Skip to main content
Environmental Science and Ecotechnology logoLink to Environmental Science and Ecotechnology
. 2026 Oct 23;34:100781. doi: 10.1016/j.ese.2026.100781

Halving PM2.5 mortality by 2040: China's tightened air quality standard and residual disease burdens

Ruohan Wang a,b, Ning Kang a,b, Yixuan Zheng c,⁎, Tao Xue a,b,d,e,⁎⁎
PMCID: PMC13769092  PMID: 43034549

Abstract

Fine particulate matter (PM2.5) remains one of the leading environmental causes of premature death worldwide, and the World Health Organization has called on countries to halve air-pollution-related mortality by 2040 (the “halving target”). China has made major progress in reducing PM2.5 pollution over the past decade, and in 2026 it tightened its National Ambient Air Quality Standards (NAAQS; GB 3095-2026) from 35 to 25 μg m−3 through phased implementation. However, how much this revision moves China toward the halving target, and how its health benefits would differ across diseases, remain unclear. Here we integrated high-resolution PM2.5 data, gridded population estimates, and disease-specific exposure–response functions within a comparative risk assessment framework to estimate PM2.5-attributable mortality in mainland China from 2013 to 2024. In 2024, full compliance with the 25 μg m−3 limit would avert 4.3 times as many deaths as full compliance with the former 35 μg m−3 limit. PM2.5-specific progress toward the halving target would increase from 66.2% under the 35 μg m−3 limit to 83.4% under the 25 μg m−3 limit. Health benefits varied markedly across diseases, with larger responses for chronic obstructive pulmonary disease, stroke, lung cancer, lower respiratory infections, and ischemic heart disease, but much smaller responses for dementia and type 2 diabetes; the dementia estimate requires caution due to weaker evidence. These findings indicate that the revised NAAQS could make a meaningful contribution to reducing PM2.5-attributable mortality and advancing progress toward the halving target, while the marked disease-specific heterogeneity highlights the need to pair future air quality governance with health-oriented strategies.

Graphical abstract

graphic file with name ga1.webp

Highlights

  • •

    China tightened its air quality limit for PM2.5 from 35 to 25 μg m−3.

  • •

    This expands the population living above the national limit from 33% to 65%.

  • •

    The stricter limit could avert 4.3 times more deaths than the old one.

  • •

    Benefits vary across diseases, with weaker gains for dementia and type 2 diabetes.

  • •

    Full compliance with the stricter limit could achieve ∼83% of progress required for the WHO halving target.

1. Introduction

Exposure to fine particulate matter with a diameter of 2.5 μm (PM2.5) or less is widely recognized as a leading environmental risk factor contributing to the global burden of disease [1]. Epidemiological studies have demonstrated a robust causal association between long-term PM2.5 exposure and all-cause mortality, as well as cause-specific outcomes, including cardiovascular and respiratory diseases [2], [3], [4]. To address this challenge, China, which bears a substantial disease burden attributable to PM2.5 exposure, revised its National Ambient Air Quality Standards (NAAQS; GB 3095-2012) in 2012. This revision incorporated PM2.5 as a routine monitoring indicator and established an annual limit of 35 μg m−3, hereinafter referred to as NAAQS-35. Subsequently, China implemented three comprehensive clean air action plans. These interventions significantly improved air quality: From 2015 to 2024, the national average PM2.5 concentration dropped by 36%, and the frequency of heavy pollution days declined by 68%. These achievements translated into measurable gains in life expectancy and reductions in medical expenditures, progress widely recognized as the “air pollution control miracle” [5], [6], [7].

Despite these improvements, air pollution governance in China faces increasingly complex challenges. Internationally, regulatory standards have become progressively more stringent. In 2021, the World Health Organization (WHO) updated its global air quality guidelines, lowering the recommended annual PM2.5 concentration from 10 to 5 μg m−3 [8]. Similarly, major developed economies, including the United States, the European Union, and Japan, have adopted annual limits of 9, 10, and 15 μg m−3, respectively [9], [10], [11]. In comparison, the NAAQS-35 standard remains substantially less stringent. Domestically, the “Beautiful China Initiative” mandates that the national average PM2.5 concentration must fall below 25 μg m−3 by 2035. However, many cities currently compliant with the NAAQS-35 standard still exceed this 2035 target, indicating a significant gap between current regulatory standards and future strategic objectives.

To bridge this gap and reinforce the guiding role of air quality standards, China's Ministry of Ecology and Environment and the State Administration for Market Regulation jointly issued the updated National Ambient Air Quality Standards (GB 3095-2026) on February 24, 2026. Officially implemented on March 1, 2026, the updated standards mark the first major revision in more than a decade, tightening the annual PM2.5 limit from 35 to 25 μg m−3. The revision outlines a stepwise implementation strategy: Phase I (2026–2030) establishes an interim target of 30 μg m−3 (NAAQS-30), followed by Phase II, beginning in 2031, when the 25 μg m−3 (NAAQS-25) limit is scheduled to take full effect.

This revision also provides an opportunity to assess China's updated PM2.5 standard in relation to broader global health objectives. During the Second Global Conference on Air Pollution and Health in 2025, the WHO established a target for nations to achieve a 50% reduction in the population-attributable fraction (PAF) of mortality derived from anthropogenic air pollution by 2040 relative to a 2015 baseline, hereinafter referred to as the “halving target” [12], [13], [14]. Given China's substantial baseline burden and demonstrated capacity for effective air pollution control, the country is poised to play a pivotal role in achieving this global goal [15]. Therefore, quantitatively evaluating the magnitude of the health benefits associated with the revised NAAQS, as well as its potential contribution to the WHO's halving target, is important. Such an assessment can provide evidence of the public health implications of China's updated regulatory framework and inform health-oriented air quality management in other developing countries.

