Medical Examiner

Autism Linked to Common Genes

Environmental causes and specific mutations aren’t as much to blame.

Stockholm day care.
The researchers looked at a sample of Swedish children with and without autism to estimate the overall influence of genes.

Photo by Melanie Stetson Freeman/The Christian Science Monitor via Getty Images

Few topics in science are as divisive as autism. The disorder affects roughly one in 68 children, and more children than ever are receiving the diagnosis. In 6- to 17-year-old children, autism diagnoses increased by almost 1 percentage point between 2007 and 2012.* Symptoms typically don’t emerge until age 2 or 3, and the diagnosis isn’t always straightforward. There is no known cause, but over the years researchers have proposed quite a range: complications during birth, pesticides, not being loved enough, not having enough worms.

However, as in most traits, environmental factors alone do not cause autism; there is also a genetic basis. In fact, autism is mostly genetic: The Population-Based-Autism Genetics and Environment Study Consortium, a multidisciplinary group of researchers, published a study in Nature Genetics on Sunday in which it found that autism is 55 percent heritable. Heritability represents the portion of a certain trait that can be attributed to genetics. One example of a highly heritable trait is height. In the United States, height is 80 percent heritable. This means that your genes explain 80 percent of how tall you end up—for example, the genes of a 5’1” and 5’7” couple are unlikely to produce a 6’1” child—and the other 20 percent is left up to environmental factors, such as nutrition. Similarly, genes are responsible for more than half of a child’s chance of having autism. “Autism is really a lot more like height than people thought,” says Kathryn Roeder, a Carnegie Mellon University statistics and computational biology professor who led the study. “Each person has a certain amount of risk in their genome.”

The most surprising finding in this study is that the genetic risk for autism lies mostly in variations of common genes, and not specific mutations. A small mutation in a single gene can cause a disease such as Huntington’s, and mutation of the BRCA1 gene increases a woman’s chance of developing breast cancer. These sorts of mutations account for only 2.6 percent of autism risk, according to the new PAGES study, compared with 52 percent accounted for by common genes. In the vast majority of cases of autism, there is no one errant gene that codes for the disease, but rather a combination of common variations predicts autism risk. “You get a lot of the bad side of the coin and eventually push you into a disease,” says Roeder.

To determine the contribution of genetics versus environment, Roeder’s team studied data collected by the Swedish government, which maintains detailed medical databases. Sweden has a complete list of every autistic person in the country, and Roeder’s team took a random sample of about 400 autistic individuals from that list and compared them with more than 2,500 individuals without autism. In these samples, some people were related but not very closely. “Third cousins have genes in common, but they don’t have environment in common—they don’t even know each other,” says Roeder. That allowed the researchers to estimate the importance of shared genes in people who didn’t grow up together. The researchers compared the genes in pairs of autistic people and homed in on any groups of people who were distantly related and seemed to have higher instances of autism than would be predicted by chance. That allowed them to estimate the overall influence of genes.

Still, 55 percent heritability isn’t 100 percent. When I asked Roeder what environmental factors contribute to autism, she said she wasn’t sure. “Virtually nothing is known about environment—everything has been conjecture,” Roeder said. “They don’t know what causes it. All we’ve shown here is that [environmental factors are] not the main cause.”

This is an important contribution to our understanding of autism, if for no other reason than showing that parents’ actions have little role in the disorder. The belief that autism could be caused by failure to bond with a child is unfortunately still prevalent in some circles, despite decades of research suggesting the contrary. Given research linking advanced parental age and autism, some parents also worry that their decision to have children after their mid-30s contributed to their child’s disorder. The PAGES consortium’s work shows that genes are a more likely explanation.

It may be difficult for some people to accept that autism is caused by a complicated mix of risk factors, and the lack of clear answers can lead parents to desperate solutions. One particularly pernicious myth some parents cling to is that the chemical thimerosal, formerly found in childhood vaccines, causes autism. (Thimerosal was removed from childhood vaccines more than a decade ago, yet autism diagnoses continued to rise.) These views are championed by high-profile but ill-informed celebrities such as Robert F. Kennedy Jr., who believes (and just won’t stop claiming) that scientists are conspiring to hide thimerosal’s danger from the public, and Jenny McCarthy, who began her anti-vaccination crusade after her son was diagnosed with autism. (It turns out that McCarthy’s son may not have autism after all—many believe that he was misdiagnosed and actually has Landau-Kleffner syndrome—and, in any case, McCarthy claims to have cured her Crystal child of autism with “a gluten-free, casein-free diet, vitamin supplementation, detox of metals, and anti-fungals.”)

Despite countless studies showing that vaccines do not cause autism, the anti-vaccination movement has still successfully persuaded too many parents to opt out of vaccinations for their children. Opting out is dangerous for everyone because vaccines lose effectiveness when less than 95 percent of the population is vaccinated. This means people lose what’s called herd immunity: Even if one person gets a disease, it’s harder for the disease to spread across the population if others are vaccinated. In communities where herd immunity to diseases has been lost, there have been dangerous outbreaks of previously defunct diseases, like measles in Texas and Ohio and whooping cough in California.

Loving parents out there may be dismayed to know that science doesn’t tell us much about how to prevent autism, short of changing your genes. But on the upside, it does tell us yet again that we should vaccinate our kids so that they don’t suffer from the diseases we already know how to prevent.

*Correction, July 21, 2014: This article originally misstated the increase in autism diagnoses among 6- to 17-year-old children, and the years under comparison. The number of diagnoses increased by almost 1 percentage point rather than by 1.8 percent, and between 2007 and 2012 rather than 2007 and 2013. (Return.)