TITLE:
Pregnancy Air Exposure: An R Package for Estimation of Exposure to Air Pollution during Critical Windows of Pregnancy
AUTHORS:
Séverine Deguen, Laure Nicollet, Morgane Gilles, Arlette Danzon, Olivier Blanchard, Géraldine Le Nir, Denis Zmirou-Navier, Wahida Kihal-Talantikite
KEYWORDS:
Air Pollution Exposure, Pregnant Women, Critical Window of Exposure, R Software
JOURNAL NAME:
Open Journal of Statistics,
Vol.7 No.3,
June
8,
2017
ABSTRACT: Today, it is well known that ambient air pollution exposures may be related with different health outcomes including adverse birth outcomes (for instance, low birth weight and preterm birth). Then, researchers must estimate exposure to air pollution during different critical windows of exposure across pregnancy. In the literature, the most commonly examined critical windows of exposure are the first, second, and third trimesters of pregnancy, although the entire pregnancy is also studied. In this context, we developed an automatic procedure which estimates individual exposures by trimester. We illustrate usage of the Pregnancy Air Exposure package for all births within the city of Paris in 2011. Nitrogen dioxide data from monitoring stations and data modeled at the census block level were available from the air quality monitoring networks of Ile de France Region. Nitrogen dioxide, NO2, has been chosen as an indicator of traffic-generated pollution. Newborn health data are available from the first birth certificate registered by the Maternal and Child Care department of Paris; more specifically, we used the date of birth and the gestational age to individually determine the beginning and the end of each trimester. The Pregnancy Air Exposure R package was created to simplify application of the procedure for non-specialists, and is freely available from the Equit’Area project website: (www.equitarea.org/documents/package_Exposure). The package contains two functions as well as the help files needed to run the procedure and yields air pollution exposure estimates by pregnancy trimester. The two functions can be used successively or independently according to the available input data.