Boys, Girls Differ in Autistic Trait Expression

For a child with autism, it may be impossible to maintain friendships or even hold a conversation. But what about children who may not meet diagnostic criteria for autism yet exhibit a range of "subthreshold" traits that cause the child to miss social cues or simply to exist on a different wavelength from everyone else?

These social deficits are relatively common in the general population, according to the results of a new study, and by studying the origins of these traits, researchers hope to unravel many of the mysteries surrounding autism.

Principal investigator John Constantino, M.D., measured the traits by administering the Social Responsiveness Scale (SRS) to the parents of 788 pairs of twins aged 7 to 15 years.

Constantino is an assistant professor of psychiatry and pediatrics at Washington University School of Medicine in St. Louis. The study was published in the May Archives of General Psychiatry.

The SRS gathers information about various aspects of social behavior in children and determines the extent to which they avoid eye contact, isolate themselves, or show "rigid or inflexible patterns of behavior that seem odd," for example.

The scale is scored from 1 to 195, with higher scores indicating significant social impairment. Although the majority of the boys and girls in the sample had scores ranging from the upper 20s to mid-30s, which indicate normal levels of social functioning, Constantino found that 1.4 percent of boys and 0.3 percent of girls had scores of 101.5 or higher, which is the cutoff for behavior deficits consistent with a diagnosis of pervasive developmental disorder not otherwise specified (PDD-NOS).

PDD-NOS is considered to be the most prevalent and also the mildest form of the pervasive developmental disorders, of which autism is the most disabling.

But even SRS scores between 80 and 100, which were more prevalent among the children in the sample than scores over 100, still indicate significant social impairment, according to Constantino.

"These [autistic] traits constitute a continuously distributed variable," he told Psychiatric News. "The distinct criteria that differentiate the person who is affected with a pervasive developmental disorder from someone who has autistic traits may be arbitrary."

Constantino made a comparison: "What is the difference between a child with an IQ of 69 and another with an IQ of 71?" Clinicians consider a person to be mentally retarded at a score of 70. "There is little difference, although those with an IQ of 69 are eligible for comprehensive public educational interventions, whereas those with a slightly higher IQ are not."

He described the prevalence of autistic traits as "common" in boys because his findings, when extrapolated to the general population, mean that at least one out of every 100 boys may have significant social impairment. Years ago, he said, it was thought that only 1 out of 10,000 children had an autistic-spectrum condition.

He analyzed the distribution of the SRS scores in the 788 twins using several mathematical models to learn more about the heritability of autistic traits in boys and girls.

Why, he wondered, were the traits so much more common among boys?

The model of best fit, according to Constantino, was a "common effects sex limitation model." Simply put, it means that rather than there being a sex-linked genetic liability to autistic traits, both boys and girls have the same genetic predisposition toward autistic traits. Girls, however, are protected from fully expressing those traits, possibly as a result of increased sensitivity to early environmental influences that operate to promote social competency. Further research is needed to discover more about environmental and genetic factors that converge to bring about autistic traits, he said.

"The findings have potential implications for environmental interventions," he added, in the early lives of affected children. "Somehow, if boys could pick up on these environmental factors, they might have better outcomes."

He is currently studying the family members of autistic children, in which subthreshold traits tend to be more common.

"The $64,000 question is whether subthreshold autistic traits are caused by the same genes that cause autism," he said.

Researchers have had a difficult time identifying the genes responsible for autistic disorders, mostly because they have limited their studies to those who are "severely affected," according to Constantino. Studying the genes of people with autistic traits "will yield larger study samples, which increases our statistical power to find the specific genetic influences on these traits," he added.

An abstract of the study, "Autistic Traits in the General Population," is posted on the Web at <http://archpsyc.ama-assn.org/cgi/content/abstract/60/5/524?>. ▪

Archives of General Psychiatry 2003 60 524