Abstract
The World Health Organization highlights the severe impact of air pollution on human health, leading to millions of excess deaths a year globally. Previous studies have suggested that switching from fossil fuels to renewable energy sources could save lives and limit climate change. However, estimates of mortality from air pollution vary widely. This study uses new risk functions, an updated global atmospheric model and exposure data to assess the health consequences of phasing out fossil fuels. The results suggest that a worldwide phase-out of fossil fuels could avoid approximately five million excess deaths a year from air pollution, particularly in the densely populated South and East Asia, where air quality is strongly compromised. The study emphasizes the urgent need to transition from fossil to clean energy sources to achieve significant health benefits.
1 Introduction
The World Health Organization (WHO) emphasizes the profound impact of air pollution on human health, likening its effects to tobacco smoking. The recently updated WHO air quality guidelines reflect this concern [1]. However, these guidelines are violated in most countries worldwide, which causes millions of excess deaths annually, as reported in the 2019 Global Burden of Disease (GBD) study [2]. Transitioning from fossil fuels to renewable energy sources could prevent many of these deaths and help limit the global temperature rise to below 2°C, aligning with the Paris Climate Agreement. However, estimating the mortality burden from degraded air quality is challenging, with varying estimates and limited studies assessing outcomes due to all pollution-related causes of death, while a larger number of analyses have addressed specific disease categories.
The present report is based on a recent study, reported in The BMJ, which evaluated the impacts of phasing out fossil fuels on disease-specific and all-cause mortality attributable to air pollution [3]. It contributed to the United Nations Climate Change Conference, i.e., the Conference of the Parties of the UNFCCC, or COP28, held in Dubai, United Arab Emirates, in 2023. The research utilized a new risk model and updated exposure estimates to fine particulate matter with a diameter of less than 2.5 µm (PM2.5) and ozone [4]. An Earth system model was employed to calculate atmospheric composition and global air quality to apportion pollution exposure and the health impacts to the source categories, notably the energy, mobility and industrial sectors that depend on fossil fuel combustion. The study assessed the health benefits from partial and full phaseouts of fossil fuels, assuming the transition into clean, renewable technologies.
2 Methods
The study employed a novel global atmospheric modelling method to compute gaseous and particulate air pollutants and attribute them to various source categories. Focusing on 2019 to align with recent disease burden assessments, the study utilized observational satellite and air quality station data to calculate long-term exposure to PM2.5 at a global horizontal resolution of 10 km (ref. 5). Based on the model results, we estimated fractional changes in PM2.5 related to emission sectors by simulating scenarios assuming that non-polluting alternatives, especially renewable energy sources, substitute different pollution categories. Emission sectors include fossil fuel combustion, e.g., energy production, industry, land transport, shipping and aviation. Other anthropogenic sources involve, e.g., residential energy from solid biofuels, waste incineration, agriculture, solvent production and use. Note that the residential combustion of solid biofuels in households significantly contributes to both indoor and ambient air pollution, with pervasive health impacts in South and East Asia and Africa.
The Earth system model employed (EMAC) [6] integrates global atmospheric, land and ocean submodels, exchange processes, and detailed air chemistry and aerosol routines. Meteorological data from 2019 are assimilated in the model using Newtonian nudging towards meteorological reanalysis data. Emission data for gaseous and particulate pollutants are derived from the Community Emission Data System [7]. Four scenarios were considered, of which three are presented here, ranging from phasing out all fossil fuel-related emission sources to removing all anthropogenic sources, also considering natural sources such as aeolian dust and wildfires. Evaluation against observational data and ground-based measurements indicated close agreement with the model’s PM2.5 and ozone (O3) results [8].
The study also calculated exposure to ozone in 2019, in addition to PM2.5, based on atmospheric model results only, even though the relatively coarse model resolution limits its ability to capture local ozone variability; however, longer-term averages are generally well captured. Note that ozone-attributable deaths account for less than 5% of those attributed to PM2.5, minimizing the impact of ozone-related uncertainties on the mortality assessment. Of the three scenarios, the first includes all emissions described above, the second simulates a complete phaseout of fossil fuels, and the third “halfway” scenario assumes that 50% of the phaseout has been achieved. Note that ref. 3 also presents a “quarterway” scenario in which 25% of the phaseout is achieved. These scenarios were applied to study potential non-linearities along the pathway toward a complete phaseout and determine the health benefits of intermediate goals.
The study employs the FUSION relative risk model to estimate all-cause and cause-specific mortality associated with long-term exposure to PM2.5 (ref. 4). Relative risk functions derived from epidemiological cohort studies have provided the basis to estimate the global number of deaths attributable to air pollution, accounting for six disease categories: ischemic heart disease (IHD), stroke (ST), diabetes mellitus type-2 (DM), lung cancer (LC), chronic obstructive pulmonary disease (COPD) and lower respiratory infections (LRI). The counterfactual concentration range for PM2.5, below which health impacts are not statistically significant, is 2.4-5.9 µg/m3. Consistency with the GBD methodology is maintained for mortality attributable to ozone, focusing on COPD-related deaths.