Critically, the protective efficacy of the revised NAAQS is unlikely to be uniform across health outcomes. Epidemiological evidence suggests that exposure–response relationships for PM2.5-related diseases, ranging from cardiovascular and respiratory conditions to metabolic and neurodegenerative disorders, exhibit distinct nonlinear patterns [16], [17], [18]. These disease-specific nonlinear patterns may translate into differences in the marginal benefits expected from the revised standard. Furthermore, in the context of China's rapidly aging society, the rising burden of age-dependent conditions may partially offset the benefits of reduced exposure [19], [20]. Therefore, a generalized assessment of total mortality is insufficient. A comprehensive, disease-specific assessment is needed to quantify how different health conditions respond to stricter PM2.5 standards and to identify where additional health gains may remain achievable. This granular approach can inform more health-oriented air quality management strategies.

Building on established methodologies for comparative risk assessment, this study integrates high-resolution air pollution data, population estimates, and exposure–response functions linking long-term PM2.5 exposure to specific health outcomes to quantify the health burden attributable to PM2.5 exposure in mainland China from 2013 to 2024. Specifically, we (1) evaluate the potential health benefits of full compliance with alternative annual PM2.5 standards, including NAAQS-35, NAAQS-30, and NAAQS-25; (2) analyze the disease-specific heterogeneity of these benefits across mainland China; and (3) assess PM2.5-specific progress toward the WHO 2040 halving target using the 2015 observed scenario as a reference. Together, these analyses provide a timely, policy-relevant evidence base to inform implementation of the new standard and optimize future air quality management strategies.

2. Methods

2.1. Environmental and population data

Grid-level PM2.5 concentrations from 2013 to 2024 were obtained from the Tracking Air Pollution (TAP) dataset (http://tapdata.org.cn/) at a spatial resolution of 1 km × 1 km [21]. The TAP framework generated this high-resolution product by fusing the earlier 10 km TAP PM2.5 product [22], 1 km satellite aerosol optical depth retrievals, and land-use parameters in a random forest framework. Validation of this 1 km resolution product showed out-of-bag cross-validation R2 values of 0.80–0.84 across years and a by-year cross-validation R2 of 0.76 for annual PM2.5 estimates. This dataset has been widely used in epidemiological and health risk assessment studies [23], [24], [25].

High-resolution gridded population data were sourced from the WorldPop dataset (https://www.worldpop.org), originally provided at a spatial resolution of 100 m × 100 m. This dataset disaggregates census data into grid cells using a random forest algorithm based on various geospatial covariates [26]. To capture accurate demographic trends, we constrained the gridded data using annual county-level population statistics from the China Statistical Yearbook (2013–2024). We then aggregated the population data and spatially aligned it with the PM2.5 dataset at a unified resolution of 1 km × 1 km.

2.2. Assessment of health burden

We followed the comparative risk assessment framework of the Global Burden of Disease (GBD) 2023 study to quantify the cause-specific mortality attributable to long-term PM2.5 exposure [1], [16]. The assessment focused on seven diseases included in the GBD 2023 particulate matter pollution risk–outcome set: ischemic heart disease (IHD), stroke, chronic obstructive pulmonary disease (COPD), lung cancer, lower respiratory infections (LRI), type 2 diabetes mellitus (T2DM), and dementia [16], [17]. Dementia is a recently incorporated outcome in this GBD risk–outcome set based on the exposure–response function reported by Huang et al. [17]. Evidence strength, as rated by the GBD burden-of-proof framework, was four stars for COPD, three stars for IHD, stroke, lung cancer, LRI, and T2DM, and two stars for dementia. The corresponding exposure–response curves are shown in Supplementary Fig. S1. The study covered the Chinese mainland, excluding the Hong Kong Special Administrative Region, the Macau Special Administrative Region, and Taiwan due to data availability.

The number of premature deaths attributable to PM2.5 exposure, referred to as the attributable number (AN), was calculated for each spatial grid, calendar year, disease type, and age–sex group as:

ANs,t,j,g=AFj(Cs,t)×Bs,t,j,g×Ps,t,g (1)

where subscripts s, t, j, and g denote the spatial grid (1 km × 1 km), calendar year (2013–2024), specific disease type, and age–sex group, respectively. Age–sex groups were defined by sex and 18 age categories (younger than 1 year, 1–4 years, five-year groupings from 5 to 9 to 75–79 years, and ≥80 years). Here, Ps,t,g represents the population of group g in grid s during year t, and Bs,t,j,g denotes the corresponding disease-specific baseline mortality rate.

The attributable fraction (AF) quantifies the fraction of deaths attributable to PM2.5 exposure and is calculated from the relative risk (RR):

AFj(Cs,t)=1−1RRj(Cs,t) (2)

where Cs,t represents the annual mean PM2.5 concentration in grid s during year t. RR represents the mortality risk at the observed concentration relative to that at the theoretical minimum risk exposure level (TMREL). It is calculated using the disease-specific exposure–response function (fj(C)), which describes the relationship between PM2.5 concentration C and mortality risk for disease j:

RRj(Cs,t)=fj(Cs,t)fj(TMREL) (3)

The exposure–response function was estimated using the meta-regression–Bayesian regularized trimmed (MR-BRT) tool [27]. Following the GBD framework, disease-specific exposure–response functions were assumed to be identical across age–sex groups. As a sensitivity analysis, we applied the global exposure mortality model (GEMM) for non-communicable diseases and lower respiratory infections (NCD + LRI) as an alternative exposure–response function to assess the robustness of the relative differences in avoided deaths across counterfactual scenarios [28].

