Using standardized diagnostic instruments to classify children with autism in the study to explore early development.

Area of Science:

• Neurodevelopmental disorders
• Pediatric psychology
• Genetics and etiology of autism spectrum disorder

Background:

• Autism spectrum disorder (ASD) diagnosis relies on behavioral assessments.
• Standardized instruments like the Autism Diagnostic Interview-Revised (ADI-R) and Autism Diagnostic Observation Schedule (ADOS) are crucial for ASD diagnosis.
• Discordance between ADI-R and ADOS can complicate classification and research.

Purpose of the Study:

• To introduce and detail the SEED algorithm for classifying children with ASD using ADI-R and ADOS data.
• To evaluate the psychometric properties of various ASD classification approaches, including the SEED method.
• To investigate the presence of restricted interests and repetitive behaviors in cases with resolved ADI-R/ADOS discordance.

Main Methods:

• Description of the SEED algorithm, a novel method for ASD classification.
• Psychometric analysis of different ASD diagnostic classification strategies.
• Examination of specific behavioral phenotypes in relation to instrument discordance resolution.

Main Results:

• The SEED algorithm provides a reliable method for classifying children with ASD.
• The SEED criteria demonstrate utility in forming well-defined cohorts for clinical and research purposes.
• Analysis revealed insights into behavioral characteristics associated with resolved diagnostic instrument discordance.

Conclusions:

• The SEED algorithm offers a robust approach to classifying autism spectrum disorder.
• The SEED criteria are valuable for ensuring homogeneity in study populations.
• This classification method supports precise research outcomes in pediatric autism studies.