Comparison with the GBD methodology shows that the FUSION relative risk model yields larger health benefits from reductions in air pollution, especially at the high and low ends of the PM2.5 concentration range. It was found that FUSION is particularly suited for global applications across various income levels and exposures [3,4].
3 Results
3.1 Global Mortality Burden
The results highlight the significant number of deaths worldwide attributable to long-term exposure to PM2.5 and O3, with a particularly large number of cases in South and East Asia, Eastern Europe, the Middle East, and West Africa. The global total all-cause excess mortality was estimated at 8.34 million per year, with a 95% confidence interval (CI) of 5.63-11.19 per year (Fig. 1). Cardiometabolic conditions, such as IHD, ST and DM contribute significantly to the mortality burden, with IHD alone contributing about 30% to the total.
While specific causes of death are identifiable, there is “another” category representing additional health outcomes from exposure to air pollution, adding up to about 20% of the total excess deaths. This indicates the need for further research, e.g., on pollution impacts through hypertension and neurological diseases [9.10]. South and East Asia, the Middle East, and Eastern Europe were shown to exhibit particularly high per-capita mortality rates. China and India stand out with the highest total attributable mortality, reflecting both the size of the population and the severity of air pollution in these regions.
3.2 Fossil Fuel-Related Mortality
Fossil fuel-related emissions, predominantly from industry, transportation, and power generation, contribute significantly to global mortality related to air pollution. Coal combustion is the leading contributor to fossil fuel-related mortality. Phasing out fossil fuels could potentially lead to significant relative reductions in attributable mortality in high-income countries that strongly rely on this energy source. In absolute numbers, fossil fuel phaseouts are highly effective in reducing attributable deaths in many low- and middle-income countries.
The global total contribution from exposure to fossil fuel-related air pollution is 61%, i.e., 5.13 (95%CI 3.63-6.32) excess deaths annually. This amounts to 82% of all anthropogenic and, hence, potentially avoidable deaths from air pollution. Smaller reductions in fossil fuel-related emissions rather than a radical phaseout still yield significant positive health outcomes. It was found that the health benefits respond relatively linearly to the lowering of exposure (Fig. 2). In high-income countries, the halfway scenario is comparatively most effective because, in some countries, the counterfactual PM2.5 level (below 5 µg/m3) can be reached under this scenario. Nevertheless, the 50% phaseout greatly improves air quality in all regions.
4 Discussion
Applying the new FUSION risk model, coupled with updated data on exposure to ambient fine particulate matter (PM2.5) and ozone (O3), we estimate global all-cause attributable mortality at about 8.3 million per year, higher than previously reported by the GBD Study for 2019 (ref. 2). The all-cause mortality related to PM2.5 exposure alone (i.e., without O3) is estimated at nearly eight million per year, slightly lower than previous estimates [3,11].
Differences with the GBD results are primarily due to the FUSION model’s optimization of the exposure-response relationship across ambient PM2.5 levels. The relatively low number estimated by the GBD is attributed to its use of integrated exposure-response functions, based on PM2.5 dosage and toxicity assumptions of second-hand smoking and indoor air pollution, used as proxies for high levels of air pollution. In contrast, FUSION only considers studies involving ambient air pollution, including those representing heavily polluted air.
The study finds high attributable mortality rates in South and East Asia, as well as in Eastern Europe and parts of the Middle East. South, East, and Southeast Asia account for 55% of the world’s population but 70% of air pollution-related mortality. Cardiometabolic diseases, including ischemic heart disease, stroke, and diabetes, contribute about 65% to the global disease-specific mortality burden from ambient air pollution.
A global phaseout of fossil fuels is projected to yield substantial health benefits, particularly in reducing mortality from cardiometabolic outcomes but also chronic obstructive pulmonary disease (COPD) and lung cancer. However, lower respiratory infections (LRIs), most frequent in low-income countries, may see fewer reductions related to the less prevalent role of fossil fuels in air pollution. The study attributes a worldwide major fraction of mortality to fossil fuel-related air pollution owing to its leading influence on exposure to PM2.5, especially in the low – to moderately high concentration ranges. By phasing out fossil fuel use in industry, energy generation and transportation, most countries will accomplish the WHO air quality guideline concentration of 5 µg/m3, i.e., at the level of the counterfactual concentration.
The FUSION relative risk model draws on various epidemiological cohort studies worldwide. However, some continents, like Africa, are underrepresented, and there is a need to develop such studies there. An analysis by Pope et al.[12] has addressed general questions about mortality risk calculations. They showed that disease burden analyses, based on growing epidemiological data and supported by numerous clinical and toxicological studies in the past 25 years, have become robust and that initial concerns raised several decades ago have been largely overcome. Nevertheless, challenges persist in fully accounting for confounding factors. While cohort studies adjust for numerous covariates, residual confounding remains a possibility.
