5 concentrations and various demographic and socioeconomic factors, indicating complex exposure patterns across communities. Urban areas, in particular, experience persistent high exposure, while significant correlations in rural areas with EJScreen factors highlight the unique vulnerabilities of these populations to smoke exposure. These results advocate for a comprehensive approach to environmental health that leverages advanced models, integrates human mobility data, and addresses socio-demographic disparities, contributing to the development of equitable strategies against the growing threat of wildfires.

Pregnancy exposure to PM2.5 from wildland fire smoke and preterm birth in California

BACKGROUND

Wildfires in the Western United States are a growing and significant source of air pollution that is eroding decades of progress in air pollution reduction. The effects on preterm birth during critical periods of pregnancy are unknown.

METHODS

We assessed associations between prenatal exposure to wildland fire smoke and risk of preterm birth (gestational age < 37 weeks). We assigned smoke exposure to geocoded residence at birth for all live singleton births in California conceived 2007-2018, using weekly average concentrations of particulate matter ≤ 2.5 µm (PM2.5) attributable to wildland fires from United States Environmental Protection Agency's Community Multiscale Air Quality Model. Logistic regression yielded odds ratio (OR) for preterm birth in relation to increases in average exposure across the whole pregnancy, each trimester, and each week of pregnancy. Models adjusted for season, age, education, race/ethnicity, medical insurance, and smoking of the birthing parent.

RESULTS

For the 5,155,026 births, higher wildland fire PM2.5 exposure averaged across pregnancy, or any trimester, was associated with higher odds of preterm birth. The OR for an increase of 1 µg/m3 of average wildland fire PM2.5 during pregnancy was 1.013 (95 % CI:1.008,1.017). Wildland fire PM2.5 during most weeks of pregnancy was associated with higher odds. Strongest estimates were observed in weeks in the second and third trimesters. A 10 µg/m3 increase in average wildland fire PM2·5 in gestational week 23 was associated with OR = 1.034; 95 % CI: 1.019, 1.049 for preterm birth.

CONCLUSIONS

Preterm birth is sensitive to wildland fire PM2.5; therefore, we must reduce exposure during pregnancy.

Spatial Heterogeneity of the Respiratory Health Impacts of Wildfire Smoke PM2.5 in California

Wildfire smoke fine particles (PM2.5) are a growing public health threat as wildfire events become more common and intense under climate change, especially in the Western United States. Studies assessing the association between wildfire PM2.5 exposure and health typically summarize the effects over the study area. However, health responses to wildfire PM2.5 may vary spatially. We evaluated spatially-varying respiratory acute care utilization risks associated with short-term exposure to wildfire PM2.5 and explored community characteristics possibly driving spatial heterogeneity. Using ensemble-modeled daily wildfire PM2.5, we defined a wildfire smoke day to have wildfire-specific PM2.5 concentration ≥15 μg/m3. We included daily respiratory emergency department visits and unplanned hospitalizations in 1,396 California ZIP Code Tabulation Areas (ZCTAs) and 15 census-derived community characteristics. Employing a case-crossover design and conditional logistic regression, we observed increased odds of respiratory acute care utilization on wildfire smoke days at the state level (odds ratio [OR] = 1.06, 95% confidence interval [CI]: 1.05, 1.07). Across air basins, ORs ranged from 0.88 to 1.57, with the highest effect estimate in San Diego. A within-community matching design and spatial Bayesian hierarchical model also revealed spatial heterogeneity in ZCTA-level rate differences. For example, communities with a higher percentage of Black or Pacific Islander residents had stronger wildfire PM2.5-outcome relationships, while more air conditioning and tree canopy attenuated associations. We found an important heterogeneity in wildfire smoke-related health impacts across air basins, counties, and ZCTAs, and we identified characteristics of vulnerable communities, providing evidence to guide policy development and resource allocation.

