Unravelling the exposome of severe exacerbations of obstructive respiratory diseases in France Authors: Colas Tcherakian, Julien Cottet, Isabella Annesi-Maesano, Christiane Pochulu, Lea Antoniali, Aurélie Chekroun Martinot, Fayssoil Fouad, Rosa Falotico, Floriane Huret, Marcade Fulcrand Veronique, Mathieu Rosé, Thomas Sejourne, Priscille de la Tour, Nicolas Molinari Published: 2025 ABSTRACT AbstractIntroductionUnderstanding the influence of environmental and socioeconomic factors, collectively known as the exposome, on severe exacerbation risks in chronic obstructive pulmonary disease (COPD) and asthma is crucial for enhancing patient care.MethodsWe assessed the impact of the exposome on COPD and asthma severe exacerbations in France between 2018 and 2022, combining data from the French National Hospital Discharge Programme de Médicalisation des Systèmes d'Information database and open data for multiple environmental exposition. Using a retrospective matched case–control design, 473 990 patients with COPD and 187 332 patients with asthma were matched for sex, birth year and date of first hospitalisation. Conditional logistic regressions were performed to quantify ORs of exacerbations within the studied population.ResultsIn a multivariate model, exposure to extreme cold temperatures and exceeding WHO standards for nitrogen dioxide (NO2), fine particulate matter and ozone increased the odds of COPD exacerbation. Only exceeding NO2 was related to an increased risk of asthma exacerbation. Patients with COPD or asthma living in urban or artificial areas or near livestock farms were much more likely to have a severe exacerbation, whereas those living near water, wetlands or green spaces were protected from this risk. Higher poverty rates and long-term tobacco smoking were linked to a greater likelihood of COPD exacerbations.ConclusionsThe study highlights how exposome factors influence bronchial diseases and underscores the benefit of open-access data for advancing research. COPD and asthma exacerbations account for numerous deaths and cost billions and will increase with global warming, identifying modifiable environmental risk factors, improving patient care and helping shape valuable public health policies.Keywords: Pulmonary Disease, Chronic Obstructive; Asthma; Respiratory Infection; COPD Exacerbations; COPD epidemiology; Asthma Epidemiology WHAT IS ALREADY KNOWN ON THIS TOPIC Although the role of the exposome in modulating exacerbation risk in chronic obstructive pulmonary disease (COPD) and asthma has been already explored in various studies, data remain limited in France, highlighting the need for context-specific research to improve patient care. Although the role of the exposome in modulating exacerbation risk in chronic obstructive pulmonary disease (COPD) and asthma has been already explored in various studies, data remain limited in France, highlighting the need for context-specific research to improve patient care. WHAT THIS STUDY ADDS This study explores for the first time at the national level in France the temporal–spatial dynamics of combined exposome factors on chronic respiratory diseases. Results showed that living in urban or artificialised areas, municipalities with low-income rates, regions with high carbon dioxide emissions, exceedances of nitrogen dioxide, particulate matter (PM)10 and PM2.5 daily limits, and extreme temperatures increase the likelihood of asthma or COPD severe exacerbations. This study explores for the first time at the national level in France the temporal–spatial dynamics of combined exposome factors on chronic respiratory diseases. Results showed that living in urban or artificialised areas, municipalities with low-income rates, regions with high carbon dioxide emissions, exceedances of nitrogen dioxide, particulate matter (PM)10 and PM2.5 daily limits, and extreme temperatures increase the likelihood of asthma or COPD severe exacerbations. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY Deepening our knowledge of the impact of environmental exposures is crucial for advancing preventive medicine, especially for patients with chronic obstructive respiratory diseases. Deepening our knowledge of the impact of environmental exposures is crucial for advancing preventive medicine, especially for patients with chronic obstructive respiratory diseases. Introduction Approximately 392 million people worldwide are affected by chronic obstructive pulmonary disease (COPD) according to WHO, which was the third leading cause of death in 2019, responsible for 3.23 million deaths.1 In France, approximately 150 000 hospital admissions occur annually due to COPD exacerbations,2 and the condition is expected to affect 2.8 million individuals by 2025.3 Asthma affected an estimated 262 million people globally in 2019, resulting in 455 000 deaths.4 France reported 4 million asthma cases and about 30 000 hospital admissions annually due to disease exacerbations, with severe asthma