The study used data from 253,347 study-site births with complete parental age information. Cases included 1,251 children aged 8 years with complete parental age information from the same birth cohort and identified as having an autism spectrum disorder. In evaluating the association between parental age and autism risk, it was important to account for other variables related to both parental age and. Birth order is a potentially confounding factor because it is positively associated with parental age and has been reported in some studies to be associated with autism risk, with at least 3 studies reporting firstborn children to be at increased risk of autism. The goal of this study was to determine, in a large, population-based cohort of US children, whether advancing maternal and paternal age each independently increase a child's risk of developing autism after controlling for the other parent's age, birth order, and other risk factors.
In unadjusted analyses, both mean maternal age and mean paternal age were significantly higher for ASD cases than for the birth cohort as a whole. With parental age 25–29 years as the reference group, the odds of developing ASD was significantly reduced for parental age <20 years and increased for maternal age 35 and paternal age 40 years. Those age cutoffs (maternal age 35, paternal age 40 years) were used to classify each parent's age as "older" versus "younger." Other significant predictors of ASD in unadjusted analyses included low birth order, male gender, advanced maternal education, and preterm birth.
Multivariable analysis of parental ages modeled as categorical variables
After the data was adjusted for the other parent's age and other covariates, the increases in ASD risk associated with maternal age 35 years and paternal age 40 years (relative to age 25–29 years) were slightly reduced compared with the unadjusted analysis. In contrast, the results for birth order suggest that the decline in ASD risk associated with increasing birth order is somewhat stronger in the adjusted analysis than in the unadjusted analysis. In addition, the apparent increase in ASD risk associated with higher levels of maternal education in the unadjusted analysis is no longer evident in the adjusted model, suggesting that the apparent maternal education effect is due to its association with parental age.
Parental ages modeled as continuous variables
In unadjusted analyses, the risk of developing ASD increased significantly with each 10-year increase in both maternal age and paternal age. After adjustment for age of the other parent and other covariates, each 10-year increase in maternal age was associated with a 20% increase in ASD risk while each 10-year increase in paternal age was associated with a 30% increase in ASD risk.
Combined effects of parental age and birth order
The risk of ASD within each of 3 parental age categories (both parents "younger," 1 parent "older," and both parents "older") was highest among firstborn children and declined with increasing birth order . Considering the combined effects of parental age and birth order, (excluding from the analysis births to mothers aged <20 years), the lowest risk group to be third- or later-born offspring of mothers aged 20–34 years and fathers aged <40 years. The risk of ASD increased with both declining birth order and increasing number of older parents. The highest risk group included firstborn offspring of mothers aged 35 years and fathers aged 40 years, with a risk 3 times that of the reference group.
The overall results of this study provide the most compelling evidence to date that ASD risk increases with both maternal and paternal age and decreases with birth order.
Source: American Journal of Epidemiology, Advanced Parental Age and the Risk of Autism Spectrum Disorder
If you're interested in reading additional articles on autism, ScientificBlogging's Autism Awareness page provides a listing of all the articles on our site that touch on the subject.