Residential greenspace counteracts PM2.5 on the risks of preterm birth subtypes: A multicenter study

Abstract

BackgroundThe association between residential greenspace and preterm birth (PTB) risk remained inconclusive. The PTB subtypes have been ignored and the effect of co-exposure of PM2.5 on PTB risk is still unclear. ObjectiveTo investigate the independent, interactive, and mixed effects of residential greenspace and PM2.5 on the risk of PTB subtypes. MethodsA total of 19,900 singleton births from 20 hospitals in Shanghai, China, from 2015 to 2017 were included. The Normalized Difference Vegetation Index (NDVI) within 500 m and 1000 m buffers of the maternal residence and a combined geoscience-statistical model-derived PM2.5 and its six components were used as the exposure measures. PTB (<37 completed weeks of gestation) were divided into early PTB (24–33 weeks) vs. late PTB (34–36 weeks) and into spontaneous PTB (sPTB), preterm premature rupture of the fetal membranes (PPROM), and iatrogenic PTB. Multivariable logistic regression models were applied to assess the independent and interactive effects of NDVI and PM2.5 on PTB in each trimester. The quantile g-computation approach was employed to explore the mixture effect of PM2.5 components and greenspace across the pregnancy and to determine the main contributors. ResultsLevels of PM2.5 and greenspace were associated with increased [aOR (95%CI) ranging from 1.18 (1.07, 1.30) to 3.36 (2.45, 4.64)] and decreased risks [aORs (95%CI) ranging from 0.64 (0.53, 0.78) to 0.86 (0.73, 0.99)] of PTB subtypes, respectively. At the same PM2.5 level, higher residential greenspace was associated with lower risks, and vice versa. All these associations were more pronounced in late pregnancy. Early PTB and PPROM were the main affected subtypes, and the main drivers in PM2.5 were black carbon and ammonium. ConclusionsResidential greenspace may mitigate the PTB risks due to PM2.5 exposure during pregnancy.