greatly increasing mortality, morbidity and healthcare use.5,7 Both patients with COPD and patients with asthma can present severe disease manifestations, such as exacerbations (ie, flareups of worsening symptoms), which are associated with acute airway inflammation and may require hospitalisation. Asthma and COPD exacerbations share common physiological pathways involving type 2 inflammation (at least in a subset of patients), which has led to the concept of ‘treatable traits’.8 This common ground allows the two diseases to be studied in parallel within the same framework, not aiming for direct comparison but rather to examine the parallel effects of shared factors on both conditions. The ultimate goal is to provide evidence within the same study to support health policies targeting these common triggers, which could prevent a substantial number of exacerbations and yield simultaneous benefits across both diseases. Exacerbations place a substantial burden on patients, families and healthcare systems, impacting follow-up care, reducing patient productivity and increasing the need for public health interventions.1 9 10 Despite distinct pathogeneses, COPD and asthma share similarities as chronic progressive inflammatory airway disease that may worsen over time.1 11 Their aetiology involves complex interactions between genetic predisposition and environmental factors. About 40–50% of people who smoke tobacco lifetime will develop COPD, with >70% of COPD cases occurring in high-income countries.12 Inhalation of toxic particles and gases from household, occupational and air pollution also contributes.13,15 Asthma, which commonly manifests in childhood, has been linked to inhaled allergens, air pollution and other airborne irritants.16 Additional factors, such as weather (eg, cold/dry air, hot/humid air), viral infections and strenuous physical activity, have been identified as potential contributors to the disease onset and exacerbations.10 Socioeconomic factors and exposure to pollutants increase the risk, progression and management of both COPD and asthma.17,20 Low socioeconomic status, in combination with increased exposure to pollutants, has been associated with a higher risk of developing COPD.13 14 Evidence consistently highlights the contribution of occupational exposures, particularly in manual labour occupations, to COPD risk.15 Low income is independently associated with faster pulmonary disease progression and greater symptom burden in smokers, both with and without COPD.17 Poor nutrition and cumulative environmental insults further exacerbate disease.15 Inadequate housing conditions have been linked to a greater frequency of COPD exacerbations.20 Previous studies have shown that patients with asthma with lower socioeconomic status have substantially higher secondary healthcare utilisation, with suggestive evidence of increased exacerbations and mortality risk. These disparities have been consistently reported even in countries with universal healthcare systems.19 Low income is also associated with an increased risk of asthma and asthmatic symptoms, with notable gender differences: women with low income appear particularly vulnerable, exhibiting a higher risk. Additionally, the findings indicate a social gradient in asthma diagnosis.18 The concept of exposome, introduced by Dr Wild in 2005,21 encompasses environmental and lifestyle factors encountered throughout life that influence COPD and asthma onset and exacerbation. These diseases are influenced by air pollution, smoking, occupational hazards, allergens, climate and socioeconomic conditions, making the exposome a comprehensive tool for studying these multiple, often interacting, risk factors. Recent research has started exploring the potential causal relationship between environmental exposures, considering climate, geography and residential setting, and health outcomes. A 2019 global study covering 204 countries identified air pollution as a major contributor to chronic respiratory diseases, with COPD as the leading cause of mortality and asthma as the most prevalent condition.22 Subsequent studies indicate that particulate matter (PM) elevates the risk of long-term health effects, including new cardiovascular and respiratory diseases, as well as premature mortality, especially among the elderly.23 24 For asthma, exacerbations have been linked to air pollution and pollen, particularly in children.25,27 Rising global temperatures and urbanisation further exacerbate mortality rates from respiratory diseases.28 29 To date, studies assessing environmental and socioeconomic factors on COPD and asthma in France remain scarce, with slightly more research on asthma than on COPD.30 To our knowledge, no studies have examined these factors’ temporal–spatial dynamics on COPD and asthma severe exacerbations at the national level. The aim of the study is to determine using freely accessible data the impact of the exposome on the occurrence of hospitalisations