Clinical characteristics associated with language regression for children with autism spectrum disorders.
Language regression in autism is common yet carries no extra burden on later development.
01Research in Context
What this study did
The team looked at kids with autism who lost words they once had.
They compared these kids to others whose language never took off and to kids whose words stayed flat.
Parents filled out forms and kids took short tests to see if regression meant bigger problems.
What they found
Kids who lost words looked almost the same as the other groups.
The only small plus went to the flat-language kids—they had slightly better social skills later on.
Regression did not signal worse medical issues, lower IQ, or more behavior trouble.
How this fits with other research
Bhaumik et al. (2009) saw that one in five kids with autism lose skills, so regression is common.
Warnes et al. (2005) already showed that regression history does not predict later IQ—our paper backs that up.
Chan et al. (2023) found that poor language links to more hitting and tantrums; our work says regression itself is not the driver, overall language level is.
Together the story is: regression happens, but it does not doom a child to tougher outcomes.
Why it matters
You can stop treating regression like a red-alert marker.
Plan goals from today’s skills, not from whether words were lost.
Tell worried parents that kids who regress can catch up just as well as peers who never had the words to begin with.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Score current language and social levels, then set goals—skip extra probes for regression history.
02At a glance
03Original abstract
We investigated correlates of language regression for children diagnosed with autism spectrum disorders (ASD). Using archival data, children diagnosed with ASD (N = 114, M age = 41.4 months) were divided into four groups based on language development (i.e., regression, plateau, general delay, no delay) and compared on developmental, adaptive behavior, symptom severity, and behavioral adjustment variables. Few overall differences emerged between groups, including similar non-language developmental history, equal risk for seizure disorder, and comparable behavioral adjustment. Groups did not differ with respect to autism symptomatology as measured by the Autism Diagnostic Observation Schedule and Autism Diagnostic Interview-Revised. Language plateau was associated with better adaptive social skills as measured by the Vineland Adaptive Behavior Scales. Implications and study limitations are discussed.
Journal of autism and developmental disorders, 2010 · doi:10.1007/s10803-009-0823-3