Case studies for the utility and validity test of the “Outdoor Air and Health Atlas”: Definition of the conceptual framework

Introduction

Air pollution is a major global public health concern and one of the leading risk factors for human health, with its burden on healthcare systems steadily increasing.^1,2 According to the World Health Organization (WHO), approximately 4.2 million premature deaths in 2019 are attributable to outdoor air pollution worldwide.³

Exceedances of air quality standards continue to represent a persistent issue, with significant health and social implications, and Italy is still among the country where air pollution levels exceed the WHO guidelines.⁴

In response to the persistent challenge of air pollution and its health impacts, the Italian Ministry of Health launched the project “ATLAS – Air and Health: an integrated Atlas to support decisions and research”. The project aims to develop a national interactive Atlas integrating air quality data and health outcomes, providing a tool to support researchers, policymakers, technical experts, and the general public. The Atlas is thought to enable the evaluation of counterfactual scenarios to estimate the potential impact of pollution reduction strategies or mitigation measures on health outcomes.

Alongside the national Atlas, the project also included the development of regional Atlases, which apply the same methodological framework at the regional scale while adapting health outcomes, pollutants, territorial units, and time periods to local priorities and data availability. Their common purpose is to support the evaluation of Regional Air Quality Plans, translating the emission reduction scenarios of each plan into estimates of the associated health benefits at the regional and sub-regional level.

Among the various aims of the ATLAS project, Objective 4 focuses on validating the Atlas and its metrics to ensure they accurately and effectively inform policymaking. To achieve this, a set of case studies has been performed, based on exemplary real-life situations, to evaluate the usability and practical utility of the Atlas in its present form and on possible future developments in terms of new metrics, outcomes, and pollutants to be included.

Specifically, two main types of case studies were defined, with distinct aims. First, some of the case studies are intended to validate the Atlas in its ability to inform decision-making. A special case of this aim is the test application of the data produced by the regional Atlas to sustain local plans for the air quality.⁵ Second, other case studies aim to explore and test novel metrics or indicators for the further development of the Atlas. 

To ensure a coherent and consistent interpretative framework across all case studies, it was necessary to adopt a conceptual model capable of positioning the indicators provided by the Atlas within a coherent sequence of causes and effects. This paper aims to describe the process underlying the selection and application of a shared conceptual framework to the case studies proposed by the partner units of the ATLAS – Air and Health project, highlighting its role in supporting comparability and interpretability of results.

Materials and methods

Selection of the DPSEEA model

There are several previous experiences in the application of conceptual frameworks for the selection and analysis of impact indicators.^6-10 In Italy, it has been adopted within the Italian National Prevention Plan to identify key risk factors and determinants to be monitored in the evaluation of the Plan’s impact on population health.^11,12

For the Italian National Prevention Plan evaluation project, a review of the existing literature was conducted, examining eleven different conceptual models to identify the most suitable framework for the development of health indicators (EHIs) to assess and monitor the impacts of preventive intervention. In the field of environmental health and climate change, a systematic review was identified.¹³ The review highlighted that no single framework can fully capture the complexity of interactions involved in assessing the health impacts of climate change. However, the Driving Force-Pressure-State-Exposure-Effect-Action (DPSEEA) model emerged as the most flexible and sufficiently complete framework for the development of EHIs and the most applicable to the context of this project. A key strength of the DPSEEA framework lies in its ability to clearly identify multiple intervention points along the environmental health causal chain.¹⁴  Moreover, it is particularly useful for addressing risks associated with environmental pollution, where the pathway from pollution sources to health outcomes can be explicitly delineated, thereby providing a robust tool to illustrate causal linkages from hazards to health effects.¹⁵  Finally, in application to the National Prevention Plan evaluation, the model proved to be particularly flexible, being adapted to almost all the topics of the plan.^11,12,16

Accordingly, at the stage of project proposal development, the DPSEEA model was selected as the reference framework, as it was considered the most flexible, comprehensive, and operationally suitable for the study objectives.^15,17

DPSEEA model

The DPSEEA framework (Figure 1) was adopted to quantify the effects of air pollution across different geographic areas and for a range of health outcomes.

