Diagnostic Stability and Developmental Profiles of Autism in Preschool Age

Background: In Autism Spectrum Disorders (ASDs) longitudinal studies show a wide range of outcomes in adaptive behavior domains (Bolte and Poustka,2002), degree of autism (Jonsdottir et al.,2006; Szatmari et al.,2003), speech (Eaves and Ho,1996; Turner et al.,2006), and IQ (Eaves and Ho,1996; Szatmari et al.,2003); diagnostic stability of autism ranged from 81 and 87,5% (Moore e Goodson,2003; Eaves e Ho,2004). In the field of ASDs, diagnostic instruments have been helpful in defining population (Beglinger,2001), merging samples (Lord,2006), and comparing results across studies (Risi,2006; Gotham,2008). Recently,Gotham et al. (2009) published calibrated severity scores for the Autism Diagnostic Observation Schedule (ADOS; Lord et al.1999). Developing these calibrated severity scores was inspired by the need in clinical practice and research for describing the severity of the behavior of children with ASDs referring to the core symptoms in the autism spectrum to try to delineate different outcome profiles.

Objectives: investigating the stability of autism diagnosis in preschoolers; describing developmental profiles of autism at preschool age; defining factors and predictive indexes of different developmental profiles.

Methods: the study has been conducted on 60 subjects who have received a clinical diagnosis of Autistic Disorder (AD; 20, 33, 3%) or PDD-NOS (40, 66,67%) at 36 months (Time 0; SD: 6,57; range 23-48) and who were re-assessed two years later (Time 1; mean: 61 months, SD: 7,75; range 47-78). The subjects were evaluated at T0 and at T1 with ADOS-G, revised diagnostic algoritms (Gotham et al,2007),the Calibrated Severity Scores (CSS; Gotham et al,2009) and psychometric tests.

Results: the FU has shown 5 different developmental profiles with regard to the clinical diagnosis received at T0 and at T1. Most children (17/20) diagnosed with autistic disorder at the age of 3, are going to keep the same diagnosis at the age of 5; some of these same children (3/20) are going to receive instead a diagnosis of PDD-NOS. None of them will exit autism. Among children diagnosed with PDD-NOS at the age of 3,in subjects who go out of autism at the age of 5 (9/40), CSS at T0 is significantly lower than in subjects who are diagnosed with AD at the age of 5 (12/40). The cognitive level (non verbal IQ - nvIQ) at 3 years is not different in the AD and in the PDD-NOS developmental profiles groups.

Conclusions: about 15% (9/60) of children early diagnosed as ASD exited the autism diagnosis at T1. CSS at the first evaluation could be hypothesized as a predictive factor of the outcome in ASD children, while nvIQ does not seem to be hypothesized as a predictive factor of the outcome. This preliminary statistical analysis is not sufficient to demonstrate the predictive value of these factors, since we should elaborate a regression model in order to confirm this correlation or to find further correlations among CSS,nvIQ and language level. Our aim is to collect further data on this sample and to study the developmental profiles of other subjects in order to confirm our preliminary results about factors and predictive indexes of the different developmental profiles of autism.