The association between ambient PM2.5 exposure and the risk of preterm birth in China: A retrospective cohort study

The relationship between PM2.5 exposure and preterm birth remains unclear.

To explore the effect of exposure to PM2.5 on preterm birth in China.

The birth outcomes of 426,246 pregnant women enrolled between January 2014 and December 2014 in NFPCP (National Free Pre-pregnancy Checkups Project) were collected, and their individual PM2.5 exposure values were estimated from the China National Environmental Monitoring Centre. The time of gestational exposure to PM2.5 was divided into four periods (the first trimester, the second trimester, the third trimester and the entire pregnancy). The average concentration and the corresponding quartiles of PM2.5 were calculated in these periods by the daily average PM2.5 data. Cox proportional hazards regression was used and he effects of maternal age, education level, occupation, second-hand smoking, alcohol use, pre-pregnancy BMI, baby's sex, number of previous pregnancies, coastal areas and season of conception were adjusted for.

A total of 426,246 singleton births were included, among which 35,261 (8.3%) were preterm birth. Effect of each 10μg/m³ increase of PM2.5 on preterm birth was most significant during the third trimester (HR, 1.06; 95%CI, 1.06-1.07), and also significant during the first trimester (HR, 1.04; 95%CI, 1.03-1.04), the second trimester (HR, 1.02; 95%CI, 1.02-1.02) and the entire pregnancy (HR, 1.06; 95%CI, 1.05-1.06). Compared with the lowest quartile of PM2.5, other quartiles increased the risk of preterm birth, and were most significant during the third trimester (HR, 1.87; 95%CI, 1.69-2.06). Subgroup analysis showed that compared with other subgroups, women who were older than 30years, had low education level, worked as farmers, had male baby, had previous pregnancies, not live in coastal areas and pregnant in winter were more sensitive to PM2.5 exposure.

Ambient PM2.5 exposure during pregnancy played an important role in the pregnancy process and increased the risk of preterm birth.