for severe exacerbation of patients with COPD and/or asthma at the national level in France. The potential effects of the geographical location in which the subjects live and changes of weather conditions, such as extreme temperatures, are considered. Methods Data sources The Programme de Médicalisation des Systèmes d'Information (PMSI)—Medicine, Obstetric Surgery, and Odontology (French acronym MCO) database, which records all hospitalisations in France in all public and private healthcare centres, was used. Hospitalisations are recorded based on the International Classification of Diseases (ICD-10) classification31 (online supplemental tables S1 and S2). National environmental and socioeconomic data measurements, hereinafter referred to as the exposome data, available in open access were extracted. Exposome factors were categorised into two groups: non-fixed factors (NFs), which vary over time and include daily measurements of weather conditions (extreme temperatures, rainfall, humidity, wind speed), air pollutants (nitrogen dioxide (NO2), ozone (O3), plus PM10 and PM2.5, namely PM with a diameter <10 and 2.5 µm, respectively), pollen (alder and birch) levels; fixed factors (F), which in general remain constant over time for long periods and include tonnes of CO2/km2 for the greenhouse gas emissions, population density, geographical characteristics, land use, poverty rate and lung cancer hospitalisations (used as an indirect indicator of long-term tobacco smoking32). In the absence of direct smoking data, lung cancer hospitalisations were employed as an indirect indicator of long-term tobacco smoking.32 This approach is supported by evidence suggesting a correlation between such outcomes and chronic smoking behaviour. Notably, Desai et al33 validated claims-based algorithms for identifying smokers, demonstrating high specificity but limited sensitivity, thereby supporting their cautious use as a proxy measure in observational studies. Fixed factors reflect the static patient environment, while NFs fluctuate with time, weather, pollination and air quality. The list of open data sources and dates of their measurements is provided (online supplemental supplementa table S3). Data linkage and data management Patient-level PMSI and exposome fixed factors were linked by municipality, while NFs were linked using the nearest neighbourhood method between patient postcode from PMSI and monitoring station’s location. Details on the data linkage, missing value imputation and data management are available in the supplemental materials (online supplemental figure S1, online supplemental tables S4 and S5). Study design This retrospective observational matched case–control study focused on patients with COPD and asthma hospitalised between 1 January 2018 and 31 December 2022 in mainland France. The index date was defined as the date of the first hospitalisation for asthma or COPD during the study period. A lookback period of a minimum of 24 months prior to the index date was defined. The follow-up period started at the index date and lasted until the end of the study period or death. Study population Three separate populations were considered using specific ICD-10 codes: COPD, asthma and paediatric asthma (3–18 years old) (figure 1 and online supplemental table S1). Three populations of matched patients analysed separately. COPD, chronic obstructive pulmonary disease; PMSI, Programme de Médicalisation des Systèmes d'Information. In each population, a matching 1–2 case/controls for COPD or 1–3 for asthma was performed between individuals with at least one exacerbation and individuals with no exacerbation, based on gender, birth year and index date. This matching process aimed to identify a comparable date to consider the NF environmental factors for patients without exacerbation, ensuring that it fell within the same time period as the exacerbation experienced by the matched patients. For the matched patients without an exacerbation, the NF factors were measured using a ‘reference date’ selected based on the time between their inclusion and the exacerbation date of their peers (period of 15 and 5 days preceding COPD or asthma exacerbations, respectively). Statistical analysis Patient sociodemographic characteristics and fixed exposome factors were described at baseline. The primary endpoint was the occurrence of the first severe exacerbation for COPD or asthma assessed by an algorithm using ICD-10 codes specific to each population (online supplementaltable S2). Univariate conditional logistic regressions were performed to assess OR of exacerbations and its 95% CI. Multivariate analysis using stepwise selection of the variables was performed. All tests were two-sided and were considered significant at p=0.05. All analyses were performed with R software (V.4.3.0). Patient and public involvement Patients and/or the public were not directly involved in this study as the analysis was based on secondary data. Results Demographic characteristics of patients with and without exacerbations are presented below. The mean age of patients with COPD was 73 years (SD 12), 46 years (SD 30) for asthma and 7 years (SD 4) for paediatric asthma. 