Validity and reliability of a new instrument for assessing mood symptomatology: The Structured Clinical Interview for Mood Spectrum (SCI-MOODS)

We evaluated the internal consistency, discriminant validity, inter-rater reliability and test-retest reliability of a new instrument for the assessment of lifetime symptoms related to mood spectrum disorders: the Structured Clinical Interview for Mood Spectrum (SCI-MOODS). We report on results obtained from: 491 subjects assessed across eight psychiatric centres in Italy who were given the SCI-MOODS and the Mini International Neuropsychiatrie Interview (MINI). The study sample consisted of four groups: 141 students, 116 gastrointestinal (GI) patients, 112 bipolar patients and 122 patients with recurrent depression. To evaluate the inter-rater reliability and the test-retest reliability, an additional group of 30 subjects (10 non-psychiatric patients enrolled in an orthopaedic clinic, 10 unipolar patients and 10 bipolar patients) was given the SCI-MOODS at a baseline assessment and six to eight days later. At the baseline assessment, these subjects were also interviewed using the Structural Clinical Interview for DSM-IV (SCID IV) and were rated for the severity of symptomatology using the Clinical Global Impression Scale (CG1). The internal consistency of the four domains and six subdomains of the interview, expressed in terms of the Kuder-Richardson coefficient, was high, ranging from 0.79 to 0.92. Correlations of each of the domains with the others ranged from 0.63 to 0.85, indicating a strong interrelationship among the domains. Moreover, the correlations of each of the three manic subdomains with the others and each of the three depressive subdomains with the others were consistently higher than those between manic and depressive subdomains. The SCIMOODS was found to discriminate between patients with mood disorders and subjects belonging to two control groups as well as between bipolar and unipolar patients. The inter-rater reliability for the four domains and the six subdomains was excellent, with the intra-class correlation coefficient being close to unity for each of them. Test-retest reliability was also excellent, ranging from 0.93 to 0.94 for the four domains. Test-retest reliability was slightly lower in the manic (0.86 to 0.89) than in the depressive subdomains (0.95 to 0.96), but well within the acceptable range. These findings provide strong support for the internal consistency, the discriminant validity and the reliability of the SCI-MOODS