The study found that rates of autism spectrum disorder were lowest among children born to young, similarly aged parents.
The study, published in Molecular Psychiatry, involved the analysis of more than 5.7 million children in Australia, Denmark, Israel, Norway and Sweden.
While previous research has established the association between older paternal age and autism spectrum disorders (ASD), there have been significant differences between studies, as well as a lack of definitive information on whether paternal and maternal ages are independent risk factors.
Co-author Michael Rosanoff, director of public health research for Autism Speaks, funders of the study, suggests that the study is unique among research conducted on ASD and parental age.
"By linking national health registries across five countries, we created the world's largest dataset for research into autism's risk factors," he explains. "The size allowed us to look at the relationship between parents' age and autism at a much higher resolution - under a microscope, if you will."
The aim of the researchers was to establish whether or not maternal and paternal ages independently increase the risk of ASD and, if so, to what extent.
Of the 5,766,794 children examined by the study, more than 30,000 had ASD. Each child included in the study was born between 1985-2004 and their development was tracked by the researchers until 2009. The researchers obtained autism diagnoses by consulting the appropriate national health records.
To separate the influence of maternal and paternal ages, the researchers adjusted their findings for the possible influence of the other parent's age, as well as controlling for other age-related influences that could influence ASD risk.
Age gaps between parents 'contribute significantly' to ASD risk
The researchers found ASD prevalence was 66% higher in children born to fathers aged older than 50 and 28% higher in children born to fathers in their 40s, compared with those born to fathers in their 20s.
Similarly, ASD prevalence among children born to mothers in their 40s was 15% higher, compared with children born to mothers in their 20s. However, ASD rates were also 18% higher among children born to teenage mothers.
ASD rates also rose when both parents were older and with widening gaps between the two parents' ages, with high rates observed whether it was the father or the mother who was oldest. ASD rates were highest when the father was aged 35-44 and his partner was 10 or more years younger.
"After finding that paternal age, maternal age and parental age gaps all influence autism risk independently, we calculated which aspect was most important," co-author Dr. Sven Sandin states. "It turned out to be parental age, though age gaps also contribute significantly."
The higher risk of ASD associated with fathers aged over 50 could be explained by the idea that genetic mutations in sperm, which become more prevalent as a man ages, could contribute to ASD. However, the association of ASD risk with maternal age and gaps in parental ages remains a mystery.
"In this study, we show for the first time that autism risk is associated with disparately aged parents," states co-author Abraham Reichenberg, a neuropsychologist and epidemiologist with the Icahn School of Medicine at Mount Sinai in New York City, NY. "Future research should look into this to understand the mechanisms."
The authors acknowledge a number of limitations for their study. There is a lack of information about some variables that could affect their findings, such as parental psychiatric history, and as an observational study, it is unable to rule out factors associated with parental age - such as birth weight and obstetric complications - as contributing to the results.
"Although parental age is a risk factor for autism," adds Dr. Sandin, "it is important to remember that, overall, the majority of children born to older or younger parents will develop normally."