Abstract
This study investigated the associations of air pollution and residential greenspace with immune thrombocytopenic purpura (ITP) risk, along with their combined effects, in a cohort of 356,482 UK Biobank participants free of ITP at baseline. Ambient nitrogen dioxide (NO2), nitrogen oxides (Nox), fine particulate matter (PM2.5), coarse particulate matter with aerodynamic diameters ranging between 2.5 and 10 μm (Pmcoarse) and particulate matter with diameters of less than 10 μm (PM10) exposures were estimated by land-use regression models and residential greenspace was calculated using land-use data, defined as the percentage of outdoor greenspace surrounding each participant’s home location. The hazard ratios (HR) and 95% confidence intervals (CI) were estimated by using Cox proportional hazard models, and non-linear relationships were assessed using restricted cubic spline (RCS) curves. A total of 500 incident ITP cases were diagnosed during a median follow- up of 13.54 years. Long-term exposure to high ambient concentrations of PM2.5 (HR=1.15, 95% CI: 1.04-1.28; P=0.007), NO2 (HR=1.23, 95% CI: 1.10-1.37; P=1.83×10-4), and NOx (HR=1.12, 95% CI: 1.03-1.21; P=0.011), as well as low residential greenspace (HR=0.77, 95% CI: 0.67-0.87; P=7.96×10-5), were associated with an increased risk of ITP. RCS curve revealed a non-linear relationship of PM10 and NOx with ITP risk (P for non-linearity=0.003 for PM10 and =0.030 for NOx). Participants with high air pollution and low residential greenspace had the highest risk of ITP, though no evidence of mediation or interaction effects were observed. In conclusion, long-term exposure to ambient PM2.5, PM10, NO2 and NOx may increase ITP risk, whereas residential greenspace may decrease this risk.
Introduction
Immune thrombocytopenic purpura (ITP), also described as idiopathic thrombocytopenic purpura, is an acquired autoimmune disease characterized by a decrease in platelet count due to excessive platelet destruction and/or impaired platelet production.1 From 1966 to 2009, the global incidence of ITP in children ranged from 0.5 to 10.5 cases per 100,000 person-years, and in adults it ranged from 1.6 to 3.9 cases per 100,000 person-years, respectively.2 Although the incidence of ITP was relatively low, it substantially impaired patients’ quality of life and negatively affected their emotional well-being, social activities, work, and productivity.3 Moreover, severe ITP can also lead to substantial bleeding, such as intracranial and gastrointestinal bleeding, increasing the risk of infection, thrombosis, and death.4-6 Given the absence of a cure, current treatment strategies for ITP primarily focus on controlling the condition. Therefore, identifying modifiable risk factors is crucial for preventing this disease.
Emerging epidemiological studies have indicated that longterm exposure to ambient air pollution facilitated the occurrence of various autoimmune diseases such as rheumatoid arthritis, chronic obstructive pulmonary disease, multiple sclerosis, and inflammatory bowel disease.7-10 In vivo and in vitro studies have also demonstrated that prolonged exposure to air pollution could trigger oxidative stress, induce inflammation, and disrupt immune regulatory pathways, thereby leading to the onset of autoimmune diseases.11 However, studies on the relationship between air pollution and ITP have remained extremely limited. To date, only one matched case-control study from Taiwan reported that maternal exposure to particulate matter with diameters of less than 10 µM (PM10) during pregnancy increased the risk of childhood ITP.12 The association between air pollution and the risk of ITP in adults remains unknown. Residential greenspace has been associated with various health outcomes by mitigating environmental hazards such as air pollution, noise and heat, as well as promoting physical activity and mental well-being.13,14 Nevertheless, currently there is no evidence linking residential greenspace to the risk of incident ITP. Despite the spatial correlation between residential greenspace and air pollution, few studies have simultaneously considered residential greenspace and air pollution exposure and explored the potential mechanisms by which they influence outcomes.15 In this study, we hypothesized that residential greenspace may influence ITP risk through multiple potential pathways: (i) as an independent protective factor; (ii) with air pollution acting as a mediator in the association between residential greenspace and ITP risk; or (iii) in synergy with air pollution to modify the overall ITP risk.