A general description of the different domains of the model has been provided in several previous studies.^18,19 Here, a brief description and some examples of what can be considered in the DPSEEA causal chain were reported.

Within this model, Driving Forces refer to the broad social, demographic, and economic processes that motivate and shape environmental change. Population growth is often considered the most influential driver, alongside technological development and economic expansion. Some policy actions may also act as driving forces, such as directives on building, transport, or economy, not intended to act on the environment.

Driving forces generate Pressures on the environment, typically through human activities that involve resource use and environmental exploitation. These pressures arise across all sectors of economic activity, including mining, energy production, manufacturing, services, transport, tourism, agriculture, and forestry. They occur throughout the entire supply chain, from resource extraction and processing to distribution, consumption, and waste generation.

In response to these pressures, the State of the environment is altered. These changes may be complex and wide-ranging, affecting multiple environmental compartments and dimensions, including the frequency and magnitude of natural hazards, the availability and quality of natural resources, and levels of environmental pollution. Such changes occur across different spatial scales: some are localized and concentrated near the source of pressure (e.g., habitat loss, urban air pollution, local water contamination), while others are more diffuse and contribute to regional or global environmental changes (e.g., desertification, marine pollution, climate change). Due to the interconnected nature of environmental systems, these changes often produce secondary and cascading effects.

Exposure represents the interface between humans and environmental hazards. It occurs when individuals come into contact with environmental contaminants or risk factors. According to the National Academy of Sciences,²⁰ exposure is defined as “an event that occurs when there is contact at the boundary between a human and the environment with a contaminant of a specific concentration for a specified period of time”.²¹ In the context of environmental pollution, exposure may occur through multiple pathways, including inhalation, ingestion, and dermal absorption, potentially affecting different organs and systems.

Finally, exposure to environmental hazards may result in a range of Effects on human health. These effects can vary in type, severity, and magnitude depending on the nature of the hazard, the level and duration of exposure, and the size and vulnerability of the exposed population.

On the other side of the model, Actions represents the interventions that can be implemented to change the direction of the driving forces, reduce the pressure, change the state, reduce exposure, or mitigate the effects. 

By its nature as a conceptual model, the application of the DPSEEA framework requires adaptation to specific contexts and, where necessary, modification to reflect local conditions and data availability. Nevertheless, it provides a structured matrix to guide the reconstruction of the environment-health relationship through the development of indicators that support contextual understanding, as well as the design and evaluation of intervention strategies.

In the field of environmental health, the components of the DPSEEA framework have been operationalized as reported in the Figure 2 by previous work done by the WHO project on Development of Environment and Health Indicators for European Union Countries (ECOEHIS) and by the work conducted for the Italian National Prevention Plan.^16,22,23

Application of the DPSEEA framework within the “Air and Health Atlas” project

The DPSEEA framework was applied within the “Air and Health Atlas” project through an iterative and participatory process involving both collective discussion and targeted expert consultation. An initial version of the model, tailored to the specific objectives of the project, was developed during a plenary meeting involving all project partners. This preliminary framework was subsequently refined through back-office activities, during which revisions were proposed, documented, and circulated among participants via email. Feedback and comments were systematically collected and integrated, leading to an updated version of the model that was further discussed, adapted, and formally validated during a second plenary meeting.

The operational application of the framework to the selected case studies was carried out through structured group work sessions during in-person meetings. These activities aimed to define a shared analytical approach and to develop a common template for the description and contextualization of each case study, ensuring consistency and comparability across different settings.

In addition, dedicated bilateral meetings were conducted involving the Objective 4 coordinator and the investigators involved in each case study to provide methodological support and to specify how the DPSEEA components should be translated into the standardised reporting format. This step allowed for the harmonisation of content across case study reports, while preserving the specificities of each local context.

Results

Within the project, an initial framework was developed for presenting the case studies, structured into the following sections:

• Rationale;
• Narrative description including:
  • description of the area and the main environmental and health critical issues;
  • environmental monitoring and exposure assessment data to be used for the case study;
  • counterfactual/intervention scenario;
  • indicators;
• Expected results;
• Actions.

