Stefania Sterling was just 21 when she had her son, Charlie. She was young and healthy, with no genetic issues apparent in either her or her husband's family, so she expected Charlie to be typical.
"It is surprising that the prevalence of a significant disorder like autism has risen so consistently over a relatively brief period."
It wasn't until she went to a Mommy and Me music class when he was one, and she saw all the other one-year-olds walking, that she realized how different her son was. He could barely crawl, didn't speak, and made no eye contact. By the time he was three, he was diagnosed as being on the lower functioning end of the autism spectrum.
She isn't sure why it happened – and researchers, too, are still trying to understand the basis of the complex condition. Studies suggest that genes can act together with influences from the environment to affect development in ways that lead to Autism Spectrum Disorder (ASD). But rates of ASD are rising dramatically, making the need to figure out why it's happening all the more urgent.
The Latest News
Indeed, the CDC's latest autism report, released last week, which uses 2016 data, found that the prevalence of ASD in four-year-old children was one in 64 children, or 15.6 affected children per 1,000. That's more than the 14.1 rate they found in 2014, for the 11 states included in the study. New Jersey, as in years past, was the highest, with 25.3 per 1,000, compared to Missouri, which had just 8.8 per 1,000.
The rate for eight-year-olds had risen as well. Researchers found the ASD prevalence nationwide was 18.5 per 1,000, or one in 54, about 10 percent higher than the 16.8 rate found in 2014. New Jersey, again, was the highest, at one in 32 kids, compared to Colorado, which had the lowest rate, at one in 76 kids. For New Jersey, that's a 175 percent rise from the baseline number taken in 2000, when the state had just one in 101 kids.
"It is surprising that the prevalence of a significant disorder like autism has risen so consistently over a relatively brief period," said Walter Zahorodny, an associate professor of pediatrics at Rutgers New Jersey Medical School, who was involved in collecting the data.
The study echoed the findings of a surprising 2011 study in South Korea that found 1 in every 38 students had ASD. That was the the first comprehensive study of autism prevalence using a total population sample: A team of investigators from the U.S., South Korea, and Canada looked at 55,000 children ages 7 to 12 living in a community in South Korea and found that 2.64 percent of them had some level of autism.
Searching for Answers
Scientists can't put their finger on why rates are rising. Some say it's better diagnosis. That is, it's not that more people have autism. It's that we're better at detecting it. Others attribute it to changes in the diagnostic criteria. Specifically, the May 2013 update of the Diagnostic and Statistical Manual of Mental Disorders-5 -- the standard classification of mental disorders -- removed the communication deficit from the autism definition, which made more children fall under that category. Cynical observers believe physicians and therapists are handing out the diagnosis more freely to allow access to services available only to children with autism, but that are also effective for other children.
Alycia Halladay, chief science officer for the Autism Science Foundation in New York, said she wishes there were just one answer, but there's not. While she believes the rising ASD numbers are due in part to factors like better diagnosis and a change in the definition, she does not believe that accounts for the entire rise in prevalence. As for the high numbers in New Jersey, she said the state has always had a higher prevalence of autism compared to other states. It is also one of the few states that does a good job at recording cases of autism in its educational records, meaning that children in New Jersey are more likely to be counted compared to kids in other states.
"Not every state is as good as New Jersey," she said. "That accounts for some of the difference compared to elsewhere, but we don't know if it's all of the difference in prevalence, or most of it, or what."
"What we do know is that vaccinations do not cause autism."
There is simply no defined proven reason for these increases, said Scott Badesch, outgoing president and CEO of the Autism Society of America.
"There are suggestions that it is based on better diagnosis, but there are also suggestions that the incidence of autism is in fact increasing due to reasons that have yet been determined," he said, adding, "What we do know is that vaccinations do not cause autism."
Zahorodny, the pediatrics professor, believes something is going on beyond better detection or evolving definitions.
"Changes in awareness and shifts in how children are identified or diagnosed are relevant, but they only take you so far in accounting for an increase of this magnitude," he said. "We don't know what is driving the surge in autism recorded by the ADDM Network and others."
