Of the 61 children who had autism at 2 years of age, 46 also had it at 4 years of age. The remaining 15 children moved off the autistic spectrum. A total of 16 children were not autistic at either age but still screened positive because they had some problems, such as global developmental and language delays. No children who were nonautistic at 2 years of age became autistic at 4 years of age.
At 2 years of age, the 15 children in that study who left the spectrum were indistinguishable on measures of communication, socialization, symptom severity, and cognitive ability from children who remained autistic. But at 4 years of age, the “recovered” children reached almost normal levels on these measures, she said.
The only major difference between these groups at 2 years of age was in motor skills, in which recovered children had significantly higher scores that reached near-normal levels at 4 years of age.
In another study that will appear in the Journal of Autism and Developmental Disorders, Dr. Fein and her associates reported on 14 children aged 5–9 years with prior ASD diagnoses. These children had IQ scores in the normal range, were placed in age-appropriate mainstream classes, and were considered to be generally functioning at the level of their normally developing peers.
They had normal performance on many measures, including different aspects of linguistic ability, expressive vocabulary, and sentence memory, and had adaptive skills in communication and socialization.
But these children still had impairments in knowing the difference between mental state verbs (know, guess, estimate) and in reasoning about animate objects (not including people). They failed to produce good narratives, judged by such factors as being less likely to discuss major events and characters' motivations. They also were not good at understanding “second order theory of mind” situations, such as knowing “what does Johnny know that Suzie knows about what Rick thinks,” Dr. Fein said.
But follow-up visits with these children at 9–12 years of age indicated that the children were “closing the gap” between themselves and normal children because their only remaining deficit was with mental state verbs.
Misdiagnosis of children who leave the spectrum appears unlikely, so the reasons why a small minority of autistic children can recover may lie in the possibility that they have a form of autism that can be alleviated with maturation or that their recovery was mediated by successful treatment or some interaction between treatment and their characteristics, she said.
These possibilities leave two big questions that remain to be answered, Dr. Fein said: Are the children for whom recovery is possible the ones with minimal structural brain abnormalities? When recovered children do a task normally, are they using the same brain systems as children who were never autistic?