Abstract
BACKGROUND: Despite a growing evidence base that internet-delivered cognitive behavioral therapy for insomnia (iCBT-I) is associated with decreased insomnia severity, its efficacy has been minimally examined in veterans. OBJECTIVE: The objective of this study was to evaluate the efficacy of an unguided iCBT-I (Sleep Healthy Using the Internet [SHUTi]) among veterans. METHODS: We conducted a single-blind, randomized controlled trial in Operation Enduring Freedom, Operation Iraqi Freedom, and Operation New Dawn veterans eligible for Veterans Health Administration care. Participants were randomly assigned (1:1) to receive SHUTi (a self-guided and interactive program) or an Insomnia Education Website (IEW) that provided nontailored and fixed insomnia information. Web-based assessments were administered at baseline, postintervention, 6 months postintervention, and 1 year postintervention. The primary outcome was self-reported insomnia severity (Insomnia Severity Index [ISI]). Secondary outcomes were self-reported mental and physical health functioning (Veterans RAND 36-item Health Survey). Exploratory outcomes comprised sleep diary parameters. RESULTS: Of the 231 randomized participants (mean age 39.3, SD 7.8 years; 170/231, 73.5% male sex; 26/231, 11.3% Black; 172/231, 74.5% White; 10/231, 4.3% multiracial; and 17/231, 7.4% other; 36/231, 15.6% Hispanic) randomized between April 2018 and January 2019, a total of 116 (50.2%) were randomly assigned to SHUTi and 115 (49.8%) to the IEW. In intent-to-treat analyses, SHUTi participants experienced significantly larger ISI decreases compared with IEW participants at all time points (generalized η(2) values of 0.13, 0.12, and 0.10, respectively; all P<.0001). These corresponded to estimated larger differences in changes of -3.47 (95% CI -4.78 to -2.16), -3.80 (95% CI -5.34 to -2.27), and -3.42 (95% CI -4.97 to 1.88) points on the ISI for the SHUTi group. SHUTi participants experienced significant improvements in physical (6-month generalized η(2)=0.04; P=.004) and mental health functioning (6-month and 1-year generalized η(2)=0.04; P=.009 and P=.005, respectively). Significant sleep parameter improvements were noted for SHUTi (all P<.05), though the pattern and magnitude of these reductions varied by parameter. No adverse events were reported. CONCLUSIONS: Self-administered iCBT-I was associated with immediate and long-term improvements in insomnia severity. Findings suggest that leveraging technology to meet insomnia treatment demands among veterans may be a promising approach. TRIAL REGISTRATION: ClinicalTrials.gov NCT03366870; https://clinicaltrials.gov/ct2/show/NCT03366870.