The Predictive Value of the AQ and the SRS-A in the Diagnosis of ASD in Adults in Clinical Practice.

The study analysed 92 of 147 patients, and 37% of the initial data were omitted because of incomplete data, non-compliance, diagnostic drop-out, or possible low intellectual functioning. The non-ASD group was smaller than the ASD group (29 patients versus 63 patients). The non-ASD group had other psychiatric diagnoses, so the study did not compare ASD with people who had no diagnosis. Not all participants had IQ testing, so low intellectual functioning was not fully ruled out. 30% of the ASD group had a co-morbid psychiatric disorder, which can affect self-report scores. The sample came from an outpatient general mental healthcare setting, and most ASD diagnoses were DSM-5 level 1. Raw scores were used to find cut-offs, while clinical practice may prefer standard scores. The questionnaires were self-report, so results depend on the patient's self-insight.

The paper names funding from the University of Groningen. The supplied text does not report any additional conflict-of-interest declaration.

Do not treat the AQ as a standalone diagnostic test. It predicted the final diagnosis better than the SRS-A, but it still missed 42% of people later diagnosed with ASD.