The Independent, Moderating, and Interaction Effects of Prenatal Exposure to Maternal Smoking and Ambient Fine Particulate Matter on Risk of Autism Spectrum Disorder

Background: Recent estimates of autism spectrum disorder’s (ASD) heritability range from 38%-87% indicating an important role for the environment in ASD’s etiology. Among the many environmental factors that have been investigated, exposure to prenatal ambient fine particulate matter (PM_2.5) and maternal smoking have been independently, albeit inconsistently, linked to ASD. Although ambient PM_2.5 and maternal smoking exposure during pregnancy may engage similar biological mechanisms to incur ASD risk (e.g. inflammation), the moderating and interaction effects of these exposures on ASD risk have yet to be examined.

Objectives: Investigate the independent, moderating and interacting effects of prenatal ambient PM_2.5 and maternal smoking exposures on ASD risk.

Methods: Children with ASD born from 2008-2010 in Salt Lake County, Utah (N=247) were identified by the Utah Registry of Autism and Developmental Disabilities and each matched to five population controls (N=1008) based on birth year and sex from the Utah Population Database. Exposure to maximum daily (split at ≤ vs >median concentration in each exposure window) and average daily (split at ≤ vs >12 µm) concentrations of ambient fine particulate matter was estimated separately for five exposure windows including the three months prior to conception, the first, second, and third trimesters, and throughout pregnancy. Prenatal maternal smoking (yes vs. no) and additional pregnancy-related covariates for ASD were acquired from birth certificates. A series of multivariable conditional logistic regression models were fit to determine the relationship between 1) prenatal maximum and average daily PM_2.5 exposures and ASD risk, and 2) prenatal maternal smoking and ASD risk. The PM_2.5 models were updated to include a prenatal PM_2.5 by maternal smoking interaction term.

Results: Average daily ambient PM_2.5 exposure was not associated with ASD risk during any of the pregnancy exposure windows. Maximum daily ambient PM_2.5 exposure during the first trimester was associated with decreased ASD risk (Odds ratio (OR): 0.743, 95% CI: 0.56-0.99). Prenatal maternal smoking exposure was associated with increased ASD risk (OR: 1.73, 95% CI: 1.05-2.84). The interaction between average daily PM_2.5 concentration throughout pregnancy and prenatal maternal smoking was associated with increased ASD risk with an OR of 4.38 (95% CI: 1.52-12.61) among children exposed to maternal smoking vs. an OR of 0.89 (95% CI: 0.635-1.24) for children not exposed to maternal smoking. (p=0.004). The relationship between maximum daily PM_2.5 exposure and ASD during the third trimester was moderated by prenatal maternal smoking exposure (OR: 2.66, 95% CI: 1.002-7.03 vs. OR: 1.21, 95% CI: 0.88-1.65 in children exposed vs. not exposed to maternal smoking), although the interaction was not significant (p=0.13).

Conclusions: Our results indicate that prenatal maternal smoking is independently associated with ASD, and suggest that the combination of prenatal maternal smoking and fine particulate matter exposure may yield interacting and moderating effects on ASD risk. Effect estimate precision is low, however, indicating that further investigation with a larger sample is warranted. Replication of study findings would suggest that interventions aimed at reducing prenatal maternal smoking might have a corresponding impact on the risk of ASD incurred through prenatal PM_2.5 exposure.