We derived the disease-specific baseline mortality rate in two steps. First, we constructed gridded estimates of baseline total mortality rates by combining annual reports from the National Disease Surveillance Program with gridded maps of sex–age–residence subpopulations. These gridded mortality estimates were validated against county-level mortality rates from census data. Second, we estimated disease-specific baseline mortality rates by applying the cause-of-death profile from the GBD study [1], [29].

Additionally, we estimated the PM2.5-attributable incidence burden for the same set of diseases. We followed the same framework described above but replaced the baseline mortality rates with cause-specific incidence rates from the GBD study [1]. The results regarding the incidence burden are provided in the Supplementary Materials.

2.3. Scenario design for potential health benefits

To quantify the potential health benefits of tightening air quality regulations, we defined counterfactual scenarios in which PM2.5 concentrations met specific regulatory targets. This framework compares the observed health burden against the hypothetical burden achievable under full compliance. We constructed three counterfactual scenarios:

  • (1)

    NAAQS-35, full compliance with the previous annual PM2.5 standard (35 μg m−3).

  • (2)

    NAAQS-30, full compliance with the Phase I interim target of the revised NAAQS (30 μg m−3).

  • (3)

    NAAQS-25, full compliance with the post-2031 target of the revised NAAQS (25 μg m−3).

In each scenario, the counterfactual concentration (Cs,t′) for a given grid cell and year was defined as:

Cs,t′=min(Cs,t,Ctarget) (4)

where Cs,t is the observed concentration, and Ctarget represents the regulatory threshold (35, 30, or 25 μg m−3 under NAAQS-35, NAAQS-30, and NAAQS-25 scenarios, respectively). This definition implies that pollution levels in noncompliant areas are reduced strictly to the target level while areas already meeting the standard remain at their observed levels.

We defined potential health benefits as the number of avoided premature deaths under the counterfactual scenarios compared with the observed scenario. For each calendar year and disease, avoided deaths were calculated as:

ΔANt,j=∑s∑g[Ps,t,g×Bs,t,j,g×(AFj(Cs,t)−AFj(Cs,t′))] (5)

To quantify uncertainty, we conducted Monte Carlo simulations with 1000 draws by sampling from the GBD 2023 exposure–response function draws and disease-specific baseline mortality-rate uncertainty distributions, and recalculating all estimates under paired draws. The 2.5th and 97.5th percentiles of the resulting empirical distributions were used to define 95% uncertainty intervals.

2.4. Assessment of progress toward the WHO 2040 halving target

The WHO 2040 halving target aims to reduce the population-attributable fraction of mortality from anthropogenic air pollution by 50% by 2040 relative to a 2015 baseline. Given the focus of this study on China's revised PM2.5 standard, we assessed PM2.5-specific progress toward this target using the fraction of mortality attributable to long-term PM2.5 exposure. For disease j in year t, the total number of PM2.5-attributable deaths was obtained by aggregating grid- and age–sex-specific estimates across all spatial grids and applicable age–sex groups:

ANt,j=∑s∑gANs,t,j,g (6)

The corresponding PAF was defined as follows:

PAFt,j=ANt,jDt,j (7)

where Dt,j denotes the total number of deaths from disease j in year t. Under the observed scenario, we calculated PAF using observed PM2.5 concentrations. Under each counterfactual scenario (NAAQS-35, NAAQS-30, and NAAQS-25), we recalculated PAF using the corresponding counterfactual concentration defined in equation (4).

Using 2015 as the baseline year, PM2.5-specific progress toward the halving target was calculated as the reduction in PAF relative to the 50% reduction required by the target:

Gt,j=PAF2015,j−PAFt,j0.5×PAF2015,j×100% (8)

where Gt,j denotes the progress toward the halving target for disease j in year t. The same framework was applied at the provincial level by aggregating grid-level estimates in each province. All data preparation, statistical analyses, and visualizations were performed using R statistical software (version 4.3.0). Specifically, we used the packages rgdal, raster, and sf for spatial data processing and grid alignment, and the packages ggplot2 and tidyterra for visualization.

3. Results

3.1. Expansion of regulatory coverage

China's air quality improved substantially between 2013 and 2024, with the population-weighted PM2.5 concentration declining from 62.8 to 29.7 μg m−3 (Fig. 1). Correspondingly, the percentage of the population exposed to concentrations exceeding the NAAQS-35 standard declined markedly, from 83% to 33%. These data indicate that, as most of the population now resides in compliant areas under the former standard, the capability of the 35 μg m−3 threshold to drive further broad-scale exposure reduction is diminishing.

Fig. 1.

Fig. 1

Temporal trends in population-weighted PM2.5 concentrations and exposure distributions relative to air quality standards in China (2013–2024). The yellow line represents the annual national population-weighted average PM2.5 concentrations (right axis). Stacked bars show the percentage of the population exposed to different PM2.5 concentration intervals: ≤5, >5–25, >25–30, >30–35, and >35 μg m−3 (left axis). These thresholds of 5, 25, 30, and 35 μg m−3 correspond to the World Health Organization air quality guideline (WHO AQG), the NAAQS-25 target, the NAAQS-30 interim target, and the NAAQS-35 target. NAAQS, National Ambient Air Quality Standards.

In contrast, approximately 65% of the population in 2024 remained exposed to PM2.5 concentrations above the NAAQS-25 standard. NAAQS-25 therefore increases the proportion of the population exposed above the regulatory threshold from 33% to 65%, substantially expanding the scope for further exposure reduction. Given that compliance with the WHO air quality guideline (5 μg m−3) remained below 1%, the 25 μg m−3 limit represents a pragmatic intermediate milestone between current pollution levels and long-term global health goals.