Subsequently, an effort was made to identify appropriate indicators according to the DPSEEA model, providing a preliminary theoretical application to the topic of air pollution. 

This approach explicitly defines the actions under evaluation, as illustrated in Table 1.

Application of the DPSEEA Model to the case studies in the Outdoor and Health Atlas project

Once the framework and guiding principles had been defined, the participating operational units identified a set of relevant case studies within their respective territories, to which the DPSEEA model was subsequently applied.

A total of nine case studies were analysed under Objective 4, covering six regions involved in the project (Table 2). The DPSEEA model was tailored to each case study, building upon the general framework presented in Table 1 and adapting it to the specific environmental and health contexts under investigation.

The case studies were further categorized according to their primary aim within the frame of the ATLAS project:

• Aim 1: validation-oriented studies, aimed at assessing and supporting the functionality of the Atlas;
• Aim 2: exploratory studies, designed to investigate and test novel metrics or indicators to inform and enhance the future development of the Atlas.

The application of the DPSEEA model to each individual case study is presented in detail in Tables 3 and 4. Each case study established specific objectives.

The Veneto case study incorporates the estimation of health impacts from the Regional Air Protection and Remediation Plan into the DPSEEA framework, identifying appropriate metrics for evaluating effectiveness;

The first Puglia and the Molise case studies, by contrast, test the effectiveness of air quality improvement targets related to biomass combustion within subregional areas, as outlined in their respective Regional Air Quality Plans;

All three regionally-focused case studies aim to address equity considerations, contributing both to the design of the Atlas and to effect modification analyses;

The first Emilia-Romagna case study similarly leverages the air pollution scenarios from the Regional Air Plan, aiming to produce impact estimates at successive implementation stages and, where feasible, differential evaluations corresponding to specific initiatives.

These case studies apply a substantially different approach to forecast the impact of changes in the health impact of pollutant reduction in the future, compared to the national Atlas. In fact, while in the national Atlas, the reduction in pollutants to reach EU limits or WHO standards is assumed to be equal in all geographic areas, in these regional pilots, the reduction in pollutant levels is set as predicted by the models produced by the regional Environmental agencies, and they may be different across the areas of the region according to the expected effect of specific policies and actions.

The Palermo case study examines the effectiveness of urban planning policies on sustainable mobility, complemented by risk perception surveys that provide insights valuable for decision-makers and for engaging citizens in defining air quality improvement scenarios. 

These five case studies answer the first aim, i.e., validating the Atlas, mostly in its regional declination, for its utility in informing decision-making.

The Taranto case study tests pollutant dispersion models specific for the steel plant emissions and capable of producing higher-resolution exposure estimates, i.e., 5x5 m definition for the level of exposure instead of the 1x1 km provided by the model adopted by the Atlas now,^24,25 and evaluates their added value for designing interventions to reduce industrial pollution, also considering stratification by socioeconomic status. This case study tests the utility of a new metric, thus answering mainly to the second aim of the case studies, but it also tests it on the decision-making process, thus participating also in the first aim.

The Northern Salento Area case study provides an updated assessment of actual wood biomass consumption, evaluating the environmental and health impacts of regional emission reduction measures associated with its use, according to aims 1 and 2.

The second Emilia-Romagna case study explores a potentially innovative aspect of impact assessment linked to the chemical characterisation of particulate matter that can attribute the particulate to different sources of emission,²⁶ investigating the feasibility of integrating these results into the Atlas.

The Milan case studies explore the feasibility of incorporating psychiatric disorders and small-for-gestational-age (SGA) outcomes into the panel of health endpoints associated with air pollution, thereby enriching the Atlas available metrics.^26,27 These studies do not propose any counterfactual scenario; the DPSEEA model serves primarily as a conceptual framework for assessing associations between exposure and less-established health outcomes, given the context defined by driving forces, pressures, and states.

These three last case studies validate new indicators, one of exposure and two outcomes.

