Melissa Smith had always wanted to have three children. Last April, as she was trying to get pregnant for a third time, she learned her youngest child was autistic. After she got over the initial shock, the news came as a relief. “I had known for months that something was wrong,” she said. “To finally have an answer, and to have a diagnosis that would afford us solutions and therapy to help, was definitely a relief.”
Still, she couldn’t help but worry that her next child would be autistic, too. She feels conflicted: She sees her son’s autism as a gift and wouldn’t change him, but caring for him takes effort. An additional child with autism could make that even harder.
Having one child with autism meant she had an increased chance of having another, because it tends to run in families. Then she saw a study linking medications called S.S.R.I.s (selective serotonin reuptake inhibitor) with autism. The connection is purely an association rather than proof the drugs cause autism. Smith, 30, had been taking an S.S.R.I. called Lexapro for the past year.
Lexapro was the only thing that alleviated her anxiety, which was so severe that it was difficult to drive a car, clean the house or shop for groceries without lashing out at her family. Her doctor couldn’t say for sure whether she should continue taking her medication, because other studies suggested stress from untreated anxiety carried its own prenatal risks, including preterm birth, which is also a risk factor for autism.
“It makes me feel stuck,” said Smith, who works as a coordinator at a clinic in Salt Lake City that tests children for genetic and metabolic disorders.
Then, at the end of January, her concerns were no longer hypothetical: She was pregnant. Smith sought the advice of genetic counselors, but none could tell her whether taking her medication would be riskier than not taking it.
“If this child does have autism, I don’t want to think, well, what if I had done this, or what if I had done that,” she said. “I just want to be able to live with myself at the end of the day and know it wasn’t something I did that caused it.”
In the past 50 years, scientists have compiled a short list of factors, including certain genes, premature birth, and some medications, that might contribute to autism. They have begun to understand which people have the greatest chances of having a child with autism, and they have identified a few things you can do to minimize those chances.
Still, there is no way to prevent a child from having autism, in large part because experts don’t actually know what causes it in most people. “We can find many, many factors that are correlated with autism,” said Dr. Katarzyna Chawarska, Ph.D., professor of child psychiatry at Yale University’s Child Study Center. “But in reality, we really don’t know if they’re involved in the cause of autism.”
It’s also important to note that many autistic people and their families see autism as something to be celebrated. And that traits of autism exist in everyone. “Many of them, whether you have autism or not, may very well give you something that’s positive,” such as an intense passion for certain subjects and a keen eye for detail, said Ed Cook, M.D., professor of psychiatry at the University of Illinois at Chicago.
One thing is certain: There’s no one single cause of autism, and every case is the product of multiple factors working together.
What’s in Your DNA?
Until the 1970s, many experts subscribed to now-discredited notions about autism being caused by “cold” parenting styles or growing up in extreme isolation (as in the famous case of the Wild Boy of Aveyron in France in the late 1700s). Since then, studies have shown autism runs in families and have put its heritability at around 80 percent, or about as heritable as height or eye color.
Scientists cannot fully explain why autism runs in families. They suspect that, like height, it has complex causes and is determined by numerous genes that act together to influence a person’s odds of having autism. Their studies suggest that, collectively, these genes could explain about half of autism’s genetic basis.
Another 10 to 20 percent can be explained by rare but powerful mutations that occur at random and are unrelated to genes that run in families. Scientists have found about 100 such genes that, when mutated, can substantially raise a person’s odds of having autism.
Even when these mutations do occur, they don’t cause autism every time. “None of these things are perfectly predictive of an autism diagnosis, and each one of them is individually quite rare,” said Dr. Jeremy Veenstra-VanderWeele, M.D., a child and adolescent psychiatrist at Columbia University. “We can’t really put our finger on exactly what the risk is going to be for an individual.”
What Else Raises the Risk?
So, autism is highly genetic, but that doesn’t mean environmental factors are unimportant, said Dr. Brian Lee, Ph.D., associate professor of epidemiology and biostatistics at the A.J. Drexel Autism Institute in Philadelphia. Height, for example, is influenced by environmental factors like malnutrition. “You can have a genetic underlying factor that puts you at risk, but there may need to be some sort of an environmental condition or trigger,” Lee said.
S.S.R.I.s are strong candidates for factors that could trigger autism. They act on the brain chemical serotonin, which is important for social function and is found at high levels in some autistic people, and many of them cross from a woman’s blood into the womb.
But Dr. Alan Brown, M.D., professor of psychiatry and epidemiology at Columbia University, said that it’s difficult to separate the effects of a medication from a pregnant woman’s reason for taking them — her underlying mental health condition. Several recent studies suggest that the autism risk associated with S.S.R.I. use is very small, perhaps nonexistent. “The literature is confusing on S.S.R.I.s and autism,” Brown said. “Some studies show the association, others don’t.”
Researchers also have looked into prenatal exposure to toxins, such as pesticides or air pollution, and have come up with similarly inconclusive results. They have found no increased autism risk related to prenatal smoking, cesarean section, fertility treatments or vaccines.
The most reliable risk factor by far is having had a previous child with autism. About 1 in 5 families like Smith’s go on to have another child on the spectrum. Biological sex matters, too: males are about three times as likely as females to have autism, although the reasons for this are not fully understood. The chances of having an autistic child also increase slightly and steadily with a parent’s age.
“Most of the potential risk factors for autism are things that happen around birth, when early brain development is happening,” said Dr. Paul Carbone, M.D., a pediatrician and associate professor of pediatrics at the University of Utah. “But do they by themselves cause autism? Or is it far more common for our environment to somehow interact with genetic susceptibility to increase a person’s risk?”
Risk Is Not Destiny
It’s important to recognize that most families with these risk factors do not have children with autism. And autism remains relatively rare: about 1.8 percent of children in the United States have a diagnosis, according to the latest estimates by the Centers for Disease Control and Prevention. Autism rates have risen over the past few decades, but almost all of that is because of increased awareness and shifts in diagnostic criteria that qualify more people for a diagnosis, said Dr. Eric Fombonne, M.D., professor of psychiatry, pediatrics and behavioral neuroscience at Oregon Health & Science University. “A lot of the increase is accounted for by the fact that what we call autism today is much larger than what we called autism 50 years ago,” he said.
And, perhaps most importantly, most of autism’s influences remain outside of our control anyway. “There’s certainly a lot of things we don’t know about the causes of autism,” Dr. Carbone said. “But one thing that we absolutely are sure of is that parents don’t cause autism.”
Smith said her doctor understood her concerns about taking an S.S.R.I., and put her on an alternative medication called Buspar. But after taking it for about a week in early March, Smith had severe nausea and migraines, possible side effects of the drug. So she stopped taking it, and since then she has been managing her anxiety with meditation and breathing exercises.
In the first week of April, Smith learned the baby was a boy. This worried her at first, because males have a higher risk of autism than females. But then she came to terms with it. “It’s not up to me,” she said. “I’ll just do anything I can to try and make sure this child is healthy, and that’s all I can do.”
Nicholette Zeliadt is a freelance health and science reporter.