3.2. Potential health benefits and disease-specific heterogeneity

The potential health benefits of alternative PM2.5 standards varied over time, as the marginal benefits of the former NAAQS-35 standard diminished with declining ambient PM2.5 concentrations (Fig. 2a). By 2024, nationwide compliance with NAAQS-35 would avert an estimated 53,985 premature deaths (95% uncertainty interval [UI]: 45,992–62,021), corresponding to a 3.7% reduction (95% UI: 3.3–4.1%) relative to the observed scenario. By contrast, compliance with the more stringent NAAQS-25 would avert an estimated 233,668 premature deaths (95% UI: 198,354–270,734), equivalent to a 16.1% reduction (95% UI: 14.3–17.7%). Thus, the NAAQS-25 scenario would yield approximately 4.3 times as many avoided deaths as the NAAQS-35 scenario in 2024. A sensitivity analysis using the GEMM NCD + LRI exposure–response function showed a similar pattern, with the NAAQS-25 scenario averting approximately 3.9 times as many deaths as the NAAQS-35 scenario in 2024 (Supplementary Fig. S2).

Fig. 2.

Fig. 2

Potential health benefits of tightening China's PM2.5 air quality standard (2013–2024). a, Temporal trends in total (summed across the seven modeled disease categories) and disease-specific PM2.5-attributable deaths under the observed scenario and three counterfactual scenarios of full compliance with alternative annual PM2.5 standards. Grey bars show the observed scenario, based on actual PM2.5 concentrations; light, medium, and dark blue markers represent NAAQS-35, NAAQS-30, and NAAQS-25, respectively, defined here as full compliance with annual mean PM2.5 limits of 35, 30, and 25 μg m−3. b, Corresponding relative reductions in estimated PM2.5-attributable mortality under each counterfactual scenario, expressed as a percentage reduction from the observed scenario. Error bars show 95% uncertainty intervals. NAAQS, National Ambient Air Quality Standards; COPD, chronic obstructive pulmonary disease.

Disease-specific analyses revealed substantial heterogeneity in the potential benefits (Fig. 2b). Under the NAAQS-25 scenario, mortality reductions in 2024 were estimated at 22.8% (95% UI: 22.0–23.4%) for COPD, 18.4% (17.3–19.5%) for stroke, 18.4% (17.6–19.2%) for lung cancer, 18.0% (17.0–19.1%) for LRI, and 16.0% (15.3–16.7%) for IHD. In contrast, the corresponding reductions were only 1.4% (1.2–1.6%) for dementia and 1.1% (1.0–1.2%) for T2DM. This sharp divergence underscores that while tighter standards are highly effective for cardiovascular and respiratory health, supplementary strategies may be needed to address the burden of diseases heavily influenced by population aging.

PM2.5-attributable incidence followed trends similar to those observed for mortality (Supplementary Fig. S3). Consistent with the mortality results, the protective effect of the former NAAQS-35 scenario showed clear signs of saturation after 2020, whereas stricter counterfactual standards produced progressively larger reductions in attributable incident cases. The disease composition of incidence differed from that of mortality. PM2.5-attributable incidence burden was dominated by LRI, COPD, and stroke in the earlier years, while dementia became increasingly prominent over time and ranked third by 2024. PM2.5-attributable incident dementia cases continued to rise despite declining PM2.5 exposure, increasing by 60.6% relative to 2013, suggesting that population aging partly offset the benefits of improved air quality for this outcome.

3.3. PM2.5-specific progress toward the WHO 2040 halving target

We further evaluated China's PM2.5-specific progress toward the halving target using the 2015 observed scenario as the reference (Fig. 3). At the national level, the total PM2.5-attributable fraction of mortality declined from 32.5% (95% UI: 25.6–37.6%) in 2015 to 22.6% in 2024, corresponding to a 30.6% relative reduction from the 2015 observed scenario. This indicates that substantial progress has already been achieved by the observed improvement in ambient PM2.5 concentrations.

Fig. 3.

Fig. 3

PM2.5-specific progress and disease-specific gaps toward the halving target in China. a, Total (aggregated across the seven modeled disease categories) and disease-specific PM2.5-attributable fractions of mortality in 2015 and 2024 under the observed scenario. Bars show point estimates, and error bars denote 95% uncertainty intervals. b, Relative reduction in PAF from the 2015 observed scenario. Green bars indicate the relative reduction already achieved under the 2024 observed scenario; light, medium, and dark blue markers represent NAAQS-35, NAAQS-30, and NAAQS-25, respectively, defined here as full compliance with annual mean PM2.5 limits of 35, 30, and 25 μg m−3. The red dashed line represents the halving target, defined as a 50% reduction in the PM2.5-attributable fraction of mortality relative to the 2015 observed scenario. PAF, population attributable fraction; NAAQS, National Ambient Air Quality Standards.

Counterfactual analyses showed that full compliance with progressively stricter annual PM2.5 standards in 2024 would further narrow the remaining gap. Under the NAAQS-35, NAAQS-30, and NAAQS-25 counterfactual scenarios, the relative reduction in the total PM2.5-attributable fraction compared with the 2015 observed scenario would increase to 33.1%, 36.3%, and 41.7%, respectively. These reductions correspond to 66.2%, 72.6%, and 83.4% of the progress required to meet the 50% halving target.

Regional analyses further showed heterogeneity in PM2.5-specific progress across regions (Supplementary Fig. S4). At the urban–rural level, the relative reduction in the PM2.5-attributable fraction of mortality from the 2015 observed scenario was slightly greater in urban areas than in rural areas under the 2024 observed scenario (29.5% versus 27.6%); under full compliance with NAAQS-25, these reductions would increase to 41.5% and 38.4%, respectively. Province-level differences were more pronounced, with progress under the 2024 observed scenario ranging from 15.1% in Xinjiang to 72.6% in Xizang. Provinces with greater observed progress were mainly in low-exposure western or coastal regions, whereas several northwestern and central provinces showed slower progress. Full compliance with NAAQS-25 would narrow the remaining distance to the halving target in most provinces, with particularly large gains in historically polluted and densely populated regions, such as Hebei, Tianjin, Beijing, Shandong, and Henan.

