Back to search

National and Kapodistrian University of Athens

Μακροχρόνια έκθεση στην ατμοσφαιρική ρύπανση και θνησιμότητα στην Ελλάδα

Abstract

dc:description

Background: Epidemiological studies have documented the adverse effects of long-term exposure to fine particulate matter (PM2.5) on health, while there has been less research on the effects of nitrogen dioxide (NO2), black carbon (BC) and especially of ozone (O3). Simultaneously, the evidence regarding the exposure to the aforementioned exposures on diagnoses-specific mortality outcomes is limited. Furthermore, there is growing interest in the effects of green and blue spaces and their possible synergistic effects with air pollutants. We investigated the association of long- term exposure to air pollution, green and blue spaces with total mortality excluding external causes and cause-specific mortality and mortality by diagnoses-specific outcomes in Greece, using an ecological study design. Methods: We collected socioeconomic and mortality data for 1,035 Municipal Units (MU) from the 2011 Census from the Hellenic Statistical Authority (ELSTAT). Annual PM2.5, NO2, BC and O3 concentrations for 2010 were predicted at 100x100m grids by hybrid land use regression (LUR) models and a weighted average based on the built-up area percentage of each MU was estimated. The normalized difference vegetation index (NDVI) and the percentage of water bodies were used to assess exposure to green and blue spaces. The exposure assessment was conducted through Geographic Information System (ArcGIS). We assessed the shape of the exposure-outcome association by applying a natural cubic spline with 3 degrees of freedom for each outcome-exposure pair. We applied single and two-exposure Poisson regression models on standardized mortality rates, accounting for spatial autocorrelation and for possible confounders. We further assessed interactions between the pollutants, blue spaces and greenness. Extensive sensitivity analysis was performed on the models applied, the confounders adjusted for in the models, but also on the shape of the underlying association between exposures and outcomes. Results: The results provided support for the linearity of the vast majority of the associations with the exception of exposure to blue space were a three level categorical variable was used (in tertiles). An interquartile range (IQR) increase in PM2.5, NO2 and BC was associated with increases in total mortality (Relative Risk (RR) = 1.09, 95% confidence interval (CI): 1.08, 1.11; RR = 1.03 (95% CI: 1.03, 1.04) and RR = 1.02 (95% CI: 1.02, 1.03), respectively) while greenness was associated with lower mortality (RR = 0.95, 95% CI: 0.94, 0.96 per IQR). All associations regarding the exposure to air pollutants were stronger in urban areas and especially for PM2.5 (urban areas: RR = 1.10, 95% CI: 1.07, 1.12, per IQR; the rural ones RR = 1.05, 95% CI: 1.03, 1.07 per IQR). The exclusion of the MUs of the Athens and Thessaloniki areas resulted in slightly lower effect estimates while the adverse effects of air pollution were higher in mainland Greece compared to the islands. We found higher effect estimates in MUs that are in the coastline of mainland Greece while the exposure to blue space was protective to those specific areas. The results were mostly robust to co-exposure adjustment, especially when the correlation between the exposures was small to moderate (r ≤ 0.57). We did not find any statistically significant interaction between pollutants and NDVI. On the contrary, statistically significant interactions were observed between exposure to air pollutants and blue space (p < 0.02), where the adverse effects of PM2.5, NO2 και BC were higher in areas with higher levels of blue space. Furthermore, the statistically significant interaction between the exposure to green and blue spaces indicated that the protective effect of exposure to greenness was higher in areas with higher levels of blue space.We observed increased risk of cardiovascular mortality after long- term exposure to PM2.5 and NO2 (RR = 1.10, 95% CI: 1.08, 1.13; RR = 1.03 95% CI: 1.02, 1.04 per IQR, respectively), while the results were almost robust in the two- exposure models with the exception of NO2 and BC were a high correlation was observed (r = 0.78). We estimated a negative interaction between PM2.5 and NDVI for cardiovascular mortality (p < 0.001), implying lower PM2.5 effects in areas with higher greenness, while the effects of air pollutants were higher in areas with higher levels of blue space.Long- term exposure to PM2.5 and NO2 was associated with respiratory mortality (RR = 1.11, 95% CI: 1.06, 1.16; RR = 1.03, 95% CI: 1.01, 1.04 per IQR, respectively). The results were almost robust to co-exposure adjustment with the exception of NO2 and BC. We detected significant positive interactions for respiratory mortality between NDVI and NO2, BC and O3 (p < 0.001), indicating higher pollution effects in areas with higher greenness. Furthermore, the effects of air pollutants were higher in areas with higher levels of blue space.An IQR increase in PM2.5, NO2 and BC was associated with increased risk in mortality from diseases of the nervous system (RR = 1.14, 95% CI: 1.01, 1.28, RR = 1.03, 95% CI: 0.99, 1.07, RR = 1.05, 95% CI: 1.00, 1.10 respectively) while the exposure to green and blue spaces was protective. Furthermore, we found positive and statistically significant interactions between PM2.5 and NDVI with nervous system related mortality (p = 0.002), while negative ones between NDVI and O3 (p = 0.007). The results were robust in the two- exposures models. Aggravating effects of PM2.5, NO2 and BC were estimated for cerebrovascular mortality, while exposure to greenness was protective. Adjustment for O3 resulted to null associations for NO2 and BC. Exposure to PM2.5 was associated with mortality from ischemic heart disease (RR = 1.05, 95% CI: 1.01, 1.10 per IQR) and diabetes, but the latter was not statistically significant. Our results support the aggravating effects of PM2.5, ΝΟ2 and ΒC in chronic obstructive pulmonary disease (COPD) mortality, though none of the associations reached the nominal level of statistical significance. The protective association between long- term exposure to O3 and mortality may be partly attributed to the small exposure contrast of the O3 levels, as well as to complex non-linear interactions between Ο3 and the other pollutants. As the exposure assessment models for O3 had a low R2 over Greece, it is possible that the measurement error in the O3 estimates, along with residual confounding, may have also impacted our effect estimates.Conclusions: In conclusion, our findings support the associations of long-term exposure to environmental factors with total and cause-specific mortality. More specifically, we estimated an increased risk of mortality after long- term exposure to PM2.5, NO2 and BC and found protective effects with greenness, while we detected adverse estimates regarding the exposure to blue spaces. Our results on the interactions between pollutants and greenness as well as between pollutants and blue space may imply differential biological mechanisms for mortality and warrant further investigation. However, as our study used an ecological design, more studies with individual-level data are needed to confirm our results and further elaborate the synergistic interplay of environmental exposures.

Degree

thesis:*
Grantor dc:publisher
National and Kapodistrian University of Athens
Year dc:date
2022

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Kasdagli, Maria-Iosifina
  • Κάσδαγλη, Μαρία-Ιωσηφίνα

Subjects

dc:subject × 16

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/51263

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Kasdagli, Maria-Iosifina; Κάσδαγλη, Μαρία-Ιωσηφίνα. Μακροχρόνια έκθεση στην ατμοσφαιρική ρύπανση και θνησιμότητα στην Ελλάδα. National and Kapodistrian University of Athens, 2022. http://hdl.handle.net/10442/hedi/51263