• prodrome;
  • psychosis;
  • schizophrenia;
  • ultra-high risk;
  • validation



The study aimed to examine whether the Arabic version of the Comprehensive Assessment of At Risk Mental States (CAARMS) has good construct validity, concurrent validity and reliability.


Validity was established on a sample of 58 Tunisian adolescents and young adults aged between 16 and 30 years. These subjects were divided into three groups according to the CAARMS scores: ultra-high risk positive subjects (UHR (+) ) (n = 22), ultra-high risk negative subjects (UHR (–) ) (n = 25) and subjects meeting the criteria of a first-episode psychosis (FEP) (n = 11). For construct validity, we used the convergent validity. We used the Positive and Negative Syndrome Scale (PANSS) concomitantly with the CAARMS. For concurrent validity, we studied the correlation between symptoms of the CAARMS and their equivalents in the PANSS. The CAARMS reliability was conducted by the study of interrater reliability.


The UHR (+) group was shown with intermediate scores of PANSS between the two groups UHR (–) and FEP. That confirms a good construct validity of the Arabic version of the CAARMS. We noted a correlation between the scores in positive and negative sections measured by the CAARMS and their corresponding level of the PANSS. These results show that the CAARMS has a good concurrent validity with the PANSS. For the reliability study, we noted a good correlation between the two raters with a Pearson coefficient ranging from 0.55 to 0.90.


Analysis of the results of construct validity, concurrent validity and reliability of the CAARMS indicates that this version translated into Arabic is valid and reliable.