Psychometric properties of the recovery measurement in homeless people with severe mental illness

Abstract

ObjectiveThe Recovery Assessment Scale (RAS) is one of the most widely used measurements of recovery in mental health research. To date, no data have been available concerning the psychometric characteristics of the RAS in homeless people with severe mental illness. The aim of this study was to provide new data regarding the psychometric properties of the RAS in homeless people with schizophrenia and bipolar disorder. MethodsThis multi-center study was conducted in 4 French cities. In addition to the RAS, data on sociodemographic information, disease severity using the Modified Colorado Symptom Index — MCSI, and the number of mental health comorbidities, care characteristics and quality of life (S-QoL-18) were collected. The RAS was tested for construct validity, reliability, external validity, sensitivity to change and acceptability. ResultsSix hundred fifty-eight homeless patients participated in this study. The five-factor structure was confirmed by confirmatory factor analysis (RMSEA=0.043, CFI=0.95, NFI=0.94 and SRMR=0.063). The internal item consistency (from 0.40 to 0.80) and reliability (Cronbach's alpha from 0.79 to 0.87) were satisfactory for all dimensions. External validity testing revealed that the dimension scores were correlated significantly with the MCSI and S-QoL 18 scores. Significant associations with age, disease severity, psychiatric comorbidities and care characteristics showed good discriminant validity. The percentage of missing data (<14.4%) and sensitivity to change were satisfactory. ConclusionsOur study demonstrated the satisfactory acceptability and psychometric properties of the RAS, supporting its use as a mean of recovery measurement for homeless patients.