36.3% of patients with COPD were female, 41.0% for asthma and 59.5% for paediatric asthma. No difference in demographic characteristics was observed between patients with and without exacerbations. Geographical analysis revealed distinct patterns of exacerbations across French departments (figure 2). COPD exacerbations were concentrated in the north of the country, with the highest concentration in Normandy, followed by Île-de-France, Brittany and Hauts-de-France. Asthma and paediatric asthma exacerbation rates were predominantly observed in Île-de-France and Hauts-de-France. Departmental distribution of chronic obstructive pulmonary disease (COPD) and asthma severe exacerbation rate over 2018–2022. Rates of occurrence of the first hospitalisation for COPD or asthma exacerbation. The impact of exposome factors on COPD and asthma exacerbations is presented in tables1 2. Table 1Univariate analysis of the impact of exposome factors on COPD or asthma exacerbations: fixed factorsFixed factorsCOPD (n=473 990)Asthma (n=187 332)Paediatric asthma (n=52 045)%OR (95% CI)%OR (95% CI)%OR (95% CI)Social factorsEPCI poverty rate (%) <1013.7Ref14.1Ref15.0Ref ≥1086.31.06 (1.04 to 1.08)***85.91.27 (1.24 to 1.30)***85.01.36 (1.30 to 1.42)***Long-term tobacco smoking indicator No92.1Ref99.0RefRefNA Yes7.91.13 (1.1 to 1.15)***1.01.35 (1.22 to 1.5)*** NAGeographical locationPopulation density/km2 <50 inhab/km223.8Ref18.4Ref15.6Ref 50–10013.11.00 (0.98 to 1.02)12.11.02 (0.98 to 1.06)11.61.10 (1.01 to 1.19)*** 100–100031.40.98 (0.96 to 0.99)**29.81.07 (1.04 to 1.11)***30.41.16 (1.09 to 1.24)*** >100031.81.08 (1.06 to 1.09)***39.71.40 (1.35 to 1.44)***42.41.53 (1.44 to 1.63)***Living areas Rural47.8Ref40.4Ref37.4Ref Urban52.21.03 (1.02 to 1.04)***59.61.27 (1.24 to 1.30)***62.61.36 (1.30 to 1.42)***Municipality land typology Agricultural land50.1Ref45.8Ref45.1Ref Artificial areas33.01.05 (1.03 to 1.06)***40.51.26 (1.23 to 1.29)***42.71.29 (1.23 to 1.34)*** Forest and semi-natural areas16.50.96 (0.94 to 0.97)***13.50.95 (0.91 to 0.98)**12.00.91 (0.85 to 0.98)** Water and wetland areas0.50.81 (0.74 to 0.89)***0.21.05 (0.84 to 1.31)0.20.77 (0.46 to 1.29)Specialisation in agricultural production Mixed farming and/or livestock34.5Ref32.0Ref30.5Ref Livestock18.41.03 (1.01 to 1.05)**14.21.03 (0.99 to 1.07)12.20.99 (0.93 to 1.06) Crop production30.51.05 (1.03 to 1.06)***34.11.09 (1.06 to 1.12)***36.31.15 (1.09 to 1.20)*** Viticulture6.70.95 (0.92 to 0.97)***6.40.95 (0.91 to 0.99)*6.50.99 (0.91 to 1.08) No exploitation9.91.11 (1.09 to 1.14)***13.21.27 (1.23 to 1.32)***14.41.27 (1.20 to 1.36)*** Not classified0.11.30 (1.03 to 1.64)*0.10.78 (0.49 to 1.25)0.10.60 (0.27 to 1.35)Altitude (m) <20 6.3Ref6.5Ref7.1Ref 20–250 69.31.03 (1.01 to 1.06)**76.01.04 (0.99 to 1.09)78.90.99 (0.92 to 1.08) 250–500 16.40.97 (0.94 to 0.99)*11.90.95 (0.90 to 1.00)9.80.83 (0.75 to 0.92)*** 500–1000 6.50.98 (0.95 to 1.01)4.60.94 (0.88 to 1.00)3.40.82 (0.71 to 0.93)** >1000 1.41.03 (0.98 to 1.09)1.00.94 (0.84 to 1.06)0.80.77 (0.60 to 0.98)*Municipality with massif (ref: no) No84.1Ref88.7Ref90.9Ref Yes15.90.94 (0.93 to 0.96)***11.30.87 (0.84 to 0.90)***9.10.80 (0.74 to 0.86)***Air pollutionYearly total greenhouse gas emissions (CO2 equivalent/km2) <50030.6Ref24.7Ref21.8Ref 500–300032.81.01 (0.99 to 1.03)30.91.10 (1.06 to 1.13)***30.91.12 (1.06 to 1.19)*** 3000–10 000 00018.61.03 (1.01 to 1.05)***20.31.33 (1.29 to 1.38)***21.41.37 (1.28 to 1.45)*** >10 00018.01.12 (1.10 to 1.14)***24.01.48 (1.44 to 1.52)***25.81.62 (1.53 to 1.72)***Univariate models for COPD, asthma and paediatric asthma populations were adjusted for age, gender and date of exacerbation (index date).*p<0.05, **p<0.01, ***p<0.001.CO2, carbon dioxide; COPD, chronic obstructive pulmonary disease; EPCI, Établissement Public de Coopération Intercommunale (Public Establishment for Intercommunal Cooperation); NA, not applicable; Ref, reference. Univariate analysis of the impact of exposome factors on COPD or asthma exacerbations: fixed factors Univariate models for COPD, asthma and paediatric asthma populations were adjusted for age, gender and date of exacerbation (index date). *p<0.05, **p<0.01, ***p<0.001. CO2, carbon dioxide; COPD, chronic obstructive pulmonary disease; EPCI, Établissement Public de Coopération Intercommunale (Public Establishment for Intercommunal Cooperation); NA, not applicable; Ref, reference. Table 2Univariate analysis of impact of exposome factors on COPD or asthma exacerbations: non-fixed factorsNon-fixed factors†COPD (n=473 990)Asthma (n=187 332)Paediatric asthma (n=52 045) % OR (95% CI) % OR (95% CI) % OR (95% CI) Air pollutionExceeding NO2 WHO threshold No47.8Ref56.6Ref55.1Ref >25 µg/m3 24 hours52.21.06 (1.04 to 1.07)***43.41.17 (1.15 to 1.20)***44.91.27 (1.22 to 