A full description of the case studies is reported in Appendix 1 (see online Supplementary Materials).

Discussion

For all proposed case studies, representation through the DPSEEA model was feasible and played a crucial role in guiding the design of the study activities. The model contributed to the conceptualisation of the most appropriate methods and indicators, tailored to the specific context and the interventions under evaluation, based on a clearly defined rationale.

The nine case studies were conceived to contribute simultaneously to multiple dimensions of the project: the construction and validation of the integrated Atlas, the development of health impact assessment algorithms and counterfactual scenarios, and the exploration of new exposure metrics and health outcomes. This cross-cutting structure reflects the overall ambition of the ATLAS project – to make health impact assessment of air pollution a routinely exercisable capacity across all participating Regions – and the DPSEEA framework, by providing a coherent causal structure applicable across heterogeneous contexts, was instrumental in making this function possible.

Six of the nine case studies were developed in direct connection with regional or local air quality planning processes. In Italy, Regional Air Quality Plans are the primary regulatory instrument through which Regions implement Legislative Decree 155/2010, transposing European Directive 2008/50/EC, to achieve air quality standards for key pollutants within defined timeframes. Their effectiveness has historically been assessed almost exclusively in terms of air quality improvement, with limited systematic quantification of the corresponding health benefits. The Atlas, by providing nationally consistent, spatially resolved estimates linking pollution concentrations to health outcomes, offers the infrastructure needed to translate emission reduction scenarios into quantified health gains – avoided deaths, hospitalisations, and chronic disease burden – directly relevant to public health governance. The case studies demonstrate concretely how this translation can be operationalised: the Veneto case study quantifies health events potentially avoided under the final air quality plan implementation scenario; the first Emilia-Romagna case study tests the Atlas usability against regional model outputs; the Puglia and Molise case studies assess health benefits expected from biomass combustion reduction targets in subregional critical areas; the Taranto case study explores higher-resolution dispersion modelling at sub-municipal scale within the integrated environmental authorization framework, thus contributing to both the validation and exploratory aims; the Palermo case study evaluates the health impact of sustainable urban mobility policies under the local urban mobility plan. By mapping each intervention onto the Pressure, State, and Exposure nodes of the DPSEEA chain, the framework made explicit which levers a given policy acts upon and through which pathway health benefits are expected – a distinction that is essential when comparing measures acting at different points of the causal chain. Some of these case studies directly informed the design of the regional Atlases developed within the project. In Veneto and Emilia-Romagna, in particular, the case study design stage made clear the value of a tool able to communicate the health benefits anticipated from the measures planned and implemented by the Regions and this requirement was incorporated into the structure of the corresponding regional Atlas; in Puglia, the regional Atlas focuses on the health benefits expected from the biomass combustion reduction targets, an objective shared by the Molise case study for the same emission source. The national and regional Atlases are also effective tools to implement what the EU’s directive 2024/28811 requires for member states, i.e., to ensure that the public as well as stakeholders, decision makers and healthcare professional organizations and other relevant health-care bodies and the relevant industrial federations are informed, adequately and in good time, among the others, of the air quality, air quality plans and air quality roadmaps, and the effects of exceedances of limit values, target values, average exposure reduction obligations, average exposure concentration objectives, alert thresholds and information thresholds in a summary assessment.²⁸

Four exploratory case studies addressed possible expansions of the Atlas scope. The already reported case study investigates the utility and validity of a more detailed spatial discrimination in the Taranto area. The Milan case studies investigate associations between air pollution and psychiatric disorders and small-for-gestational-age births – outcomes of growing scientific interest not yet part of the standard health impact assessment core set – using the DPSEEA model as a conceptual anchor in the absence of a defined counterfactual scenario. The second Emilia-Romagna case study explores the integration of particulate matter chemical characterization into the Atlas, with implications for improving the epidemiological validity of exposure estimates. Together, these case studies trace a development trajectory for the Atlas toward finer-scale exposure assessment and a broader range of health endpoints.

