Late referral for autism assessement: an investigation of adolescents and young adults admitted to a psychiatric hospital

Abstract

This study explored the characteristics of adolescents and young adults referred for testing to rule out Autism Spectrum Disorder (ASD) while admitted to a psychiatric hospital for inpatient or day treatment between 2012 and 2017. Data regarding social-developmental, psychiatric, and ASD-symptom variables were collected from an archival database. Seventy-five files belonging to individuals between the ages of 13 and 25 at the time of admission were included. Forty-six participants (61%) were male, and 29 participants (39%) were female. Forty-two patients (56%) were diagnosed with ASD at time of discharge, and 33 patients (44%) were not. Regression analyses were conducted to compare the ASD and non-ASD groups. Participants in the non-ASD group were more likely to be the first-born child in their family and had a greater number of symptoms related to obsessive-compulsiveness and psychosis. Individuals in the ASD group were more likely to use stereotyped or idiosyncratic language, display poor social response, use few emphatic and/or emotional gestures, and demonstrate limited acceptance of responsibility for their own behavior. Participants diagnosed with ASD were shown to be more impaired in terms parent-reported social awareness and self-reported social motivation compared to those in the non-ASD group. Qualitatively, a discharge diagnosis of ADHD was common for the ASD group, while substance use disorders were prevalent in the non-ASD group. For the total sample, the most common reason for hospital admission was risk of self-harm. Individuals in both groups had extensive psychiatric histories and had previously accessed various treatments. Most participants had been tried on four or more psychotropic medications, and nearly half received three or more diagnoses at discharge. The most common diagnoses both prior to admission and at discharge were depressive and anxiety disorders. These patients were typically referred for ASD evaluation due to social interaction difficulties and rigidity. The retrospective design of this study presents multiple limitations. The data collected were limited to the information available in patients' charts, and the researchers were unable to address inconsistencies in these files. Future studies using prospective designs would allow researchers to obtain more detailed information and clarification on study variables.