Collectively, this study aimed to utilize the UK Biobank (UKB) to investigate the individual or combined associations of long-term exposure to ambient air pollutants and residential greenspace with the risk of incident ITP.
Methods
Study design and population
UKB is a cohort study that enrolled approximately 0.5 million participants from 2006 to 2010.16 Participants completed touchscreen questionnaires, physical measurements and provided biological samples.17 Ethical approval was obtained from the North West Multi-center Research Ethics Committee, and detailed information on study design and data collection is available in the Online Supplementary Appendix.
Assessment of air pollution and residential greenspace
The concentrations of fine particulate matter (PM2.5), coarse particulate matter with aerodynamic diameters ranging between 2.5 and 10 µm (PMcoarse), PM10, nitrogen dioxide (NO2), and nitrogen oxide (NOx) were estimated by the UK Small Area Health Statistics Unit using land use regression (LUR) models.18,19 Residential greenspace exposure for each participant was determined by calculating the percentage of greenspace within 300 m and 1,000 m buffers around their homes using the 2005 Generalized Land Use Database (GLUD) for England. Following Natural England’s guideline for greenspace access within 300 m, this study focused mainly on the 300 m buffer (field ID: 24503).20 Further information on the LUR models and greenspace assessments is provided in Online Supplementary Appendix.
Ascertainment of immune thrombocytopenic purpura
Incident ITP was defined by the International Classification of Diseases 10th Revision (ICD-10) code D69.3. Participants were followed from their initial visit to the assessment centers until the time of diagnosis of ITP, loss to follow-up, death, or the end of the follow-up (Oct 31, 2022), whichever came first.
Statistical analysis
Baseline characteristics were summarized as numbers (percentages) for categorical variables and as mean ± standard deviation (SD) or median ± interquartile range (IQR) for continuous variables. Pearson correlation analysis was used to explore the correlations of air pollutants with greenspace. The associations of environmental factors with ITP were evaluated using multivariable Cox proportional hazard models. The assumption of proportional hazards was tested using Schoenfeld residuals (Online Supplementary Figure S2). Covariates were selected based on biological plausibility and prior studies,21-23 with details provided in Online Supplementary Table S1. Model A was adjusted for age and sex, and Model B was additionally adjusted for ethnicity, body mass index (BMI), education, household income, alcohol consumption, smoking, and physical activity. Non-linear dose-response relationships were explored using restricted cubic spline (RCS) models with three knots, selected based on Akaike Information Criterion.24 Interactions between air pollutants and greenspace were assessed using multiplicative and additive models, and additive interactions were evaluated by relative excess risk due to interaction (RERI) and the attributable proportion due to interaction (AP).25,26 Mediation analysis examined air pollution’s role in greenspace and ITP associations. Details of interaction and mediation analysis are provided in the Online Supplementary Appendix. We also conducted stratification analyses and sensitivity analyses excluding participants who developed ITP within the first year, resided at the same location for <10 years, or had prior diagnoses of other purpura and hemorrhagic conditions (ICD10: D69.0-D69.2, D69.4-D69.9). Analyses also incorporated residential greenspace buffer at 1,000 m (field ID: 24500) to examine broader residential greenspace coverage and ITP risk.
Two-sided P values of <0.05 were considered statistically significant, and all analyses were performed in R (v4.3.1) using the survival, interactionR, and regmedint packages.
