The Development of a Brief Version of the Insomnia Severity Index (ISI-3) in Older Adult Veterans with Posttraumatic Stress Disorder

Abstract

ABSTRACT Objective/Background: The Insomnia Severity Index (ISI) is an insomnia self-report measure used to identify individuals at risk for insomnia disorder. Although the full ISI is only seven questions, a briefer version would allow more efficient and pragmatic administration in routine practice settings. Reliable and valid brief measures can support measurement-based care. The present study was a proof-of-concept study that developed a brief version of the ISI, the ISI-3, in a sample of older adult veterans with posttraumatic stress disorder (PTSD), a group which is at increased risk for insomnia. Patients/Methods: Participants included 86 older veterans with combat- or military-related PTSD. Veterans completed a clinician-administered PTSD diagnostic interview, self-report measures of insomnia and other psychosocial questionnaires, and two neuropsychological assessments. The factor structure of the ISI was examined to reduce the measure into a brief version. The reliability and validity of the ISI-3 were examined. Results: Principal axis factoring yielded a one-factor solution, which reproduced 59% of the item variance. Item reduction procedures resulted in three items, which best represented this factor (“Insomnia Impact;” ISI-3). For the ISI-3, internal consistency was good (α =.89). Convergent validity was demonstrated via moderate to high positive correlations between the ISI-3 and other measures of sleep disturbance. Divergent validity was demonstrated via non-significant correlations between the ISI-3 and unrelated measures and moderate correlations with self-reported depression. Conclusions: The ISI-3 is a psychometrically valid brief version of the ISI. Clinicians can administer the ISI-3 to screen for insomnia and monitor changes in insomnia during treatment.