Can a child recover from Autism?

Yesterday I wrote about how, when people consider individual cases, the possibility of improvement for children with Autism might make otherwise inert therapies appear to be beneficial. In yesterday’s post I referred to research by Molly Helt and colleagues (2008) about recovery among individuals with Autism, and I hinted about an important recent study by Deborah Fein and her colleagues (2013) related to that phenomenon. Today I discuss that second study.

The more recent study is just another among many by Professor Fein, who was a principal author on the Helt et al. (2008) study, and who has been doing exemplary work about Autism and Autism Spectrum Disorders (ASD) for many years. In this one she provides new data about “recovery,” a word they rarely use in the course of their article.

The more recent study by Professor Fein (of the University of Connecticut) and her colleagues is entitled “Optimal Outcome in Individuals with a History of Autism.” Professor Fein and her team compared three relatively high-functioning (Full Scale IQ > 77) groups of boys or young men and girls or young women (ages ranged from ~8 to ~22 years; about 13% female) who were matched on age, gender, and nonverbal IQ. They started by screening 267 individuals and, after various evaluations, they compared 112 of them divided into three relatively similar groups The three groups were characterized as

  1. Those who had been diagnosed as having ASD and subsequently no longer met the criteria for diagnosis (“Optimal Outcomes” OO);
  2. Those had been diagnosed as ASD and still met the criteria as having “High Functioning Autism” (HFA);
  3. Those identified as “Typically Developing” (TD).

Because they were matched on age, gender, and non-verbal IQ, there were no significant differences on those variables, but the three groups differed on verbal IQ; the mean for the HFA group was about a half standard deviation lower than the means for the OO and TD groups.

There are many other details about the characteristics of the participants (did some individuals have first-degree relatives with ASD?). One can read them in the full report, which is available for free on line (link follows). Suffice it to say that this is a no-nonsense piece of research.

The researchers reported results on a host of measures that most experts consider the best in their classes. They included the Autism Diagnostic Observation Schedule (ADOS), Autism Diagnostic Interview-Revised (ADI-R), Social Communication Questionnaire (SCQ), Wechsler Abbreviated Scale of Intelligence (WASI), Vineland Adaptive Behavior Scales (VABS), Benton Facial Recognition Test (Benton), Clinical Evaluation of Language Fundamentals-IV (CELF), and the Edinburgh Handedness Inventory (Edinburgh). To be sure, some of us would like to see more behavioral measures, but one cannot expect to collect sure measures in a study of this sort where the focus is at a less idiopathic level, where researchers need to focus on norms rather than individual performance.

O.K. That’s the Set-up

What did Fein et al. find? In a nutshell, their OO participants who had a well-documented diagnosis of ASD but—as near-adolescents, adolescents, or young adults—no longer met criteria for an ASD, did not differ from TD students on socialization, communication, face recognition, or most language competence subscales.

On average, the OO group displayed milder symptoms than the HFA group in the social domain during early development, but they had equally severe problems with communication and repetitive behaviors at that stage. Still, they had later outcomes that differed from their peers, including IQs in the high-average range!

There are many, many other intriguing outcomes in this study. Read the report to learn more.

So, What’s it Mean?

The journal included a set of “key points.” I’m inserting them here, verbatim.

Key points

  • Autism outcomes vary widely. Moving off the autism spectrum into normal functioning has been suggested as a possibility.
  • The current study documents a group of individuals with clear early histories of autism who currently show normal language, face recognition, communication and social interaction, and no autism symptoms.
  • Results have implications for prognosis by widening the range of possible outcomes for autism.

I encourage interested readers to read the reports. This is accessible science. Both the earlier Helt et al. (2008) discussion paper (abstract only via my link) and the Fein et al. (2013) full report (thanks to the publisher!) are valuable pieces of the puzzle that are emerging.


Fein, D., Barton, M., Eigsti, I.-M., Kelley, E., Naigles, L.,Schultz, R. T., . . . Tyson, K. (2013). Optimal outcome in individuals with a history of autism. Journal of Child Psychology and Psychiatry 54, 195–205. doi:10.1111/jcpp.12037

Helt, M., Kelley, E., Kinsbourne, M., Pandey, J., Boorstein, H., Herbert, M., & Fein, D. (2008). Can children with autism recover? If so, how? Neuropsychology Review, 18, 339-366. doi:10.1007/s11065-008-9075-9

0 Responses to “Can a child recover from Autism?”

Comments are currently closed.