Assessing Mortality Burden from Coal Electricity-Generating Units: A Comprehensive Study on Individual-Level Health Records in the United States


Air pollution is a significant global concern associated with adverse health effects and an increased risk of mortality. Among the major contributors to poor air quality are coal electricity-generating units (EGUs), commonly known as power plants. The debate surrounding the continued use of coal for electricity generation persists, fueled by both economic considerations and growing concerns about public health and climate change.

While coal use is projected to rise globally, European nations have faced challenges related to energy stability, leading to increased reliance on coal. Despite a decline in coal EGU emissions in the United States in recent decades, understanding the health burden associated with coal EGUs and the impact of emission reduction measures is crucial for shaping effective public health, climate, and energy policies.


Previous studies assessing the mortality burden from coal EGUs in the US relied on estimated concentration response functions (CRFs), assuming that fine particulate matter (PM2.5) from coal emissions has the same toxicity as PM2.5 from all sources. However, emerging evidence suggests that exposure to sulfur, sulfates, or PM2.5 from coal emissions may carry higher relative morbidity or mortality risk compared to other PM2.5 constituents or sources. Limited regional and temporal scope, along with the absence of coal-specific exposure estimates, has impeded the use of coal-specific PM2.5 CRFs, likely resulting in underestimations of the mortality burden.


This study aimed to estimate the number of deaths associated with exposure to coal PM2.5 from EGUs. Using a national-scale approach, individual-level health records covering over 650 million person-years in the US Medicare population (≥65 years of age) from 1999 to 2016 were analyzed. The study focused on “coal PM2.5,” defined as PM2.5 specifically originating from coal EGU sulfur dioxide (SO2) emissions.


To estimate coal PM2.5 exposure, the study employed the HYSPLIT with Average Dispersion (HyADS) model. This model accounts for date-specific atmospheric transport of PM2.5, providing insights into exposure from individual EGUs. HyADS, a reduced complexity model, enabled the estimation of 22 years of coal PM2.5 exposure (1999 to 2020) from 480 US EGUs, a task requiring significantly less computation time than a typical full-scale chemical transport model.

Fig. 1. ZIP code–level coal PM2.5 over time.

Box plots (median, first, and third quartiles are shown as horizonal lines and outliers as dots) summarize the distribution of ZIP code levels of coal PM2.5. Map areas shown in white do not have ZIP codes. Plots were produced in R using ggplot2; spatial information comes from the USAboundaries package.


The study made several significant contributions:

  • Comparative Mortality Risk Analysis:
    • Estimated and compared mortality risks associated with exposure to coal PM2.5 versus total PM2.5 from all sources, revealing that prior analyses had underestimated the mortality burden from coal EGUs in the US.
  • Evaluating Individual Coal EGUs:
    • Calculated the number of deaths linked to each of the 480 coal EGUs, ranking their contributions to the mortality burden.
    • Tracked the contribution of each EGU over time, considering the implementation of emissions controls and retirements.
  • Spatial Distribution Analysis:
    • Documented the spatial distribution of the mortality burden across the US, providing insights into geographical variations in health impacts associated with coal EGUs.

Discussion: Understanding the Implications of Long-Term Exposure to Coal Emissions on Mortality

This study represents a significant milestone in the investigation of the health impacts associated with exposure to sulfur dioxide (SO2) emissions from coal electricity-generating units (EGUs) in the United States. The primary innovation lies in the integration of coal EGU-specific exposure estimates with individual-level health data over a prolonged period, a method hindered by historical limitations in large-scale health databases and source-specific exposure information. The study underscores the importance of utilizing advanced modeling techniques to derive air pollution exposure for well-characterized sources, providing critical insights for epidemiological and risk assessment purposes.

