Utility of the Brief Symptom Inventory (BSI) in psychiatric outpatients with intellectual disabilities

Background Diagnostics and care for people with intellectual disabilities (ID) and psychiatric disorders need to be improved. This can be done by using assessment instruments to routinely measure the nature and severity of psychiatric symptoms. Up until now, in the Netherlands, assessment measures are seldom used in the psychiatric care for this population. The objective of the present paper is to evaluate the use of the Brief Symptom Inventory (BSI), a widely used standardised questionnaire in general psychiatry, in a well-defined sample of people with borderline intellectual functioning or mild ID diagnosed with one or more psychiatric disorders. Methods A total of 224 psychiatric outpatients with either borderline intellectual functioning or mild ID participated in this study. All participants were new patients of Kristal, Centre for Psychiatry and Intellectual Disability in the Netherlands, in the period between 1 April 2008 and 1 October 2009. All participants were assessed by a multidisciplinary team, including a certified psychiatrist. Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR) criteria were applied. The mean total intelligence quotient was measured with the Wechsler Adult Intelligence Scale (WAIS-III). The BSI was administered in an assisted fashion. Utility and psychometric properties of the BSI were investigated. Internal consistency coefficients (Cronbach's alphas) were computed. Bivariate correlations between the sub-scales were computed to assess differentiation between the scales. Mean sub-scale scores were compared between different DSM-IV-TR subgroups to investigate the discriminant abilities of the scales. A confirmatory factor analysis was conducted. Results The results suggest that the BSI is practically useful. Internal consistencies ranged from 0.70 to 0.96 and thus are considered good to adequate. Sub-scale inter-correlations showed there is a degree of differentiation between the sub-scales. Discriminant validity was shown for the sub-scales depression, anxiety and phobic anxiety. Confirmatory factor analysis showed that the underlying structure of the BSI could be described by the same nine-factor model as reported in previous studies. Conclusions As a result of the psychometric properties illustrated, this study supports the use of the BSI as a screener for psychopathology and a general outcome measure in people with ID.