3111 — The Validity of Using Patient Self-report to Assess Psychotic Symptoms in Schizophrenia

Young AS (Greater Los Angeles VA & UCLA), Niv N (Greater Los Angeles VA & UCLA), Cohen AN (Greater Los Angeles VA), Mintz J (University of Texas at San Antonio), Ventura J (UCLA)

Objectives:
Organizational efforts to improve healthcare quality rely on accurate and routinely available data regarding patients' clinical needs. Obtaining these data has been a particular problem in specialty mental health. In schizophrenia, it has been shown that medical records contain little valid and reliable data on patients' clinical status. Efforts to have providers routinely enter these data into the medical record have had only limited success. Using patient self-report is a promising alternative. Patient self-report has been shown to be feasible at VA clinics in patients with schizophrenia. While self-report is valid for mood symptoms, it has been unclear if self-report is accurate for the assessment of psychotic symptoms. We examined the concurrent validity and receiver operating characteristics of the psychosis scale of an established self-report scale, the Revised Behavior and Symptom Identification Scale (BASIS-R).

Methods:
We randomly sampled patients with schizophrenia from a large VA mental health clinic (n=71). Participants were assessed using a self-report measure (the BASIS-R), and a validated, clinician-rated measure administered by highly trained research staff (the UCLA Brief Psychiatric Rating Scale [BPRS]).

Results:
Correlations between psychosis items on the BASIS-R and BPRS were all in the moderate range (range of r=.44 to r=.62). Using the BPRS as the gold standard for measuring psychosis, receiver operating characteristics indicate that the BASIS-R psychosis scale adequately identifies psychosis of moderate or greater severity. A cutoff score of 0.5 best detects moderate or greater psychosis, and a cutoff score of 1.0 best detects severe psychosis. The performance of the weighted and unweighted versions of the BASIS-R were similar.

Implications:
Assessment of psychosis by self-report produces valid data, though they are somewhat different from expert clinical assessment. Self-report does not need to be performed by clinicians, can be performed by computerized assessment, and can be entered electronically into the medical record using existing software.

Impacts:
In specialty mental health, self-report and the BASIS-R have the potential to be useful for clinical monitoring, performance indicators, and quality improvement. Also, self-report provides assessment from the patient’s perspective, which is an inherently valuable addition to clinician assessment.