California scientists have discovered clusters of autism, largely in the Los Angeles and San Francisco areas, where children are twice as likely to have autism as children in surrounding areas.

The 10 clusters were found mostly among children with highly educated parents, leading researchers to report that they probably can be explained by better access to medical experts who diagnose the disorder.

Because of the strong link to education, the researchers from University of California at Davis said the new findings do not point to a localized source of pollution, such as an industry, near the clusters.

“I suspect access to services plays the major role,” said Irva Hertz-Picciotto, senior author of the study published Tuesday in the journal Autism Research.

She added, however, that there could be other reasons why higher-educated parents lead to more autism. Environmental exposures, such as chemicals from consumer products, could be more common in those households, she said.

“Certainly there may be some consumer products to which more educated persons are more likely to be exposed. There is undoubtedly a possibility of higher exposures in the more educated,” said Hertz-Picciotto, a professor of public health sciences and an autism expert at the UC Davis MIND Institute.

For the study, the researchers analyzed the birth records of about 2.5 million babies born in California between 1996 and 2000. Nearly 10,000 were later diagnosed with an autism spectrum disorder.

Autism is a neurological disorder involving impaired social development and communication skills. Experts believe that it is caused during pregnancy or early infancy since symptoms often arise by age two or three.

The new research is the first time that anyone has looked for geographic areas with high incidence, or clusters, of autism, according to lead author Karla Van Meter, an epidemiologist.

The 10 clusters were located in: the Torrance, Beverly Hills, Van Nuys and Calabasas areas of Los Angeles County; the Laguna Beach/Mission Viejo area of Orange County; the La Jolla/Del Mar area of San Diego County; San Francisco; the Sunnyvale/Santa Clara area; the Redwood City area; and Fresno. Two other possible clusters were also found in the Norwalk/Cerritos area of Los Angeles County and the Modesto area.

Although children within these cluster areas were more likely to have autism, the majority of California’s autistic children were born outside of them.

Bernard Weiss, a professor of environmental medicine at University of Rochester who was not involved in the study, said the research “confirms an association between parental education and autism risk” that had been found in earlier studies.

“I would agree that their findings indicate a higher likelihood of seeking services by educated parents,” he said.

In contrast, other research has shown that attention deficit disorders are less likely in children of high socioeconomic status, Weiss said.

The UC Davis study also found a link between autism and older age of the parents, but it was much weaker than the link to education.

One in every 110 U.S. children is diagnosed with an autism disorder, an increase of 57 percent between 2002 and 2006, according to a report by the U.S. Centers for Disease Control and Prevention.

One California study found a sevenfold increase in autistic children between 1990 and 2006.

Many experts say that improved diagnoses by doctors can explain only a portion of the increase, so environmental causes may be responsible for the rest. Hertz-Picciotto said the new study cannot explain the rising rate of autism since education levels have not changed.

CDC behavioral health scientist Catherine Rice, an autism expert, would not comment Tuesday on the new study, but she said that "much work needs to be done to understand how complex biologic and environmental factors interact to result in the symptoms which make up autism spectrum."

The CDC and medical institutes in six states are working on the Study to Explore Early Development, or SEED, the largest study in the United States to identify the genetic factors, environmental exposures and pregnancy factors that may increase risk of autism.

Dozens of chemicals in the environment are neurodevelopmental toxicants, which means they alter how the brain grows. Mercury, polychlorinated biphenyls, lead, brominated flame retardants and pesticides are examples. Mothers of autistic children were twice as likely to use pet flea shampoos, which contain pesticides, according to one study that has not yet been published.

Another new study found a link between autism and phthalates, which are used in vinyl and cosmetics.

When looking for clusters of diseases, the aim “is to look for possible evidence of localized sources of exposures,” such as a local industry, Hertz-Picciotto said.

“Since the [autism] clusters seemed to be associated, quite consistently, with parental education, the conclusion here is that environmental factors that are involved in autism are not likely to be localized sources of contaminants. This means we need to look within the homes, or for widely dispersed types of environmental factors,” she said.

This article originally ran at Environmental Health News, a news source published by Environmental Health Sciences, a nonprofit media company.