The validity of using patient self-report to assess psychotic symptoms in schizophrenia.

Brief, reliable and valid measures of psychosis can be very useful in both clinical practice and research, and for identifying unmet treatment needs in persons with schizophrenia. This study examines the concurrent validity and receiver operating characteristics of the psychosis scale of the Revised Behavior and Symptom Identification Scale (BASIS-R). The study was conducted with 71 adults with schizophrenia who were randomly sampled from a large mental health clinic. Study participants at the West Los Angeles Veterans Healthcare Center were assessed using the BASIS-R, a subjective, self-report measure, and the UCLA Brief Psychiatric Rating Scale (BPRS), a clinician-rated measure administered by highly trained research staff. The psychosis scale of the BASIS-R shows good concurrent validity with the psychosis items on the BPRS. Using the BPRS as the gold standard for measuring psychosis, receiver operating characteristics suggest that both the weighted and unweighted versions of the BASIS-R psychosis scale adequately identify psychosis that is moderate or greater or severe. The performance of the two versions was similar. Unweighted scores are easier to calculate, and we therefore recommend cutoff scores based on the unweighted BASIS-R. We identified a cutoff score of 0.5 to best detect moderate or greater psychosis, and a cutoff score of 1.0 to best detect severe or extremely severe psychosis. The BASIS-R has potential as an assessment tool and screening instrument in schizophrenia.