Validating the Child Behavior Checklist 1.5-5 as a screening tool for autism spectrum disorder.

Area of Science:

• Child Psychology
• Developmental Pediatrics
• Psychometric Validation

Background:

• Early identification of autism spectrum disorder (ASD) is critical, particularly in low- and middle-income countries (LMICs) where resources are scarce.
• The Child Behavior Checklist (CBCL) 1.5-5, utilizing Autism Spectrum Problems (ASP) and Withdrawn Syndrome (WS) subscales, presents potential for ASD screening.
• However, its construct validity in LMIC contexts requires thorough investigation.

Purpose of the Study:

• To validate the CBCL 1.5-5 as an ASD screening tool in a representative Brazilian sample.
• To assess the reliability and construct validity of the ASP and WS subscales for ASD identification.

Main Methods:

• Confirmatory factor analysis (CFA) was employed to evaluate model fit and item-subscale correlations.
• Receiver operating characteristic (ROC) curves were used to determine optimal cutoff scores.
• The study included 1292 typically developing Brazilian children (3-5 years) and 70 children diagnosed with ASD (1-5 years).

Main Results:

• The ASP model showed good fit (CFI=0.96, RMSEA=0.037) and reliability (ω=0.869).
• The WS model also demonstrated good fit (CFI=0.974, RMSEA=0.034) and reliability (ω=0.776).
• Optimal cutoffs identified were 6 for ASP (82.5% sensitivity, 83.4% specificity) and 4 for WS (87.9% sensitivity, 82.2% specificity).

Conclusions:

• The CBCL 1.5-5 ASP and WS scales are reliable and valid level 1 ASD screeners for Brazilian children.
• The findings support the use of CBCL 1.5-5 for cost-effective early ASD identification in LMICs.
• Further validation in diverse populations is recommended to confirm its broad applicability.