Harm to human health from air pollution in Europe: burden of disease status, 2025

This briefing provides the latest estimates of population health impacts caused by long-term exposure to fine particulate matter, nitrogen dioxide and ozone. It is based on data up to the year 2023.

Key messages

While significant progress has been made to reduce air pollution, 95% of urban Europeans remain exposed to air pollutant concentrations above World Health Organization (WHO) recommendations. 

Reducing air pollution to WHO guideline levels could have prevented 182,000 deaths attributable to fine particulate matter (PM_2.5) exposure, 63,000 to ozone (O₃) exposure and 34,000 to nitrogen dioxide (NO₂) exposure in the EU in 2023.

For some diseases caused and/or aggravated by air pollution, such as asthma, the main impact is poorer health. For others, such as ischemic heart disease and lung cancer, it is premature death.

New evidence suggests that air pollution may also cause dementia. Dementia’s disease burden is estimated to be higher than that of other relevant diseases.

Eastern and south-eastern European countries suffer the most significant health impacts from air pollution due to high pollution levels.

This briefing is the 2025 update of the burden of disease series. It is also part of the air quality in Europe package, which includes briefings on air quality status and air pollution in Europe.

Box 1. Measuring impacts on health: understanding the ‘environmental burden of disease’ concept

‘Burden of disease’ describes the impact of a disease on the health of a population. The burden of disease can generally be measured using four indicators including mortality and morbidity (see Figure 1):

Morbidity Indicator — years lived with disability (YLD), which quantifies the number of years a population has lived in reduced health due to a particular health outcome. YLD is estimated by multiplying the number of prevalent cases of a particular health outcome with a disability weight – a factor indicating the severity of the health outcome on a scale from 0 (full health) to 1 (most severe health state).
Mortality Indicator — number of attributable deaths (AD) that have occurred because of a specific disease or group of diseases and is attributable to a certain risk.
Mortality Indicator — years of life lost (YLL), defined as the number of years of potential life lost due to death caused by a disease or group of diseases. YLL is an estimate of the average number of additional years that people in a population could have statistically lived if they had not died before reaching a certain statistical life expectancy. Mortality data at the national level are the baseline input used to estimate AD and YLL.
Combined Mortality and Morbidity Indicator — disability-adjusted life years (DALY). A DALY corresponds to one lost year of a healthy life due to disease or injury. DALY are obtained by adding YLL and YLD for the same disease or group of diseases. Therefore, DALY is a standardised indicator for health effects resulting from both the number of people affected by a disease and the number of people dying from it.

These indicators are used to estimate a share attributable to a certain risk, resulting in attributable deaths (AD) and attributable YLL, YLD and DALY.

The ‘environmental burden of disease’ concept quantifies the shares of the total burden of disease that can be statistically attributed to environmental risk factors, such as the population’s exposure to air pollution. The attribution is generally based on evidence of a causal link between a risk factor and a health outcome. The attributable burden is considered preventable if its cause can be eliminated or reduced.

The ‘attributable number of deaths per 100,000 inhabitants at risk’ and ‘YLL per 100,000 inhabitants at risk’ calculations are also used in this briefing as rate indicators that allow comparisons across countries. ‘At risk’ means the population above certain age considered in the associated epidemiological studies. These rates are provided in the corresponding tables.

You can find additional information on how air pollution is related to mortality in EEA’s web article Why do we not see ‘air pollution’ on death certificates, if it causes premature mortality?

Figure 1. Burden of disease as a sum of YLD and YLL

Please select a resource that has a preview image available.

Estimating the burden of disease due to air pollution

Geographical coverage

The analysis in this briefing refers to 41 countries. These are the 27 Member States of the European Union (EU-27), other EEA member and cooperating countries and additional European microstates. Türkiye is not included in the PM_2.5 estimations as the number of background monitoring stations from which data are available was too low to produce concentration maps for fine particulate matter. Consequently, PM_2.5 estimations were made for 40 countries. 

Concentration-response functions

This briefing estimates the health impacts — both mortality and morbidity — attributable to air pollution exposure. Concentration-response functions are an essential component to attribute disease burden to air pollution exposure. They are based on epidemiological studies and estimate the increase in risk for a particular health outcome per unit increase of concentration of a certain air pollutant.

All-cause natural mortality (referred hereafter to as all-cause mortality) represents an estimate of the total mortality due to all potential natural causes linked to a specific pollutant. It does not disaggregate mortality between different diseases. The EEA uses the concentration-response functions for all-cause mortality as set out in the WHO’s 2021 Global Air Quality Guidelines.

