Feasibility of psychosis risk assessment for adolescents diagnosed with autism

Autism and psychosis share overlapping clinical features and can occur comorbidly. Given growing recognition that early identification of psychosis risk symptoms may lead to better functional outcomes, the field needs valid tools for use in the assessment of psychosis risk symptoms within autism. This study employed a multi-method approach to evaluate the utility of a psychosis risk assessment tool, the Structured Interview for Psychosis-Risk Syndromes, for use with adolescents with autism. A total of 43 adolescents (N = 21 with autism, N = 22 typically developing) were interviewed using the positive symptom domain of the Structured Interview for Psychosis-Risk Syndromes. Study participant answers to interview questions were coded for verbal and behavioral responses, and error rates were compared between groups. Results indicate that adolescents with autism (all of whom had intact language skills) did not significantly differ from typically developing peers when answering questions about positive psychosis risk symptoms. A majority of verbal responses (93%) and behavioral responses (89%) to Structured Interview for Psychosis-Risk Syndrome items were rated as adequate for both groups, suggesting that the positive domain items from the Structured Interview for Psychosis-Risk Syndromes can be used with adolescents with autism. Regardless of diagnosis, higher rates of response errors were significantly correlated with greater difficulty understanding ambiguous language and increased interview times. Structured Interview for Psychosis-Risk Syndrome interviewers are cautioned to use follow-up probes to clarify items that might be confusing to participants and/or have higher response error rates for all adolescents, irrespective of autism status, to prevent false-positive responses for people with and without autism. Lay abstract Individuals with autism may experience a variety of psychiatric symptoms that may cause distress and difficulty functioning. The tools that exist to help evaluate symptoms for psychosis for individuals with autism are limited. We investigated whether a specialized interview for symptoms of psychosis risk could be used for adolescents with autism. We recruited 21 adolescents with autism and 22 typically developing adolescents and interviewed them using the Structured Interview for Psychosis-Risk Syndromes. Participants were asked to rephrase interview questions as a way to understand how they interpreted the question. Their responses were evaluated by clinicians and third-party raters to determine potential response errors. Results of the study showed that youth with autism who have intact language skills are able to answer questions about psychosis risk symptoms as well as their typically developing peers. In general, adolescents across both groups who had more difficulty with nonliteral language (understanding words with multiple meanings) had more difficulty completing the Structured Interview for Psychosis-Risk Syndromes. Problematic items that required more clarification by the clinician involved misinterpretation of words/phrases or questions. Care should be taken to ensure adolescents understand the intent of interviewer questions when assessing risk of psychosis.