OBJECTIVE. To investigate prenatal, perinatal, and neonatal risk factors for autism spectrum disorders by using participants identified through broad ascertainment and reliable classification methods.
METHODS. The targeted population was 8-year-old children born in 1994 and residing in 1 of the 3 most populous counties in Utah who were identified as having an autism spectrum disorder on the basis of methodology used by the 2002 Autism and Developmental Disabilities Monitoring Network. Of those identified, 132 children (115 boys, 17 girls) had birth certificate records available. Each child was matched by gender and birth year to 100 controls (11 500 boys, 1700 girls) from the birth certificate database in a nested case-control design. Birth certificate records of participants and controls were surveyed for 23 potentially pathologic prenatal, perinatal, and neonatal factors.
RESULTS. The prenatal factors that occurred significantly more frequently among children with autism spectrum disorders were advanced maternal age and parity. Increased duration of education among mothers of children with autism spectrum disorders was small but statistically significant. Significant perinatal factors were breech presentation and primary cesarean delivery. When corrected for breech presentation, a known indication for cesarean delivery, the association between primary cesarean delivery and autism spectrum disorders was eliminated. There were no significant associations found between autism spectrum disorders and neonatal factors.
CONCLUSIONS. In the absence of other complications suggesting fetal distress, the association between breech presentation and autism spectrum disorders in this study suggests a shared etiology rather than causal relationship. Additional investigation focused on both genetic and environmental factors that link these autism spectrum disorder risk factors individually or collectively is needed.