[Psychometric properties of the French version of the signs and symptoms of psychotic illness (SSPI) scale].

OBJECTIVE

This report describes the psychometric evaluation of the French translation of the Signs and Symptoms of Psychotic Illness (SSPI) scale. The SSPI scale was designed to assess the five main clusters of symptoms of people suffering from psychotic disorders (psychomotor poverty, reality distortion, disorganisation, depression, and psychomotor excitation) across diagnostic entities. This new tool has been built by Liddle because, in the existing scales assessing psychotic symptoms, individual items cover symptoms that belong to different pathophysiological processes. The SSPI scale comprises 20 items. Its interview is semi-standardised and typically lasts around 25 min. The English version of this scale has shown good psychometric properties (inter-rater reliability, factor structure).

METHOD

We used the SSPI ratings of 81 patients with psychotic symptoms to assess its factor structure and concurrent validity with the Clinical Global Impressions (CGI) scale. Twenty-eight videotaped ratings were used to calculate the intra-class correlation coefficient (ICC) as a measure of inter-rater reliability.

RESULTS AND DISCUSSION

The sample was composed of 46 schizophrenic subjects, 14 with schizoaffective disorder, three with major depressive episode with psychotic features, nine with manic episode with psychotic features and nine with other psychotic disorders. A principal component analysis was conducted to determine the factor structure. Using the Cattell test, we retained a five-factor solution. This solution explained 56.9% of the variance. After varimax rotation, 18 items were attributed to a unique factor. The five factors were: a psychomotor poverty factor, a reality distortion factor, a disorganised factor, an anxious/depressive factor and a psychomotor excitation factor. This structure is close to the original one. The inter-rater reliability of the French version of the SSPI was satisfactory for 18 items, with a mean ICC of 0.64 for the individual items, and an ICC of 0.76 for the global scale. Only two items had an unsatisfactory ICC. This scale showed a good correlation with the CGI scale, with a correlation coefficient between CGI score and SSPI global score of 0.64. Among the factor scores, reality distortion, disorganisation and depression factor scores exhibited a significant correlation with the CGI score.

CONCLUSIONS

The French version of the SSPI scale has good psychometric properties, similar to the English version. Furthermore, its factor structure is similar to the English one. This scale is a robust instrument to rate psychotic symptoms and dimensions across diagnosis entities.