3.4. Disease-specific gaps to fulfilling the halving target

Disease-specific analyses revealed marked heterogeneity in PM2.5-specific progress toward the halving target (Fig. 3). Under the NAAQS-25 scenario, COPD showed the greatest progress. The relative reduction in the PM2.5-attributable fraction of mortality from the 2015 observed scenario increased from 39.9% under the 2024 observed scenario to 53.6%, exceeding the 50% halving target. Stroke, LRI, and lung cancer also showed substantial progress under NAAQS-25, with relative reductions reaching 44.7%, 44.4%, and 43.5%, respectively, while IHD reached 39.9%.

In contrast, dementia and T2DM showed limited progress. Their relative reductions from the 2015 observed scenario were only 11.5% and 7.4% under the 2024 observed scenario, and increased only modestly to 12.7% and 8.4%, respectively, under NAAQS-25. These findings indicate that stricter enforcement of PM2.5 standards can substantially advance PM2.5-specific progress for major cardiopulmonary outcomes, whereas neurodegenerative and metabolic conditions may account for much of the remaining gap.

4. Discussion

This study provides a comprehensive quantitative assessment of the potential health benefits associated with China's revised PM2.5 air quality standards. Using a comparative risk assessment framework, we found that full compliance with the NAAQS-25 threshold under 2024 exposure conditions could substantially reduce PM2.5-attributable mortality compared with the former NAAQS-35 threshold. The revised standard would also increase the proportion of the population exceeding the regulatory threshold and contribute to PM2.5-specific progress toward the halving target. These findings show the public health implications of China's updated regulatory framework and highlight the importance of health-oriented evaluations of air quality standards.

4.1. Alignment of the standard revision with domestic and global health targets

The revised NAAQS represents a significant update to air quality standards and is consistent with the domestic “Beautiful China Initiative.” It also marks China's transition in annual PM2.5 limits from the WHO Interim Target 1 (IT-1, 35 μg m−3) toward IT-2 (25 μg m−3). Over the past decade, as successive clean air action plans have been implemented and ambient PM2.5 concentrations have continued to decline, the additional health benefits achievable under the former NAAQS-35 threshold have gradually diminished. Our analysis estimates that the NAAQS-25 threshold would expand regulatory coverage by approximately twofold and increase potential health benefits by more than fourfold. Furthermore, full compliance with NAAQS-25 would reduce the PM2.5-attributable fraction of mortality by 41.7% relative to the 2015 observed scenario, corresponding to 83.4% of the progress required to meet a 50% reduction target. This indicates that the revised standard could make a substantial contribution to PM2.5-specific progress toward the halving target under a full-compliance counterfactual scenario. However, because the WHO target refers to mortality from anthropogenic air pollution more broadly, this estimate should be interpreted as a PM2.5-specific indicator rather than a complete assessment of the full target.

Regional analyses further illustrate how full compliance with the revised standard may shape PM2.5-specific health progress across China. In historically polluted and densely populated provinces, such as Henan, Hebei, and Shandong, full compliance with NAAQS-25 would substantially narrow the remaining distance to the halving target. In moderately polluted provinces, including Jiangsu, Anhui, and Hunan, NAAQS-25 would provide additional benefits beyond those achievable under the former NAAQS-35 threshold. Because these regions account for a large share of China's population and disease burden, achieving further PM2.5 reductions there will be important for sustaining national progress.

4.2. Disease heterogeneity in the health benefits of standard tightening

The potential health benefits of stricter air quality standards varied substantially across diseases, driven by differences in the shapes of disease-specific nonlinear exposure–response relationships. For IHD, stroke, COPD, lung cancer, and LRI, the exposure–response curves remain relatively steep within the 25–35 μg m−3 concentration range [30], [31]. Consequently, tightening the annual limit to 25 μg m−3 generates substantial marginal health gains, with reductions in PM2.5-attributable mortality for these conditions ranging from approximately 16% to 23%.

By contrast, the estimated potential mortality reductions for T2DM and dementia are much smaller, at 1.1% and 1.4%, respectively. For these diseases, the exposure–response relationships exhibit more pronounced sublinear patterns at policy-relevant concentrations [17], [32], [33]. The exposure–response curve for T2DM begins to flatten at approximately 25 μg m−3, whereas the dementia-related relative risk increases most rapidly below around 15 μg m−3 (Supplementary Fig. S1). As a result, reducing the standard from 35 to 25 μg m−3 yields only limited additional reductions in disease risk. However, dementia-related estimates should be interpreted cautiously because the evidence supporting the PM2.5–dementia association is weaker than that for the other disease outcomes assessed in this study.

From a policy prioritization perspective, cardiovascular and respiratory diseases account for a substantially larger share of China's overall disease burden [34], [35]. Therefore, tightening the air quality standard is likely to generate substantial population health gains at the national level. However, in regions with particularly high burdens of metabolic or neurodegenerative diseases, such as major urban agglomerations with elevated diabetes prevalence or rapidly aging regions with high dementia risk, additional PM2.5 reductions may need to be complemented by broader disease prevention and health protection strategies to achieve further health gains [36], [37], [38].