Measuring long-term exposure to wildfire PM2.5 in California: Time-varying inequities in environmental burden

Wildfires have become more frequent and intense due to climate change and outdoor wildfire fine particulate matter (PM2.5) concentrations differ from relatively smoothly varying total PM2.5. Thus, we introduced a conceptual model for computing long-term wildfire PM2.5 and assessed disproportionate exposures among marginalized communities. We used monitoring data and statistical techniques to characterize annual wildfire PM2.5 exposure based on intermittent and extreme daily wildfire PM2.5 concentrations in California census tracts (2006 to 2020). Metrics included: 1) weeks with wildfire PM2.5 < 5 μg/m3; 2) days with non-zero wildfire PM2.5; 3) mean wildfire PM2.5 during peak exposure week; 4) smoke waves (≥2 consecutive days with <15 μg/m3 wildfire PM2.5); and 5) mean annual wildfire PM2.5 concentration. We classified tracts by their racial/ethnic composition and CalEnviroScreen (CES) score, an environmental and social vulnerability composite measure. We examined associations of CES and racial/ethnic composition with the wildfire PM2.5 metrics using mixed-effects models. Averaged 2006 to 2020, we detected little difference in exposure by CES score or racial/ethnic composition, except for non-Hispanic American Indian and Alaska Native populations, where a 1-SD increase was associated with higher exposure for 4/5 metrics. CES or racial/ethnic × year interaction term models revealed exposure disparities in some years. Compared to their California-wide representation, the exposed populations of non-Hispanic American Indian and Alaska Native (1.68×, 95% CI: 1.01 to 2.81), white (1.13×, 95% CI: 0.99 to 1.32), and multiracial (1.06×, 95% CI: 0.97 to 1.23) people were over-represented from 2006 to 2020. In conclusion, during our study period in California, we detected disproportionate long-term wildfire PM2.5 exposure for several racial/ethnic groups.

Exploring spatial heterogeneity in synergistic effects of compound climate hazards: Extreme heat and wildfire smoke on cardiorespiratory hospitalizations in California

Extreme heat and wildfire smoke events are increasingly co-occurring in the context of climate change, especially in California. Extreme heat and wildfire smoke may have synergistic effects on population health that vary over space. We leveraged high-resolution satellite and monitoring data to quantify spatially varying compound exposures to extreme heat and wildfire smoke in California (2006-2019) at ZIP Code Tabulation Area (ZCTA) level. We found synergistic effects between extreme heat and wildfire smoke on daily cardiorespiratory hospitalizations at the state level. We also found spatial heterogeneity in such synergistic effects across ZCTAs. Communities with lower education attainment, lower health insurance coverage, lower income, lower proportion of automobile ownership, lower tree canopy coverage, higher population density, and higher proportions of racial/ethnic minorities experienced higher synergistic effects. This study highlights the need to incorporate compound hazards and environmental justice considerations into evidence-based policy development to protect populations from increasingly prevalent compound hazards.

Health Effects of Wildfire Smoke Exposure

We review current knowledge on the trends and drivers of global wildfire activity, advances in the measurement of wildfire smoke exposure, and evidence on the health effects of this exposure. We describe methodological issues in estimating the causal effects of wildfire smoke exposures on health and quantify their importance, emphasizing the role of nonlinear and lagged effects. We conduct a systematic review and meta-analysis of the health effects of wildfire smoke exposure, finding positive impacts on all-cause mortality and respiratory hospitalizations but less consistent evidence on cardiovascular morbidity. We conclude by highlighting priority areas for future research, including leveraging recently developed spatially and temporally resolved wildfire-specific ambient air pollution data to improve estimates of the health effects of wildfire smoke exposure.