Since most epidemiological cohort studies have been performed in high- and middle-income countries, additional focus on low-income countries is needed to understand better the health impacts of air pollution on a global scale. Despite becoming part of regular disease burden updates, assessing long-term health impacts still faces methodological heterogeneity, particularly concerning exposure data and relative risk functions.
Finally, toxicological analyses indicate variations in the oxidative potential of PM2.5 from different sources, challenging the assumption of equal toxicity of all particulate components in relative risk models. Oxidative stress has been implicated as a driver of health impacts, e.g., leading to epithelial dysfunction and inflammatory responses [13,14]. Relatively high toxicity has been attributed to combustion sources of PM2.5, both from fossil fuel and residential energy use [15,16]. In the future, it may be helpful to additionally account for the oxidative potential of particulate pollution in disease burden analyses, provided that such data can be derived from air quality measurement networks.
5 Conclusions
Air pollution remains a leading public health hazard, leading to more than eight million excess deaths annually. A substantial portion of these deaths are preventable, stemming from human activities, particularly the use of fossil fuels, globally accounting for more than five million excess deaths per year. This represents about 82% of the exposure to anthropogenic emissions. While a complete fossil-fuel phaseout is not a realistic objective in the near term, our results emphasize the tremendous health benefits of a world that would have adopted a sustainable pathway of generating energy for industry, air conditioning (heating, cooling) and transportation. Aligning with the objectives of the Paris Climate Agreement to achieve climate neutrality by 2050, transitioning to clean, renewable energy sources presents a remarkable opportunity for improving public health and mitigating climate change.
References
1. World Health Organization. WHO global air quality guidelines. Geneva: WHO 2021.
2. Murray, C.J.L., A.Y. Aravkin, P. Zheng, et al. Global burden of 87 risk factors in 204 countries and territories, 1990-2019: A systematic analysis for the Global Burden of Disease Study 2019. The Lancet 396, 1223-1249, 2020.
3. Lelieveld, J., A. Haines, R. Burnett, C. Tonne, K. Klingmüller, T. Münzel and A. Pozzer. Air pollution deaths attributable to fossil fuels: observational and modelling study. The BMJ 383, e077784, 2023.
4. Burnett, R., J.V. Spadaro G.R. Garcia et al. Designing health impact functions to assess marginal changes in outdoor fine particulate matter. Environ. Res., 204, 112245, 2022.
5. van Donkelaar, A., M.S Hammer, L. Bindle et al. Monthly global estimates of fine particulate matter and their uncertainty. Environ. Sci. Technol., 55, 15287-15300, 2021.
6. Lelieveld, J., K. Klingmüller, A. Pozzer et al. Effects of fossil fuel and total anthropogenic emission removal on public health and climate. Proc. Natl. Acad. Sci. USA., 116, 7192-7197, 2019.
7. McDuffie, E.E., R.V. Martin J.V. Spadaro et al. Source sector and fuel contributions to ambient PM2.5 and attributable mortality across multiple spatial scales. Nat. Comm., 12, 3594, 2021.
8. Pozzer, A., S.F. Reifenberg, V. Kumar et al. Simulation of organics in the atmosphere: evaluation of EMACv2.54 with the Mainz Organic Mechanism (MOM) coupled to the ORACLE (v1.0) submodel. Geosci. Model Dev., 15, 2673-2710, 2022.
9. Hahad, O., J. Lelieveld, F. Birklein et al. Cerebrovascular and neurological disorders from ambient air pollution – Inflammation and oxidative stress as common features. Int. J. Molec. Sci., 21, 4306, 2020.
10. Hahad, O, S. Rajagopalan, J. Lelieveld et al. Noise and air pollution as risk factors for hypertension: Part I – Epidemiology. Hypertension, 80, 1375-1383, 2023.
11. Burnett, R, Chen H, Szyszkowicz M, et al. Global estimates of mortality associated with long-term exposure to outdoor fine particulate matter. Proc Natl Acad Sci USA. 2018;115:9592–7.
12. Pope, C.A., N. Coleman Z.A. Pond et al. Fine particulate air pollution and human mortality: 25+ years of cohort studies. Environ. Res., 183, 108924, 2020.
13. Münzel, T., T. Gori, S. Al-Kindi et al. Effects of gaseous and solid constituents of air pollution on endothelial function. Eur. Heart J., 39, 3543-3550, 2018.
14. Park, M., H.S. Joo, K. Lee et al. Differential toxicities of fine particulate matter from various sources. Sci. Rep., 8, 1-11, 2018.
15. Bates, J.T., T. Fang V. Verma et al. Review of acellular assays of ambient particulate matter oxidative potential: Methods and relationships with composition, sources, and health effects. Environ. Sci. Technol., 53, 4003-4019, 2019.
16. Chowdhury, S., A. Pozzer, A. Haines et al. Global health burden of ambient PM2.5 and the contribution of anthropogenic black carbon and organic aerosols. Environ. Int., 159, 107020, 2022.