Measurement of Anxiety and Depression Among Autistic Adults: Concordance Between Diagnostic and Screening Instruments in a Feasibility Study

The current study aims to report on the concordance between the PHQ-9, GAD-7, Adult Self Report (ASR), Anxiety and Related Disorders Interview Schedule for DSM-5 (ADIS-5), and best estimate clinical diagnosis (BECD) among autistic adults. The current analyses leverage baseline data from a feasibility treatment study of autistic adults receiving group behavioral activation therapy to address anxiety and depressive symptoms. Interclass correlation coefficients, area under the curve, sensitivity, and specificity were computed to compare the tools to one another at various clinical cut-off levels. Post-intervention outcome data were used for internal replication. The ADIS-5 classifications of Major Depressive Disorder (MDD) and Generalized Anxiety Disorder (GAD) had strong concordance in comparison to the BECD in the Baseline comparisons, but more variable in the Outcome data. For the GAD-7, the mild cut-off had the highest concordance with BECD of any anxiety condition. PHQ-9 moderate cut-off performed most consistently with BECD and ADIS-5 MDD diagnosis. The present analyses further highlight the need for ongoing systematic investigation of the validity of anxiety and depression screening and diagnostic tools in autistic adults. Brief screening tools appear to serve their function well; more research on clinical diagnostic interviews (e.g., ADIS-5) will be needed to standardize diagnostic practices and inclusion in clinical trials. For now, BECDs of co-occurring conditions should remain the gold standard in clinical research aimed at understanding and treating co-occurring conditions in autistic adults.