Search | Search by Center | Search by Source | Keywords in Title
Todd Arnedt J, Conroy DA, Stewart H, Yeagley E, Bowyer G, Bohnert KM, Ilgen MA. Cognitive behavioral therapy for insomnia to reduce cannabis use: Results from a pilot randomized controlled trial. Drug and Alcohol Dependence. 2023 May 1; 246:109835.
BACKGROUND: Individuals with regular cannabis use demonstrate adverse health outcomes, yet infrequently seek treatment. Insomnia, a common co-occurring complaint, could be targeted to reduce cannabis use and improve functioning in these individuals. In an intervention development study, we refined and tested the preliminary efficacy of a telemedicine-delivered CBT for insomnia tailored to individuals with regular cannabis use for sleep (CBTi-CB-TM). METHODS: In this single-blind randomized trial, fifty-seven adults (43 women, mean age 37.6 ± 12.8 years) with chronic insomnia and cannabis use for sleep = 3 times/week received CBTi-CB-TM (n = 30) or sleep hygiene education (SHE-TM, n = 27). Participants completed self-reported assessments of insomnia (Insomnia Severity Index [ISI]) and cannabis use (Timeline Followback [TLFB] and daily diary data) at pre-treatment, post-treatment, and 8-week follow-up. RESULTS: ISI scores improved significantly more in the CBTi-CB-TM compared to SHE-TM condition (ß = -2.83, se = 0.84, P = 0.004, d = 0.81). At 8-week follow-up, 18/30 (60.0 %) CBTi-CB-TM compared to 4/27 (14.8 %) SHE-TM participants were in remission from insomnia (X = 12.8, P = 0.0003). The TLFB showed a small reduction in past 30-day cannabis use for both conditions (ß = -0.10, se = 0.05, P = 0.026); CBTi-CB-TM participants demonstrated greater post-treatment reductions in the % of days cannabis was used within 2 h of bedtime (-29.1 ± 7.9 % fewer days vs. 2.6 ± 8.0 % more days, P = 0.008). CONCLUSIONS: CBTi-CB-TM is feasible, acceptable, and demonstrated preliminary efficacy for improving sleep and cannabis-related outcomes among non-treatment-seeking individuals with regular cannabis use for sleep. Although sample characteristics limit generalizability, these findings support the need for adequately powered randomized controlled trials with longer follow-up periods.