Abstract
Background Insomnia is the most prevalent sleep disorder, but it is widely untreated and under-diagnosed in Saudi Arabia. Moreover, no tool to screen insomnia has been validated in the Saudi population in general or nurses in particular. This study, therefore, assessed the psychometric validity of the Insomnia Severity Index (ISI) in Saudi nurses. Methods A cross-sectional study with purposive sampling was performed with nurses (n = 134, age = 21-48 years) from Al Majmaah, Saudi Arabia. Both conventional and online survey methods using the ISI, a brief measure of metacognition, and a socio-demographics questionnaire were employed. Results No ceiling or floor effects were found in the ISI total score (8.84 +/- 5.07) or the factor scores, but the floor effect was found in the item scores. A two-factor model showed the best fit (Pclose .97, comparative fit index [CFI] 1.00, root mean square error of approximation [RMSEA] 0.00, non-significant chi(2)test, chi(2)/df 0.904). This model showed favorable configural, metric, scalar, and partial strict invariance across gender groups (CFI > .95, RMSEA < .05, chi(2)/df < 3, non-significant Delta chi(2), Delta CFI <= .01). The internal consistency was adequate (Cronbach's alpha = 0.75, 0.78 for the two factors of the ISI). The lack of correlations between the ISI scores and the scores of the brief measure of the metacognition favored its divergent validity. Conclusion The ISI showed adequate psychometric validity for screening insomnia among Saudi nurses.