1.33)***Exceeding PM10 WHO threshold No87.0Ref94.6Ref94.9Ref >45 µg/m3 24 hours13.01.06 (1.04 to 1.08)***5.41.02 (0.98 to 1.07)5.11.17 (1.07 to 1.28)***Exceeding PM2.5 WHO threshold No42.8Ref66.0Ref66.9Ref >15 µg/m3 24 hours57.21.07 (1.05 to 1.08)***34.01.01 (0.99 to 1.04)33.11.06 (1.02 to 1.12)**Exceeding O3 100 µg/m3 24 hours threshold‡ Normal21.0Ref25.8Ref27.4Ref Pollution peak8.51.11 (1.08 to 1.14)***3.00.929 (0.87 to 1.00)*2.20.79 (0.68 to 0.92)** Neutral§70.51.06 (1.03 to 1.09)***71.20.95 (0.91 to 1.00)*70.40.90 (0.83 to 0.98)*WeatherTemperature Normal T° (0–30°C)87.3Ref94.7Ref95.3Ref Extreme cold<0°C11.61.07 (1.05 to 1.10)***5.10.98 (0.93 to 1.03)4.50.98 (0.88 to 1.09) Extreme heat>30°C1.01.05 (0.99 to 1.12)0.20.94 (0.75 to 1.17)0.20.87 (0.53 to 1.43) Extreme temperature<0°C or >30°C12.71.07 (1.05 to 1.09)***5.31.01 (1.00 to 1.019)4.70.99 (0.98 to 1.02)Precipitation Daily average (in cm)¶0.20.996 (0.966 to 1.028)0.21.07 (1.04 to 1.11)***0.21.076 (1.002 to 1.155)*Humidity Daily average (%)¶75.61.004 (1.003 to 1.005)***76.21.0007 (0.9995 to 1.0019)76.90.999 (0.997 to 1.002) Daily average (10%)¶7.61.04 (1.04 to 1.05)***7.61.007 (0.995 to 1.019)7.70.99 (0.98 to 1.02)Wind speed Daily average maximum wind speed (>10 min in m/s)§,¶6.61.008 (1.004 to 1.011)***6.61.015 (1.010 to 1.021)***6.61.012 (1.002 to 1.022)*PollenExceeding alder pollen threshold Normal47.5Ref48.4Ref42.8Ref >50 grains/m3 24 hours11.00.99 (0.97 to 1.02)5.80.96 (0.91 to 1.02)5.10.93 (0.83 to 1.04) Neutral††41.50.95 (0.92 to 0.98)***45.70.90 (0.86 to 0.95)***52.10.99 (0.89 to 1.11)Exceeding birch pollen threshold Normal49.8Ref48.4Ref43.0Ref >80 grains/m3 24 hours8.60.99 (0.97 to 1.02)5.81.02 (0.96 to 1.07)4.91.01 (0.91 to 1.13) Neutral††41.50.95 (0.92 to 0.98)45.70.90 (0.86 to 0.96)***52.10.99 (0.89 to 1.11)Univariate models for COPD, asthma and paediatric asthma populations were adjusted for age, gender and date of exacerbation (index date).*p<0.05, **p<0.01, ***p<0.001.†Daily measurements during the follow-up.‡The O3 100 μg/m3 threshold was applied as a conservative approximation of the WHO 8-hour guideline, given the limitations in available data.§Neutral period: O3 measurement outside the potential O3 pollution period June–September.¶Average over 15 days for COPD and 5 days for asthma and paediatric asthma.††Neutral period: pollen measurement outside the pollination period August–December.COPD, chronic obstructive pulmonary disease; NA, not applicable; NO2, nitrogen oxide; O3, ozone; PM, particulate matter; Ref, reference. Univariate analysis of impact of exposome factors on COPD or asthma exacerbations: non-fixed factors Univariate models for COPD, asthma and paediatric asthma populations were adjusted for age, gender and date of exacerbation (index date). *p<0.05, **p<0.01, ***p<0.001. †Daily measurements during the follow-up. ‡The O3 100 μg/m3 threshold was applied as a conservative approximation of the WHO 8-hour guideline, given the limitations in available data. §Neutral period: O3 measurement outside the potential O3 pollution period June–September. ¶Average over 15 days for COPD and 5 days for asthma and paediatric asthma. ††Neutral period: pollen measurement outside the pollination period August–December. COPD, chronic obstructive pulmonary disease; NA, not applicable; NO2, nitrogen oxide; O3, ozone; PM, particulate matter; Ref, reference. Considering the geographical location, living in high-density areas (>1000 inhabitants/km²) and artificial environments increased COPD exacerbation odds by 8% and 5%, while proximity to forests/semi-natural areas and wetlands lowered the odds by 4% and 19%, respectively. In rural areas, crop and livestock production raised the odds by 3% and 5%, whereas minimal agricultural activity increased the odds by 11%. Urban residency and living in flat areas (20–250 m) increased the odds by 3%, while higher altitudes (250–500 m) reduced them by 3%. For asthma, higher population density (>1000 inhabitants/km²) and artificial areas increased exacerbation odds by 40% and 26%, with stronger effects in children (53% and 29%). Living in crop production areas raised exacerbation odds by 9% overall and 15% in children, while minimal agricultural activity increased odds by 27% overall. Urban residency increased odds by 27% overall and 36% in children. In contrast, living near mountainous regions had a protective effect, reducing exacerbation odds by 13% overall and 20% in children. Elevated CO2 (>10 000 equivalent/km2) was linked to a 12% increase in COPD exacerbation odds. Exceeding thresholds for NO2, PM10 and PM2.5 raised the odds by 6–7% while ozone peaks were associated with an 11% increase in COPD exacerbation odds. In asthma, increased CO2/km2 was linked to higher exacerbation odds, starting with a 10% increase at 500–3000 tonnes of CO2 equivalent, 12% in paediatric asthma. Odds increased to 33% at 3000–10 000 tons and 48% above 10 000 tonnes, with stronger effects in paediatric asthma. Peaks in PM10 