Autism Cases on the Rise; Reason for Increase a Mystery

Scientists are scouring genetic and evironmental data to find a cause for the rise in autism.

Medically Reviewed by Louise Chang, MD on March 28, 2008
9 min read

The number of children diagnosed with autism or related disorders has grown at what many call an alarming rate. In the 1970s and 1980s, about one out of every 2,000 children had autism.

Today, the CDC estimates that one in 150 8-year-olds in the U.S. has an autism spectrum disorder, or ASD. This expanded definition refers not only to autism but also to a collection of brain development disorders such as Asperger's syndrome and a condition known as pervasive developmental disorder -- not otherwise specified (PDD-NOS). Though all the disorders share some symptoms, they are different in other ways, including the timeline of symptoms and the severity, according to the CDC.

The apparent rise in cases triggers two burning questions for parents, physicians, and scientists:

  • Is autism truly on the rise, or do the new statistics simply reflect the growing awareness of the condition, the expanded definition, and other factors?
  • If autism is on the rise, as most experts believe, what is causing the increase?

The jump in autism cases has spawned not only alarm but also debate about whether the number of children with autism could have increased that much in a relatively brief time.

"There's a lot of controversy about that," says Jeff Milunsky, MD, director of clinical genetics and associate director of the Center for Human Genetics at Boston University.

Two researchers who tracked the rate of autism in children born in the same area of England from 1992 to 1995 and then from 1996 to 1998 found that the rates were comparable, and concluded that the incidence of autism was stable. The study was published in the American Journal of Psychiatry in 2005.

But, Milunsky says, several studies have documented an increase in the U.S.

In a recent report in the journal Archives of Disease in Childhood, Milunsky and his colleagues point to several studies finding an increase in autism rates. In 2003, for instance, a large study conducted in Atlanta found that one in 166 to one in 250 children had autism, according to a report published in the Journal of the American Medical Association.

Another study conducted by the CDC in 14 states found an overall prevalence of one in 152, which Milunsky and others say is the generally accepted figure today.

Other experts say autism is on the increase but that factors other than more children being diagnosed play a role. Some of the increase in reported cases is because of "diagnostic substitution," says Paul Shattuck, PhD, assistant professor of social work at Washington University in St. Louis and an autism researcher.

"A kid labeled autistic today could have been labeled mentally retarded 10 years ago in the same school system," Shattuck says. It wasn't until 1992 that schools began to include autism as a special education classification.

Today, children diagnosed as having autism spectrum disorder are often more mildly affected than the classic "Rain Man" stereotype some people associate with the disorder, Shattuck says. After autism was first identified in 1943, some of the first studies found most of the children intellectually disabled. "Today the minority of kids [with ASD are intellectually disabled],'' Shattuck tells WebMD.

The debate about whether the reported increase in autism is affected by factors such as more awareness misses the point, says Isaac Pessah, PhD, a professor of toxicology, director of the Center for Children's Environmental Health Sciences, and a member of the MIND Institute at the University of California Davis. Rather than argue about whether the increase is because of some children being reclassified or other factors, he says, "We need to understand why it's one in 150."

Focusing on the actual numbers -- rather than the debate -- is wise, says Craig Newschaffer, PhD, chairman and professor of the department of epidemiology and biostatistics at Drexel University School of Public Health in Philadelphia. "We thought autism was a very rare occurrence, and it's clear that it's not."

Getting to the cause -- or, more accurately, causes -- of autism will be more difficult than unraveling the causes of cancer, says Gary Goldstein, MD, president and CEO of Kennedy Krieger Institute in Baltimore, a facility that helps children with autism and other developmental disorders.

"This is harder than cancer because in cancer you can biopsy it; you can see it on an X-ray," Goldstein says. "We don't have a blood test [for autism]. There is no biomarker, no image, no pathology."

"There won't be one single explanation,'' says Marvin Natowicz, MD, PhD, a medical geneticist and vice chairman of the Genomic Medicine Institute at the Cleveland Clinic.

"There's been a lot of progress in the last few years in terms of understanding the causes of autism," Natowicz says. "We know a lot more than we did." Still, he says, research has a long way to go. "One number you see often is that about 10% of those with autism have a definitive diagnosis, a causative condition." The other 90% of cases are still a puzzle to the experts.

Often, a child with autism will have a co-existing problem, such as a seizure disorder, depression, anxiety, or gastrointestinal or other health problems. At least 60 different disorders -- genetic, metabolic, and neurologic -- have been associated with autism, according to a report published in The New EnglandJournal of Medicine.

On one point most agree: A combination of genetics and environmental factors may play a role. Scientists are looking at both areas.

Some evidence that genetics plays a role in autism and ASD is provided by research on twins. According to the CDC, if one identical twin has autism, there's a 75% chance the other twin will be affected, too. If a fraternal twin is affected, the other twin has a 3% chance of having autism.

Parents who give birth to a child with ASD have up to an 8% chance of having another child who is also affected, the CDC estimates.

Many U.S. couples have delayed childbearing, and the older ages of both the mother and the father have been linked with a higher risk of having children with ASD, according to a report in the journal Pediatrics. With age could come increased risk for genetic mutations or other genetic problems.

