The Clinical Validity of Childhood Disintegrative Disorder

Background: "Childhood Disintegrative Disorder" (CDD) is a rare ASD in which typically developing children over two years old undergo a severe and mostly irreversible regression of developmental gains.  Like the other ASDs, CDD is characterized by social and language disabilities, as well as rigid and repetitive patterns of behavior, and is estimated to occur in 1-2 children per 100,000.  With the categorization of CDD as an entity distinct from other ASDs, it became important to determine whether CDD cases were merely examples of late onset  "regressive" AD. Volkmar and Cohen have compared a group of children with CDD diagnoses to a group of children with autism that had been recognized after two years of age, and found that the children with CDD had accumulated more skills prior to regression, and ultimately had significantly greater impairment in speech and intellectual disabilities than children with regressive autism. They concluded that the 'clinical features at the time of regression and various outcome measures support the validity of the diagnostic concept, particularly when such cases are compared to "late onset" autistic ones.'

Objectives: The objective of this study is to compare these groups in terms of long-term outcomes with the intention of gathering further information on whether there is any prognostic validity to the distinction between CDD and other ASDs. Previous studies have examined level of function immediately subsequent to regression, and found that children with CDD fare much worse in multiple domains. However it is not clear that these differences are sustained over longer periods.  A secondary purpose is to further characterize this rarely studied disorder in a sample of children larger than any previously described.
Methods: We compare two groups of children, one with CDD and one with Autistic Disorder (AD) using, at baseline, a series of measures including the ABC, ADOS, and CARS and the Vineland. We follow up with this group on average 8 years subsequent to the initial visit to determine outcome in terms of social, cognitive, linguistic and behavioral function.

Results: Preliminary trends corroborate previous findings:  1) Children with CDD experience deterioration across multiple domains, particularly noticeable in the decay of self-help skills, whereas in AD, regression is often isolated to language. 2) The onset of CDD is rapid, whereas with regressive AD is insidious. 3) Children with CDD more frequently have seizure disorders. 4) Children with CDD have higher levels of anxiety and stereotypies. Data on the long-term outcomes of the group with CDD is still being collected.

Conclusions: In that these variables are not diagnostic criteria for CDD, but distinguish CDD from regressive AD, they support the clinical validity of the disorder. The presence of both CDD-like regressions among children falling into the AD category and of subtle, early developmental delay among some CDD cases could be taken as evidence of continuity between CDD and AD. On the other hand, this evidence could also support the diagnostic concept of CDD, marking the presence of a distinct and recognizable clinical entity which occasionally defies its' diagnostic tethers.