For cause-specific mortality and morbidity, attributable deaths, YLL, YLD and DALY were all estimated for six pollutant-disease pairs for PM_2.5 and five pairs for NO₂. Scientific evidence of an association between exposure to a specific air pollutant and the disease is robust for these pairs and enough data are available for estimations. The concentration-response functions applied in these cases can be found in the ETC/HE Report 2025/8 (Soares et al, 2025). For PM_2.5, mortality and morbidity were estimated for:

asthma in children/adolescents (<19 years); 
chronic obstructive pulmonary disease (in adults ≥25 years);
diabetes mellitus (in adults ≥25 years);
ischemic heart disease (in adults ≥25 years); 
lung cancer (in adults ≥25 years); 
stroke (in adults ≥25 years).

Furthermore, morbidity from dementia (in adults ≥60 years) attributable to exposure to PM_2.5 was also estimated for the first time (currently, mortality from dementia is not included due to the lack of a specific concentration-response function).

For NO₂, mortality and morbidity were estimated for:

asthma in children/adolescents (<19 years); 
asthma (in adults ≥19 years); 
chronic obstructive pulmonary disease (in adults ≥25 years)
diabetes mellitus (in adults ≥25 years);
stroke (in adults ≥25 years).

For asthma, previous analyses considered children (<15 years) and adults (≥15 years). This year, adolescents were moved from the adults group (defined now as ≥19 years) into the children/adolescents group (<19 years). The selected concentration-response functions for children and adults are derived from these specific age groups (<19 and ≥19 years), ensuring greater consistency. For diabetes mellitus, the effect measure for morbidity refers only to type 2 diabetes mellitus. Mortality also includes type 1 diabetes. 

For O₃, no morbidity-related burden of disease was estimated as the scientific evidence points more to an association between long-term O₃ exposure and respiratory mortality. Therefore, only mortality for chronic obstructive pulmonary disease (in adults ≥25 years) was estimated.

The uncertainties

One main source of uncertainty comes from relative risks. This uncertainty is expressed as confidence intervals (CIs). The upper and lower boundaries of the 95% CI mark the interval in which there is 95% confidence in finding the real value. The main text does not include these 95% CI values for readability reasons, but they are provided for all estimations in the corresponding tables, in addition to each mean value.  

Other possible uncertainties (for instance, the one related to exposure assessments) are not quantified.

Counterfactual concentrations and sensitivity analyses

A counterfactual concentration is one for which the hypothetical exposure is associated with a lower risk or ‘no-risk’ level. This is the concentration against which the estimated population exposure is compared and above which impacts are estimated. The main analysis presented in this briefing only calculates the health impact of concentrations above the WHO’s 2021 air quality guideline levels for long-term exposure to PM_2.5 (annual mean of five micrograms per cubic metre of air (μg/m³)), NO₂ (annual mean of 10μg/m³) and O₃ (peak season value of 60μg/m³). In other words, the results indicate the attributable impacts that could have been avoided if countries had achieved WHO guideline values in 2023. 

There is a higher level of uncertainty around concentration-response for lower concentrations. However, there is no scientific evidence of a threshold below which these air pollutants do not impact health. Therefore, a sensitivity analysis of attributable deaths due to all-cause mortality considering the full exposure range for PM_2.5 and NO₂(i.e. for counterfactual concentrations of 0µg/m³) is also presented. This assumes the uncertainty around these estimates is higher. For O₃, additional scenarios were not included due to the absence of any meaningful counterfactual concentration besides 60µg/m³.

The revised Ambient Air Quality Directive (AAQD) establishes updated EU air quality standards for these pollutants to be attained by 2030. Therefore, a second sensitivity analysis assessed the number of deaths due to all-cause mortality attributable to exposure above the 2030 annual limit values. These are annual PM_2.5 concentrations above 10µg/m³ and NO₂ concentrations above 20µg/m³. O₃ is not included in this sensitivity analysis because there is no EU air quality standard linked to the peak season.

Comparison with other estimations

Estimates presented here may differ from others calculated at national, European and global levels. This is due to differences in the estimation process, which include: the use of different concentration maps, where air pollutant concentrations are estimated at different scales; the use of more detailed demographic data; the use of different relative risks or different forms of concentration-response functions; and the use of different assumptions in the range of concentrations considered (the counterfactual concentrations).

