Early life environmental factors associated with autism spectrum disorder symptoms in children at age 2 years: A birth cohort study

Abstract

Mounting evidence finds that early life environmental factors increased the probability of autism spectrum disorder. We estimated prospective associations between early life environmental factors and autism spectrum disorder symptoms in children at the age of 2 years in a population-derived birth cohort, the Barwon Infant Study. Autism spectrum disorder symptoms at the age of 2 years strongly predicted autism spectrum disorder diagnosis by the age of 4 years (area under curve = 0.93; 95% CI (0.82, 1.00)). After adjusting for child’s sex and age at the time of behavioural assessment, markers of socioeconomic disadvantage, such as lower household income and lone parental status; maternal health factors, including younger maternal age, maternal pre-pregnancy body mass index, higher gestational weight gain and prenatal maternal stress; prenatal alcohol; environmental air pollutant exposures, including particulate matter < 2.5 µm at birth, child secondhand tobacco smoke exposure at 12 months, dampness/mould and home heating with oil, kerosene or diesel heaters at 2 years postnatal. Lower socioeconomic indexes for area, later birth order, higher maternal prenatal depression, and maternal smoking frequency had a dose-response relationship with autism spectrum disorder symptoms. Future studies on environmental factors and autism spectrum disorder should consider the reasons for the socioeconomic disparity and the combined impact of multiple environmental factors through common mechanistic pathways.

Lay abstract

Mounting evidence indicates the contribution of early life environmental factors in autism spectrum disorder. We aim to report the prospective associations between early life environmental factors and autism spectrum disorder symptoms in children at the age of 2 years in a population-derived birth cohort, the Barwon Infant Study. Autism spectrum disorder symptoms at the age of 2 years strongly predicted autism spectrum disorder diagnosis by the age of 4 years (area under curve = 0.93; 95% CI (0.82, 1.00)). After adjusting for child’s sex and age at the time of behavioural assessment, markers of socioeconomic disadvantage, such as lower household income and lone parental status; maternal health factors, including younger maternal age, maternal pre-pregnancy body mass index, higher gestational weight gain and prenatal maternal stress; maternal lifestyle factors, such as prenatal alcohol and environmental air pollutant exposures, including particulate matter < 2.5 μm at birth, child secondhand tobacco smoke at 12 months, dampness/mould and home heating with oil, kerosene or diesel heaters at 2 years postnatal. Lower socioeconomic indexes for area, later birth order, higher maternal prenatal depression and maternal smoking frequency had a dose-response relationship with autism spectrum disorder symptoms. Future studies on environmental factors and autism spectrum disorder should consider the reasons for the socioeconomic disparity and the combined impact of multiple environmental factors through common mechanistic pathways.