Autism
The Centers for Disease Control and Prevention's new report on autism spectrum disorder (ASD), released Thursday, offers little good news. It estimates that about one in every 59 children nationwide has the complex developmental disorder, a figure that seems unchanged from its previous estimate a year ago. But, for a few reasons, it's likely that this is an underestimate.

Autism spectrum disorder is a complicated condition, one with several contributing causes. Genes likely play a role, as do environmental factors like air pollution or having older parents. Its symptoms are complex too, but generally include problems communicating, a sensitivity to touch and other sensory inputs, and digestive issues.

For more than a decade, the CDC has estimated the autism rate in children through its Autism and Developmental Disabilities Monitoring network, or ADDM. The ADDM pools together health and special education records from children living in various states to come up with its estimate. Traditionally, this estimate-including the one released in 2018- has relied on data from 8-year-olds diagnosed with ASD. But the new report is based on data from 4-year-olds.

"The 4-year-old estimate is definitely a provisional one, with the expectation that it will actually be greater when you survey children at an older age," study author Walter Zahorodny, an associate professor of pediatrics at Rutgers, told Gizmodo. "But the advantage of studying younger children is that it's an early signal into whether the trend [of autism rates] is changing or not."

The report collectively looked nearly at 200,000 4-year-old children from seven states in the years 2010, 2012, and 2014. Only three states-New Jersey, Missouri, and Arizona-had data for all the years, though. Like other estimates, it found that the overall national prevalence of autism has increased, from 13.4 per 1,000 children in 2010 to 17.4 per 1,000 in 2014. That roughly amounts to one in every 59 children.

It's a tricky matter trying to figure out exactly how common autism is, though. Doctors, parents, and schools have generally gotten better at spotting ASD, for one. But according to Zahorodny, there has also been a real rise in autism.

"We've been monitoring for approximately 20 years, and it's clear that this increase is not all better monitoring," he said, referencing research showing around that 80 percent of the increase can be tied to outside factors like the environment. And while the new report came up with the same one in 59 figure as the CDC's earlier estimate did, it's likely that as these children get older, more will end up being diagnosed with autism.

"What these findings indicate is that we have yet to see a point where the autism rate has plateaued. It's increasing, and when we look at this same group four years from now, we'll probably see the extent to which [this rate] was an underestimate, too," said Zahorodny.

There's also the possibility that the current national estimate might be off right now. Looking at the states individually, New Jersey was the only one to show an increase in the autism rate through 2014, which was one in every 35 children. But there are several reasons to think the rate in New Jersey is the best barometer of autism's prevalence, according to Zahorodny.

"The quality of information we have in New Jersey is very good, and it's that information that surveillance studies like this need," he said. "The resources for identifying these children are really here."

New Jersey, for instance, has a relatively large and diverse population. Its public education system might be able to better connect children and parents to special education services than those in other states (the state is consistently ranked near the top in education by various organizations). And it's in the New York Metropolitan area, where families have access to plenty of doctors and specialists.

Something else that will complicate how we talk about the autism rate in the years to come is how it's now diagnosed. Previously, clinicians relied on the fourth edition of the Diagnostic And Statistical Manual Of Mental Disorders, or DSM, to diagnose a case of autism. But the DSM-V revised its criteria of ASD. That included doing away with the label of Asperger's disorder, which was considered a milder form of autism, and creating a new diagnosis-social communication disorder-that will often be applied to people with Asperger's-like symptoms.

When Zahorodny and his co-authors used both sets of criteria from the DSM-IV and DSM-V in the 2014 group of children, for instance, they found that roughly 20 percent fewer children were considered to have ASD under DSM-V. So scientists (and journalists) will need to find a way to account for these differences and accurately portray the picture of ASD and similar disorders in America.

"It's not like these individuals go away when we start categorizing them differently," Zahorodny said. "But it is possible-and very desirable-to maintain a consistent method in how we define autism."

Of course, the current estimates from the CDC are somewhat dated, given that they only look as recently as 2014. But Zahorodny says the CDC is the middle of tracking the prevalence of autism in both 4- and 8-year-olds with more recent data, and his own team is already at work sifting through some of that data. (Research elsewhere has continued to find growing rates of autism with other, more updated sources of data). The ADDM will also begin studying how children at age 16 are coping with ASD, including whether any behavioral interventions have helped.

In the meantime, we're still struggling to help young children with ASD. We know, for instance, that the earlier ASD is diagnosed in someone, the more effective interventions like communication therapy can be. But the average age of diagnosis, around 4.5 years, hasn't shifted at all in recent years, according to the CDC data. Untangling the genetic and environmental causes of autism (which do not include vaccines), and how we can prevent them is also slow-going.

"We don't really have any idea of why this increase has occurred, and what the risk factors and triggers for this increase are. And as importantly, we have not made any real progress in improving early detection," Zahorodny said. "So please, let's do everything possible to enhance the early screening of autism-at 18 to 24 months preferably-and let's keep watching and trying to understand the factors behind the rise of autism."