4.3. Challenges to realizing the potential health benefits of revised NAAQS

Despite the substantial potential health benefits expected from the revised NAAQS, sustaining the pace of improvement required to achieve the WHO halving target will face two major challenges: increasing abatement difficulty and rapid population aging. First, China's air pollution control has entered a more challenging phase, in which further emission reductions are becoming increasingly difficult, and the pace of concentration decline is slowing. As a result, both marginal abatement costs and the overall social costs of improving air quality are likely to rise relative to earlier stages of pollution control. Previous studies estimated that reducing national population-weighted PM2.5 exposure by 1 μg m−3 required an investment of approximately RMB 49 billion during 2013–2017, whereas the marginal cost increased to around RMB 100 billion per μg m−3 during 2018–2020 [39]. Rising abatement costs may therefore constrain the pace of achieving NAAQS-25. Looking ahead, advancing China's carbon peaking and carbon neutrality (“dual-carbon”) goals may provide an important pathway to mitigating these constraints [40], [41]. Structural transitions in energy and industrial systems could unlock substantial co-benefits for air pollution reduction and enable more efficient, targeted emission reductions.

Population aging represents another critical challenge [20], [42]. Rapid demographic aging influences the health benefits of air pollution control through two main pathways. First, an expanding elderly population increases the baseline mortality rates of age-related diseases, implying that the absolute number of PM2.5-attributable deaths may remain high even as exposure levels continue to decline. A previous decomposition study showed that population aging can substantially offset the health benefits of air quality improvement, reporting that population aging alone contributed approximately 342,000 additional PM2.5-related deaths in China during 2012–2017 [43]. Second, demographic aging increases the share of metabolic and neurodegenerative diseases in the overall disease burden. Because the health improvements associated with NAAQS-25 for these conditions are comparatively modest, this shift may slow the overall pace of population health improvement. In regions experiencing rapid population aging, additional measures to protect older adults from air pollution exposure, such as strengthened public health advisories and improved healthcare support, may therefore be needed to maximize the health gains from stricter PM2.5 standards.

4.4. Limitations

This study has several limitations. First, the counterfactual scenarios assumed full compliance with the specified PM2.5 standards, implying a uniform reduction of pollution levels to the regulatory threshold in all noncompliant areas. This idealized assumption simplifies complex implementation dynamics, including technical feasibility, temporal progression, and regional heterogeneity in the effectiveness of emission control measures. Consequently, the estimated health benefits should be interpreted as hypothetical full-compliance estimates rather than a precise forecast of near-term outcomes.

Second, the health impact assessment relied on exposure–response relationships derived primarily from large-scale epidemiological meta-analyses within the GBD framework. Although these functions incorporate evidence from Chinese cohort studies and represent the best available evidence for global assessments, they may not fully capture regional differences in population susceptibility or healthcare access within China. In addition, although we used GEMM to assess the robustness of relative differences in total PM2.5-attributable mortality estimates across counterfactual scenarios, alternative exposure–response functions suitable for disease-specific sensitivity analyses remain unavailable.

Third, uncertainty analysis incorporated uncertainty from the disease-specific exposure–response functions and baseline mortality rates, while exposure concentrations, population estimates, and the TMREL were treated as fixed inputs. Therefore, the reported uncertainty intervals may not capture the full range of uncertainty.

Finally, the analysis focused exclusively on PM2.5 and did not account for potential co-benefits or trade-offs associated with simultaneous changes in gaseous co-pollutants, such as ozone, or climate-related factors, which may further influence population health outcomes.

4.5. Conclusion

This study highlights the substantial health gains potentially achievable under full compliance with the revised PM2.5 standards while revealing important disease heterogeneity in these benefits. The NAAQS-25 scenario could substantially advance PM2.5-specific progress toward the WHO 2040 halving target. Future air pollution governance may require moving beyond a purely compliance-oriented framework toward a more health-oriented approach that maximizes the public health benefits of air quality standards.

CRediT authorship contribution statement

Ruohan Wang: Writing – original draft, Methodology, Investigation, Formal analysis, Data curation. Ning Kang: Writing – review & editing, Data curation. Yixuan Zheng: Writing – review & editing, Supervision, Funding acquisition. Tao Xue: Writing – review & editing, Supervision, Funding acquisition.

Data and code availability statement

The analysis code and non-restricted processed data generated in this study have been deposited in Zenodo: https://doi.org/10.5281/zenodo.21877076. The PM2.5 concentration data used in this study were obtained from the Tracking Air Pollution in China (TAP) dataset (http://tapdata.org.cn/). Gridded population data were obtained from WorldPop (https://www.worldpop.org/) and constrained using county-level population statistics from the China Statistical Yearbook. Baseline mortality rates were estimated using data from the National Disease Surveillance Program, while cause-of-death profiles and incidence rates were obtained from the Global Burden of Disease Study 2023 through the GBD Results Tool (https://vizhub.healthdata.org/gbd-results/). Exposure–response functions were obtained from the GBD 2023 air pollution risk curve release (https://ghdx.healthdata.org/record/ihme-data/gbd-2023-air-pollution-exposure-estimates-1990-2023). Raw third-party datasets were not redistributed because they are subject to the terms of use of the original data providers.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (42422507, 42375191), the Beijing Natural Science Foundation (JQ24052), and the Science and Technology Projects of Xizang Autonomous Region, China (XZ202501ZY0072).

Footnotes

Appendix A

Supplementary data to this article can be found online at https://doi.org/10.1016/j.ese.2026.100781.

Contributor Information

Yixuan Zheng, Email: zhengyx@caep.org.cn.

Tao Xue, Email: txue@hsc.pku.edu.cn.

