Autism risk grows steadily with fathers' increasing age, but rises rapidly with mothers' age after 30, a new study has found.

Researchers found that older parents are more likely to have a child who develops an autism spectrum disorder (ASD) than are younger parents.

The study by researchers from the Drexel University School of Public Health in Philadelphia and Karolinska Institute in Sweden found that fathers' and mothers advancing ages have different impacts on their child's autism risk.

The rise in ASD risk with parental age was greater for older mothers as compared to older fathers.

Researchers found that the risk of having a child with ASD had a more complicated relationship to age in women than in men - whose risk of fathering a child with ASD increased linearly with age across their lifespan.

Among women giving birth before the age of 30, the risk of ASD in the child showed no association with age - it was simply very low. But for babies born to mothers aged 30 and older, the chance of developing ASD rose rapidly with the mother's age.

Brian K Lee, an assistant professor in the Drexel University School of Public Health and research fellow of the AJ Drexel Autism Institute noted that the non-linear maternal age effect that is relatively stronger than the paternal age effect on ASD risk has been observed in previous studies, but has not received much attention.

Multiple mechanisms could be in play to account for the different patterns of risk, including environmental risk factors occurring in women after age 30. Factors such as complications in pregnancy could also underlie the effect of mothers' ages on a child's ASD risk but not a paternal age effect.

The linear, steady increase in risk associated with fathers' ages is consistent with the hypothesis of increased genomic alterations over the father's lifespan that can increase risk of ASD, said Lee, senior author of the study.

In the study, Lee and colleagues analysed a large population registry sample of 417,303 children born in Sweden between 1984 and 2003 and adjusted for numerous possible factors that could vary with parental age and also influence risk, such as family income and each parent's psychiatric history.