Reliability and validity of the Turkish versionof comprehensiveassessment of at risk mental states

Abstract

Aims To perform the validation of the Comprehensive Assessment at Risk Mental State (CAARMS) in Turkish. Methods Sixty-five volunteers (15-24 years) were enrolled in this study. Concurrent validity was performed with Spearman's Correlation Test using Brief Psychiatric Rating Scale (BPRS). Median scores of the groups were compared using Mann-WhitneyUTest. Interrater reliability was assessed by intragroup correlation coefficient method. Internal consistency was studied by the calculation of Cronbach Alfa Coefficient. Results The correlation of the severity scores of the CAARMS with unusual thought content, suspiciousness, hallucinations and conceptual disorganization items of the BPRS showed that the concurrent validity was good. ROC analysis revealed that CAARMS could discriminate between individuals with UHR and healthy volunteers well. We found a good correlation between the raters. Internal consistency was at very high level. Conclusion Analyses of concurrent validity, criterion validity, interrater reliability and internal consistency indicate that the Turkish version is valid and reliable.