Specific genetic problems help explain only a small percentage of autism cases so far. "We know that major chromosomal abnormalities are identified in about 5% of ASD," says Milunsky of Boston University. "We know that Fragile X syndrome is responsible for about 3%." Fragile X syndrome, a family of genetic conditions, is the most common cause of inherited mental impairment, and also the most common known cause of autism or autism-like behaviors.

"Hot spots" of genetic instability may play a role, researchers say. For instance, a team of researchers reported in The New England Journal of Medicine that duplications and deletions on a specific chromosome seem to be associated with some cases of autism.

Specific genes or problems on chromosomes are implicated in a small number of ASD cases, Milunskey writes in a report on autism research published in the Archives of Disease in Childhood. For instance, maternal duplication on a specific chromosome region has been linked to about 1% of those with ASD.

"We are homing in on those 'hotspot' regions and identifying some of the single genes involved in either the direct causation or the susceptibility to ASD," Milunsky says.

But genetics is not the whole story, he and other experts say.

A variety of environmental triggers is under investigation as a cause or contributing factor to the development of ASD, especially in a genetically vulnerable child.

Exposure to pesticides during pregnancy may boost risk. In a study published in Environmental Health Perspectives, researchers compared 465 children diagnosed with ASD with nearly 7,000 children without the diagnosis, noting whether the mothers lived near agricultural areas using pesticides.

The risk of having ASD increased with the poundage of pesticides applied and with the proximity of the women's homes to the fields.

Besides pesticide exposure, exposure to organic pollutants that have built up in the environment are another area of concern, says Pessah of UC Davis. For instance, polychlorinated biphenyls or PCBs, substances previously found in electrical equipment, fluorescent lighting and other products, are no longer produced in the U.S. but linger in the environment, he says. "Particular types of PCBs are developmental neurotoxins," he says.

Another toxin to the brain is mercury in its organic form. But according to a report published in Pediatrics, there is no evidence that children with autism in the U.S. have increased mercury concentrations or environmental exposures. Though many parents of children with ASD believe their child's condition was caused by vaccines that used to contain thimerosal (a mercury-containing preservative), the Institute of Medicine concludes there is no causal association.

Even so, many autism organizations remain convinced there is a link. The vaccine-autism debate reignited in early March 2008, after federal officials conceded to award compensation to the family of a 9-year-old Georgia girl who developed autism-like symptoms as a toddler after getting routine childhood vaccinations. Officials said the childhood vaccines given to the girl in 2000, before thimerosal was phased out, aggravated a pre-existing condition that then manifested as autism-like symptoms. The pre-existing condition was a disorder of the mitochondria, the "power sources" of the cell, according to the family.

More answers are coming. Pessah of UC Davis is one of the researchers in the CHARGE Study (Childhood Autism Risks from Genetics and the Environment), an ongoing study of 2,000 children. Some of the children have autism, some have developmental delay but not autism, and some are children without developmental delays.

Pessah and other researchers are focusing on how the interaction of genes and the environment play a role in autism.

Among the findings so far, he says, is that the immune system functioning of the mother may play a role in the child's later development of autism. Pessah and his colleagues took blood samples from 163 mothers in the CHARGE study -- 61 had children with autism, 62 had normally developing children, and 40 had children with non-autistic developmental delays. Then they isolated immune system antibodies, called IgG, from the blood of all the mothers. They took the blood samples and exposed them in the laboratory to fetal brain tissue obtained from a tissue bank.

Antibodies from the mothers of children with autism were more likely than antibodies from the other two groups to react to the fetal brain tissue, Pessah says, and there was a unique pattern to the reaction.

In an animal study, the UC Davis team then injected the antibodies into animals. The animals getting the IgG antibodies from mothers of children with autism displayed abnormal behavior, while the animals given antibodies from the mothers of normally developing children did not exhibit abnormal behaviors.

In another study, the UC Davis team found that levels of leptin, a hormone that plays a role in metabolism and weight, was much higher in children with autism than in normally developing children, especially if their autism was early in onset.

Another study, just launched by the CDC and now enrolling children, will track genetic and environmental factors that may increase risk for ASD.

Called SEED -- the Study to Explore Early Development -- the five-year study will follow more than 2,000 children at six sites across the U.S., says Newschaffer of Drexel, a co-principal investigator of the study. Some will have been diagnosed with ASD, some will have a developmental problem other than ASD, and a third group will be children without developmental problems.

Researchers will collect a host of genetic and environmental information, Newschaffer tells WebMD. They will find out about medical and genetic histories of the children and their parents, exposures during pregnancy to potential toxins, information about behavior, sleep problems, gastrointestinal problems, and other facts.

The hope, he says, is to find things that "stand out" -- early exposure to certain substances, for instance, or certain genetic information or a specific behavior pattern -- that might turn out to be markers for ASD.

Even if some environmental exposures or other findings do stand out, he says, "we are going to have to resist the temptation to say, 'This is it,'" Newschaffer says.

Natowitz of Cleveland Clinic agrees. "There won't be one single explanation."