Ozone and Preterm Birth in Women With Gestational Diabetes
Ozone and Preterm Birth in Women With Gestational Diabetes
The distribution of maternal and infant characteristics in the study population is presented in Table 1. More cases than controls were male (57%), had mothers less than 20 years of age or 35 years of age or older, and were born into a lower socioeconomic group (<25th percentile). The odds of preterm delivery were higher for those mothers who experienced pregnancy complications, including gestational diabetes mellitus, gestational hypertension, and preeclampsia (Table 1).
The distributions of monthly mean air pollutant concentrations and correlations for the 3 trimesters of the pregnancy are presented in Table 2 and Table 3. The average concentrations of nitrogen dioxide and carbon monoxide were highly correlated in all 3 trimesters (correlation coefficients ranged from 0.83 to 0.84), reflecting their common emission source (motor vehicles). The concentrations of ozone were moderately associated with PM10 concentration (r = 0.52–0.53) and positively but weakly correlated with sulfur dioxide concentration (r = 0.16–0.17). Ozone was negatively correlated with the mainly traffic-related pollutants (Table 3).
Figure 1 shows the association estimates for preterm birth during the 3 trimesters in the single-pollutant model. A 10-ppb increase in ozone level was significantly positively associated with preterm birth; the adjusted odds ratio was 1.03 (95% confidence interval (CI): 1.02, 1.04) for the first trimester, 1.02 (95% CI: 1.01, 1.02) for the second trimester, and 1.02 (95% CI: 1.01, 1.03) for the third trimester. Each 10-µg/m increase in PM10 exposure was slightly associated with a 1% increase in the risk of preterm birth in the first trimester but was negatively associated during the second and the third trimesters. Negative associations were found for traffic-related pollutants (carbon monoxide and nitrogen dioxide), but no association was found with sulfur dioxide in any of the trimesters (Figure 1).
(Enlarge Image)
Figure 1.
Association between air pollution and preterm birth from single-pollutant models, Taiwan, 2001–2007. Unit changes in air pollutants are as follow: 10 ppb for ozone (O3) and nitrogen dioxide (NO2); 1 ppb for sulfur dioxide (SO2); 100 ppb for carbon monoxide (CO); and 10 µg/m for particulate matter with an aerodynamic diameter of 10 µm or less (PM10). All models were adjusted for maternal age, sex of the infant, household income, season of conception, and year of conception. Squares represent the first trimester, circles represent the second trimester, and diamonds represent the third trimester. Bars, 95% confidence intervals.
Figure 2 presents the relationship between ozone concentration and preterm birth in 9 multipollutant models. In the 2-pollutant models (models 1, 2, and 3), the association estimates for ozone while controlling for other pollutants ranged from a 1% to 4% increase in the risk of preterm birth, and the relationship remained significant during each trimester. We found a similar association during all 3 trimesters; the association estimates of ozone were stable with combinations of the other 2 pollutants in the 3-pollutant models (model 5 and model 6). However, the associations between ozone concentration and preterm birth were diluted in models 7, 8, and 9 (Figure 2).
(Enlarge Image)
Figure 2.
Association between air pollution and preterm birth from multipollutant models in Taiwan, 2001–2007. Unit changes in air pollutants are as follow: 10 ppb for ozone (O3) and nitrogen dioxide (NO2); 1 ppb for sulfur dioxide (SO2); 100 ppb for carbon monoxide (CO); and 10 µg/m for particulate matter with an aerodynamic diameter of 10 µm or less (PM10). All models were adjusted for maternal age, sex of the infant, household income, season of conception, and year of conception. Model 1 included O3 and PM10. Model 2 included O3 and SO2. Model 3 included O3 and NO2. Model 4 included O3 and CO. Model 5 included O3, PM10, and SO2. Model 6 included O3, PM10, and NO2. Model 7 included O3, PM10, and CO. Model 8 included O3, SO2, and CO. Model 9 included O3, SO2, and NO2. Model 10 included O3, PM10, SO2, and CO. Model 11 included O3, PM10, SO2, and NO2. Bars, 95% confidence intervals.
Table 4 shows the association of ozone exposure with preterm birth for 2 different levels of pregnancy complications. In a stratified analysis that was adjusted for maternal age, infant sex, season of conception, and year of conception, the estimates for the association of ozone exposure with preterm birth among women who developed gestational diabetes mellitus (in the second trimester, adjusted odds ratio (OR) = 1.15, 95% CI: 1.05, 1.27; in the third trimester, adjusted OR = 1.12, 95% CI: 1.01, 1.23) were higher than for women who did not develop gestational diabetes mellitus. We found an apparent effect modification (P for interaction <0.01) between ozone exposure and gestational diabetes mellitus on the risk of preterm birth during the second and third trimesters. However, we did not find any interaction with gestational hypertension and preeclampsia (Table 4).
