Psychometric Evaluation of a New Schizophrenia Patient Relapse Risk Screener
This poster was presented at the 29th Annual U.S. Psychiatric & Mental Health Congress, held October 21-24, 2016, in San Antonio, Texas.
Objective: Few self-report tools are available to assess relapse risk in schizophrenia. The purpose of this study was to evaluate the psychometric properties of a new self-report tool, Relapse Assessment for Schizophrenia Patients (RASP).
Methods: The initial RASP (k= 22 items) was administered to N=166 patients with schizophrenia from three outpatient community treatment centers. The RASP and PANSS were administered concurrently at baseline, 3 months, and 6months. Relapse was measured using: 1) 9 month chart review; 2) psychiatric crisis database; 3) increase in PANSS total score. Test-retest reliability, internal consistency (Cronbach's alpha), item response characteristics, and factor analysis of items were analyzed. Concurrent validity was assessed through correlation of RASP items and factors with subscales from PANSS, and criterion validity through relationship between RASP item endorsement and evidence of subsequent relapse.
Results: Test-retest reliability was good (k> 0.4) for all but 3 items and Cronbach's alpha among the items was good (.76). Exploratory factor analysis revealed 4 factors accounting for 67% of the variance. Correlations with the PANSS subscales were significant for all factors in the expected direction. Eliminating weak and redundant items, the final screener included 6 items from 2 domains (Anxiety and Social isolation) and 2 disease insight items. Internal consistency, concurrent validity, and criterion validity of the final instrument demonstrated its ability to adequately identify relapse when assessing chart review and crisis data.
Conclusions: The RASP has good reliability, concurrent validity and criterion validity, and may be a useful instrument to measure relapse within clinical settings.