However, because the case studies are being conducted in parallel with the other objectives of the project, this paper documents the application of the DPSEEA framework at their design stage: it establishes the feasibility of using the model to structure heterogeneous case studies and to identify appropriate indicators, but its empirical performance in supporting decision-making will be assessed only once the case studies are completed, when their full results will allow a more complete evaluation. The uneven availability of data across contexts conditioned which indicators could be populated at each node of the model, without preventing the consistent mapping of all nine case studies onto the framework. A further consideration is that the operationalisation of the indicators relied on expert consensus and on a case-specific adaptation of the model, so that the configuration adopted is one of several possible. The participatory and iterative process adopted for the application of the DPSEEA framework – involving plenary meetings, back-office revisions, and bilateral expert consultations – proved effective in generating a shared analytical vocabulary across heterogeneous case studies spanning different regions, pollutants, health outcomes, and intervention types. The common reporting template enhanced comparability while preserving the specificity of each local context.

The application of the DPSEEA model to the case studies supporting the national Air and Health Atlas has proven both feasible and analytically productive. By providing a shared conceptual structure, the model enabled a coherent and comparable representation of diverse environmental health scenarios across Italian Regions. It supported the identification of relevant indicators, the positioning of policy actions within the environment-health causal chain, and the structured engagement with regional and local air quality planning processes. Looking ahead, the continued refinement of the Atlas – enriched by the methodological contributions of the case studies here presented – holds the potential to strengthen the evidence base available to decision-makers and to consolidate the role of health impact assessment within Italian environmental governance.

Conflicts of interest: none declared. 

Funding: project implemented with the technical and financial support of the Ministry of Health – PNC PREV-A-2022-12376. 

