Interrater reliability of the Japanese version of the Positive and Negative Syndrome Scale and the appraisal of its training effect.

The purpose of the present study is to test interrater reliability of the Japanese version of the Positive and Negative Syndrome Scale (PANSS) and to examine factors possibly affecting the reliability. The study group conducted the PANSS rating on 20 patients with DSM-IV schizophrenia. For the analysis of interrater reliability, intraclass correlation coefficient (ICC) was calculated. The ICC for individual items of the PANSS ranged from 0.26 to 0.92, and those for the positive, negative, and general psychopathology subscales were 0.85, 0.83 and 0.75, respectively. The Cronbach's alpha coefficient for the subscales were 0.84, 0.87 and 0.76, respectively. The interrater reliability and the internal consistency were satisfactory and similar to those obtained in the antecedent studies. No salient training effect was found in a sequential analysis of the concordance rate. It is concluded that the Japanese version of the PANSS is a reliable and efficient tool for comprehensive assessment of the schizophrenic syndrome.