Appendix A. Supplementary data

The following is the Supplementary data to this article:

Multimedia component 1
mmc1.docx (638KB, docx)

References

  • 1.Hay S.I., Ong K.L., Santomauro D.F., et al. Burden of 375 diseases and injuries, risk-attributable burden of 88 risk factors, and healthy life expectancy in 204 countries and territories, including 660 subnational locations, 1990-2023: a systematic analysis for the global burden of disease study 2023. Lancet. 2025;406(10513):1873–1922. doi: 10.1016/S0140-6736(25)01637-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.U.S. Environmental Protection Agency . 2019. Integrated Science Assessment (ISA) for Particulate Matter. [PubMed] [Google Scholar]
  • 3.Yang Y., Deng J., Zhou P., et al. Short-term PM2.5 exposure and cardiovascular mortality: a global exposure-response analysis to inform alert thresholds. JACC (J. Am. Coll. Cardiol.) 2026;88(9):914–930. doi: 10.1016/j.jacc.2025.11.055. [DOI] [PubMed] [Google Scholar]
  • 4.Lu H., Kang N., Tong M., Xue T., Liu J. Explanatory machine learning model to associate all-cause mortality with PM2.5 by sources and components: a nationwide difference-in-differences study. Environ. Sci. Technol. 2026;60(16):12109–12121. doi: 10.1021/acs.est.6c00976. [DOI] [PubMed] [Google Scholar]
  • 5.Xue T., Wang R., Wang M., et al. Health benefits from the rapid reduction in ambient exposure to air pollutants after China's clean air actions: progress in efficacy and geographic equality. Natl. Sci. Rev. 2024;11(2) doi: 10.1093/nsr/nwad263. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Xue T., Zhu T., Peng W., et al. Clean air actions in China, PM2.5 exposure, and household medical expenditures: a quasi-experimental study. PLoS Med. 2021;18(1) doi: 10.1371/journal.pmed.1003480. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Wang R., Lu H., Kang N., et al. Estimating infant and adult mortality burden attributable to fine particulate matter in China from 2013 to 2023: an environmental justice perspective. Environ. Sci. Technol. 2026;60(26):18456–18465. doi: 10.1021/acs.est.5c18497. [DOI] [PubMed] [Google Scholar]
  • 8.World Health Organization . 2021. WHO Global Air Quality Guidelines: Particulate Matter (PM2.5 and PM10), ozone, nitrogen dioxide, sulfur dioxide and carbon monoxide. [PubMed] [Google Scholar]
  • 9.U.S. Environmental Protection Agency . 2024. National Ambient Air Quality Standards. [Google Scholar]
  • 10.European Union . 2024. Directive (EU) 2024/2881 of the European Parliament and of the Council of 23 October 2024 on Ambient Air Quality and Cleaner Air for Europe. [Google Scholar]
  • 11.Ministry of the Environment Government of Japan . 2009. Environmental Quality Standards in Japan: Air Quality. [Google Scholar]
  • 12.World Health Organization . 2025. Second Global Conference on Air Pollution and Health. [Google Scholar]
  • 13.World Health Organization . 2025. Updated Road Map for an Enhanced Global Response to the Adverse Health Effects of Air Pollution: Report by the Director-General. [Google Scholar]
  • 14.Du P., Du H., Liu Y., Wang S., Li T. Targeting 50% air-pollution mortality reduction by 2040 requires transformative, health-oriented precision emission reduction. Innovation. 2026;7(3) doi: 10.1016/j.xinn.2026.101276. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Kang N., Wang R., Li P., et al. Beyond the halving commitment: China can do more to achieve clean air for global health. Innovation. 2025;6(10) [Google Scholar]
  • 16.Zheng P., Afshin A., Biryukov S., et al. The burden of proof studies: assessing the evidence of risk. Nat. Med. 2022;28(10):2038–2044. doi: 10.1038/s41591-022-01973-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Huang X., Steinmetz J., Marsh E.K., et al. A systematic review with a burden of proof meta-analysis of health effects of long-term ambient fine particulate matter (PM2.5) exposure on dementia. Nat. Aging. 2025;5(5):897–908. doi: 10.1038/s43587-025-00844-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Li P., Ni X., Wu J., Guo C., Xue T., Zhang L. Global and national burden of chronic kidney disease attributable to fine particles: an updated exposure-response meta-analysis. Innovation. 2026;7(8) doi: 10.1016/j.xinn.2026.101429. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Kang N., Deng J., Li P., et al. Trends of PM2·5-attributable dementia deaths in the context of population ageing in China, 2000–60: a health impact assessment. Lancet Healthy Longev. 2026;7(4) doi: 10.1016/j.lanhl.2026.100841. [DOI] [PubMed] [Google Scholar]
  • 20.Xu F., Huang Q., Yue H., et al. The challenge of population aging for mitigating deaths from PM2.5 air pollution in China. Nat. Commun. 2023;14(1):5222. doi: 10.1038/s41467-023-40908-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Xiao Q., Geng G., Liu S., Liu J., Meng X., Zhang Q. Spatiotemporal continuous estimates of daily 1 km PM2.5 from 2000 to present under the tracking air pollution in China (TAP) framework. Atmos. Chem. Phys. 2022;22(19):13229–13242. [Google Scholar]
  • 22.Geng G., Xiao Q., Liu S., et al. Tracking air pollution in China: near real-time PM2.5 retrievals from multisource data fusion. Environ. Sci. Technol. 2021;55(17):12106–12115. doi: 10.1021/acs.est.1c01863. [DOI] [PubMed] [Google Scholar]
  • 23.Xiao Q., Geng G., Xue T., et al. Tracking PM2.5 and O3 pollution and the related health burden in China 2013-2020. Environ. Sci. Technol. 2022;56(11):6922–6932. doi: 10.1021/acs.est.1c04548. [DOI] [PubMed] [Google Scholar]
  • 24.Xue T., Liu J., Zhang Q., et al. Rapid improvement of PM2.5 pollution and associated health benefits in China during 2013–2017. Sci. China Earth Sci. 2019;62(12):1847–1856. [Google Scholar]