Results
Characteristics of Cases and Control Subjects
The distribution of maternal and infant characteristics in the study population is presented in Table 1. More cases than controls were male (57%), had mothers less than 20 years of age or 35 years of age or older, and were born into a lower socioeconomic group (<25th percentile). The odds of preterm delivery were higher for those mothers who experienced pregnancy complications, including gestational diabetes mellitus, gestational hypertension, and preeclampsia (Table 1).
Air Pollution
The distributions of monthly mean air pollutant concentrations and correlations for the 3 trimesters of the pregnancy are presented in Table 2 and Table 3. The average concentrations of nitrogen dioxide and carbon monoxide were highly correlated in all 3 trimesters (correlation coefficients ranged from 0.83 to 0.84), reflecting their common emission source (motor vehicles). The concentrations of ozone were moderately associated with PM10 concentration (r = 0.52–0.53) and positively but weakly correlated with sulfur dioxide concentration (r = 0.16–0.17). Ozone was negatively correlated with the mainly traffic-related pollutants (Table 3).
Air Pollution and Preterm Births
Figure 1 shows the association estimates for preterm birth during the 3 trimesters in the single-pollutant model. A 10-ppb increase in ozone level was significantly positively associated with preterm birth; the adjusted odds ratio was 1.03 (95% confidence interval (CI): 1.02, 1.04) for the first trimester, 1.02 (95% CI: 1.01, 1.02) for the second trimester, and 1.02 (95% CI: 1.01, 1.03) for the third trimester. Each 10-µg/m increase in PM10 exposure was slightly associated with a 1% increase in the risk of preterm birth in the first trimester but was negatively associated during the second and the third trimesters. Negative associations were found for traffic-related pollutants (carbon monoxide and nitrogen dioxide), but no association was found with sulfur dioxide in any of the trimesters (Figure 1).
(Enlarge Image)
Figure 1.
Association between air pollution and preterm birth from single-pollutant models, Taiwan, 2001–2007. Unit changes in air pollutants are as follow: 10 ppb for ozone (O3) and nitrogen dioxide (NO2); 1 ppb for sulfur dioxide (SO2); 100 ppb for carbon monoxide (CO); and 10 µg/m for particulate matter with an aerodynamic diameter of 10 µm or less (PM10). All models were adjusted for maternal age, sex of the infant, household income, season of conception, and year of conception. Squares represent the first trimester, circles represent the second trimester, and diamonds represent the third trimester. Bars, 95% confidence intervals.
Figure 2 presents the relationship between ozone concentration and preterm birth in 9 multipollutant models. In the 2-pollutant models (models 1, 2, and 3), the association estimates for ozone while controlling for other pollutants ranged from a 1% to 4% increase in the risk of preterm birth, and the relationship remained significant during each trimester. We found a similar association during all 3 trimesters; the association estimates of ozone were stable with combinations of the other 2 pollutants in the 3-pollutant models (model 5 and model 6). However, the associations between ozone concentration and preterm birth were diluted in models 7, 8, and 9 (Figure 2).
(Enlarge Image)
Figure 2.
Association between air pollution and preterm birth from multipollutant models in Taiwan, 2001–2007. Unit changes in air pollutants are as follow: 10 ppb for ozone (O3) and nitrogen dioxide (NO2); 1 ppb for sulfur dioxide (SO2); 100 ppb for carbon monoxide (CO); and 10 µg/m for particulate matter with an aerodynamic diameter of 10 µm or less (PM10). All models were adjusted for maternal age, sex of the infant, household income, season of conception, and year of conception. Model 1 included O3 and PM10. Model 2 included O3 and SO2. Model 3 included O3 and NO2. Model 4 included O3 and CO. Model 5 included O3, PM10, and SO2. Model 6 included O3, PM10, and NO2. Model 7 included O3, PM10, and CO. Model 8 included O3, SO2, and CO. Model 9 included O3, SO2, and NO2. Model 10 included O3, PM10, SO2, and CO. Model 11 included O3, PM10, SO2, and NO2. Bars, 95% confidence intervals.
Table 4 shows the association of ozone exposure with preterm birth for 2 different levels of pregnancy complications. In a stratified analysis that was adjusted for maternal age, infant sex, season of conception, and year of conception, the estimates for the association of ozone exposure with preterm birth among women who developed gestational diabetes mellitus (in the second trimester, adjusted odds ratio (OR) = 1.15, 95% CI: 1.05, 1.27; in the third trimester, adjusted OR = 1.12, 95% CI: 1.01, 1.23) were higher than for women who did not develop gestational diabetes mellitus. We found an apparent effect modification (P for interaction <0.01) between ozone exposure and gestational diabetes mellitus on the risk of preterm birth during the second and third trimesters. However, we did not find any interaction with gestational hypertension and preeclampsia (Table 4).
Source...