Efficacy and safety of a novel digital cognitive behavioral therapy for insomnia application versus zolpidem: An exploratory randomized controlled trial

21/08/2026

AI Summary

5-week dCBT-I app improved subjective sleep onset latency at Week 5 comparable to zolpidem; no significant between-group difference.
At Week 10, dCBT-I maintained and further improved sSOL, while zolpidem showed partial symptom recurrence; differences not statistically significant.
dCBT-I produced greater improvement in dysfunctional sleep beliefs and had no treatment related adverse events; one zolpidem patient reported daytime somnolence.

Sleep Med. 2026 Aug 18;148:109222. doi: 10.1016/j.sleep.2026.109222. Online ahead of print.

ABSTRACT

OBJECTIVE: To evaluate the efficacy and safety of a novel, concise 5-week digital cognitive behavioral therapy for insomnia (dCBT-I) application, “IIIP MED: Sleepy Med,” compared to zolpidem.

METHODS: This exploratory randomized controlled trial enrolled patients with chronic sleep-onset insomnia who remained symptomatic after a 4-week screening period following sleep hygiene education. Participants were randomized to either the fully unguided dCBT-I group or the zolpidem (5 mg/day) group for a 5-week treatment period, followed by a 5-week follow-up. The primary outcome was the change in subjective Sleep Onset Latency (sSOL) from baseline at Week 5. Secondary outcomes included objective sleep parameters via portable electroencephalography (EEG) and safety assessments.

RESULTS: At Week 5, the mean sSOL change was -20.65 min (95% CI: [-32.05, -9.65]) in the dCBT-I group (n = 11) and -22.42 min (95% CI: [-29.37, -15.46]) in the zolpidem group (n = 13), with no significant difference between groups (p = 0.7729). At the Week 10 follow-up, the dCBT-I group (n = 9) demonstrated sustained improvement (-29.02 min, 95% CI: [-51.49, -6.54]), whereas the zolpidem group (n = 12) showed a slight recurrence of symptoms (-16.19 min, 95% CI: [-27.08, -5.31], p = 0.2634). The dCBT-I group showed significantly greater improvement in Dysfunctional Beliefs and Attitudes about Sleep at Week 10. No treatment-related adverse events occurred in the dCBT-I group, while one case of daytime somnolence was reported in the zolpidem group.

CONCLUSION: The dCBT-I application was associated with an improvement in sSOL over 5 weeks, with no statistically significant between-group difference compared with zolpidem. Furthermore, the improvement in the dCBT-I group was maintained during the 5-week post-treatment follow-up. Given the exploratory design and small sample size (N = 24), the study was underpowered, and these findings should be interpreted with caution. Larger adequately powered confirmatory trials are warranted.

TRIAL REGISTRATION: jRCT2032230353.

PMID:42623797 | DOI:10.1016/j.sleep.2026.109222

Document this CPD