The effectiveness of cognitive behavioral therapy for insomnia on sleep outcomes in the context of pain among older adult veterans

Background: Cognitive behavioral therapy for insomnia (CBT-I) is the gold-standard treatment for insomnia disorder in adults. Compared to young adults, older adults have increased risk for the development of conditions associated with chronic pain, which may impact the efficacy of CBT-I in improving insomnia symptoms in older adults. This study evaluated the effect of participant-rated pain on sleep-related outcomes of a supervised, non-clinician administered CBT-I program in older adult patients with chronic insomnia disorder. Methods: Secondary analysis was conducted using data from a randomized controlled trial among 106 community-dwelling older adult veterans (N = 106; mean age 72.1 years, 96% male, 78.3% White, 6.6% Hispanic, 5.7% African American) with chronic (>= 3 months) insomnia disorder. Participants engaged in five sessions of manual-based CBT-I in individual or group format within one Department of Veterans Affairs healthcare system, provided by non-clinician "sleep coaches" who had weekly telephone supervision by behavioral sleep medicine specialists. Insomnia symptoms (Insomnia Severity Index), perceived sleep quality (Pittsburgh Sleep Quality Index), fatigue (Flinder's Fatigue Scale), daytime sleepiness (Epworth Sleepiness Scale), and perceived pain severity (items from the Geriatric Pain Measure) were assessed at 4 time points: baseline, one-week posttreatment, 6-month follow-up, and 12-month follow-up. Mixed effects models with time invariant and time varying predictors were employed for analyses. Results: CBT-I improved insomnia symptoms, perceived sleep quality, fatigue, and daytime sleepiness among older veterans with chronic insomnia. Participant-reported pain was associated with greater improvements in insomnia symptoms following CBT-I. Pain did not affect improvements in other sleep-related outcomes (-0.38 <= b <= 0.07, p > 0.05). Between-subjects differences in pain, but not within-subject changes in pain over time, appeared to play a central role in insomnia symptom improvement at posttreatment, with individuals with higher-than-average pain showing greater insomnia symptom improvement (ISI score reduction; -0.32 <= b <= -0.28, p <= 0.005). Conclusions: Pain did not meaningfully hinder the effects of CBT-I on sleep outcomes. Among older veterans with chronic insomnia disorder, individuals with higher pain exhibited slightly greater improvement in insomnia than those with lower levels of pain. These findings suggest that experiencing pain does not impair treatment response and should not preclude older adults with insomnia from being offered CBT-I.