Combining Cognitive-behavioral Therapy For Insomnia With Exercise Training In Adults With Insomnia And Short Sleep Duration: Impact On Sleep Outcomes

Abstract

Cognitive-behavioral therapy for insomnia (CBT-I) is the first-line treatment for insomnia; however, there is mixed evidence as to whether CBT-I efficacy is blunted in adults with insomnia and short sleep duration. Exercise training may be a useful adjunct treatment to CBT-I for this population. PURPOSE: To explore changes in actigraphic- and polysomnographic-based sleep outcomes from a CBT-I intervention augmented by exercise training in a sample of adults with insomnia and short sleep duration. METHODS: Eight adults (50% female, 62.5% white) with insomnia disorder and short sleep duration (self-reported sleep duration <6 h) completed a 12-week single-arm trial. Participants were guided through the online “Sleep Healthy Using the Internet” (SHUT-I) CBT-I program while completing a supervised exercise program (EX; 150 min/wk of moderate-intensity aerobic exercise and 2 d/wk of resistance training). Prior to and following the intervention, participants completed 1 wk of home-based actigraphy monitoring and 1 night of laboratory-based polysomnography (PSG). Post-intervention PSG data were missing for 1 participant due to COVID lockdown restrictions. Differences between baseline and post-intervention sleep were analyzed using Cohen’s d effect size calculations and paired t-tests. RESULTS: At baseline, participants had actigraphy-assessed total sleep time (TST) of 6.9 ± 0.6 h and sleep efficiency (SE) of 86.6 ± 4.1%; PSG-based TST and SE were 6.9 ± 0.9 h and 87.6 ± 6.1%, respectively. At post-intervention, actigraphy-assessed TST decreased by 15.9 min (d = 0.40, p = .29) and SE increased by 1.4% (d = 0.92, p = .08) relative to baseline. For PSG-based outcomes, TST decreased by 27 min (d = -1.48, p = .08) and SE increased by 3.9% (d = 0.91, p = .05) at post-intervention. In addition, the proportion of time spent in non-REM (NREM) stage 1 sleep decreased by 2.0% (d = -1.05, p = .03) and the proportion of time spent in NREM stage 2 sleep increased by 3.0% (d = 0.82, p = .07) at post-intervention. CONCLUSIONS: These preliminary data suggest improvements in sleep continuity following the combination of CBT-I and exercise training, accompanied by small decreases in sleep duration. Future research should disentangle the independent effects of CBT-I and exercise on sleep outcomes in this population. Support: NIH K23 HL118318 (Kline)