This study investigated the factor structure of the Brief Symptom Inventory (BSI; Derogatis, 1992) for adult and adolescent psychiatric inpatients. The BSI was administered to 217 adults and 188 adolescents at admission and discharge from a private psychiatric hospital. Principal components factor analyses revealed that most variance among dimension scores was accounted for by one unrotated factor. Factorial invariance was evident across adult and adolescent samples for admission and discharge scores. Our findings are consistent with previous research on the BSI and Symptom Checklist-90-R (Derogatis, 1977), suggesting that both instruments measure primarily a unidimensional construct of general psychological distress.

In this study the Brief Symptom Inventory (BSI; Derogatis & Spencer, 1982) was administered to 89 males and 128 females at admission and discharge from a private psychiatric hospital. For mean scores, statistically significant decreases were observed on all BSI scales and global indices. Effect sizes ranged from high medium to large. Also, when clinical significance indices were calculated with regard to changes made by individual patients, we found that approximately 50% of all inpatients evidenced a decrease on the Global Severity Index (GSI), which meets the dual criteria of significant change and posttreatment functioning in the normal range. Compared with inpatient norms (Derogatis & Spencer, 1982), our sample generally scored higher at admission and lower at discharge.

The utility of the Brief Symptom Inventory (BSI) as an outcome measure for adolescent psychiatric inpatients was evaluated. The BSI was administered to 88 male and 100 female psychiatric inpatients at admission and discharge. There were statistically significant mean score changes from pretest to posttest on most BSI indices, with effect sizes ranging between small and medium. Statistical analyses produced generally equivalent results, whether T scores or raw scores were used as dependent variables. However, the ability of the BSI to detect pretest to posttest change reliably for individual patients was quite limited. The degree of symptom distress reported by adolescent inpatients, moreover, did not differ greatly from those found in normative data for adolescent non-patients. It is possible that this relative lack of discrimination between adolescent inpatients and non-patients may be associated with limitations inherent in using self-report inventories with adolescent populations.