Defining the Term

The Centers for Disease Control and Prevention (CDC) define autism spectrum disorders as developmental disabilities comprising not only classic autism, but also Asperger's disorder and a cluster of other conditions grouped under the rubric "pervasive development disorder-not otherwise specified" (PDDNOS). A range of social and behavioral symptoms characterize ASDs, including fearfulness, withdrawal, impaired ability to communicate, and rigid or repetitive behaviors. The severity of the condition varies greatly, as does the exact constellation of symptoms. Some autistic people have intellectual disabilities, some don't. Some have linguistic impairments, some don't. Intense sensitivities-..-to light or noise, for instance--are a common feature. Some researchers have suggested that immune and digestive problems are part of the syndrome.

"We call all these conditions 'autism,'" said Julie Daniels, assistant professor of epidemiology and maternal and child health at the University of North Carolina Gillings School of Global Public Health, "but they are probably not caused by the same thing, although they might have similar components. They might all have gene x and gene y, but one condition requires gene z, and another also requires environmental condition x."

In other words, ASDs are probably not "one disease."

Looking for Causes--A Perennial Source of Controversy Discussion of autism has long been plagued by the simplistic models of causality that inform so many nature-nurture debates. Because the condition involves social symptoms, mid-20th century research tended to look for social-psychologica..l causes such as bad parenting, early-childhood trauma, or stress. When evidence began to accumulate of family inheritance separate from upbringing, the pendulum swung to genetic determinism. But no one gene has been identified as "causing" autism. While the evidence still points to a genetic component, an apparent increase in the incidence of ASDs has prompted some researchers to look again at the possibility of environmental influences on the disease.

Of course, nothing is simple here. The increase in the number of children diagnosed with ASDs almost certainly reflects changes in diagnostic criteria, as well as the younger age at which autism diagnoses are now made. Also, the availability of funding and support services has increased the motivation to obtain an autism diagnosis. But do those factors account for all of the increase? A paper recently published in the journal Epidemiology finds, according to authors Irva Hertz-Picciotto and Lora Delwiche, that "younger ages at diagnosis, differential migration, changes in diagnostic criteria, and inclusion of milder cases do not fully explain the observed increases." They also note, though, that "other artifacts have yet to be quantified, and as a result, the extent to which the continued rise represents a true increase in the occurrence of autism remains unclear."

A wide range of environmental exposures have been proposed as triggers or exacerbators of ASDs: heavy metals (including lead, mercury, cadmium, and arsenic); flame retardants; insecticides; phthalates used in vinyl and cosmetics; household cleaning products that could affect the immune system, such as antibacterial soaps; and in utero exposure to thalidomide, valproic acid, and bacterial or viral infections. Strangely enough, research published last November in the Archives of Pediatrics & Adolescent Medicine even found autism prevalence rates to be positively associated with levels of precipitation. The authors discuss the possibility that increased precipitation could entail increased television viewing, higher rates of vitamin D deficiency, or increased exposure to household cleaners.

Other studies continue to find strong indication of genetic origins. In one recent study, Daniels and her co-authors found that parents of people with autism are more likely than average to have been hospitalized for psychiatric conditions such as schizophrenia, depression, and personality disorders. That suggests, she told JEH, that genes that contribute in one combination to a psychiatric disorder might in a different combination contribute to autism. It does not mean that parents of autistic patients are "damaging" their children psychologically.

The Great Vaccine Debate

The environmental hypothesis has in popular discussion largely been associated with a crusade against childhood vaccinations. The case against vaccines is sometimes based on the presence of mercury in thimerosal, a compound used as a preservative in some vaccines. The use of thimerosal in the manufacture of early-childhood vaccines ended in 1999, and according to the National Network for Immunization Information, the last lots of vaccines made with it expired in 2003. (Thimerosal is still used in some other vaccines, including influenza and tetanus-diphtheria vaccines.) One CDC-funded study has also considered the possibility that the measles virus in the MMR vaccine might lead to autism by causing a persistent intestinal measles infection.

Daniels thinks it's time to move on.

"The bulk of evidence does not suggest that more research on vaccines is going to be useful. People have gotten this tunnel vision on vaccines, and they've failed to look at other environmental hypotheses," she said.

In the meantime, the debate has sowed doubt about the safety of vaccinations among the general public.

"Vaccination rates are dropping," Daniels said. "In the wake of all the media attention, people quite naturally think: Well maybe vaccination is not the cause--but maybe it is. And do I want to take the chance?" But, she said, "the risks associated with not being vaccinated are actually much higher than the risk of autism."

And it puts other people at risk. Ultimately, "herd immunity" could be lost.

But Martha Herbert, who is assistant professor of neurology at