raised exacerbation odds by 2% and 17% in asthma and paediatric asthma. During pollen season, alder and birch pollen showed no significant impact on exacerbations. Considering meteorological factors, COPD exacerbation odds increased by 4% with high humidity and 7% with temperatures below 0°C. In asthma, maximum wind speed was linked to a 1.5% increase in exacerbation odds. While precipitation showed no impact on COPD exacerbations, it was associated with a 7% increase in asthma exacerbations and a 7.6% increase in paediatric cases. Long-term tobacco smoking was associated with a 13% increase in the odds of COPD exacerbations and 35% in the asthmatic population. Among the socioeconomic factors, municipalities with a poverty rate exceeding 10% were linked to a 6% increase in COPD exacerbation, with a stronger effect for asthma (27%) and paediatric asthma (36%). Figure 3 presents the relationships between the exposome and COPD or asthma exacerbations based on multivariate analysis. This confirmed a positive association between COPD exacerbations and NO2, PM2.5 and O3 excessive concentrations. Living in artificial areas, near crop and livestock production or in regions with minimal agricultural activity increased the likelihood of COPD exacerbations. Whereas living close to forest/semi-natural and wetland areas reduced the odds. Extreme cold temperatures were confirmed favouring COPD exacerbations. Among the socioeconomic factors considered, the likelihood of developing a COPD exacerbation was higher in a municipality with a poverty rate exceeding 10% and in cases of lung cancer. Relationship of exposome factors to chronic obstructive pulmonary disease (COPD) or asthma exacerbations in the multivariate model. ORs with 95% CIs. For COPD and asthma, analysis adjusted on average humidity (%), average maximum wind speed (>10 min in m/s) and average precipitation (in mm). For paediatric asthma, analysis adjusted on average maximum wind speed (>10 min in m/s). 1>25 μg/m3 24 hours, 2>15 μg/m3 24 hours, 3>100 μg/m3 24 hours. CO2, carbon dioxide; EPCI, Établissement Public de Coopération Intercommunale (Public Establishment for Intercommunal Cooperation); NO2, nitrogen dioxide; O3, ozone; PM, particulate matter. In asthma, only excessive concentrations of NO₂ were linked to increased exacerbation odds. Artificial area, including crop/livestock regions, minimally agricultural areas and urban areas, increased the likelihood of asthma exacerbations, whereas proximity to forest/semi-natural areas reduced the odds. Municipalities with poverty rates >10% were associated with a 10% increase in odds, and lung cancer was linked to a 35% increase. This association between NO2 and exacerbations was also found in asthmatic children. Excess of CO₂ increased exacerbation odds proportionally, reaching 50% increase at emissions exceeding 10 000 tonnes/km2, whereas living in mountainous municipalities reduced the odds. Discussion We identified key exposome factors (air pollution, artificial spaces and socioeconomic deprivation) associated with severe exacerbations of COPD and asthma in mainland France, highlighting the combined effects of various high-risk and protective factors. Data back our hypothesis that some triggers may be common in both asthma and COPD exacerbation, as they can trigger T2 inflammation. Short-term air pollution exposure was a consistent risk factor for both conditions, while extreme cold temperatures impacted only patients with COPD. Long-term tobacco smoking also contributed to the odds of exacerbation. Socioeconomic factors, particularly higher poverty rates, were associated with increased exacerbation odds. Geographical analysis revealed that COPD exacerbations were concentrated in northern regions, while asthma exacerbations were more prevalent in major urban areas. In rural areas, proximity to crop and livestock production increased the odds for both conditions. Conversely, certain geographical and environmental contexts acted as protective factors. Living near forests, semi-natural and wetland areas was protective against COPD exacerbations, while mountainous regions appeared to protect patients with paediatric asthma. Green spaces appear to reduce the odds of exacerbations in COPD, while also offering protection for patients with asthma, despite the potential higher seasonal pollen peaks. Our results align with previous regional studies emphasising the protective effects of green space. Notably, no significant impact was observed for the analysed allergens on asthma exacerbations, likely due to the limited pollen types considered. This supports the hypothesis that allergens play a role in the genesis of the disease rather than in the flares and that viral infections play a crucial role in exacerbating asthma symptoms, often more important than that of allergens such as pollen.34 This is the first study to present a comprehensive analysis of the impact of exposome on the occurrence of severe exacerbation