Final consideration

Estimations for the environmental burden of disease are made individually for the three pollutants and therefore are presented separately. They cannot be combined as they exhibit a degree of correlation. This is especially the case for the burden of disease attributable to PM_2.5 and NO₂, for which the WHO estimated a possible overlap and double counting of mortality effects of 30%. In other words, some deaths attributable to PM_2.5 exposure are also partly attributable to NO₂ exposure. Therefore, it is impossible to sum the deaths attributable to these two pollutants without risking an overestimation of the total number of attributable deaths due to double counting. 

The source of all tables and figures is Burden of disease from air pollution (EEA).

Additional information, including the burden of disease for individual countries, can be found in the Eionet ETC/HE Report 2025/8 (Soares et al, 2025).

Policy context 

The EU set air quality standards for key air pollutants several decades ago, as outlined in the AAQ directives. These standards have been recently updated in the AAQD (2024/2881), adopted in October 2024. The revised standards, to be achieved by 2030, are aligned more closely with the 2021 WHO air quality guidelines. 

Under the European Green Deal, the zero pollution action plan outlines a target of improving air quality by 2030, with a focus on PM_2.5. It aims to reduce the number of premature deaths attributable to air pollution in the EU by a minimum of 55%, relative to those attributable in 2005.

How air pollution harmed human health in Europe in 2023

To estimate the total harm from air pollution on health, the ‘environmental burden of disease concept’ is used as described above. This section presents the individual estimates of the environmental burden of disease for several specific health conditions, considering all countries included in the analysis.  

The impacts of air pollution were calculated for each of the three considered pollutants in two ways. First, ‘all-cause mortality’ estimated the total number of deaths attributable to air pollution without providing any breakdown of the individual diseases associated with these deaths. Second, ‘cause-specific mortality and morbidity’ were calculated to determine the deaths and health effects associated with individual diseases. This means that both mortality (deaths due to a specific disease, expressed as AD and YLL) and morbidity (the state of having a disease or disability, expressed as YLD) were estimated. YLL and YLD were then combined into the DALY. 

The sections below therefore present mortality figures from all natural causes (i.e. excluding accidental and other non-natural causes) and mortality and/or morbidity figures for specific causes attributable to long-term exposure to PM_2.5, NO₂ and O₃.

Estimations for the environmental burden of disease were made individually for the respective air pollutants and are presented separately. They cannot be added together as they exhibit a degree of correlation that could lead to an overestimate.

Fine particulate matter (PM_2.5)

In 2023, deaths due to all-cause mortality were as follows: 

In the EU-27, 182,000 deaths were attributable to exposure to PM_2.5 concentrations above the WHO guideline level of 5µg/m³. This resulted in 1,914,000 YLL.
In all 40 countries considered for PM_2.5, 206,000 deaths were attributable to exposure above the WHO guideline level of 5µg/m³. This resulted in 2,160,000 YLL.

Table 1 provides confidence intervals and rates per 100,000 inhabitants aged 30 years and above.

As shown in the EEA’s indicator Premature deaths due to exposure to fine particulate matter in Europe,  premature deaths attributable to PM_2.5 exposure above the WHO guideline level of 5µg/m³ fell by 57% in the EU between 2005 and 2023. This indicates that the zero pollution action plan’s target was achieved for 2023. 

The highest absolute numbers of attributable deaths due to all-cause mortality in 2023 occurred in Italy, Poland and Germany (in decreasing order) (see Table 1 and Map 1). However, the highest relative impacts (YLL per 100,000 inhabitants 30 years and above) were observed in south-eastern European countries, both for the analysis of all countries (North Macedonia, Bosnia and Herzegovina and Albania, in decreasing order) and when considering only EU-27 countries (Bulgaria, Greece and Romania, in decreasing order). 

The lowest relative impacts (YLL per 100,000 inhabitants 30 years and above) due to PM_2.5 exposure occurred in countries located in the north and north-west of Europe, including Iceland, Finland, Sweden, Estonia and Norway (in order of increasing burden of disease).

Map 1. Mortality attributable to long-term exposure to PM_2.5, 2023

For the seven studied diseases, the total burden of disease (DALY, including its rate) and corresponding mortality (both as AD and YLL) and morbidity (YLD) estimations are presented in table 2 (for dementia, only YLD are presented, see above), ordered by decreasing number of DALY. 

The highest attributable burden of disease results from ischemic heart disease (IHD), with a total of 579,409 DALY for the EU-27 and 637,073 DALY for the 40 countries considered. In this case, mortality effects (YLL) represent by far most of the burden of disease (97.2% and 96.9% for the EU-27 and 40 countries, respectively).  

