Background: Autism is considered to have a genetic basis, although exposure to certain stimuli in the prenatal period has been implicated to be causal in some cases. Some investigations have shown an association with obstetric complications but findings have been inconsistent owing to differences in sampling and methods.Objective: To examine the association of obstetric factors with autism spectrum disorders for a cohort of children, using obstetric data contained in a statutory database collected at the time of birth.Design: Subjects born in Western Australia between 1980 and 1995 and diagnosed with an autism spectrum disorder by 1999 were included as cases (n = 465). Siblings of the cases (n = 481) and a random population-based control group (n = 1313) were compared with the cases on obstetric information contained in the Maternal and Child Health Research Database of Western Australia.Results: Compared with control subjects, cases had significantly older parents and were more likely to be firstborn. Case mothers had greater frequencies of threatened abortion, epidural caudal anesthesia use, labor induction, and a labor duration of less than 1 hour. Cases were more likely to have experienced fetal distress, been delivered by an elective or emergency cesarean section, and had an Apgar score of less than 6 at 1 minute. Cases with a diagnosis of autism had more complications than those with pervasive developmental disorder not otherwise specified or Asperger syndrome. Nonaffected siblings of cases were more similar to cases than control subjects in their profile of complications.Conclusions: Autism is unlikely to be caused by a single obstetric factor. The increased prevalence of obstetric complications among autism cases is most likely due to the underlying genetic factors or an interaction of these factors with the environment.