Association of ozone with mortality and emergency room visits for cardiovascular and respiratory diseases in Seoul, Korea

Abstract

Background: Ground-level ozone is known to cause acute health effects on mortality, emergency room (ER) visits and hospital admission. Many previous studies have shown significant positive associations with non-accidental, cardiovascular and respiratory mortality and relatively fewer studies have examined the effect of ozone on ER visits in Seoul, Korea. Further, the shapes of concentration-response (C-R) relationship with mortality showed inconsistent results and the relationship with ER visits has never been explored. Objectives: The purpose of this study is to examine the effect of ozone on cardiovascular and respiratory diseases and to explore the shapes of the C-R relationship with mortality and ER visits. Methods: Time series analyses with generalized linear model (GLM) were used to estimate the effect of ozone. We constructed natural spline model of ozone to explore the shapes of the associations and compared Quasi-Akaike Information Criterion (QAIC) between natural spline model and linear model. We chose either a spline model or a linear model for each health outcome with the shape of the C-R relationship and QAIC. To estimate the effect of short-term exposure to ozone, excessive relative risk (ERR) was computed. We also performed age- and sex-stratified analyses and respiratory-specific diseases analyses. Results: We found a significant non-linear relationship of ozone with cardiovascular mortality, while a linear relationship with respiratory ER visits. Ozone showed adverse effects on cardiovascular mortality (ERR = 1.36%, 95% CI: 0.422 - 2.311), respiratory ER visits (1.869%, 95% CI: 0.739 - 3.011), asthma ER visits (6.267%, 95% CI: 3.846 - 8.744), COPD ER visits (3.707%, 95% CI: 1.673 - 5.781) and pneumonia ER visits (3.150%, 95% CI: 1.847 - 4.469). Moreover, we found cardiovascular mortality and ER visits had significantly higher risk on people above 65 years old (1.170%, 95% CI: 0.098 - 2.258, 1.769, 95% CI: 0.340 - 3.218) and respiratory ER visit also had higher risk on people above 65 years (1.845%, 95% CI: 0.316 - 3.397) and male (1.884%, 95% CI: 0.735 - 3.047). Conclusion: Our results confirmed the non-linear relationship between ozone and cardiovascular mortality with the existence of a threshold. However, the association between ozone and ER visits showed a linear relationship rather than non-linear. Further, the findings of estimated risk indicate that the impact of ozone on respiratory diseases could trigger ER visits, while leading to death on cardiovascular diseases.