Results
As shown in Online Supplementary Figure S1, a total of 356,482 participants free of ITP at baseline were included in this study. Table 1 presents the baseline characteristics of study participants. The mean age was 56.21 years (SD 8.08), and 52.5% (N=187,218) were female. The majority of the participants reported being previous or current consumers of alcohol (96.3%) and engaging in regular physical activity (71.5%). Over a median follow-up time of 13.54 years (4,711,602 person-years), 500 new-onset cases of ITP were identified. The median concentrations of ambient PM2.5, PMcoarse, PM10, NO2, and NOx were 9.92 µg/m3 (IQR, 9.28-10.55), 6.10 µg/m3 (IQR, 5.84-6.62), 16.02 µg/m3 (IQR, 15.23-16.98), 26.04 µg/ m3 (IQR, 21.28-31.20), and 42.07 µg/m3 (IQR, 34.04-50.64), respectively. Additionally, residential greenspace values were higher within a larger buffer. Specifically, within the 300 m and 1,000 m buffer, the median residential greenspace was 29.81% (IQR, 17.32-48.86) and 42.05% (IQR, 27.61-60.43), respectively. Strong correlations were observed for NO2 and NOx with PM2.5 (r=0.87 for NO2, r=0.85 for NOx, respectively), and moderate correlations were found with PM10 (r=0.51 for NO2, r=0.52 for NOx, respectively). Residential greenspace buffers at 300 m and 1,000 m were highly correlated (r=0.85). Meanwhile, moderately negative correlations were observed for ambient PM2.5, NO2, NOx with residential greenspace in different buffer sizes (Table 2).
Table 3 summarizes the associations between air pollutant concentrations, residential greensapce and the risk of incident ITP in multivariable-adjusted models. Per IQR increase in the concentrations of ambient PM2.5 (HR=1.15, 95% CI: 1.04-1.28; P= 0.007), NO2 (HR=1.23, 95% CI: 1.10-1.37; P=1.83×10-4), and NOx (HR=1.12, 95% CI: 1.03-1.21; P=0.011) were associated with an increased risk of incident ITP, while no statistically significant associations were observed for PMcoarse (HR=0.98, 95% CI: 0.91-1.06; P=0.679) and PM10 (HR=1.04, 95% CI: 0.96-1.13; P=0.300). Long-term exposure to residential greenspace buffer at 300 m was associated with a reduced risk of incident ITP, with each IQR increase in residential greenspace linked to a 23% (95% CI: 13-33; P=7.96×10-5) decrease in the risk of ITP. Similarly, using a wide-area residential greenspace buffer at 1,000 m showed a 24% (95% CI: 12-34; P=2.54×10-4) reduction in the risk of incident ITP (Online Supplementary Table S2). Consistent associations were also observed in the analysis of per 10 µg/m3 increase in air pollutant concentrations or per 10% increase in residential greenspace buffer at 300 m (Online Supplementary Table S3). Stratified analyses by age, sex, BMI, household income, alcohol drinking status and physical activity showed that none of the baseline characteristics significantly modified the associations of ambient air pollutants or residential greenspace buffer at 300 m with the risk of incident ITP (all P for heterogeneity >0.05) (Online Supplementary Figure S3). Additionally, a series of sensitivity analyses further supported the primary findings, reinforcing the robustness of the study (Online Supplementary Tables S4-S6).
Table 1.Baseline characteristics of participants from UK Biobank at baseline.
Table 2.Descriptive statistics and Pearson correlation matrix for pollutants and residential greenspace.
Table 3.The associations between five air pollutants and residential greenspace buffer at 300 m with per interquartile range increase and the risk of immune thrombocytopenic purpura.
Restricted cubic spline models were used to evaluate potential non-linear associations between these environmental factors and ITP risk (Figure 1). Statistically significant dose-response associations were found for PM2.5, PM10, NO2, NOx, and residential greenspace with ITP risk (all P for overall <0.05). Notably, PM10 and NOx demonstrated non-linear associations with ITP risk (P for non-linearity=0.003 for PM10 and =0.030 for NOx), with PM10 showing an inverted U-shaped curve and NOx displaying a L-shaped curve, indicating potential differences in risk across different concentration levels.
To further investigate these non-linear associations, we categorized participants into quartiles based on their baseline PM10 and NOx exposure levels. As shown in Online Supplementary Table S7, compared with participants in the lowest quartile of PM10 exposure group (Q1), those in the second (Q2), the third (Q3) and the highest exposure group (Q4) showed 1.46, 1.38 and 1.32-fold increased risk of ITP, respectively. Although the trend test did not reach statistical significance (P for trend =0.098), these quartile-based results suggest a concentration-specific risk pattern for PM10. Similarly, for NOx, participants in the third quartile (42.10-50.60 µg/m3) showed the strongest association with ITP risk (HR=1.54, 95% CI: 1.20-1.98; P=7.29×10-4) (P for trend =0.003).
