Predictive model using autism diagnostic observation schedule, second edition for differential diagnosis between schizophrenia and autism spectrum disorder.

Abstract

Background: Although schizophrenia and autism spectrum disorder (ASD) are currently conceptualized as distinct disorders, the similarity in their symptoms often makes differential diagnosis difficult. This study aimed to identify similarities and differences in the symptoms of schizophrenia and ASD to establish a more useful and objective differential diagnostic method and to identify ASD traits in participants with schizophrenia.

Methods: A total of 40 participants with schizophrenia (13 females, mean age: 34 ± 11 years) and 50 participants with ASD (15 females, mean age: 34 ± 8 years) were evaluated using the Autism Diagnostic Observation Schedule, Second Edition (ADOS-2) and other clinical measures.

Results: ADOS-2 Module 4 original and revised algorithms did not significantly discriminate schizophrenia and ASD, whereas the "Predictive Model" combining the A7, A10, B1, B6, B8, and B9 showed superior accuracy in differentiating both disorders. Both algorithms in the ADOS-2 had high schizophrenia false-positive rates, and significant positive correlations were observed between all domains and the total scores of both algorithms in the ADOS-2 and Positive and Negative Syndrome Scale (PANSS) negative scale scores in the schizophrenia group. The PANSS negative-scale scores were significantly higher in patients positive for autism spectrum cut-offs (CutOff-POS) than in patients negative for autism spectrum cut-offs (CutOff-NEG) for both algorithms in the ADOS-2. Logistic regression analysis revealed that the positivity for both algorithm scales in the ADOS-2 was predicted using only the PANSS negative scale scores.

Conclusions: This study showed that a combination of several items in the ADOS-2 is useful for discriminating between ASD and schizophrenia. The study's findings could help develop strategies benefiting ASD and schizophrenia treatments.