Emergency department visits associated with wildfire smoke events in California, 2016-2019

Wildfire smoke has been associated with adverse respiratory outcomes, but the impacts of wildfire on other health outcomes and sensitive subpopulations are not fully understood. We examined associations between smoke events and emergency department visits (EDVs) for respiratory, cardiovascular, diabetes, and mental health outcomes in California during the wildfire season June-December 2016-2019. Daily, zip code tabulation area-level wildfire-specific fine particulate matter (PM2.5) concentrations were aggregated to air basins. A "smoke event" was defined as an air basin-day with a wildfire-specific PM2.5 concentration at or above the 98th percentile across all air basin-days (threshold = 13.5 μg/m3). We conducted a two-stage time-series analysis using quasi-Poisson regression considering lag effects and random effects meta-analysis. We also conducted analyses stratified by race/ethnicity, age, and sex to assess potential effect modification. Smoke events were associated with an increased risk of EDVs for all respiratory diseases at lag 1 [14.4%, 95% confidence interval (CI): (6.8, 22.5)], asthma at lag 0 [57.1% (44.5, 70.8)], and chronic lower respiratory disease at lag 0 [12.7% (6.2, 19.6)]. We also found positive associations with EDVs for all cardiovascular diseases at lag 10. Mixed results were observed for mental health outcomes. Stratified results revealed potential disparities by race/ethnicity. Short-term exposure to smoke events was associated with increased respiratory and schizophrenia EDVs. Cardiovascular impacts may be delayed compared to respiratory outcomes.

The Burden of Wildfire Smoke on Respiratory Health in California at the Zip Code Level: Uncovering the Disproportionate Impacts of Differential Fine Particle Composition

Wildfires constitute a growing source of extremely high levels of particulate matter that is less than 2.5 microns in diameter (PM2.5). Recently, toxicologic and epidemiologic studies have shown that PM2.5 generated from wildfires may have a greater health burden than PM2.5 generated from other pollutant sources. This study examined the impact of PM2.5 on hospitalizations for respiratory diseases in California between 2006 and 2019 using a health impact assessment approach that considers differential concentration-response functions (CRF) for PM2.5 from wildfire and non-wildfire sources of emissions. We quantified the burden of respiratory hospitalizations related to PM2.5 exposure at the zip code level through two different approaches: (a) naïve (considering the same CRF for all PM2.5 emissions) and (b) nuanced (considering different CRFs for PM2.5 from wildfires and from other sources). We conducted a Geographically Weighted Regression to analyze spatially varying relationships between the delta (i.e., the difference between the naïve and nuanced approaches) and the Centers for Disease Control and Prevention's Social Vulnerability Index (SVI). A higher attributable number of respiratory hospitalizations was found when accounting for the larger health burden of wildfire PM2.5. We found that, between 2006 and 2019, the number of hospitalizations attributable to PM2.5 may have been underestimated by approximately 13% as a result of not accounting for the higher CRF of wildfire-related PM2.5 throughout California. This underestimation was higher in northern California and areas with higher SVI rankings. The relationship between delta and SVI varied spatially across California. These findings can be useful for updating future air pollution guideline recommendations.

Wildfire Exposure and Health Care Use Among People Who Use Durable Medical Equipment in Southern California

BACKGROUND

People using electricity-dependent durable medical equipment (DME) may be vulnerable to health effects from wildfire smoke, residence near wildfires, or residence in evacuation zones. To our knowledge, no studies have examined their healthcare utilization during wildfires.

METHODS

We obtained 2016-2020 counts of residential Zip Code Tabulation Area (ZCTA) level outpatient, emergency department (ED), and inpatient visits made by DME-using Kaiser Permanente Southern California members 45+. We linked counts to daily ZCTA-level wildfire particulate matter (PM) 2.5 and wildfire boundary and evacuation data from the 2018 Woolsey and 2019 Getty wildfires. We estimated the association of lagged (up to 7 days) wildfire PM 2.5 and residence near a fire or in an evacuation zone and healthcare visit frequency with negative binomial and difference-in-differences models.

RESULTS

Among 236,732 DME users, 10 µg/m 3 increases in wildfire PM 2.5 concentration were associated with the reduced rate (RR = 0.96; 95% confidence interval [CI] = 0.94, 0.99) of all-cause outpatient visits 1 day after exposure and increased rate on 4 of 5 subsequent days (RR range 1.03-1.12). Woolsey Fire proximity (<20 km) was associated with reduced all-cause outpatient visits, whereas evacuation and proximity were associated with increased inpatient cardiorespiratory visits (proximity RR = 1.45; 95% CI = 0.99, 2.12, evacuation RR = 1.72; 95% CI = 1.00, 2.96). Neither Getty Fire proximity nor evacuation was associated with healthcare visit frequency.

