Panel 1
Massimiliano BULTRINI, Simona BUSCEMI, Giorgio CATTANI, Alessandra GAETA, Federica NOBILE (Dipartimento di Epidemiologia del Servizio Sanitario Regionale-Regione Lazio ASL Roma 1), Maria Antonietta REATINI, Massimo STAFOGGIA (Dipartimento di Epidemiologia del Servizio Sanitario Regionale-Regione Lazio ASL Roma 1)
Estimates of annual population-weighted exposure (PWE) to PM2.5 aggregated at municipal level, were produced using an integrated approach combining measurements and statistical modelling. The resulting data, covering the period 2016–2023, allow for comparison of PWE across all municipalities and provide a comprehensive national overview of average exposure levels.
In 2023, 100% of the population was exposed to concentrations above the WHO air quality guideline (5 µg/m3).
The national average in 2023 was 12 µg/m³ (minimum–maximum range: 4–25 µg/m³).
Between 2016 and 2023, the national average decreased from 15 µg/m³ in 2016 to 12 µg/m³ in 2023, corresponding to a reduction of 20%.
The most authoritative health estimates attribute a significant share of premature deaths and reduced life expectancy to exposure to air pollutants (WHO, 2016; EEA, 2022; WHO, 2021).
Assessing population exposure and its spatial and temporal variability in outdoor environments is a key step in epidemiological studies that investigate the relationship between air pollution and health effects (Caplin et al., 2019).
This indicator provides an estimate of the annual population-weighted exposure (PWE) to PM2.5, aggregated at the municipality level, enabling comparisons between cities and offering a comprehensive national overview of average exposure.
It is representative of chronic exposure and is useful for estimating long-term health impacts.
Exposure levels are compared with the reference values (guideline levels and interim targets) set by the World Health Organization (WHO, 2021).
Estimate the average population exposure to the main airborne pollutants, assess exposure levels against WHO reference values, and evaluate the temporal trends in population exposure.
Legislative Decree No. 155 of 13 August 2010, as amended, transposing Directive 2008/50/EC and Directive 2004/107/EC.
WHO (2021), Global Air Quality Guidelines. Particulate matter (PM10, PM2.5), ozone, nitrogen dioxide, sulfur dioxide and carbon monoxide. Geneva, World Health Organization, 2021.
A Clean Air Programme for Europe, Brussels, 18.12.2013, COM(2013) 918 final.
A Europe that protects: Clean air for all, Brussels, 17.5.2018, COM(2018) 330 final.
Pathway to a Healthy Planet for All – EU Action Plan: “Towards Zero Pollution for Air, Water and Soil”, Brussels, 12.5.2021, COM(2021) 400 final.
As part of the revision of its Air Quality Guidelines, the World Health Organization (WHO) has provided quantitative risk estimates for a range of health effects associated with an increase of 10 µg/m³ in the annual average concentration of various pollutants.
In its assessments, WHO does not establish a threshold value below which no risk is observed. However, it identifies 5 µg/m³ as the lower limit for annual mean exposure to PM2.5, referred to as the Air Quality Guideline level (“reference level”): this is the lowest concentration at which an increase in total mortality, cardiopulmonary mortality, and lung cancer mortality has been observed, with a 95% confidence interval. WHO has also defined a set of interim targets, i.e. higher reference levels to be achieved in successive steps through the implementation of air quality improvement policies. The interim targets for PM2.5 proposed by WHO are 10 µg/m³ (IT-4), 15 µg/m³ (IT-3), 25 µg/m³ (IT-2), and 35 µg/m³ (IT-1).
EU policy objectives for clean air aim to progressively align with WHO air quality guidelines, which are significantly stricter than current EU air quality standards and are based on levels deemed necessary to ensure effective protection of human health.
The EU Clean Air Policy Package and the Zero Pollution Action Plan set the objective of reducing the number of premature deaths attributable to air pollution by more than 50% by 2030, compared to 2005 levels.
Panel 2
Caplin A., Ghandehari M., Lim C., Glimcher P., Thurston G.(2019), Advancing environmental exposure assessment science to benefit society. Nature communication (2019) 10:1236. https://doi.org/10.1038/s41467-019-09155-4.
EEA (2022), Health Risk Assessment of Air Pollution and the Impact of the New WHO Guidelines. Eionet Report – ETC/ATNI 2022/10.
EUROSTAT (2022), Sustainable development in the European Union — Monitoring report on progress towards the SDGs in an EU context - 2022 edition. (DOI): 10.2785/313289.
Stafoggia M, Bellander T, Bucci S et al. Estimation of daily PM10 and PM2,5 concentrations in Italy, 2013-2015, using a spatiotemporal land-use random-forest model. Environ Int 2019;124:170-79.
Stafoggia M, Cattani G, Ancona C, Ranzi A. La valutazione dell’esposizione della popolazione italiana all’inquinamento atmosferico nel periodo 2016-2019 per lo studio della relazione tra inquinamento atmosferico e COVID-19. Epidemiol Prev. 2020 Sep-Dec;44(5-6 Suppl 2):161-168. Italian. doi: 10.19191/EP20.5-6.S2.115. PMID: 33412807.
WHO (2016), Ambient air pollution: A global assessment of exposure and burden of disease. Geneva, World Health Organization, 2016.
WHO (2021), Global Air Quality Guidelines. Particulate matter (PM10, PM2,5 ), ozone, nitrogen dioxide, sulfur dioxide and carbon monoxide. Geneva, World Health Organization, 2021.
The population considered is the one distributed across the 2021 census sections, which represent the most recent territorial reference available.