He suggested that the increase in prevalence could be due to non-genetic environmental triggers or risk factors we do not yet know about, citing possibilities including parental age, prematurity, low birth rate, multiplicity, breech presentation, or C-section delivery. It may not be one, but rather several factors combined, he said.
"Increases in ASD prevalence have affected the whole population, so the triggers or risks must be very widely dispersed across all strata," he added.
There are studies that find new risk factors for ASD almost on a daily basis, said Idan Menashe, assistant professor in the Department of Health at Ben-Gurion University of the Negev, the fastest growing research university in Israel.
"There are plenty of studies that find new genetic variants (and new genes)," he said. In addition, various prenatal and perinatal risk factors are associated with a risk of ASD. He cited a study his university conducted last year on the relationship between C-section births and ASD, which found that exposure to general anesthesia may explain the association.
Whatever the cause, health practitioners are seeing the consequences in real time.
"People say rates are higher because of the changes in the diagnostic criteria," said Dr. Roseann Capanna-Hodge, a psychologist in Ridgefield, CT. "And they say it's easier for children to get identified. I say that's not the truth and that I've been doing this for 30 years, and that even 10 years ago, I did not see the level of autism that I do see today."
Sure, we're better at detecting autism, she added, but the detection improvements have largely occurred at the low- to mid- level part of the spectrum. The higher rates of autism are occurring at the more severe end, in her experience.
A Polarizing Theory
Among the more controversial risk factors scientists are exploring is the role environmental toxins may play in the development of autism. Some scientists, doctors and mental health experts suspect that toxins like heavy metals, pesticides, chemicals, or pollution may interrupt the way genes are expressed or the way endocrine systems function, manifesting in symptoms of autism. But others firmly resist such claims, at least until more evidence comes forth. To date, studies have been mixed and many have been more associative than causative.
"Today, scientists are still trying to figure out whether there are other environmental changes that can explain this rise, but studies of this question didn't provide any conclusive answer," said Menashe, who also serves as the scientific director of the National Autism Research Center at BGU.
"It's not everything that makes Charlie. He's just like any other kid."
That inconclusiveness has not dissuaded some doctors from taking the perspective that toxins do play a role. "Autism rates are rising because there is a mismatch between our genes and our environment," said Julia Getzelman, a pediatrician in San Francisco. "The majority of our evolution didn't include the kinds of toxic hits we are experiencing. The planet has changed drastically in just the last 75 years –- it has become more and more polluted with tens of thousands of unregulated chemicals being used by industry that are having effects on our most vulnerable."
She cites BPA, an industrial chemical that has been used since the 1960s to make certain plastics and resins. A large body of research, she says, has shown its impact on human health and the endocrine system. BPA binds to our own hormone receptors, so it may negatively impact the thyroid and brain. A study in 2015 was the first to identify a link between BPA and some children with autism, but the relationship was associative, not causative. Meanwhile, the Food and Drug Administration maintains that BPA is safe at the current levels occurring in food, based on its ongoing review of the available scientific evidence.
Michael Mooney, President of St. Louis-based Delta Genesis, a non-profit organization that treats children struggling with neurodevelopmental delays like autism, suspects a strong role for epigenetics, which refers to changes in how genes are expressed as a result of environmental influences, lifestyle behaviors, age, or disease states.
He believes some children are genetically predisposed to the disorder, and some unknown influence or combination of influences pushes them over the edge, triggering epigenetic changes that result in symptoms of autism.
For Stefania Sterling, it doesn't really matter how or why she had an autistic child. That's only one part of Charlie.
"It's not everything that makes Charlie," she said. "He's just like any other kid. He comes with happy moments. He comes with sad moments. Just like my other three kids."
Scientists have been using MRI images (like the ones pictured here) to train artificial intelligence to detect cancers earlier and with more accuracy. Here, MIT's AI system, MIRAI, looks for patterns in a patient's mammograms to detect breast cancer earlier than ever before. news.mit.edu 