This is also the case for lung cancer (99.0% and 99.1%), stroke (81.7% and 76.7%), diabetes mellitus (76.5% and 76.7%) and COPD (62.8% and 63.2%). In contrast, morbidity represents the main burden of disease for asthma in children/adolescents (99.2% in both cases).

Table 2. Burden of disease for individual diseases attributable to PM_2.5, 2023 

EU-27

YLL

YLD

DALY

DALY rate

IHD

58,720

(31,622-83,808)

563,314

(303,315-804,090)

16,095

(6,614-25,421)

579,409

(309,929-829,511)

175.3

(93.8-251.0)

Stroke

44,622

(18,440-67,825)

401,507

(165,806-610,681)

89,956

(10,014-168,018)

491,463

(175,820-778,700)

148.7

(53.2-235.6)

Diabetes

30,697

(16,673-42,124)

299,123

(162,459-410,453)

91,755

(47,431-133,287)

390,878

(209,891-543,740)

118.3

(63.5-164.5)

Lung cancer

22,340

(8,526-34,886)

331,630

(126,485-518,202)

3,201

(1,349-5,104)

334,831

(127,834-523,307)

101.3

(38.7-158.3)

COPD

15,286

(271-28,001)

154,038

(2,719-283,417)

91,062

(36,118-138,762)

245,101

(38,838-422,180)

74.2

(11.7-127.7)

Asthma in children/adolescents

(1-5)

268

(95-420)

32,997

(11,931-50,945)

33,266

(12,027-51,366)

39.4

(14.3-60.9)

Dementia

293,195

(151,603-417,445)

Total (six diseases, without dementia)

171,668

(75,532-256,649)

1,749,880

(760,879-2,627,263)

325,066

(113,457-521,537)

2,074,948

(874,339-3,148,804)

40 countries considered 

YLL

YLD 

DALY 

DALY rate 

IHD

64,332

(34,721-91,625)

617,186

(333,078-879,127)

19,887

(8,204-31,290)

637,073

(341,282-910,418)

179.9

(96.4-257.1)

Stroke

50,247

(20,850-76,092)

452,106

(187,487-685,038)

97,263

(10,861-181,118)

549,369

(198,349-866,156)

155.1

(56.0-244.6)

Diabetes

34,333

(18,780-46,835)

333,405

(182,326-454,889)

101,318

(52,426-147,044)

434,723

(234,753-601,933)

122.7

(66.3-170.0)

Lung cancer

25,108

(9,618-39,075)

371,451

(142,187-578,506)

3,536

(1,496-5,618)

374,988

(143,684-584,125)

105.9

(40.6-164.9)

COPD

16,829

(300-30,676)

170,148

(3,023-311,363)

99,002

(39,427-150,335)

269,150

(42,451-461,699)

76.0

(12.0-130.4)

Asthma in children/adolescents

(1-5)

278

(99-436)

35,404

(12,838-54,528)

35,682

(12,938-54,965)

39.4

(14.3-60.6)

Dementia

315,390

(163,573-447,862)

Total (six diseases, without dementia)

190,852

(84,269-284,308)

1,944,574

(848,200-2,909,359)

356,410

(125,252-569,933)

2,300,985

(973,457-3,479,296)

Notes: Parentheses indicate the 95% confidence interval. IHD, ischemic heart disease. Diabetes, type 1 and 2 diabetes mellitus. COPD, chronic obstructive pulmonary disease. AD, attributable deaths. YLL, years of life lost. YLD, years lived with disability. DALY, disability-adjusted life years. DALY rate, DALY/100,000 inhabitants at risk.

Maps 2 and 3 show the burden of disease at national level for two of the analysed diseases: IHD, with the highest DALY value; and COPD, a respiratory disease with a high burden ofdisease due to morbidity and therefore impact on overall health. Both maps show the differences across the 40 countries considered.

Map 2. Burden of disease for ischemic heart disease attributable to long-term exposure to PM_2.5, 2023 

Map 3. Burden of disease for COPD attributable to long-term exposure to PM_2.5, 2023

The total number of deaths due to the six diseases (dementia is not included) attributable to exposure to PM_2.5 is 172,000 for the EU-27 and 191,000 for the 40 considered countries (Table 2). These numbers are lower than the deaths attributable to all natural causes, due to deaths related to these and other diseases.  

