Comparative Effectiveness of an FDA-Authorized Digital Therapeutic to Medications and Cognitive Behavioral Therapy Treating Chronic Insomnia in Adults

Abstract

Introduction

Chronic insomnia affects the physical and mental health, quality of life, and productivity of 6 to 10% of the adult population (15-25 million U.S. adults). Available treatments include guideline-recommended first-line cognitive behavioral therapy for insomnia (CBT-I) and medications. However, limitations such as patient access to CBT-I and limited efficacy, the presence of significant side effects, as well as safety concerns about medications limit favorable outcomes. Somryst is an FDA-authorized prescription digital therapeutic for the treatment of chronic insomnia in adults. The purpose of this analysis is to compare the effectiveness of the digital therapeutic vs CBT-I and medications for primary insomnia.

Methods

Chronic insomnia trials focused on digital therapeutic, CBT-I, or medication were identified in a systematic literature review. Studies using a comparator arm that cannot be considered clinically equivalent to other treatments in the network were excluded (eg, meaningfully different definition of placebo arm). A Bayesian network meta-analysis was performed in R on the mean change from baseline and the proportion of remitters using the insomnia severity index (ISI) endpoint with follow-up timepoints between 6 and 12 weeks. Mean change in ISI score from baseline was analyzed as a continuous endpoint while comparisons of the proportion of remitters were performed using odds ratios. The analysis used a random-effects model for the base case analysis. A surface under the cumulative ranking curve (SUCRA) analysis was performed to rank the treatments on each endpoint.

Results

In total, 13 studies reported ISI mean change from baseline data. Only the digital therapeutic and CBT-I were significantly different than placebo. The digital therapeutic had the greatest mean change from baseline in ISI from placebo (−5.77 points, 95% Credible Interval (CrI) [−8.53, −3.07]), followed by CBT-I (−4.3 points, 95% CrI [−6.32, −2.39]). In the SUCRA analysis, the digital therapeutic had the highest probability (56%) of being the most effective treatment based on ISI mean change from baseline. Only 8 studies reported the proportion of ISI remitters. Only the digital therapeutic showed a statistically significant difference in remission vs placebo and had the highest odds ratio for remission vs placebo (12.33 95% CrI [2.28, 155.91]). The odds ratio for remission vs placebo in CBT-I was not statistically significant (4.08 95% CrI [0.45, 45.58]). The digital therapeutic had the highest probability (64%) of being the most efficacious for inducing remission per ISI.

Conclusions

Somryst was projected to be the most effective therapy on both mean change in ISI and ISI remission within 6 to 12 weeks of treatment start vs either CBT-I or medications. Further investigation should be performed to demonstrate the long-term effectiveness of all chronic insomnia treatments.

Funding

Pear Therapeutics