Prenatal exposure to multiple persistent organic pollutants and childhood BMI trajectories- a comparison of three different methods for exposure mixture analysis in a mixed model framework
Year: 2020
Type: conference-abstract
Abstract: Background: Prenatal exposure to organochlorine compounds has been associated with increased risk for pediatric obesity; however, evidence is still inconclusive and most previous studies did not include repeated outcome measures.Aim: To determine the association between prenatal exposure to three organochlorine pesticide and six polychlorinated biphenyl concentrations in 2nd pregnancy trimester maternal serum and childhood body mass index (BMI) trajectories across 4 to 11 years of age in 280 child-mother dyads from the prospective mother-child RHEA study.Methods: We first applied independent linear mixed effect models to evaluate the association between individual compounds and BMI trajectories, using a Bonferroni adjustment to account for multiple comparisons. As an alternative, we applied the Bayesian Weighted Quantile Sum Regression (BWQSR) to model all exposures jointly and to provide an overall association of all compounds on BMI trajectories. We also used the Bayesian varying coefficient kernel machine regression (BVCKMR) to model the exposure mixture jointly while allowing for non-linear and non-additive associations. All results were adjusted for socio-demographic information.Results: From the linear mixed model, a 10-fold increase of HCB and PCB-118 was significantly associated with 0.5 kg/m2 (95%CI: 0.3, 0.7) and 0.3 kg/m2 (95%CI:0.2, 0.6) increase in BMI trajectories across 4-11 years respectively. The BWQSR found that an increase in the chemicals mixture was associated with an increase of 0.12 kg/m2 (95%CI: 0.02, 0.24) in childhood BMI trajectories; HCB, DDE, and PCB-156 were important drivers of this association, as indicated by posterior inclusion probabilities. The BVCKMR identified positive associations of HCB, DDT, and PCB-118 with BMI trajectories and a negative association between DDE and BMI trajectories.Conclusions: Despite some inconsistencies, all three analyses demonstrated a positive association between HCB and BMI trajectories. Novel methods offer new opportunities for assessing exposure mixture effects on outcome trajectories, but results should be interpreted considering each method’s strengths and limitations.
Source: ISEE Conference Abstracts
Language: English
Subfield: Health, Toxicology and Mutagenesis
Field: Environmental Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W3171942868