Association Between Gestational Exposure to Toxicants and Autistic Behaviors Using Bayesian Quantile Regression

Area of Science:

• Environmental Health
• Neurodevelopmental Disorders
• Epidemiology

Background:

• Autism spectrum disorder (ASD) affects 1%-2% of children, characterized by social communication impairments and stereotypic behaviors.
• Prenatal exposure to toxicants is linked to autistic behaviors, but research often focuses on average effects, not the full spectrum.

Purpose of the Study:

• To investigate associations between prenatal toxicant exposure and autistic behaviors across the entire distribution of behaviors.
• To apply Bayesian quantile regression for a more nuanced understanding of these relationships.

Main Methods:

• Utilized data from the pan-Canadian Maternal-Infant Research on Environmental Chemicals cohort (2008-2011).
• Measured concentrations of metals, pesticides, PCBs, phthalates, bisphenol-A, and triclosan in maternal samples during the first trimester.
• Assessed autistic behaviors in 478 children aged 3-4 years using the Social Responsiveness Scale (SRS).

Main Results:

• Lead, cadmium, and phthalates showed mild increases in SRS scores at the 90th percentile.
• Manganese and some pesticides were linked to mild decreases in SRS scores at the 90th percentile.
• Identified monotonic trends where associations changed across the SRS distribution.

Conclusions:

• Quantile regression offers a more detailed view of toxicant-environment-autism spectrum disorder behavior associations than traditional methods.
• Findings highlight the importance of considering the full behavior distribution when assessing environmental risk factors for ASD.
• Suggests wider adoption of quantile regression in epidemiological studies of neurodevelopmental outcomes.