The indicator expresses a municipality-level average of population exposure. It is well known that within a single municipality, there may be spatial variability in pollutant concentrations, with some areas experiencing higher levels than the municipal average or background concentrations. However, the assessment of local- or microscale-level exposure within individual municipalities falls outside the scope of this indicator. It is important to note that areas with higher population density within a municipality are given greater weight in the calculation of the Population Weighted Exposure (PWE).
In any case, the exposure values obtained should not be compared with legal limit values: for the purpose of compliance assessment with legal thresholds, the highest value measured by individual air quality monitoring stations should be used instead.
Application of high-resolution spatiotemporal models, using spatiotemporal predictors, satellite data, and air quality monitoring data.
Data quality assessment
DEP: Dipartimento di Epidemiologia del SSR Regione Lazio - ASL RME ISPRA (Istituto Superiore per la Protezione e la Ricerca Ambientale)
ISTAT (Istituto Nazionale di Statistica)
SNPA (Sistema Nazionale per la protezione dell’ambiente)
ISPRA - Database InfoARIA;
ISTAT - www.demo.istat.it
National, Regional and Municipal level (7,901 out of 7,914 municipalities)
2016-2023
Indicator assessment
The model uses daily average concentrations measured at air quality monitoring stations across the national territory as the response variable, and a selection of spatial and temporal predictors as explanatory variables. Monitoring data are collected and stored by ISPRA in the InfoAria database, in accordance with Directive 2008/50/EC (and its national transposition by Legislative Decree No. 155/2010) and Decision 2011/850/EU.
A machine learning model—random forest—was developed to relate daily pollutant concentrations from monitoring stations to spatial predictors (such as population density, road network, land cover, altitude, artificial surfaces, etc.) and spatiotemporal predictors (dispersion modelling, meteorological data), in order to estimate daily mean PM2.5 concentrations for each km² grid cell across the Italian territory.
Based on the daily concentration estimates from the random forest model, the annual population-weighted exposure to PM2.5 was calculated by weighting the concentration values in each grid cell by the population residing in that cell.
The “national average concentration value” is computed as the population-weighted mean of the estimated concentrations at the municipal level, using the resident population as weights.
Lastly, the percentage distribution of the resident population exposed to different PM2.5 concentration classes is estimated.
In 2023, 100% of the population was exposed to concentrations above the WHO guideline value (5 µg/m³); 85% was exposed to levels exceeding interim target IT-4 (10 µg/m³); and 33% of the population was exposed to concentrations above IT-3 (15 µg/m³). In the latter case, the exposed population was mainly concentrated in the Po Valley, particularly in Veneto, Lombardy, Piedmont and Emilia-Romagna (Figure 1 – Table 1). No municipality recorded PWE levels above IT-2 (25 µg/m³).
The national average of PWE was 12 µg/m³ (minimum–maximum range: 4–25 µg/m³) (Figure 2).
Over the eight-year period considered (2016–2023), PWE values show an overall decreasing tendency, with the sole exception of 2017, when a higher average value (16 µg/m³) was observed (Figure 2). The national average declined from 15 µg/m³ in 2016 to 12 µg/m³ in 2023, corresponding to a 20% reduction. However, the observation period is too short to allow a reliable estimation of a trend and its statistical significance, which in most cases can only be identified when data are analysed over a time horizon of 10 years or more.
From 2016 to 2023, 100% of the population continued to be exposed to PM2.5 concentrations above the WHO guideline value. By contrast, a decreasing tendency was observed in the share of the population exposed to levels above IT-3 (15 µg/m³), which fell from 52% to 33%. In all years considered, PWE values never exceeded IT-1 (35 µg/m³), and in 2023, IT-2 (25 µg/m³) was not exceeded, confirming the pattern already observed in the five previous years (2018–2022) (Figure 3).
Data
Figure 1: Population-Weighted Exposure to PM2.5 at Municipal Level (2023)
ISPRA processing based on data from the Epidemiology Department of the Lazio Regional Health Service
Figure 2: Annual distribution of Population-Weighted Exposure (PWE) to PM2.5 at Municipal Level
ISPRA processing based on data from the Epidemiology Department of the Lazio Regional Health Service
The box represents the interquartile range (IQR), which is the difference between the third quartile (Q3, 75th percentile) and the first quartile (Q1, 25th percentile) of the data. The line inside the box indicates the median (Q2, 50th percentile), while the dot represents the mean value.
Figure 3: Percentage of population exposed to PM2.5 by exposure range and year
ISPRA processing based on data from the Epidemiology Department of the Lazio Regional Health Service
c= PWE (µg/m3)
The six PWE classes correspond respectively to the WHO air quality guideline value and the four interim targets.
Table 1: PM2.5 - Population-Weighted Exposure (PWE) by region: Percentage distribution by exposure range (2023)
ISPRA processing based on data from the Epidemiology Department of the Lazio Regional Health Service
PWE: Population Wheigted Exposure
The six PWE classes correspond respectively to the WHO air quality guideline value and the four interim targets.
The indicator is consistent with those used at European level by the EEA and at global level by the World Health Organization (WHO) to monitor exposure. It also provides the basis for exposure assessment to support health risk assessment and burden of disease estimates, in line with the approaches adopted by these two international organisations.
With regard to the United Nations Sustainable Development Goals (SDGs), the methodological approach is consistent with that used for the calculation of the indicator designed to monitor progress towards Goal 11 (Make cities and human settlements inclusive, safe, resilient and sustainable), target 11.6 (By 2030, reduce the adverse per capita environmental impact of cities, including by paying special attention to air quality and municipal and other waste management).
From a methodological perspective, the exposure estimates also provide the basis for one of the indicators used to monitor progress towards Goal 3 (Ensure healthy lives and promote well-being for all at all ages), target 3.9 (By 2030, substantially reduce the number of deaths and illnesses from hazardous chemicals and air, water and soil pollution and contamination)