We also observed that the risk of incident ITP exhibited a dose-response pattern in relation to the joint effects of air pollutants and the residential greenspace buffer at 300 m. As illustrated in Figure 2, the trend of increased risk of incident ITP persisted with higher air pollution levels or decreased residential greenspace (all P for trend <0.05). Compared to the reference group characterized by low air pollution concentrations and high residential green-space buffer at 300 m, the group with high air pollution concentrations and low residential greenspace buffer at 300 m had an increased risk of incident ITP. Specifically, the risk of incident ITP increased by 44% (HR=1.44, 95% CI: 1.17-1.78; P=0.001), 36% (HR=1.36, 95% CI: 1.09-1.70; P=0.006), 56% (HR=1.56, 95% CI: 1.26-1.92; P=3.01×10-5), and 58% (HR=1.58, 95% CI: 1.28-1.95; P=2.25×10-5) in the groups with high concentrations of PM2.5, PM10, NO2, and NOx and low residential greenspace buffer at 300 m, respectively. However, we did not observe any statistically significant interaction between air pollution or residential greenspace and the risk of ITP (Table 4). The mediation analysis indicated that PM2.5, PM10, NO2, and NOx mediated 13.30%, -1.49%, 18.53%, and -0.18% of the association between residential greenspace and incident ITP, respectively. However, the natural indirect effects (NIE) were not statistically significant (Figure 3).
Discussion
In this study, we observed that long-term exposure to high ambient concentrations of PM2.5, PM10, NO2, and NOx, as well as lower residential greenspace, was associated with an increased risk of incident ITP. Furthermore, individuals with high air pollution concentrations and low residential greenspace coverage experienced the highest relative increase in ITP risk.
Scarce studies have explored the associations between air pollution and the risk of incident ITP. A case-control study involving 917,359 children from Taiwan revealed positive associations of prenatal exposure to PM10 (odds ratio [OR] =1.001, 95% CI: 1.000-1.002; P=0.039 per 10 µg/m3) and the Pollution Standard Index (PSI) (OR= 1.016, 95% CI: 1.001-1.031; P=0.032) with the risk of childhood incident ITP.12 Our study demonstrated that for per IQR increment, PM2.5, NO2, and NOx were associated with a 15%, 23%, and 12% increased risk of incident ITP, respectively. For PM10, we observed an inverted U-shaped relationship with ITP, with the most prominent effect observed in the second quartile (15.20-16.00 µg/m3), where the risk of ITP increased by 46%. The second quartile appeared to be a critical concentration range where PM10 exerted its most significant impact on adult ITP risk. This nuanced relationship between PM10 and ITP might suggest that different biological mechanisms are triggered at varying levels of exposure. Notably, this critical concentration range falls below the World Health Organization guideline value for PM10 (20 µg/m3), suggesting that PM10 may pose a considerable risk for ITP even at relatively low levels.27 These findings underscore the importance of maintaining stringent air quality standards to mitigate potential health risks.
Figure 1.Dose response curves of associations between air pollutants and residential greenspace using restrict cubic splines. (A-F) The model adjusted for age, sex, ethnicity, body mass index, education level, household income, smoking status, alcohol drinking status, and physical activity. NO2: nitrogen dioxide; NOx: nitrogen oxides; PM2.5: fine particulate matter; PMcoarse: coarse particulate matter with aerodynamic diameters ranging between 2.5 and 10 µm; PM10: particulate matter with diameters of less than 10 µm.
