The potential for high-functioning autistic children to lose both their autism spectrum disorder diagnosis and to achieve typical, nonautistic social and communication functioning was demonstrated in a recent study in the Journal of Child Psychology and Psychiatry.
The participants initially met criteria for an autism spectrum disorder (ASD) diagnosis, but they have since lost all ASD symptoms and diagnosis based on clinical judgment and on assessments in social cognition (face recognition), language, and social interaction as measured on the Vineland Adaptive Behavior Scales (VABS) and the Autism Diagnostic Observation Schedule (ADOS).
The small study, involving 34 formerly ASD children who achieved "optimal outcomes" (OO) and their matched cohorts, includes a number of limitations and was conducted largely to "demonstrate the existence" of a group that clearly had autism previously and now no longer does, reported Dr. Deborah Fein of the University of Connecticut, Storrs, and her associates (J. Child Psychol. Psychiatry 2013;54:195-205).
Dr. Deborah Fein
The researchers matched the OO participants by gender, age, and nonverbal IQ to 44 high-functioning autism (HFA) participants and 34 typically developing (TD) participants. OO participant eligibility required a documented ASD diagnosis before age 5 years, which was confirmed independently with a study coinvestigator using only behavior notes in the child’s records. Additionally, OO participants had to have typically developing friends, could not currently meet the criteria for any ASD diagnosis (also independently confirmed), needed at least a 77 on the communication and socialization domains of the VABS, and could not be receiving any special education services related to autism.
Children in the HFA group had to meet ASD criteria clinically and with the ADOS. Children in the TD group could not have ever met criteria for ASD (by parent report) or have a first-degree relative with ASD, and had to have at least a 77 on the communication and socialization VABS domains. All group participants were excluded if they had a debilitating active psychotic disorder, severe visual or hearing impairments, a seizure disorder, fragile X syndrome, or any significant head trauma.
During approximately 6-hour testing sessions, the study participants underwent assessments using the ADOS, the VABS, the Benton Facial Recognition Test, the Clinical Evaluation of Language Fundamentals-IV, the Wechsler Abbreviated Scale of Intelligence for verbal and nonverbal IQ, and the Edinburgh Handedness Inventory. The latter test was used because "left-handedness or delayed maturation of handedness is overrepresented in autism." Parent interviews were used to establish the severity of the children’s initial ASD diagnosis, using the Autism Diagnostic Interview-Revised (ADI-R), and all parents were interviewed using the Social Communication-Questionnaire (SCQ).
No OO or TD participants met ASD diagnostic criteria currently. Although seven OO participants showed some mild social impairment, it was determined to be nonautistic and related to anxiety, depression, embarrassment, inattention, or related issues. TD and OO participants had nearly identical and high average verbal IQ scores, which were an average 7 points higher than the HFA verbal IQ scores. The HFA group was below average on the facial recognition scores; in the OO and TD groups, facial recognitions scores were average and similar.
On the ADOS communication items, 21 TD and 20 OO participants had straight zeroes, which indicates most typically functioning; none of the HFA participants scored straight zeros on these items. Also, 22 TD and 16 OO participants had straight zeroes on the ADOS social items, but none of the HFA participants did. Scores on the VABS communication (OO, 98.30; TD, 93.44), socialization (OO, 102.03; TD, 101.74), and daily living (OO, 92.30; TD, 88.76) scales were similar between OO and TD groups. The HFA group’s mean scores were significantly lower in the same domains (82.70, 75.51, and 75.40, respectively).
Although the OO participants had shown less impaired lifetime socialization scores on the ADI-R than the HFA group (15.24, compared with 20.30; P less than .001), the two groups’ communication (OO, 14.30; HFA, 15.51) and repetitive behaviors (OO, 5.85; HFA, 6.19) scores were similar. Yet the OO participants’ autism symptoms were, on average, a bit milder than those of the HFA group, according to comparisons of parent reports. Indeed, the OO individuals’ milder childhood autism is one limitation of the study, and the seemingly similar reports of communication and repetitive behaviors between the OO and HFA individuals could be biased by parent report.
Dr. Fein and her associates concluded their "results clearly demonstrate the existence of a group of individuals with an early history of ASD, who no longer meet criteria for any ASD, and whose communication and socialization skills ... are on par with that of TD individuals."