Bayesian multinomial probit modeling of daily windows of susceptibility for maternal PM2.5 exposure and congenital heart defects
Authors:
- Joshua L. WarrenJWJoshua L. Warren
- Jeanette A. StingoneJSJeanette A. Stingone
- Thomas J. LubenThomas J. Luben
- National Birth Defects Prevention StudyNSNational Birth Defects Prevention Study
- 7 February 2016
- journal article
- research article
- Published by Wiley in Statistics in Medicine
- Vol. 35 (16) , 2786-2801
Abstract
Epidemiologic studies suggest that maternal ambient air pollution exposure during critical periods of pregnancy is associated with adverse effects on fetal development. In this work, we introduce new methodology for identifying critical periods of development during post‐conception gestational weeks 2–8 where elevated exposure to particulate matter less than 2.5 µm (PM2.5) adversely impacts development of the heart. Past studies have focused on highly aggregated temporal levels of exposure during the pregnancy and have failed to account for anatomical similarities between the considered congenital heart defects. We introduce a multinomial probit model in the Bayesian setting that allows for joint identification of susceptible daily periods during pregnancy for 12 types of congenital heart defects with respect to maternal PM2.5 exposure. We apply the model to a dataset of mothers from the National Birth Defect Prevention Study where daily PM2.5 exposures from post‐conception gestational weeks 2–8 are assigned using predictions from the downscaler pollution model. This approach is compared with two aggregated exposure models that define exposure as the average value over post‐conception gestational weeks 2–8 and the average over individual weeks, respectively. Results suggest an association between increased PM2.5 exposure on post‐conception gestational day 53 with the development of pulmonary valve stenosis and exposures during days 50 and 51 with tetralogy of Fallot. Significant associations are masked when using the aggregated exposure models. Simulation study results suggest that the findings are robust to multiple sources of error. The general form of the model allows for different exposures and health outcomes to be considered in future applications. Copyright © 2016 John Wiley & Sons, Ltd.Keywords
Funding Information
- National Institute of Environmental Health Sciences (5R01ES014843-02, R01ES020619, T32ES007018)
This publication has 27 references indexed in Scilit:
- Air pollution and congenital heart defectsEnvironmental Research, 2013
- Air Pollution Metric Analysis While Determining Susceptible Periods of Pregnancy for Low Birth WeightISRN Obstetrics and Gynecology, 2013
- Ambient Air Pollution and Risk of Congenital Anomalies: A Systematic Review and Meta-analysisEnvironmental Health Perspectives, 2011
- Estimating Error in Using Ambient PM2.5 Concentrations as Proxies for Personal Exposures: A ReviewEpidemiology, 2010
- A Spatio-Temporal Downscaler for Output From Numerical ModelsJournal of Agricultural, Biological and Environmental Statistics, 2010
- Diabetes mellitus and birth defectsAmerican Journal of Obstetrics and Gynecology, 2008
- Equivalence of prospective and retrospective models in the Bayesian analysis of case-control studiesBiometrika, 2004
- General Methods for Monitoring Convergence of Iterative SimulationsJournal of Computational and Graphical Statistics, 1998
- Bayesian Analysis of Binary and Polychotomous Response DataJournal of the American Statistical Association, 1993
- Inference from Iterative Simulation Using Multiple SequencesStatistical Science, 1992