Figure 2.The joint effects between air pollution and residential greenspace buffer at 300 m on incident immune thrombocytopenic purpura. (A) PM25 and residential greenspace buffer al 300 m. (B) PM10 and residential greenspace buffer at 300 m. (C) NO2 and residential greenspace buffer at 300 m. (D) NOx and residential greenspace buffer at 300 m. *Reference group: high residential greenspace buffer at 300 m and low air pollution; group 1: low residential greenspace buffer at 300 m and low air pollution; group 2: high residential greenspace buffer at 300 m and high air pollution; group 3: low residential greenspace buffer at 300 m and high air pollution. NO : nitrogen dioxide; NOX : nitrogen oxides; PM2.5 : fine particulate matter; PM : particulate matter with diameters of less than 10 μm.
Previous studies have proposed potential mechanisms by which air pollution can trigger autoimmune diseases. Given the multiple immune dysregulations involved in ITP, such as abnormalities in various immune-related factors and effector cells, air pollution may induce ITP by stimulating chronic systemic inflammation and inducing oxidative stress through direct oxidation of proteins and lipids or activation of oxidative cell pathways.28,29 Additionally, it might regulate the function and phenotype of dendritic cells, leading to an imbalance between dendritic cells and T cells. These mechanisms aligned with the observed abnormalities in precursor helper T cells, HLA-DR+ T cells, and the elevated levels of soluble interleukin-2 receptors in ITP patients.30,31
Table 4.The interaction effects between air pollutants and the residential greenspace buffer at 300 m.
In addition to the negative association between residential greenspace buffer at 300 m and the risk of incident ITP, our analysis extended to a larger buffer at 1,000 m. As expected, we found a similar protective effect against incident ITP for both buffer sizes, suggesting that the extent of residential greenspace did not significantly impact its ability to mitigate the risk of incident ITP. Although previous studies have hinted at greenspace to potentially mediate the risk reduction related to reduced air pollution, our study did not find evidence supporting greenspace as a mediator in the associations between air pollution and the risk of ITP.32,33 The mediation analysis in this study suggested that residential greenspace primarily reduced the risk of ITP through direct effects rather than through the mediating effects of air pollutants. This result underscored the value of increasing and maintaining greenspace in urban planning, particularly in densely populated areas with limited greenspace resources, providing a scientific foundation for enhancing residents’ health. Furthermore, this study investigated the combined effects of air pollution and residential greenspace on incident ITP. The findings demonstrated an increased risk of incident ITP when air pollution levels were high and the residential greenspace was low, compared to the reference group. However, our study did not find any interaction between air pollution and residential greenspace.
To the best of our knowledge, our study is the inaugural large-scale prospective cohort investigation of the associations of air pollution and residential greenspace with the risk of ITP. However, several potential limitations warrant attention. Firstly, the measurements of certain air pollutant concentrations such as PM2.5 and residential greenspace exposure were limited to a single baseline year, hindering the capture of dynamic changes before and after the baseline. We assumed that spatial patterns of air pollution exposure in the UK, a highly urbanized and industrialized country with slow-changing land use characteristics like greenspace, remained relatively stable over the years; however, exposure misclassification potential may still exist.34,35 Secondly, the measurement data relied on participants’ residential addresses, overlooking their activity patterns and residential mobility, which might not reflect their actual exposure levels. However, we performed a sensitivity analysis limited to participants who had resided in the same location for over 10 years, and the findings showed no substantial changes. Future cohort studies are encouraged to conduct multiple measurements of air pollution and residential greenspace across different times and locations to better understand their impact on the risk of incident ITP. Thirdly, the study was limited in its analysis of the associations between different types of greenspaces and the risk of incident ITP due to insufficient information about greenspace types, quality, and specific species from GLUD. Therefore, more advanced techniques and rigorous methods are needed for accurate measurement and characterization of greenspace. Fourthly, although stringent control of potential confounding factors in this study, the associations might be influenced by other unknown or unmeasured confounding factors. Lastly, our study’s participant demographics were restricted to middle-aged and elderly individuals of European descent, which may introduce a bias in ITP prevalence estimates and limit generalizability to the broader UK population and other ethnicities and regions.
In this large-scale population-based cohort study, we observed positive associations between long-term exposure to high levels of air pollutants such as PM2.5, PM10, NO2, and NOx and an increased risk of incident ITP, while long-term exposure to high residential greenspace showed a negative association with ITP risk. Our research provided evidence for preventing ITP by improving environmental conditions, including increasing residential greenspace and mitigating air pollution. Further research is needed to delve deeper into the potential mechanisms between environmental factors and incident ITP.