of chronic bronchial disease at the national level in France. The main strengths stem from the large population from a wide geographical area, including >23 000 participants using standardised air pollution estimates at the municipality level for multiple air pollutants. Additionally, integrating various open-access datasets with nationwide hospitalisation data strengthens the analysis, capturing the full range of geographical disparities across the country. This national approach was enabled by the growing availability and quality of freely accessible environmental and socioeconomic data and the opportunity to enrich the PMSI databases. While this study provides valuable insights, several limitations should be mentioned. PMSI data are collected through standardised information systems implemented in hospitals. Open data, in contrast, are generated from various devices or platforms that capture information at the geographical level or based on patient self-reporting. The authors are aware of the quality and limitations of the data used and selected official sources, primarily published by public institutions, which support the robustness of the current dataset. Nevertheless, the authors acknowledge that environmental data remain sparse and emphasise the need for higher-quality, more integrated data to further enrich research in this field. Our analysis focused on the first hospitalisation for exacerbation, without evaluating subsequent potential relapses. Assessing the impact of exposome on the dynamics of repeated exacerbations is a key area for future research to enrich its understanding, as patients with multiple exacerbations represent a high-risk subgroup with poorer prognosis. A key limitation of this study is the absence of detailed, individual-level socioeconomic and behavioural data, which likely attenuates the observed effect estimates. Collecting such data would require privacy-sensitive procedures and highly personalised measurements—types of data that are not currently available as open resources and that raise General Data Protection Regulation compliance concerns. This limitation is explicitly acknowledged, and future research should prioritise the use of consented, privacy-preserving approaches, as well as open-access, high-quality datasets when they become available. More broadly, many epidemiological studies are constrained by the lack of accessible, granular data, and advancing the field will require improved data sharing and infrastructure. Limitations inherent to the PMSI database should be acknowledged. While comprehensive, it lacks precise address data, making the geographical distribution of COPD and asthma exacerbations an estimation. PMSI data primarily focus on hospital stays. Information on treatment patterns or use of bronchodilators to prevent acute crisis is not available. The database captures diagnoses, but does not differentiate phenotypes, which may influence the response to environmental factors. Reliance on lung cancer hospitalisations as a proxy for long-term tobacco smoking partially addresses tobacco-related risks but does not account for early-life or youth smoking history,35 which is a notable limitation given its role in the development of COPD and asthma in adulthood. The limited availability of structured and comprehensive smoking data necessitated the use of proxy measures, such as disease-specific hospitalisations. While these proxies provide useful indirect insights into cumulative tobacco exposure, they may not fully reflect the duration or intensity of individual smoking behaviours, potentially leading to exposure underestimation and misclassification, which could in turn attenuate the observed associations. Our reliance on open-access data also imposed constraints. Pollen data were limited to a few species (birch and alder). Key asthma-inducing pollens were not studied (hazel, olive and cypress), as well as herbaceous and urticaceous plants (grass, mugwort and ambrosia). Additionally, pollens known to have a potentially greater effect on asthma acute outcomes (grasses, ragweed)36 37 were also not available. Last, whereas we wanted to outline the role of a comprehensive exposome, we did not study a potential synergistic effect of all pollutants together, as a ‘cocktail effect’. Regarding ozone (O3), the decision to retain the 100 µg/m³ threshold was constrained by data availability. Hourly O3 concentrations were not accessible, preventing the computation of 8-hour rolling averages as recommended by WHO. In the absence of sub-daily measurements, we conservatively applied the WHO 8-hour guideline value (100 µg/m³) to the available 24-hour mean concentrations. We acknowledge that this approach represents a stricter criterion than the official metric; however, it ensures a transparent and health-protective interpretation of our findings. It should also be acknowledged that population mobility represents an important limitation, as individuals may relocate