Fig. 2. Annual number of excess deaths attributable to coal PM2.5, estimated using the RR for coal PM2.5 from this study and RRs for total PM2.5 from the literature.
All excess deaths are estimated relative to zero coal PM2.5. The area filled by horizontal hashing indicates deaths estimated using RRs derived from this study (bounds represent 95% CI). Areas filled by vertical and diagonal hashing correspond to deaths estimated using RRs for total ambient PM2.5 exposure from the literature (4, 33). The gray shaded region from 2017 to 2020 represents years for which ZIP code–specific baseline death rates were assumed from the 2014 to 2016 average. This figure was produced in R using ggplot2.

Magnitude of the Mortality Burden:

Over the past two decades, the study reveals a substantial association between exposure to coal PM2.5 and a staggering 460,000 excess deaths in the United States, constituting over 22% of the total excess deaths attributable to PM2.5. This emphasizes the significant role of coal EGUs in contributing to the overall mortality burden associated with fine particulate matter. Notably, the findings challenge traditional impact assessments that rely on concentration response functions (CRFs) for total PM2.5 mass, suggesting an underestimation of the mortality burden related to coal PM2.5.

Comparative Mortality Risk:

The elevated mortality risk observed in association with annual exposure to coal PM2.5 aligns with previous evidence indicating increased relative health risks related to coal-specific PM2.5 or exposure to sulfur and sulfate. However, the study acknowledges conflicting results from other research, emphasizing the complex nature of understanding the distinct health risks associated with various PM2.5 species emitted from coal EGUs.

Successes of Regulatory Measures:

The study highlights a significant decrease in annual deaths over the study period, attributing this success to emissions reductions driven by regulations under the 1990 Clean Air Act Amendments. While the use of coal in the US has remained low, the global trajectory suggests a potential resurgence in coal use, posing concerns about future mortality costs if stringent measures are not implemented globally.

Challenges in Disentangling PM2.5 Species:

The study utilizes SO2 emissions from coal to estimate coal PM2.5, acknowledging the challenge of conclusively determining the relative harm of SO2-associated PM2.5 compared to other species emitted from coal power plants. The high correlation between various species emitted from coal EGUs makes it challenging to isolate the specific health impacts of individual constituents. Nevertheless, the study’s indication of a higher mortality risk associated with exposure to coal PM2.5 relative to other PM2.5 sources suggests potential population health benefits from reducing SO2 emissions through interventions like emissions control devices or facility closures.

Advancements in Modeling Techniques:

The study employs the HYSPLIT with Average Dispersion (HyADS) model, benefitting from well-characterized source locations and emissions. The use of reduced complexity models and incorporation of information from observation and chemical transport models offer promising avenues for improving linkages between emitted species, atmospheric processes, exposure, and health outcomes. Advanced sensitivity approaches within chemical transport models and computational advancements open opportunities for more explicit incorporation of atmospheric chemistry and physics in future studies.

Policy Implications and Future Considerations:

These results contribute to the evolving understanding of the varying toxicity of PM2.5 originating from different sources. They underscore the importance of reevaluating current air pollution risk assessments, which assume equal toxicity for ambient PM2.5 from all sources. Regulatory efforts could benefit from the research platform used in this study, offering targeted evidence on how individual EGU sources contribute to the existing health burden and supporting more efficient policy interventions. As countries, including the United States, continue to regulate ambient PM2.5 concentrations, incorporating insights into changes in PM2.5 composition and sources over time becomes crucial for informed policy assessments.


This comprehensive study fills critical gaps in understanding the mortality burden associated with coal EGUs in the United States. By utilizing individual-level health records and employing advanced modeling techniques, the research sheds light on the nuanced health risks posed by coal-specific PM2.5 exposure. The findings underscore the importance of considering the individual contributions of coal EGUs in informing policies aimed at mitigating the adverse health effects of air pollution, addressing climate concerns, and shaping the future of energy generation.

reference link :


Please enter your comment!
Please enter your name here

Questo sito usa Akismet per ridurre lo spam. Scopri come i tuoi dati vengono elaborati.