Outcome at 7 years of children diagnosed with autism at age 2: predictive validity of assessments conducted at 2 and 3 years of age and pattern of symptom change over time

Objective: To examine the predictive validity of symptom severity, cognitive and language measures taken at ages 2 and 3 years to outcome at age 7 in a sample of children diagnosed with autism at age 2. Method: Twenty-six children diagnosed with autism at age 2 were re-assessed at ages 3 and 7 years. At each age symptom severity, cognitive and language assessments were completed. Results: The pattern of autistic symptom severity varied over time by domain. Across time, children moved across diagnostic boundaries both in terms of clinical diagnosis and in terms of instrument diagnosis on the Autism Diagnostic Interview-Revised (ADI-R). On all measures group variability in scores increased with age. Although non-verbal IQ (NVIQ) for the group as a whole was stable across the 3 assessments, this masked considerable individual instability. Standard assessments at age 2 did not predict outcome at age 7 even within the same domain of functioning. In contrast, standard assessments at age 3 did predict outcome. However, a measure of rate of non-verbal communicative acts taken from an interactive play-based assessment at age 2 was significantly associated with language, communication and social outcomes at age 7. Conclusions: The trajectory of autism symptoms over time differed in different domains, suggesting that they may be, at least in part, separable. Variability in language, NVIQ and symptom severity increased over time. Caution is required when interpreting the findings from assessments of children with autism at age 2 years. At this age measures of rate of non-verbal communication might be more informative than scores on standard psychometric tests. Predictive validity of assessments at age 3 years was greater.