Objective: To investigate associations between six of the Symptom Check-List 90R (SCL-90R) subscales and specific DSM-IV symptom disorders in a sample of patients with high comorbidity of axis I and axis II disorders.
Method: SCL-90R questionnaires from 1202 patients admitted for treatment in the Norwegian Network of Psychotherapeutic Day Hospitals. Mean score differences on subscales among diagnostic categories were investigated. With diagnoses as external criteria, cut-off scores for different diagnostic groups were used to calculate diagnostic efficacy. Multiple regressions were conducted in order to disentangle variance of the subscales accounted for by symptom disorders. Frequency distributions of subscales for patients with and without symptom disorders were compared.
Results: Poor diagnostic efficacy was found for most of the subscales, except for the phobic anxiety subscale. The strongest association with each subscale was with its associated symptom disorder, except for dysthymia.
Conclusion: Diagnostic inferences about DSM-IV symptom disorders based on SCL-90R should be conducted with care.