Objectives: To determine the prevalence of regression in children with DS with or without ASD, and examine its relationship to medical illness and developmental level.
Methods: Children with DS (ages 3-14 years) in NY state were recruited to examine their medical and behavioral characteristics. The Autism Diagnostic Interview-Revised (ADI-R) was completed on 174 participants and responses were used to identify regression. Medical and behavioral characteristics were described in terms of means and proportions.

Results: Nineteen children (11%; 95% confidence interval (CI) = 7-17) were identified with regression; 6 with loss of language, 3 with loss of language and other skills, and 10 with loss of other skills only. Eleven of the 19 subjects (58%; CI = 34-79) related loss to medical illness, the most common being infantile spasms/seizures. The average age at regression of language and other skills was 44 (standard deviation (SD) = 19) and 27 (SD = 25) months, respectively. Eleven of 19 children with regression (58%; CI = 34-79) were positive on the ADI-R diagnostic algorithm. Regression was reported in 11 of 70 children (16%; CI = 8-27) with positive ADI-R. The Vineland ABC was 63 in children with regression as compared to 68 in those without (p-value for difference = 0.052).

Conclusions: There appears to be greater variability in age of regression in children with DS relative to what has been reported for children with ASD only. Further research is needed to understand the nature and timing of regression, potential differences in the phenotype, and medical influences on regression in children with DS with or without ASD.

Supported by AUCD/CDC RTOI 2005-1/2-08