The EEA indicator on premature deaths due to exposure to fine particulate matter in Europe indicates a decreasing trend in the burden of disease attributable to PM_2.5 in Europe that continued between 2022 and 2023.  The continuation is partially due to a sharp decrease in the population’s exposure to the pollutant, with the population-weighted concentration dropping from 11.4µg/m³ in 2022 to 10.2µg/m³ in 2023. This represents a decrease of 1.2µg/m³ or 10.5%. It is important however to note an interannual variation in PM_2.5 levels, as observed in the indicator's trend. 

This decrease is a result of EU, national and local policy implementation to improve air quality (e.g. the EU AAQD and the plans and measures derived from them) and to reduce emissions of air pollutants (e.g. the National Emission Reduction Commitments Directive). These policies contributed to reductions in fine particulate matter emissions by 38% between 2005 and 2023. These reductions were especially important in electricity and heat production, residential fuel use and road transport. Emissions from agriculture of ammonia, a secondary PM precursor, have also fallen to a lesser extent.

Sensitivity analyses of deaths attributable to PM_2.5

If all concentrations of PM_2.5 are considered (above 0µg/m³) rather than only those above the 2021 WHO recommendation (5µg/m³), the potential impact would be as follows (see also Table 3):

An estimated 164,000 potential additional attributable deaths in the EU-27, resulting in 346,000 total attributable deaths overall. This is equivalent to 3,625,000 YLL. 
For the 40 considered countries, an estimated 176,000 potential additional attributable deaths, resulting in 382,000 total attributable deaths overall. This is equivalent to 3,996,000 YLL.

Finally, if only the impact above the annual limit value (ALV) in the revised AAQD 2024/2881 to be achieved by 2030 (10µg/m³) was considered (see also Table 4): 

for the EU-27, 55,000 deaths would be attributable to concentrations above the revised ALV, equivalent to 577,000 YLL; 
for the 40 countries, 69,000 deaths would be attributable to concentrations above the revised ALV, equivalent to 720,000 YLL.

  Estimations of these sensitivity analyses at country level can be found at Table publisher.

Nitrogen dioxide (NO₂)

In 2023, deaths due to all-cause mortality were as follows: 

In EU-27, 34,000 deaths were attributable to exposure to NO₂ concentrations above the WHO guideline level of 10µg/m³, equivalent to 351,000 YLL.
In the 41 countries considered, 56,000 deaths were attributable to exposure to NO₂ concentrations above the WHO guideline level of 10µg/m³, equivalent to 641,000 YLL.

Table 5 provides confidence intervals and the rates per 100,00 inhabitants aged 30 years and above. 

Map 4. Mortality attributable to long-term exposure to NO₂, 2023

The highest absolute numbers of attributable deaths in 2023 occurred in Türkiye, Italy and Germany (in decreasing order). However, the highest relative impacts (YLL per 100,000 inhabitants 30 years and above) were observed in Türkiye, Cyprus, Greece, Serbia and Italy (in decreasing order) (see Table 5 and Map 4). 

The lowest relative impacts (YLL per 100,000 inhabitants 30 years and above) due to NO₂ exposure occurred in countries located in the north and north-west of Europe, including Iceland, Sweden, Finland, Estonia and Denmark (in order of increasing impact). 

The number of deaths since 2005 and between 2022 and 2023 attributable to exposure to NO₂, fell both for the EU-27 and all considered countries. This fall was linked to a decrease in population exposure from a population-weighted concentration of 14µg/m³ in 2022 to 12.5µg/m³ in 2023 for the EU-27 (-10.7%) and from 16.3µg/m³ to 15.3µg/m³ (-6.1%) for all countries. 

This is due to a reduction in EU-27 NOx emissions in 2023 of 53% since 2005 and about 5.3% between 2022 and 2023. The highest reductions occurred in electricity and heat production and road transport.

For the five studied diseases, the total burden of disease (DALY) and the corresponding mortality (both as AD and YLL) and morbidity (YLD) are as presented in Table 6, ordered by decreasing number of DALY.

