Time-series analysis of association between ambient ozone and premature birth in Taiyuan, 2013-2015

Abstract

Objective: To understand the correlation between ambient ozone (O(3)) and premature birth in Taiyuan. Methods: The data of O(3) daily average concentrations and maximum concentration of 8 hours, PM(2.5) daily average concentrations, meteorological factors (including the average temperature and relative humidity) and daily premature birth numbers during 2013-2015 were collected in Taiyuan from China National Environmental Monitoring Center and China Meteorological Administration. The models were developed by using generalized linear model. The gender of the premature birth was analyzed by subgroup analysis. Sensitive analysis were used to estimate the model stability. Results: Ambient O(3) daily average concentrations and maximum concentration of 8 hours in Taiyuan of 2013 to 2015 were 45.35 μg/m(3) and 71.33 μg/m(3) respectively. In single pollutant model, the effects of different window exposures on premature birth were different. The RR of preterm birth about an increase of 10 μg/m(3) of O(3) (lag 3 weeks, lag 4 weeks) were 1.090 (95%CI: 1.042-1.139), 1.095 (95%CI: 1.032-1.163) respectively. According to the infant gender subgroup analysis, male infants were more likely to be affected by ambient O(3). The average daily of O(3) concentration increased by 10 μg/m(3), and the RR for male infants and female infant were 1.013 (95%CI: 0.997-1.029), 0.996 (95%CI: 0.983-1.010) respectively. By comparing the daily average concentration of O(3) with the maximum daily concentration of O(3) for 8 hours as an indicator of exposure, it was found that the daily average concentration of O(3) was more sensitive. The RR of preterm birth about an increase of 10 μg/m(3) of average daily of O(3) concentration and the maximum daily concentration of O(3) for 8 hours. were 1.006 (95%CI: 0.994-1.017) and 0.997 (95%CI: 0.988-1.005) respectively. As for sensitive analysis, when O(3) introduced PM(2.5) or change the degree of model variables, the association of premature birth was still constant. Conclusion: Prenatal exposure to O(3) may increase the risk of premature birth, with different window exposures premature births have different risks and this risk is more pronounced in male infants.