Predictors of Inpatient Admission for Adults with and without Autism Spectrum Disorders

Background: Compared to the general population and to individuals with Intellectual Disabilities (ID), individuals with Autism Spectrum Disorders (ASD) have increased rates of mental health disorders, and thus might be at greater risk for psychiatric hospitalization. Previous research indicates that a diagnosis of autism is one of the strongest predictors of psychiatric hospitalization in adults with ID (Cowley, et al., 2005). Additionally, studies have shown that specific demographic factors (e.g., age, residence) and clinical variables (e.g., age of ASD diagnosis, co-morbid psychiatric diagnoses, and aggression) are strongly associated with psychiatric hospitalization in children with ASD (Mandell, 2008). However, no studies have considered predictors of admission in adults with ASD.  Thus, the present study aims to provide further insight into what demographic and clinical variables predict inpatient admission in a Canadian ASD sample compared to a non-ASD sample.   

Objectives: The present study has two objectives: 1) To determine whether a diagnosis of ASD predicts inpatient admission in a clinical sample of adults with ID; and 2) To determine whether predictors of admission in adults with ASD differ from predictors for adults without ASD in the same inpatient unit

Methods: Data was obtained from an electronic clinical database that contains information on clients who were referred to a tertiary level mental health service for people with ID. The present study utilized information available at intake for patients referred during the period from April 2006 - April 2011. During this period, 326 adults, 73 with ASD, were seen in this mental health service. In addition to ASD status, variables predicting admission will include, demographic variables (e.g., age, gender, referral source, reasons for referral, residence at the time of referral, as well as forensic involvement, severity of ID, and verbal ability), as well as clinical variables (e.g., psychiatric and medical diagnoses, psychotropic medications, and risk behaviours). Scores on screening measures of maladaptive behaviour, the Reiss Screen of Maladaptive Behaviour and Aberrant Behaviour Checklist, will also be used to predict inpatient admission in the ASD group. 

Results: Data are collected and currently being analyzed. In those with ASD, predictors of admission include age, gender, and use of antipsychotic medication, with level of ID approaching significance. Further analyses are currently being conducted to determine predictors of admission in those without ASD. A final analysis will be conducted to explore whether ASD predicts admission, when other variables are taken into account. 

Conclusions: The findings from the current study will aid in the identification of the individual variables that predict inpatient admission in adults with ID, and specifically with ASD and ID. Furthermore, it will provide a better understanding of the specific treatment and support needs of this population in order to prevent future inpatient admission.