Footnotes
- Received July 24, 2024
- Accepted November 27, 2024
Correspondence
Disclosures
No conflicts of interest to disclose.
Funding
This study was jointly supported by grants from National Natural Science Foundation of China (82174208, 81973663 and 82274273), Natural Science Foundation of Zhejiang Province (LY22H260005) and Natural Science Exploration Project of Zhejiang Chinese Medical University (2023JKZ-KTS14). The funding body had no role in the study design, data analysis, interpretation of data, or preparation of the manuscript.
Acknowledgments
The authors sincerely thank the individuals for their participation and researchers for their managements in the UK Biobank.
References
- Rodeghiero F, Stasi R, Gernsheimer T. Standardization of terminology, definitions and outcome criteria in immune thrombocytopenic purpura of adults and children: report from an international working group. Blood. 2009; 113(11):2386-2393. Google Scholar
- Terrell DR, Beebe LA, Vesely SK, Neas BR, Segal JB, George JN. The incidence of immune thrombocytopenic purpura in children and adults: a critical review of published reports. Am J Hematol. 2010; 85(3):174-180. Google Scholar
- Cooper N, Kruse A, Kruse C. Immune thrombocytopenia (ITP) World Impact Survey (I-WISh): Impact of ITP on health-related quality of life. Am J Hematol. 2021; 96(2):199-207. Google Scholar
- Kohli R, Chaturvedi S. Epidemiology and clinical manifestations of immune thrombocytopenia. Hamostaseologie. 2019; 39(3):238-249. Google Scholar
- Frederiksen H, Maegbaek ML, Nørgaard M. Twenty-year mortality of adult patients with primary immune thrombocytopenia: a Danish population-based cohort study. Br J Haematol. 2014; 166(2):260-267. Google Scholar
- Moulis G, Rueter M, Duvivier A. Difficult-to-treat primary immune thrombocytopenia in adults: prevalence and burden. Results from the CARMEN-France registry. Br J Haematol. 2024; 204(4):1476-1482. Google Scholar
- Adami G, Pontalti M, Cattani G. Association between long-term exposure to air pollution and immune-mediated diseases: a population-based cohort study. RMD open. 2022; 8(1):e002055. Google Scholar
- Zhang J, Fang XY, Wu J. Association of combined exposure to ambient air pollutants, genetic risk, and incident rheumatoid arthritis: a prospective cohort study in the UK Biobank. Environ Health Perspect. 2023; 131(3):37008. Google Scholar
- Li FR, Wu KY, Fan WD, Chen GC, Tian H, Wu XB. Long-term exposure to air pollution and risk of incident inflammatory bowel disease among middle and old aged adults. Ecotoxicol Environ Saf. 2022; 242:113835. Google Scholar
- Gu X, Jing D, Xiao Y. Association of air pollution and genetic risks with incidence of elderly-onset atopic dermatitis: a prospective cohort study. Ecotoxicol Environ Saf. 2023; 253:114683. Google Scholar
- Glencross DA, Ho TR, Camiña N, Hawrylowicz CM, Pfeffer PE. Air pollution and its effects on the immune system. Free Radic Biol Med. 2020; 151:56-68. Google Scholar
- Yen HC, Lin CH, Lin MC, Hsu YC, Lin YH. Prenatal exposure to air pollution and immune thrombocytopenia: a nationwide population-based cohort study. Front Pediatr. 2022; 10:837101. Google Scholar
- Yang BY, Zhao T, Hu LX. Greenspace and human health: an umbrella review. Innovation (Camb). 2021; 2(4):100164. Google Scholar
- Markevych I, Schoierer J, Hartig T. Exploring pathways linking greenspace to health: theoretical and methodological guidance. Environ Res. 2017; 158:301-317. Google Scholar
- Hystad P, Davies HW, Frank L. Residential greenness and birth outcomes: evaluating the influence of spatially correlated built-environment factors. Environ Health Perspect. 2014; 122(10):1095-1102. Google Scholar
