Validation of the Phenomenon of Autistic Regression Using Home Videotapes

Emily Werner, Geraldine Dawson, UW Autism Center and Depts. of Psychology and Psychiatry and Behavioral Science, Univ. of Washington, Seattle, WA.

Abstract: Archives of General Psychiatry. 2005 Aug;62(8):889-95

Findings

A subset of children with autism spectrum disorder (ASD) appear to experience a loss of previously acquired skills in social and/or communication skills early in life.  The evidence for this phenomenon, known as regression, has been based largely on parental report.  This study sought to objectively confirm these parental reports by coding behaviors seen on home videotapes of children with ASD vs. typical development taken at 12 and 24 months of age.  The children were assigned to one of three groups: those with ASD with a history of regression as reported by parent on the Autism Diagnostic Interview – Revised (ADI-R) (15 children), those with ASD with early-onset autism (21 children) and those with typical development (20 children).  Parents were also interviewed about early symptoms, from birth onward.

The researchers found that at 12 months of age, infants with ASD used complex babbling, words and declarative pointing less frequently than typically developing infants of the same age. In contrast, 12 month old infants with regression used complex babble and words more frequently than their same-age typically developing peers and did not differ in their use of joint attention such as declarative pointing.  Although the infants with regression did not display the core autism symptoms at 12 months of age, the parent interview indicated that those infants had regulatory difficulties (for example, sleeping difficulties and oversensitivity to sensory stimulation).  While these symptoms are not diagnostic criteria for autism and are not highly atypical at 9 -12 months of age, they occurred more frequently than seen with typically developing infants and could provide evidence for early regulatory abnormalities, before the onset of autism specific symptoms.

At 24 months of age, both groups of toddlers with ASD – with and without regression – used words and vocalizations less frequently, and showed fewer instances of declarative pointing, social gaze and orienting to name than their 24 month old typically developing peers.  In addition, while the typically developing children showed a dramatic increase in the use of complex babble and words between 12 – 24 months, both groups of children with ASD did not make significant gains.

Conclusions

The authors conclude that their study validates the phenomenon of early autistic regression as reported by parents.  The clinical implications of early variations in the course of autism are unclear.  Indeed, a larger study conducted by the authors found no outcome differences at ages 3 and 4 years in children with or without regression, as measured by severity of symptoms, adaptive behavior, and intelligence and neuropsychological testing.