References

1. Sierra‐Vargas MP, Teran LM. Air pollution: impact and prevention. Respirology 2012;17(7):1031-38. doi:10.1111/j.1440-1843.2012.02213.x
2. Cohen AJ, Brauer M, Burnett R, et al. Estimates and 25-year trends of the global burden of disease attributable to ambient air pollution: an analysis of data from the Global Burden of Diseases Study 2015. Lancet 2017;389(10082):1907-18. doi:10.1016/S0140-6736(17)30505-6
3. World Health Organization. Ambient (outdoor) air pollution. Key facts. 24.10.2024. Available from: https://www.who.int/news-room/fact-sheets/detail/ambient-(outdoor)-air-quality-and-health (last accessed: 17.03.2026).
4. European Environment Agency. Air quality status report 2025. EEA, 09.04.2025. Available from: https://www.eea.europa.eu/en/analysis/publications/air-quality-status-report-2025 (last accessed: 17.03.2026).
5. Trozzi C, Vaccaro R. Regional Air Quality Management Plans in Italy: Balances and Perspectives. Conference: Air & Waste Management Association - 94th Annual Conference & Exhibition. Orlando (Florida, USA), 24-28 June 2001.
6. Farchi S, Molino N, Giorgi Rossi P, et al. Defining a common set of indicators to monitor road accidents in the European Union. BMC Public Health 2006;6:183. doi:10.1186/1471-2458-6-183
7. Edokpolo B, Allaz-Barnett N, Irwin C, et al. Developing a Conceptual Framework for Environmental Health Tracking in Victoria, Australia. Int J Environ Res Public Health 2019;16(10):1748. doi:10.3390/ijerph16101748
8. Barbosa Pereira B, Brandão da Cunhada P, Gomes Silva G, et al. Integrated monitoring for environmental health impact assessment related to the genotoxic effects of vehicular pollution in Uberlândia, Brazil. Environ Sci Pollut Res Int 2017;24(3):2572-77. doi:10.1007/s11356-016-8039-5
9. Liu HY, Bartonova A, Pascal M, Smolders R, Skjetne E, Dusinska M. Approaches to integrated monitoring for environmental health impact assessment. Environ Health 2012;11:88. doi: 10.1186/1476-069X-11-88
10. von Schirnding YE. Health-and-Environment Indicators in the context of sustainable development. Can J Public Health 2002;93 Suppl 1:S9-15. doi:10.1007/BF03405112
11. Broccoli S, Venturelli F, Bevere F, et al. Sviluppo di un modello concettuale di riferimento per l’interpretazione degli indicatori di prevenzione dell’obesità infantile e nell’adolescenza nel Piano nazionale della prevenzione. Epidemiol Prev 2020;44(4):243-53. doi: 10.19191/EP20.4.P243.054
12. Zappa M, Bevere F, Braga M, Broccoli S, Cosentino M, Galeone D, et al. Sviluppo di un modello concettuale di riferimento per l’interpretazione degli indicatori di monitoraggio degli screening oncologici nel Piano nazionale della prevenzione. Epidemiol Prev 2019;43(5-6):354-63. doi: 10.19191/EP19.5-6.P354.105
13. Hambling T, Weinstein P, Slaney D. A review of frameworks for developing environmental health indicators for climate change and health. Int J Environ Res Public Health 2011;8(7):2854-75. doi:10.3390/ijerph8072854
14. Briggs D (ed). Making a Difference: Indicators to Improve Children’s Environmental Health. London: Imperial College London; 2003. Available from: https://fr.ircwash.org/sites/default/files/Briggs-2003-Making.pdf
15. Briggs D. Environmental health indicators: framework and methodologies. Geneva: WHO; 1999. Available from: https://iris.who.int/server/api/core/bitstreams/454708e8-44e0-4838-9b35-6e65d7a9c2d5/content
16. Giorgi Rossi P, Broccoli S, Angelini P, et al. Politiche e interventi per un piano nazionale. Ecoscienza 2017;1.
17. Waheed B, Khan F, Veitch B. Linkage-Based Frameworks for Sustainability Assessment: Making a Case for Driving Force-Pressure-State-Exposure-Effect-Action (DPSEEA) Frameworks. Sustainability 2009;1(3):441-63. doi: 10.3390/su1030441
18. Corvalán CF, Kjellström T, Smith KR. Health, environment and sustainable development: identifying links and indicators to promote action. Epidemiology 1999;10(5):656-60.
19. Briggs D, Corvalán C, Kjellstrom T, Nurminen MM. Linkage methods for environment and health analysis. Geneva: WHO; 1996.
20. National Academies of Sciences. Human Exposure Assessment for Airborne Pollutants. Advances and Opportunities. 1991.
21. National Research Council (US) Committee on Environmental Epidemiology. Environmental Epidemiology: Volume 1. Public Health and Hazardous Wastes. Washington D.C.: National Academies Press (US); 1991.
22. Grant Agreement SPC 2002300 Between the European Commission, DG Sanco and the World Health Organization, Regional Office for Europe. Development of environment and health indicators for European Union countries. ECOEHIS. Final Report. Geneva: WHO; 2004.
23. Farchi S, Molino N, Giorgi Rossi P, et al. Defining a common set of indicators to monitor road accidents in the European Union. BMC Public Health 2006;6:183. doi:10.1186/1471-2458-6-183
24. Maio S, Gariazzo C, Stafoggia M, et al. BIGEPI project: environmental and health data. Epidemiol Prev 2023;47(6) Suppl 3:8-18. doi: 10.19191/EP23.6.S3.003
25. Gariazzo C, Carlino G, Silibello C, et al. Impact of different exposure models and spatial resolution on the long-term effects of air pollution. Environ Res 2021;192:110351. doi: 10.1016/j.envres.2020.110351
26. Ottone M, Broccoli S, Parmagnani F, et al. Source-related components of fine particulate matter and risk of adverse birth outcomes in Northern Italy. Environ Res 2020;186:109564. doi: 10.1016/j.envres.2020.109564
27. Monti P, Iodice S, Capuzzi E, et al. Short-term exposure to particulate matter and apparent temperature is associated with schizophrenia severity in Italy. Psychiatry Res 2025;353:116735. doi: 10.1016/j.psychres.2025.116735
28. European Union. 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. 20.11.2024. Available from: https://eur-lex.europa.eu/eli/dir/2024/2881/oj/eng