What Are My Chances of Having Another Child with Autism?

I have a 3.5-year-old who was diagnosed as mildly on the autism spectrum, with mainly social challenges. We’ve been thinking about having another baby. My husband just told me he’s fearful that our genes cause autism and, if we have another child, they may not only have autism but be very severely on the spectrum.

What are the chances that a second child of ours will also have autism? Is there any link between siblings and the autism spectrum, and also levels on the spectrum? I’m unable to find any clear and compelling research or data about this.

—Anne

We’re in complicated territory here, so it’s not surprising that this is difficult to parse. There is a lot in your question — on the autism spectrum, on different types — that is fodder for a full article.

But your sibling question is one that can be answered, at least partially, with data. To do so, researchers use various approaches to evaluate how an autism diagnosis in one sibling predicts diagnosis in others.

One large study comes from Denmark, where researchers used registry data — basically, medical records covering the entire country — to answer this question. Overall, in this study, only a very small share of autism diagnoses are uncovered. However, those with a sibling who is diagnosed had a risk 22 times as high. This still put the share only around 2%, though.

Other studies have pointed to a larger risk. This paper followed siblings of children with an autism diagnosis through the age of 3 and found that 20% of them were themselves diagnosed. Generally, figures quoted range from 3% to 20%, which is a huge range and reflects the variation in the data.

What is less well understood is the relationship in degrees; if one child has mild autism, is that informative about the other? This isn’t something we know well from the data. Even with the data we have, it can be hard to interpret these facts. Autism diagnosis is likely more common among families who seek it out. Having one child with a diagnosis is likely to mean a family is better informed about this possibility, and more likely to identify behaviors as possibly indicating something to discuss with their doctor.

Where does this leave you? There is an elevated risk, although certainly that risk is not close to 100%, and the majority of the time children with autism have siblings who do not have it. To your question on severity, the data simply isn’t compelling in either direction.