over time, which implies that their actual exposure to environmental factors cannot be measured with complete accuracy. Consequently, conclusions regarding the long-term effects of these exposures cannot be definitively extrapolated. Indoor air quality (inhalation of toxic particles and gases from household air pollution) and professional exposome exposures, both critical risk factors for respiratory disorders, were not available. Finally, although data from literature and current pathophysiological knowledge show that asthma exacerbations are more frequently correlated with viral exposure than with pollen exposure, the results of our study are limited as viral incidence was not analysed. Existing public health initiatives, such as the Recosanté digital service, help French citizens, particularly those sensitive to poor air quality, understand their environment and take daily protective measures.38 Our study identified potential actionable insights that could enhance public health strategies. These findings could be beneficial for both patients and clinicians, especially in primary preventive care and exacerbation trigger avoidance. This evidence could inform the development of real-time alert systems for patients with chronic obstructive respiratory disease, providing warnings about the potentially dangerous environmental conditions, including extreme temperatures, air pollution peaks (NO2, PM2.5, O3) and pollen peaks. Better-informed patients could actively manage their condition by addressing preventable factors affecting symptoms and exacerbations. Several research teams worldwide have developed methods to calculate the ‘environmental burden’ of diseases, including at the WHO, the Organisation for Economic Co-operation and Development (OECD) and the University of Washington School of Medicine.39 Our study could support such approaches for chronic respiratory diseases and then quantify the overconsumption of hospital care attributable to environmental factors. This analysis could quantify the health and economic cost of actions (eg, a new therapeutic approach) or inaction (eg, not taking measures). Our study focuses on exacerbations requiring hospitalisation as PMSI is limited to information relating to the patient’s hospital stay. A complementary study including outpatient data could collect information on medication use or retail consultations, building endpoints to assess the exposome’s impact on patients with symptoms other than severe exacerbations. Conclusion Exposome influences the health and well-being of individuals and populations. Deepening our knowledge of the nature and impact of environmental exposures is crucial for advancing preventive medicine, especially for patients with chronic obstructive respiratory diseases. Reducing exposure to air pollution through stricter environmental regulations and improved urban planning is essential to prevent COPD and asthma exacerbations. Reduced indoor pollutants can help limit respiratory risks and support disease prevention. Addressing socioeconomic deprivation by ensuring equitable access to healthcare, education and healthier living conditions is crucial to reduce disparities in COPD and asthma outcomes. With climate change, the frequency and intensity of extreme temperature variations will increase. This opens avenues for research on future developments based on the Intergovernmental Panel on Climate Change (IPCC) recommendations, environmental and health prediction models, and preparing for the impacts of climate change. Supplementary material Acknowledgements We thank Sanofi and Regeneron Pharmaceuticals Inc. for subsidising this study. The funders did not influence the results/outcomes of the study despite author affiliations with the funders. Footnotes Funding: The study was sponsored and funded by Sanofi and Regeneron Pharmaceuticals Inc. Provenance and peer review: Not commissioned; externally peer reviewed. Patient consent for publication: Not applicable. Ethics approval: This study was conducted according to the current legislation and national recommendations. The use of data from the PMSI was authorised by the National Commission for Data Processing and Civil Liberties (CNIL). Sanofi has a compliance commitment to MR-006, which exempts from committee. Data availability free text: The datasets generated for this study can be found in the PMSI database upon request to regulatory authorities. Map disclaimer: The depiction of boundaries on this map does not imply the expression of any opinion whatsoever on the part of BMJ (or any member of its group) concerning the legal status of any country, territory, jurisdiction or area or of its authorities. This map is provided without any warranty of any kind, either express or implied. Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research. Data availability statement No data are available. References Associated Data Supplementary Materials Data Availability Statement No data are available.