Meteorological Conditions and Term Birthweight in the French EDEN-PELAGIE Consortium: Identifying Critical Exposure Windows Using Distributed Lag Models
Access StatusFull text of the requested work is not available in DASH at this time ("dark deposit"). For more information on dark deposits, see our FAQ.
Jakpor, Otana Agape
MetadataShow full item record
CitationJakpor, Otana Agape. 2019. Meteorological Conditions and Term Birthweight in the French EDEN-PELAGIE Consortium: Identifying Critical Exposure Windows Using Distributed Lag Models. Doctoral dissertation, Harvard Medical School.
AbstractIntroduction: Heat stress during pregnancy may limit fetal growth, which has ramifications for health outcomes throughout the life course. However, current evidence on weather conditions and birthweight is mixed. This may be partly due to lack of temporal and spatial resolution in how exposures are measured and modelled. For example, prenatal weather exposure data are often obtained from monitoring stations (which may be sparse) and averaged over trimesters (which are fairly arbitrary time windows). We aim to clarify the impacts of mean temperature and relative humidity on term birthweight, by using a fine spatio-temporal model to assess temperature exposure and building distributed lag models to identify critical exposure windows during gestation. We also intend to investigate the role of weather variability by assessing the effects of standard deviation of temperature and relative humidity.
Methods: We analyzed data collected from a consortium of two French mother-child cohorts, EDEN and PELAGIE (n = 4771), between 2002 and 2006. Temperature exposure data were obtained from a satellite-based model with fine spatial resolution (1 km2), and humidity data were obtained from Météo France monitors. Distributed lag models were constructed for term births, using weekly exposure data from the first 37 weeks of pregnancy. This analysis was also repeated with stratification by the sex of the infant. Results for each exposure (mean temperature, standard deviation of temperature, mean relative humidity, and standard deviation of relative humidity) were adjusted for the other exposures, gestational age, season and year of conception, recruitment center, and individual confounders.
Results: For standard deviation of temperature (1 °C increase), there was a critical exposure window between weeks 6 and 20, with a cumulative change in term birthweight of -65.2 g (95% CI: -101.9, -6.4). Upon stratification, the relationship between standard deviation of temperature and term birthweight was nonsignificant in girls, but in boys there was a critical exposure window from week 1 to 21 (-5.4 g, 95% CI: -10.7, -0.2). For mean humidity (5% increase), there was a critical exposure window between weeks 26 and 37, with a cumulative change in term birthweight of -28.2 g (95% CI: -49.2, -7.1). The critical exposure window occurred later and the association was stronger in boys (weeks 29 to 37; -37.3 g, 95% CI: -63.4, -11.1) than in girls (weeks 13 to 15; -5.4 g, 95% CI: -10.7, -0.2). There was no week with a statistically significant association between term birthweight and mean temperature or standard deviation of humidity.
Conclusions: We identified critical windows in gestation in which temperature variability and mean humidity were associated with decreased term birthweight. These windows varied by sex of the infant, suggesting possible differences in mechanism. Little is currently known about the impacts of temperature variability and humidity on fetal growth, and our findings indicate that more research is needed.
Citable link to this pagehttp://nrs.harvard.edu/urn-3:HUL.InstRepos:40620138