  • 25.Zheng Y., Xue T., Zhang Q., et al. Air quality improvements and health benefits from China's clean air action since 2013. Environ. Res. Lett. 2017;12(11) [Google Scholar]
  • 26.Tatem A.J. WorldPop, open data for spatial demography. Sci. Data. 2017;4(1) doi: 10.1038/sdata.2017.4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Zheng P., Barber R., Sorensen R.J.D., Murray C.J.L., Aravkin A.Y. Trimmed constrained mixed effects models: formulations and algorithms. J. Comput. Graph Stat. 2021;30(3):544–556. [Google Scholar]
  • 28.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. 2018;115(38):9592–9597. doi: 10.1073/pnas.1803222115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Schumacher A.E., Zheng P., Barber R.M., et al. Global age-sex-specific all-cause mortality and life expectancy estimates for 204 countries and territories and 660 subnational locations, 1950-2023: a demographic analysis for the global burden of disease study 2023. Lancet. 2025;406(10513):1731–1810. doi: 10.1016/S0140-6736(25)01330-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Huang W., Zhou Y., Chen X., et al. Individual and joint associations of long-term exposure to air pollutants and cardiopulmonary mortality: a 22-year cohort study in northern China. Lancet Reg. Health West. Pac. 2023;36 doi: 10.1016/j.lanwpc.2023.100776. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Tang Z., Ku P.-W., Xia Y., Chen L.-J., Zhang Y. Preexisting multimorbidity predicts greater mortality risks related to long-term PM2.5 exposure. Environ. Pollut. 2025;368 doi: 10.1016/j.envpol.2025.125762. [DOI] [PubMed] [Google Scholar]
  • 32.Rajagopalan S., Brook R.D., Salerno P., et al. Air pollution exposure and cardiometabolic risk. Lancet Diabetes Endocrinol. 2024;12(3):196–208. doi: 10.1016/S2213-8587(23)00361-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Bowe B., Xie Y., Li T., Yan Y., Xian H., Al-Aly Z. The 2016 global and national burden of diabetes mellitus attributable to PM2.5 air pollution. Lancet Planet. Health. 2018;2(7):e301–e312. doi: 10.1016/S2542-5196(18)30140-2. [DOI] [PubMed] [Google Scholar]
  • 34.Zhou M., Wang H., Zeng X., et al. Mortality, morbidity, and risk factors in China and its provinces, 1990-2017: a systematic analysis for the global burden of disease study 2017. Lancet. 2019;394(10204):1145–1158. doi: 10.1016/S0140-6736(19)30427-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Zhang T., Jiang H., Xu X., Zhao Z., Zhou M. Non-communicable disease burden in China, 1990-2023: evidence from the global burden of disease study 2023. Chin. Med. J. 2026;139(1):48–57. doi: 10.1097/CM9.0000000000003898. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Liang J., Jia R., Wang Y., Qian S., Lin L., Xu Y. Prevalence, trends and distribution of dementia among older adults from 1990 to 2023 in China: a comprehensive systematic review and meta-analysis. Int. J. Nurs. Stud. 2026;173 doi: 10.1016/j.ijnurstu.2025.105247. [DOI] [PubMed] [Google Scholar]
  • 37.Liu Y., Gao X., Zhang Y., et al. Geographical variation in dementia prevalence across China: a geospatial analysis. Lancet Reg. Health West. Pac. 2024;47 doi: 10.1016/j.lanwpc.2024.101117. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Zhou Y., Liu J., Zhao Z., Zhou M., Ng M. The national and provincial prevalence and non-fatal burdens of diabetes in China from 2005 to 2023 with projections of prevalence to 2050. Milit. Med. Res. 2025;12(1):28. doi: 10.1186/s40779-025-00615-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Geng G., Liu Y., Liu Y., et al. Efficacy of China's clean air actions to tackle PM2.5 pollution between 2013 and 2020. Nat. Geosci. 2024;17(10):987–994. [Google Scholar]
  • 40.Zhang Q., Yin Z., Lu X., et al. Synergetic roadmap of carbon neutrality and clean air for China. Environ. Sci. Ecotechnol. 2023;16 doi: 10.1016/j.ese.2023.100280. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Cheng J., Tong D., Liu Y., et al. A synergistic approach to air pollution control and carbon neutrality in China can avoid millions of premature deaths annually by 2060. One Earth. 2023;6(8):978–989. [Google Scholar]
  • 42.Livingston G., Huntley J., Sommerlad A., et al. Dementia prevention, intervention, and care: 2020 report of the lancet commission. Lancet. 2020;396(10248):413–446. doi: 10.1016/S0140-6736(20)30367-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Geng G., Zheng Y., Zhang Q., et al. Drivers of PM2.5 air pollution deaths in China 2002–2017. Nat. Geosci. 2021;14(9):645–650. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Multimedia component 1
mmc1.docx (638KB, docx)

Data Availability Statement

The analysis code and non-restricted processed data generated in this study have been deposited in Zenodo: https://doi.org/10.5281/zenodo.21877076. The PM2.5 concentration data used in this study were obtained from the Tracking Air Pollution in China (TAP) dataset (http://tapdata.org.cn/). Gridded population data were obtained from WorldPop (https://www.worldpop.org/) and constrained using county-level population statistics from the China Statistical Yearbook. Baseline mortality rates were estimated using data from the National Disease Surveillance Program, while cause-of-death profiles and incidence rates were obtained from the Global Burden of Disease Study 2023 through the GBD Results Tool (https://vizhub.healthdata.org/gbd-results/). Exposure–response functions were obtained from the GBD 2023 air pollution risk curve release (https://ghdx.healthdata.org/record/ihme-data/gbd-2023-air-pollution-exposure-estimates-1990-2023). Raw third-party datasets were not redistributed because they are subject to the terms of use of the original data providers.


Articles from Environmental Science and Ecotechnology are provided here courtesy of Elsevier

RESOURCES