| Objectives:Currently the findings of ambient air pollutants and preterm birth is inconsistent, and our research in this area is still rare. This study is to understand the epidemiological characteristics of preterm birth in Xuhui District and to quantitative evaluate the air pollutants (PM10, SO2, and NO2) in premature with establishing the appropriate exposure response relationship to atmospheric pollutants and preterm.Methods:We collected the information of all the babies of Xuhui District born in hospitals in Shanghai, during January 1,2006 and December 31,2012, while collecting the data of air pollutants, temperature and humidity in the same period of time. We undertook a case-control study to realize the risk factors for preterm newborns, using one-factor chi-square test, analyses of one-factor and multiple unconditional logistic regression.We used a generalized additive model (GAM) with penalized splines to analysis the association of air pollutants (PM10, SO2, and NO2) and preterm delivery, including the covariates of temperature, humidity and day of week (DOW).Results:A total of 38,585 births were included from 2006 to 2012. Of these,5.81% were premature. The rate of preterm birth was between 5% and 7% in those years. The average age of preterm delivery mother was 29.74 years old, while the age of others was 28.91 on average. The variables that were statistically significantly associated with preterm birth were birth defects, type of delivery, gestation, parity, number of child, maternal age, maternal education, paternal age and paternal education, while the infant’s sex and abortion has no statistical significance, by the one-factor chi-square test. Furthermore, one-factor unconditional logistic regression suggested birth defects, type of delivery, gestation, parity, number of child, maternal age, and paternal age as a possible risk factor for preterm delivery; whereas the others were meaningless. Finally, using the step forward method of multiple unconditional logistic regression, the seven factors, probably thought to be the risk factors, were put in the model step by step. The result turned out to be significant for the factors of maternal age, birth defects, and number of child, with the respective odds rations of 1.041[95% confidence interval (CI):1.029-1.054],2.362 (95% CI:1.621 3.442),0.032 (95% CI: 0.0280.037).The daily mean concentrations of PM10, SO2, NO2, temperature, and humidity were 82.65ug/m3,43.55ug/m3,59.21ug/m3,17.44 ℃ and 69.41%, respectively. According to the expose-response curve, the curve was linear monotonically increasing for PM10 and NO2, and for SO2, was not clear. We observed a significant effect of PM10 only with 1-week exposure before preterm births. An increase of 10ug/m3 of 1-week average PM10 corresponded to 1.868%(95% CI:0.339%-3.397%) increases of preterm births. NO2, with 1-week,2-week and 3-week exposure before preterm deliveries, were statistically significant with preterm birth, with the influence of rising every 10ug/m3, incidence of preterm birth were up 6.241% (95% CI:2.294%-10.134%),7.098%(95% CI:2.100%-12.096%) and 6.498%(95% CI:0.344%-12.652%). Otherwise, the 4-week prenatal cumulative exposure of NO2 was not meaningful. The impact on the premature occurrence of SO2, cumulative exposure of prenatal 1-week,2-week,3-week or 4-week, has no statistical significance (P> 0.05).After stratifying by maternal age, birth defects, and number of child, we considered the stronger associations with PM10 in the levels of single births, infants who had birth defects, maternal age< 24 years and maternal age≥ 35. An increase of 10ug/m3 of 1-week average PM10 corresponded to respectively 0.799% (95%CI: 0.035%-1.563%),0.667%(95%CI:0.216%-1.118%)、1.364%(95%CI:0.619%-2.109%) and 1.154%(95%CI:0.292%-2.016%) increases of preterm births. NO2 was only related to the level of maternal age 35 years older, while the 1-week average NO2 concentration increased 10ug/m3, the incidence of preterm birth was increased by 3.680%(95%CI:95%1.210%). The effect of SO2 and premature was not observed within each group.Studies of acute effects were also overtaken. In the analysis of pollutants only NO2 with Iag4 and Iag5 exhibited an effect on preterm births. Other lags of NO2 and other pollutants in any lag day were not observed to have a statistically significant effect.Conclusion:Increased maternal age, multiple births, and birth defects will increased the risk of preterm birth. The influences of air pollutants PM10 and NO2 were statistically significant on preterm birth, especially the cumulative exposure concentrations of NO2 lag time was longer. The effect of exposure of 3 weeks before births was more meaningful than acute effect. The impact of SO2 and premature was not found. Our study may provide some important bases in the prevention of preterm birth and the protection of air pollution. |
PDF Full Text Request