Bridging Languages, Broadening Access: Examining an Observation-Based Autism Assessment with a Latinx Sample

Standardized observational tools are part of the gold standard for autism assessment, leading to the most reliable diagnoses. Widely used tools are often costly, require extensive training, and lack validation for use with multilingual and low-income populations, factors that contribute to prolonged diagnostic wait times.

This study examined psychometric properties of the Brief Observation of Symptoms of Autism (BOSA), a 12- to 14-min semi-structured observation designed as an autism assessment tool for both virtual and in-person administration. We evaluated the BOSA's sensitivity and specificity in both English and Spanish within a Latinx, predominantly low-income sample (N = 98), among other psychometric properties.

Findings indicate that the BOSA is a promising tool that can be used as a screener or as a part of a comprehensive evaluation administered across languages, settings (home, clinic, community), and interactants (caregivers, clinicians) for individuals with limited verbal abilities, though further research is needed to optimize its use with more verbally fluent populations.

These results add to the literature, positioning the BOSA as a promising, affordable, and adaptable tool for improving timely access to high-quality autism assessments in culturally and linguistically diverse, underserved communities. Additional research is needed to assess its usefulness in different circumstances while aiming to increase ease and efficiency of coding. The BOSA's suitability for use by non-specialists in intervention and school-based settings could help reduce diagnostic delays that disproportionately affect families of color from non-English-speaking households, making its optimization an important future goal.