0325 Reducing nocturnal cognitive arousal and rumination mediate treatment effects of digital cognitive behavioral therapy for insomnia

Abstract

Abstract Introduction High cognitive arousal—i.e., heightened cognitive activity often in the form of perseverative thinking—is a central feature insomnia. Emerging evidence from randomized controlled trials (RCTs) support alleviating cognitive arousal as a mechanism by which cognitive behavioral therapy for insomnia (CBTI) reduces symptoms of insomnia and depression. However, few RCTs have published CBTI data on cognitive arousal outcomes. The present study tested whether CBTI alleviates cognitive arousal symptoms, and whether reducing cognitive arousal symptoms mediate CBTI effects on insomnia and depression. Methods 434 patients with clinical insomnia symptoms (Insomnia Severity Index [ISI] ≥ 15) were randomized to digital CBTI or digital sleep education control. Study outcomes included insomnia (ISI), depression (quick inventory of depressive symptomatology self-report 16-item survey, QIDS-SR16), and two indices of cognitive arousal (pre-sleep arousal scale’s cognitive factor [PSASC] and daytime insomnia symptoms response scale [DISRS]). Treatment effects and mediation analyses were conducted using multivariate linear regression. Results While controlling for baseline symptoms, CBTI patients, relative to controls, reported lower levels of insomnia (b=-2.45, p<.001), depression (b=-1.13, p<.001), nocturnal cognitive arousal (b=-2.37, p<.001) and insomnia-focused rumination (b=-3.43, p<.001) after treatment. Posthoc independent samples t-tests showed that CBTI produced small-medium effects on nocturnal cognitive arousal (Cohen’s d=.40) and insomnia-focused rumination (Cohen’s d=.31). Mediation analyses showed that reducing nocturnal cognitive arousal (PSASCT2-T1) mediated 29.0% of CBTI effects on insomnia (αβ=-.709, 95%CI=-1.09, -0.38), and 35.1% of CBTI effects on depression (αβ=-.397, 95%CI=-0.65, -0.19). Mediation analyses showed that reducing insomnia-focused rumination (DISRST2-T1) mediated 10.6% of CBTI on insomnia (αβ=-.261, 95%CI=-0.50, -0.08), but did not mediate antidepressant effects (αβ=-.300, 95%CI=-1.46, 0.72). Conclusion Digital CBTI effectively alleviated nocturnal cognitive arousal and insomnia-focused rumination. Moreover, alleviating nocturnal cognitive arousal was a mechanism by which CBTI reduced symptoms of insomnia and depression. By comparison, reducing insomnia-focused rumination mediated a small proportion of CBTI effects on insomnia and was uninvolved in depression effects. Although reducing cognitive arousal facilitated insomnia and depression outcomes, CBTI is likely limited by its modest effects on cognitive arousal symptoms. Improving insomnia therapy benefits for cognitive arousal may enhance patient outcomes. Support (if any) This RCT (NCT03322774) was supported by NIMH R01-MH122636.