- Collins R. What makes UK Biobank special?. Lancet. 2012; 379(9822):1173-1174. Google Scholar
- Sudlow C, Gallacher J, Allen N. UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 2015; 12(3):e1001779. Google Scholar
- Eeftens M, Beelen R, de Hoogh K. Development of land use regression models for PM2.5, PM2.5 absorbance, PM10 and PMcoarse in 20 European study areas; results of the ESCAPE project. Environ Sci Technol. 2012; 46(20):11195-11205. Google Scholar
- Beelen R, Hoek G, Vienneau D. Development of NO2 and NOx land use regression models for estimating air pollution exposure in 36 study areas in Europe - the ESCAPE project. Atmos Environ. 2013; 72:10-23. Google Scholar
- Annerstedt van den Bosch M, Mudu P, Uscila V. Development of an urban green space indicator and the public health rationale. Scand J Public Health. 2016; 44(2):159-167. Google Scholar
- Li F-R, Zhu B, Liao J. Ambient air pollutants and incident microvascular disease: a cohort study. Environ Sci Technol. 2022; 56(12):8485-8495. Google Scholar
- Boudier A, Markevych I, Jacquemin B. Long-term air pollution exposure, greenspace and health-related quality of life in the ECRHS study. Sci Total Environ. 2022; 849:157693. Google Scholar
- Hu Y, Chen Y, Liu S. Higher greenspace exposure is associated with a decreased risk of childhood asthma in Shanghai - a megacity in China. Ecotoxicol Environ Saf. 2023; 256:114868. Google Scholar
- Núñez E, Steyerberg EW, Núñez J. [Regression modeling strategies]. Rev Esp Cardiol. 2011; 64(6):501-507. Google Scholar
- Skrivankova VW, Richmond RC, Woolf BAR. Strengthening the reporting of observational studies in epidemiology using Mendelian randomization: the STROBE-MR statement. JAMA. 2021; 326(16):1614-1621. Google Scholar
- Knol MJ, van der Tweel I, Grobbee DE, Numans ME, Geerlings MI. Estimating interaction on an additive scale between continuous determinants in a logistic regression model. Int J Epidemiol. 2007; 36(5):1111-1118. Google Scholar
- Brunekreef B, Künzli N, Pekkanen J. Clean air in Europe: beyond the horizon?. Eur Respir J. 2015; 45(1):7-10. Google Scholar
- Deane S, Teuber SS, Gershwin ME. The geoepidemiology of immune thrombocytopenic purpura. Autoimmun Rev. 2010; 9(5):A342-349. Google Scholar
- Semple JW, Milev Y, Cosgrave D. Differences in serum cytokine levels in acute and chronic autoimmune thrombocytopenic purpura: relationship to platelet phenotype and antiplatelet T-cell reactivity. Blood. 1996; 87(10):4245-4254. Google Scholar
- Piovani D, Brunetta E, Bonovas S. UV radiation and air pollution as drivers of major autoimmune conditions. Environ Res. 2023; 224:115449. Google Scholar
- Zhao CN, Xu Z, Wu GC. Emerging role of air pollution in autoimmune diseases. Autoimmun Rev. 2019; 18(6):607-614. Google Scholar
- Yang T, Gu T, Xu Z, He T, Li G, Huang J. Associations of residential green space with incident type 2 diabetes and the role of air pollution: a prospective analysis in UK Biobank. Sci Total Environ. 2023; 866:161396. Google Scholar
- Zhu Z, Yang Z, Yu L. Residential greenness, air pollution and incident neurodegenerative disease: a cohort study in China. Sci Total Environ. 2023; 878:163173. Google Scholar
- European Environment Agency. The European environment-state and outlook 2020. 2024. Publisher Full TextGoogle Scholar
- Department for Environment Food and Rural Affairs. Emissions of air pollutants in the UK, 1970 to 2022. 2024. Publisher Full TextGoogle Scholar
Data Supplements
Figures & Tables
Article Information
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.