Abstract Background Autism spectrum disorder (autism) describes a heterogeneous neurodevelopmental phenotype arising from the interplay of environmental and genetic factors in early life. Methods In a general population birth cohort, we employed a scoping approach to identify prospective associations between prenatal and birth factors and a subsequent autism diagnosis. Results Factors associated with increased likelihood of autism included those related to i) maternal health (maternal pre-pregnancy body mass index, pre-existing maternal mental health conditions, maternal use of selective serotonin reuptake inhibitors) ii) environmental exposures (maternal passive tobacco smoke exposure, and exposure to vinyl floors) iii) demographic factors (socioeconomic disadvantage). Factors associated with a decreased likelihood of autism included maternal dietary nutrition and supplementation (higher folic acid, magnesium, and iron, as well as adherence to the Australian Dietary Guidelines). Conclusion Our findings extend the evidence that autism may have a multifactorial origin in early life. Further studies should explore the composite effects of these prenatal and birth factors on autism outcomes via shared biological pathways, such as inflammation, and oxidative stress, in concert with genetic predisposition. Impact Autism spectrum disorder (autism) is a multifactorial condition. Here we report on multiple prenatal environmental, demographic, maternal and pregnancy factors that are associated with an increased likelihood of an autism diagnosis. For example, adherence to the Australian Dietary Guidelines during pregnancy is linked to a reduced likelihood of autism in the offspring, consistent with mounting evidence that prenatal nutrition impacts brain development. We examine how the multiple risk factors, identified by our comprehensive approach, may be linked to shared biological mechanisms. Future work should examine composite exposure measures acting through shared mechanisms as a more productive approach to understanding aetiology than focusing solely on individual exposures.
Holland et al. (Thu,) studied this question.