To test the factorial structure of the Italian version of the Insomnia Severity Index (ISI) using a confirmatory approach and to assess its psychometric properties. ISI questionnaire was completed by 272 patients (average age 41.28, range 18-73) with insomnia diagnosis performed by a sleep medicine physician and retrospectively enrolled in the study. All patients underwent Cognitive Behavioral Treatment for Insomnia (CBT-I) and completed sleep diaries before starting the treatment. Data from sleep diaries were analyzed for assessing concurrent validity of the ISI. Confirmatory factor analysis (CFA) for ordinal Likert-type items was applied to compare four competing models proposed in the literature. 244 patients, out of the 272, completed the ISI at the end of CBT-I. A comparison of ISI score before and after treatment was performed. The CFA analysis confirmed the presence of three main factors conceptualized as severity and impact of the disease along with sleep satisfaction. Significant correlations of the first three items of the questionnaire, investigating three different subtypes of insomnia, and the subjective measures from the sleep diaries were found, thus supporting the concurrent validity of the test. Sleep efficiency (SE) had a significant inverse correlation with the severity and satisfaction factors and with ISI's total score. After CBT-I treatment, a significant reduction of ISI's scores was observed, thus confirming the effectiveness of the CBT-I treatment. The internal reliability coefficient was 0.75. The ISI questionnaire maintains good psychometric properties in the Italian version, thus confirming that this instrument is reliable for detecting insomnia severity and identifying patients' symptoms.
Keywords: Cognitive behavioral therapy for insomnia; Confirmatory factor analysis; Insomnia; Insomnia Severity Index; Sleep diary.