Autism, Old Grandpas and Mom's Childhood

March 21, 2013

Two new research reports shine light on multi-generational influences on autism. Here's some much needed perspective on what they mean - and don't mean...

Posted by Alycia Halladay, PhD, Autism Speaks senior director of environmental and clinical sciences

Two studies published this week by JAMA Psychiatry look at some surprising influences on autism risk that appear to cross generations. One found higher rates of autism among the grandchildren of men who were in their fifties or older when they became fathers. The other found higher rates of autism among the children of women who’d been abused as children.

To the nonscientist, these connections may sound far-fetched. But while further research is needed, these sorts of multigenerational effects have already been discovered behind other disorders such as cancer, diabetes and schizophrenia. We have good reason to believe that they may likewise play a role in autism. 

In these two studies, the researchers used large population samples that allowed them to compare the family histories of hundreds of children with autism with those of tens of thousands of children with typical development.

The increases in autism risk that they documented were small but significant.  This means that most people with autism probably don’t have older grandpas or mothers who were abused. To emphasize: These findings tell us about increased autism “risk,” not “cause.” They shouldn’t be read as reasons to assign blame. 

What makes these findings particularly interesting is that they add to a growing body of evidence showing that nongenetic autism risk factors can span multiple generations. I see the real impact in how this changes how we understand the scope of environmental factors that influence autism risk. 

Clearly, we need more research to understand the mechanisms behind the increased risk seen in these two studies. The authors suggested some possible mechanisms.  For example, the old-grandfather effect might involve the same increase in small genetic mutations previously associated with having an older father. These tiny mutations may accumulate in men’s sperm germ cells over a lifetime and, so, be passed down to future generations. (More on older dads and autism risk here.)   

The effect seen in women who were abused as children or teens points, in part, to associated increases in gestational diabetes, preeclampsia and smoking during pregnancy. These are known risk factors for autism. However, they accounted for only a small fraction of the increased autism risk documented in this study. So the search continues for other possible associations – such as the lifelong legacy of increased stress and impaired immune function that may follow childhood abuse.

Of particular interest, these studies add to a growing body of research suggesting that “epigenetic” changes can play a significant role in brain development and autism risk. Epigenetics refers to the control of when and where genes become active, or express themselves. It doesn’t involve genetic mutation (change to a gene’s DNA). Instead, epigenetic changes involve mechanisms that turn on gene expression or turn it off without changing the DNA sequence itself.    

Importantly, we now know that outside influences – such as age and exposure to toxic chemicals – can produce epigenetic changes that may contribute to disease.

In recent years, epigenetic research has produced breakthroughs in the understanding of cancer, diabetes and schizophrenia. In particular, this research suggests that environmental exposures may alter epigenetic-controlled gene expression across generations. Yet these effects have proven to be anything but straightforward. Epigenetics remains a field in its infancy.

As part of Autism Speaks great interest in this work, we’re pleased to be co-sponsoring a two-day conference this weekend: “Environmental Epigenetics: New Frontiers in Autism Research,” at the University of California, Davis, MIND Institute. Attendees will discuss research on the relationship between environmental exposures, gene expression and brain development.

The two new JAMA Psychiatry reports are sure to be a hot topic of discussion! We’ll be videotaping and uploading the presentations for viewing in the week ahead. You can find them – and more information on autism and epigenetics – at

We're Here to Help Chat with Us
Autism Response Team Chat
There are no available agents at the moment.

You can also reach the Autism Response Team by phone or email: 888-288-4762, en Espanol 888-772-7050, or