Cognitive and Behavioral Profiles of Children Who Recover From Autism

Although Autism Spectrum Disorders (ASD) are generally lifelong, scattered evidence over the past 30 years indicates that between 3% and 25% of children can lose their ASD diagnosis and enter the normal range of cognitive, adaptive and social skills. Predictors of recovery include normal intelligence, receptive language, imitation, and motor development. Earlier age of diagnosis and treatment, and PDD-NOS rather than Autistic Disorder are also favorable. Seizures, mental retardation and genetic syndromes are unfavorable signs.

The purpose of our ongoing study is to document this phenomenon in a more rigorous way, explore any residual vulnerabilities, examine history and treatment factors, and assess structural and functional brain parameters in the “optimal outcome” children to see if they still show ASD features.

We will report on some cognitive testing and psychiatric co-morbidities in three groups: The “optimal outcome” (OO) group, a group with high-functioning autism (HFA), and a control group with typical development (TD), all matched for age and performance IQ. Age range of participants is 9-18.

On the Wechsler Abbreviated Intelligence Scales, the groups were matched for PIQ; there were then no significant differences on VIQ or any subtest. Furthermore, the intellectual functioning in all 3 groups was above average (mean IQ about 115).  On the BRIEF (a parent report measure of executive functioning), the OO group was solidly average on every subscale. The TD group was also unimpaired on every scale, and tended to do somewhat better than average, consistent with their high IQ’s. The HFA group, in contrast, showed elevated scores on shifting, emotional control, initiating, working memory, and self-monitoring.

Of 23 typical controls, 3 had a disorder diagnosable by the K-SADS (2 with phobias, and one with depression, all in the past). In contrast, both the HFA and OO groups showed significant psychopathologies: 11 of the 15 HFA subjects (73%) showed a diagnosable disorder, of which 9 were current. Prominent were depression, various forms of anxiety disorder, tics, ADHD, and ODD.  In the OO group, 16 of the 22 subjects (73%) showed a diagnosable disorder, of which 8 were still current. Prominent were ADHD, ODD, tics, and phobias, with one case each of depression and ODD. Psychiatric co-morbidity was, therefore, strikingly similar between the HFA and OO groups.

Finally, on retrospective measures of head circumference growth, the rate of head growth followed by deceleration was greater in the OO and HFA groups than in the TD group, and not different from each other.

Results so far, therefore, indicate that cognitive and executive functioning in the OO group has really normalized, while the HFA group continue to show the expected executive deficits. Above average IQ may help the OO children to compensate. Early biological markers do not differentiate the OO from the HFA group, suggesting that any early structural or inflammatory process was not different for the OO than for other children with ASD. Psychiatric co-morbidity appears to persist in the OO children, perhaps suggesting that it represents an independent vulnerability and is not secondary to their autistic symptomatology.