Abstract
OBJECTIVES: Examine psychometric properties of the Insomnia Severity Index (ISI) in a sample of nurses. METHOD: In a sample of day shift nurses (N = 289), a confirmatory factor analysis (CFA), convergent and discriminant validity analyses, and a test-retest reliability analysis were performed. RESULTS: CFA showed that a two-factor model provided the best fit. The ISI had moderate to poor convergent validity with sleep diary parameters, and moderate convergent validity with the Sleep Condition Indicator (r = -.66), Pittsburgh Sleep Quality Index (r = .66), and PROMIS Sleep-Related Impairment measure (r = .67). The ISI demonstrated good discriminant validity with the measures Composite Scale of Morningness (r = -.27), Nightmares Disorder Index (r = .25), PTSD Checklist for DSM-5 (sleep items removed; r = .32), and Perceived Stress Scale (r = .43). The ISI had weaker discriminant validity with the PHQ-9 (r = .69) and Generalized Anxiety Disorder Screener (r = .51). The ISI demonstrated a good test-retest reliability (ICCs = .74-.88). CONCLUSIONS: The ISI is a psychometrically strong measure for the assessment of insomnia severity in day shift nurses. Overlap with psychological symptoms, primarily anxiety and depression, suggests caution while interpreting these constructs.