CONCLUSIONS

Our results support the hypothesis that wildfire smoke or proximity interrupts DME users' routine outpatient care, via sheltering in place. However, wildfire exposures were also associated with increased urgent healthcare utilization in this vulnerable group.

Short-Term total and wildfire fine particulate matter exposure and work loss in California

BACKGROUND

Few studies investigated the impact of particulate matter (PM2.5) on some symptom exacerbations that are not perceived as severe enough to search for medical assistance. We aimed to study the association of short-term daily total PM2.5 exposure with work loss due to sickness among adults living in California.

METHODS

We included 44,544 adult respondents in the workforce from 2015 to 2018 California Health Interview Survey data. Daily total PM2.5 concentrations were linked to respondents' home addresses from continuous spatial surfaces of PM2.5 generated by a geostatistical surfacing algorithm. We estimated the effect of a 2-week average of daily total PM2.5 exposure on work loss using logistic regression models.

RESULTS

About 1.69% (weighted percentage) of adult respondents reported work loss in the week before the survey interview. The odds ratio of work loss was 1.45 (odds ratio [OR] = 1.45, 95% confidence interval [CI]: 1.03, 2.03) when a 2-week average of daily total PM2.5 exposure was higher than 12 µg/m3. The OR for work loss was 1.05 (95% CI: 0.98, 1.13) for each 2.56ug/m3 increase in the 2-week average of daily total PM2.5 exposure, and became stronger among those who were highly exposed to wildfire smoke (OR = 1.06, 95% CI: 1.00, 1.13), compared to those with lower wildfire smoke exposure (OR = 1.04, 95% CI: 0.79, 1.39).

CONCLUSIONS

Our findings suggest that short-term ambient PM2.5 exposure is positively associated with work loss due to sickness and the association was stronger among those with higher wildfire smoke exposure. It also indicated that the current federal and state PM2.5 standards (annual average of 12 µg/m3) could be further strengthened to protect the health of the citizens of California.

Ensemble-based classification approach for PM2.5 concentration forecasting using meteorological data

Air pollution is a serious challenge to humankind as it poses many health threats. It can be measured using the air quality index (AQI). Air pollution is the result of contamination of both outdoor and indoor environments. The AQI is being monitored by various institutions globally. The measured air quality data are kept mostly for public use. Using the previously calculated AQI values, the future values of AQI can be predicted, or the class/category value of the numeric value can be obtained. This forecast can be performed with more accuracy using supervised machine learning methods. In this study, multiple machine-learning approaches were used to classify PM2.5 values. The values for the pollutant PM2.5 were classified into different groups using machine learning algorithms such as logistic regression, support vector machines, random forest, extreme gradient boosting, and their grid search equivalents, along with the deep learning method multilayer perceptron. After performing multiclass classification using these algorithms, the parameters accuracy and per-class accuracy were used to compare the methods. As the dataset used was imbalanced, a SMOTE-based approach for balancing the dataset was used. Compared to all other classifiers that use the original dataset, the accuracy of the random forest multiclass classifier with SMOTE-based dataset balancing was found to provide better accuracy.

