A Measurement Invariance Analysis of the Anxiety Scale for Autism-Adults in a Sample of Autistic and Non-Autistic Men and Women.

The sample was an online volunteer sample, not a random community sample, so people interested in mental health or autism may have been more likely to take part. The analysis excluded 110 participants who self-identified as autistic but were not seeking a diagnosis, and 75 non-cis-gender participants, so it does not tell you how those groups interpret the scale. The sample was mostly White or Caucasian (91% to 92%) and mostly based in the UK (81% of autistic participants and 69% of non-autistic participants), so cultural and geographic generalisation is limited. Many participants were highly educated (61.7% of autistic participants and 81.9% of non-autistic participants reported an undergraduate or higher degree), and very few reported intellectual disability, so findings may not apply to people with lower educational attainment or higher support needs. Autism status and mental health diagnoses were self-reported, and the anxiety cut-off was not checked against a clinical interview. The specific factor scores added little unique information beyond the total score, so they should not be treated as separate anxiety types without caution.

Do not use the ASA-A total score to compare autistic and non-autistic adults, or to screen non-autistic adults with the autism cut-off. The scale did not function the same way across autistic and non-autistic groups, so the published cut-off was not verified for non-autistic people.