Study data published in Behavioral Sleep Medicine support the validity of a brief version of the Insomnia Severity Index (ISI) as a more efficient option in routine clinical practice for older adults with posttraumatic stress disorder (PTSD). In a cohort study of older veterans with combat or military related PTSD, the brief ISI was significantly correlated with results from the original yet longer ISI.
Investigators conducted a proof-of-concept study to confirm the feasibility of a brief version of the ISI. The original ISI is a 7-item self-report measure of daytime and nighttime symptoms of insomnia. Patients rate each item for the past 2 weeks on a 5-point scale; higher numbers denote greater insomnia severity. The investigators collapsed the 7 items of the original ISI to form a 3-item ISI (ISI-3).
The ISI-3 was tested in a cohort of older adults with PTSD, a population at high risk for insomnia, recruited from a large southwestern Veteran Affairs medical center. Study participants were screened with both the original ISI and ISI-3. Results from the ISI-3 were compared against results from the original ISI, as well as other insomnia-related measures, including the Pittsburgh Sleep Quality Index (PSQI) and the insomnia items from a clinician administered PTSD scale (CAPS), the stressor specific PTSD checklist (PCL-S), and the 9-item Patient Health Questionnaire (PHQ). Patients also underwent depression screening and cognitive tests. The researchers tested internal consistency, validity, and reliability of the ISI-3. A receiver operating characteristic curve (ROC) analysis was conducted to determine the optimal ISI-3 cutoff point for clinically significant insomnia.
The cohort comprised 86 older adult men (mean age, 65.4±5.8 years; 72.1% non-Hispanic white) with combat or military related PTSD. The median annual income was $30,000 to $45,000 per year. There was a mean ISI score of 15.10±6.57, and 82.6% of patients exceeded the clinical cutoff for insomnia.
The ISI-3 was strongly correlated with the full ISI (r =.93) and moderately correlated with the PSQI (r =.68). The ISI-3 also correlated with the insomnia items from CAPS (r =.71), PHQ (r =.61), and PCL-S (r =.74; all P <.01). Internal consistency for the ISI-3 was high, with each item correlating strongly with the next (Cronbach’s α = .89). In tests of divergent validity, ISI-3 was not significantly associated with age, education, or cognitive outcomes. However, the ISI-3 was slightly correlated with reading test scores (r = -.22; P <.05) and moderately correlated with the PHQ (r =.50; P <.01). In ROC analyses, the optimal ISI-3 cutoff for insomnia was ≥6, with a sensitivity of 84.5% and specificity of 100%. The area under the ROC was 0.973, indicating extremely good prognostic capacity.
These findings support the validity and reliability of a brief version of the ISI for use in older male veterans with PTSD. As a relatively small cohort study, however, results must be extrapolated with care. The ISI-3 may not be an effective measure of insomnia in young patients, non-veterans, or patients without PTSD.
“The ISI-3 may also promote insomnia screenings in healthcare settings, such as primary care, and reduce patient and provider burden through decreased administration, scoring, and completion time,” the investigators concluded.
Wells SY, Dietch JR, Edner BJ, et al. The development of a brief version of the Insomnia Severity Index (ISI-3) in older adult veterans with posttraumatic stress disorder [published online June 1, 2020]. Behav Sleep Med. doi:10.1080/15402002.2020.1760278