Retrospective analysis of wildfire smoke exposure and birth weight outcomes in the San Francisco Bay Area of California

Despite the occurrence of wildfires quadrupling over the past four decades, the health effects associated with wildfire smoke exposures during pregnancy remains unknown. Particulate matter less than 2.5 μms (PM2.5) is among the major pollutants emitted in wildfire smoke. Previous studies found PM2.5 associated with lower birthweight, however, the relationship between wildfire-specific PM2.5 and birthweight is uncertain. Our study of 7923 singleton births in San Francisco between January 1, 2017 and March 12, 2020 examines associations between wildfire smoke exposure during pregnancy and birthweight. We linked daily estimates of wildfire-specific PM2.5 to maternal residence at the ZIP code level. We used linear and log-binomial regression to examine the relationship between wildfire smoke exposure by trimester and birthweight and adjusted for gestational age, maternal age, race/ethnicity, and educational attainment. We stratified by infant sex to examine potential effect modification. Exposure to wildfire-specific PM2.5 during the second trimester of pregnancy was positively associated with increased risk of large for gestational age (OR = 1.13; 95% CI: 1.03, 1.24), as was the number of days of wildfire-specific PM2.5 above 5 μg m-3 in the second trimester (OR = 1.03; 95% CI: 1.01, 1.06). We found consistent results with wildfire smoke exposure in the second trimester and increased continuous birthweight-for-gestational age z-score. Differences by infant sex were not consistent. Counter to our hypothesis, results suggest that wildfire smoke exposures are associated with increased risk for higher birthweight. We observed strongest associations during the second trimester. These investigations should be expanded to other populations exposed to wildfire smoke and aim to identify vulnerable communities. Additional research is needed to clarify the biological mechanisms in this relationship between wildfire smoke exposure and adverse birth outcomes.

Exposure to Air Pollution and Emergency Department Visits During the First Year of Life Among Preterm and Full-term Infants

IMPORTANCE

Previous studies have focused on exposure to fine particulate matter 2.5 μm or less in diameter (PM2.5) and on birth outcome risks; however, few studies have evaluated the health consequences of PM2.5 exposure on infants during their first year of life and whether prematurity could exacerbate such risks.

OBJECTIVE

To assess the association of PM2.5 exposure with emergency department (ED) visits during the first year of life and determine whether preterm birth status modifies the association.

DESIGN, SETTING, AND PARTICIPANTS

This individual-level cohort study used data from the Study of Outcomes in Mothers and Infants cohort, which includes all live-born, singleton deliveries in California. Data from infants' health records through their first birthday were included. Participants included 2 175 180 infants born between 2014 and 2018, and complete data were included for an analytic sample of 1 983 700 (91.2%). Analysis was conducted from October 2021 to September 2022.

EXPOSURES

Weekly PM2.5 exposure at the residential ZIP code at birth was estimated from an ensemble model combining multiple machine learning algorithms and several potentially associated variables.

MAIN OUTCOMES AND MEASURES

Main outcomes included the first all-cause ED visit and the first infection- and respiratory-related visits separately. Hypotheses were generated after data collection and prior to analysis. Pooled logistic regression models with a discrete time approach assessed PM2.5 exposure and time to ED visits during each week of the first year of life and across the entire year. Preterm birth status, sex, and payment type for delivery were assessed as effect modifiers.

RESULTS

Of the 1 983 700 infants, 979 038 (49.4%) were female, 966 349 (48.7%) were Hispanic, and 142 081 (7.2%) were preterm. Across the first year of life, the odds of an ED visit for any cause were greater among both preterm (AOR, 1.056; 95% CI, 1.048-1.064) and full-term (AOR, 1.051; 95% CI, 1.049-1.053) infants for each 5-μg/m3 increase in exposure to PM2.5. Elevated odds were also observed for infection-related ED visit (preterm: AOR, 1.035; 95% CI, 1.001-1.069; full-term: AOR, 1.053; 95% CI, 1.044-1.062) and first respiratory-related ED visit (preterm: AOR, 1.080; 95% CI, 1.067-1.093; full-term: AOR,1.065; 95% CI, 1.061-1.069). For both preterm and full-term infants, ages 18 to 23 weeks were associated with the greatest odds of all-cause ED visits (AORs ranged from 1.034; 95% CI, 0.976-1.094 to 1.077; 95% CI, 1.022-1.135).

CONCLUSIONS AND RELEVANCE

Increasing PM2.5 exposure was associated with an increased ED visit risk for both preterm and full-term infants during the first year of life, which may have implications for interventions aimed at minimizing air pollution.