Table 6. Burden of disease for individual diseases attributable to NO₂, 2023

EU-27

YLL

YLD

DALY

DALY rate

Diabetes

9,531

(5,005-13,585)

90,294

(47,405-128,747)

58,007

(34,128-87,801)

148,301

(81,534-216,549)

44.9

(24.7-65.5)

Stroke

8,736

(1,496-15,643)

76,866

(13,165-137,689)

26,422

(13,724-38,213)

103,289

(26,889-175,902)

31.2

(8.1-53.2)

COPD

6,431

(2,762-9,832)

64,032

(27,463-98,035)

24,768

(14,145-34,484)

88,800

(41,609-132,519)

26.9

(12.6-40.1)

Asthma in adults

187

(20-358)

2,056

(223-3,929)

24,721

(2,691-47,254)

26,777

(2,915-51,184)

7.5

(0.8-14.3)

Asthma in children/ 

adolescents

(0-2)

(39-128)

9,199

(3,868-12,474)

9,294

(3,908-12,603)

11.0

(4.6-14.9)

Total (five diseases)

24,886

(9,283-39,419)

233,343

(88,295-368,528)

143,117

(68,556-220,226)

376,461

(156,855-588,757)

41 countries considered 

YLL

YLD

DALY

DALY rate

Diabetes

16,013

(8,669-22,242)

167,862

(91,333-232,144)

135,037

(80,528-201,024)

302,899

(171,861-433,169)

74.3

(42.1-106.2)

Stroke

13,961

(2,429-24,642)

135,176

(23,586-237,996)

40,073

(21,026-57,424)

175,250

(44,612-295,420)

43.0

(10.9-72.4)

COPD

11,820

(5,221-17,625)

120,690

(53,311-179,986)

64,359

(37,648-87,699)

185,049

(90,960-267,686)

45.4

(22.3-65.6)

Asthma in adults

483

(54-886)

5,184

(589-9,498)

65,691

(7,498-119,853)

70,875

(8,087-129,352)

15.9

(1.8-29.0)

Asthma in children/ 

adolescents

(1-3)

219

(95-291)

23,136

(10,093-30,685)

23,355

(10,189-30,977)

20.4

(8.9-27.1)

Total (five diseases)

42,279

(16,374-65,398)

429,131

(168,914-659,915)

328,296

(156,793-496,685)

757,428

(325,709-1,156,604)

Notes: Parentheses indicate the 95% confidence interval. Diabetes, type 1 and 2 diabetes mellitus. COPD, chronic obstructive pulmonary disease. AD, attributable deaths. YLL, years of life lost. YLD, years lived with disability. DALY, disability-adjusted life years. DALY rate, DALY/100,000 inhabitants at risk.  

The highest health impact comes from diabetes mellitus, followed by stroke and COPD. In these three cases, the burden of disease results mainly from mortality (61%, 74% and 72% of the burden, respectively in the EU-27; and 55%, 77% and 65% for all countries). For asthma, the disease burden is dominated by morbidity both in adults and in children/adolescents (92% and 99%, respectively in the EU-27; and 93% and 99% for all countries). Map 5 shows the distribution by country of the burden of disease due to diabetes mellitus.

Map 5. Burden of disease for diabetes mellitus attributable to long-term exposure to NO₂, 2023

When considering only deaths due to these five diseases, 25,000 are estimated for the EU-27 and 42,000 for the 41 countries considered in the analysis. This is well below the 34,000 and 56,000 deaths, respectively, due to all natural causes. This lower number reflects other possible diseases not being included, for which the association with air pollution is currently unknown.

Finally, the estimated total burden of disease for NO₂, expressed as DALY and considering five diseases is around five and a half times lower than for PM_2.5, considering six diseases (dementia excluded): 376,461 DALY attributable to NO₂ and 2,074,948 attributable to PM_2.5 for the EU-27; and 401,044 DALY to NO₂ and 2,300,985 DALY to PM_2.5 for the 40 countries for which estimations attributable to exposure to both pollutants have been calculated.

Sensitivity analyses of deaths attributable to NO₂

If all concentrations are considered (that is, from 0µg/m³), instead of using concentrations above the 2021 WHO recommendation (10µg/m³), the potential impact would be (see also Table 3): 

For the EU-27, an estimated number of 78,000 potential additional attributable deaths, resulting in 112,000 total attributable deaths overall, equivalent to 1,161,000 YLL. 
For the 41 countries, an estimated additional 92,000 potential attributable deaths, resulting in 148,000 total attributable deaths overall, equivalent to 1,630,000 YLL.

Finally, if only the impact above the ALV in the revised AAQD 2024/2881 to be achieved by 2030 (20µg/m³) was considered (see also Table 4) the potential impact would be: 

For the EU-27, 5,300 deaths would be attributable to concentrations above the revised ALV, equivalent to 54,000 YLL. 
For the 41 countries, 17,100 deaths would be attributable to concentrations above the revised ALV, equivalent to 214,000 YLL.

The estimations of these sensitivity analyses at country level can be found at Table publisher.

Ozone (O₃) 

In 2023, deaths due to all-cause mortality were as follows: 

In the EU-27, 63,000 deaths attributable to O₃ concentration exposure above the WHO guideline level of 60µg/m³. This is equivalent to 655,000 YLL.
In the 41 countries considered, 71,000 deaths attributable to exposure to O₃ concentrations above the WHO guideline level of 60µg/m³. This is equivalent to 753,000 YLL.

Table 7 provides confidence intervals and the rates per 100,00 inhabitants aged 25 years and above.  

The countries with the highest absolute numbers of attributable deaths in 2023 were Germany, Italy and France (in decreasing order, see Table 7 and Map 6). However, when considering relative impacts, that is YLL per 100,000 inhabitants 25 years and above, the highest rates were seen in Bosnia and Herzegovina, Montenegro, Albania, Croatia and Hungary (in decreasing order).

Map 6. Mortality attributable to long-term exposure to O₃, 2023

The countries with the lowest relative impacts (YLL per 100,000 inhabitants 25 years and above) were Ireland, Türkiye, Iceland, Cyprus and Norway. 

Apart from mortality due to all natural causes, mortality was also estimated for chronic obstructive pulmonary disease, as follows (see also Table 8 for confidence intervals): 

In the EU-27, 7,000 deaths were attributable to exposure to O₃ concentrations above the WHO guideline level of 60µg/m^3. This is equivalent to 73,245 YLL.
In the 41 countries considered, 8,100 deaths were attributable to exposure to O₃ concentrations above the WHO guideline level of 60µg/m³. This is equivalent to 84,712 YLL.

Table 8. Burden of disease for chronic obstructive pulmonary disease attributable to O₃, 2023 

	AD	YLL
EU-27	7,040 (6,723-7,510)	73,245 (69,946-78,145)

All countries	8,125 (7,759-8,668)	84,712 (80,894-90,385)

Νotes: Parentheses indicate the 95% confidence interval. AD, attributable deaths. YLL, years of life lost.  

Summary of results

Tables 1, 5 and 7 present the following for every pollutant, country and country grouping:  

the total population at risk (those above a certain age determined by the epidemiological studies from which the relative risks were derived) for all-cause mortality, for PM_2.5 and NO₂ (30 years and above) and O₃  (25 years and above); 
the population-weighted mean concentrations (as an indication of exposure); 
the estimated number of attributable deaths, the attributable deaths per 100,000 inhabitants at risk, the YLL and YLL per 100,000 inhabitants at risk in 2023, together with their confidence intervals.

Table 3 presents the potential additional attributable deaths and overall total attributable deaths by pollutant for the EU-27 and for the larger number of European countries included in the assessment when considering the whole range of concentrations for PM_2.5 and NO₂.Table 4 shows the attributable deaths estimated for PM_2.5 and NO₂ for concentrations above the annual limit values to be attained in 2030 according to AAQD (EU) 2024/2881.

Table 3. Potential additional attributable deaths and total attributable deaths in 2023 when considering the full concentration range

Geographical scope	Pollutant	Potential additional attributable deaths	Overall total attributable deaths
EU-27	PM_2.5	164,000	46,000 (95% CI: 265,000-386,000)
EU-27	NO₂	78,000	112,000 (95% CI: 57,000-218,000)
40 European countries	PM_2.5	176,000	382,000 (95% CI: 293,000-426,000)
41 European countries	NO₂	92,000	148,000 (95% CI: 75,000-288,000)

Notes: Türkiye is not included in the PM_2.5 estimations as the number of background monitoring stations from which data are available was too
low to produce concentration maps for fine particulate matter. CI, confidence interval.

Table 4. Attributable deaths in 2023 when considering the concentration range above the 2030 ALV 

Geographical scope	Pollutant	Potential additional attributable deaths
EU-27	PM_2.5	55,000 (95% CI: 42,000-62,000)
EU-27	NO₂	5,300 (95% CI: 2,700-10,500)
40 European countries	PM_2.5	69,000 (95% CI: 53,000-77,000)
41 European countries	NO₂	17,100 (95% CI: 8,700-33,300)

Notes: Türkiye is not included in the PM_2.5 estimations as the number of background monitoring stations from which data are available was too low to produce concentrationmaps for fine particulate matter. CI, confidence interval.

Figures 2 and 4 show, for all countries considered, the total burden of disease (with mortality and morbidity components) attributable to long-term exposure to PM_2.5 and NO₂, respectively, for the specific diseases.  

Figures 3 and 5 show, at country level, the percentage contribution of each disease to the total burden of disease (DALY) estimated for the six and five diseases considered for PM_2.5 and NO₂ respectively.

Figure 2. Burden of disease (YLL, YLD, DALY) attributable to PM_2.5, differentiated by diseases

Figure 3. Contribution of each disease to the total burden of disease (DALY) attributable to PM_2.5

Figure 4. Burden of disease (YLL, YLD, DALY) attributable to NO₂ , differentiated by diseases 

Figure 5. Contribution of each disease to the total burden of disease (DALY) attributable to NO₂  

EEA Briefing 16/2025:

Title: Harm to human health from air pollution in Europe: burden of disease status, 2025

HTML: TH-01-25-033-EN-Q - ISBN: 978-92-9480-739-7 - ISSN: 2467-3196 - doi: 10.2800/8961999

Soares, J., et al.,2025, Assessing the environmental burden of disease related to air pollution in Europe in 2023, Eionet Report – ETC HE 2025/8, European Topic Centre on Human Health and the Environment (https://doi.org/10.5281/zenodo.17658760)

Harm to human health from air pollution in Europe: burden of disease status, 2025

Key messages

Box 1. Measuring impacts on health: understanding the ‘environmental burden of disease’ concept

Figure 1. Burden of disease as a sum of YLD and YLL

How air pollution harmed human health in Europe in 2023

Fine particulate matter (PM2.5)

Map 1. Mortality attributable to long-term exposure to PM2.5, 2023

Table 2. Burden of disease for individual diseases attributable to PM2.5, 2023

Map 2. Burden of disease for ischemic heart disease attributable to long-term exposure to PM2.5, 2023

Map 3. Burden of disease for COPD attributable to long-term exposure to PM2.5, 2023

Sensitivity analyses of deaths attributable to PM2.5

Nitrogen dioxide (NO2)

Map 4. Mortality attributable to long-term exposure to NO2, 2023

Table 6. Burden of disease for individual diseases attributable to NO2, 2023

Map 5. Burden of disease for diabetes mellitus attributable to long-term exposure to NO2, 2023

Sensitivity analyses of deaths attributable to NO2

Ozone (O3)

Map 6. Mortality attributable to long-term exposure to O3, 2023

Table 8. Burden of disease for chronic obstructive pulmonary disease attributable to O3, 2023

Summary of results

Table 3. Potential additional attributable deaths and total attributable deaths in 2023 when considering the full concentration range

Table 4. Attributable deaths in 2023 when considering the concentration range above the 2030 ALV

Figure 2. Burden of disease (YLL, YLD, DALY) attributable to PM2.5, differentiated by diseases

Figure 3. Contribution of each disease to the total burden of disease (DALY) attributable to PM2.5

Figure 4. Burden of disease (YLL, YLD, DALY) attributable to NO2 , differentiated by diseases

Figure 5. Contribution of each disease to the total burden of disease (DALY) attributable to NO2

Fine particulate matter (PM_2.5)

Map 1. Mortality attributable to long-term exposure to PM_2.5, 2023

Table 2. Burden of disease for individual diseases attributable to PM_2.5, 2023 

Map 2. Burden of disease for ischemic heart disease attributable to long-term exposure to PM_2.5, 2023 

Map 3. Burden of disease for COPD attributable to long-term exposure to PM_2.5, 2023

Sensitivity analyses of deaths attributable to PM_2.5

Nitrogen dioxide (NO₂)

Map 4. Mortality attributable to long-term exposure to NO₂, 2023

Table 6. Burden of disease for individual diseases attributable to NO₂, 2023

Map 5. Burden of disease for diabetes mellitus attributable to long-term exposure to NO₂, 2023

Sensitivity analyses of deaths attributable to NO₂

Ozone (O₃) 

Map 6. Mortality attributable to long-term exposure to O₃, 2023

Table 8. Burden of disease for chronic obstructive pulmonary disease attributable to O₃, 2023 

Table 4. Attributable deaths in 2023 when considering the concentration range above the 2030 ALV 

Figure 2. Burden of disease (YLL, YLD, DALY) attributable to PM_2.5, differentiated by diseases

Figure 3. Contribution of each disease to the total burden of disease (DALY) attributable to PM_2.5

Figure 4. Burden of disease (YLL, YLD, DALY) attributable to NO₂ , differentiated by diseases 

Figure 5. Contribution of each disease to the total burden of disease (DALY) attributable to NO₂  