3 - (IOP 14) Effectiveness of Digital CBT for Generalized Anxiety Disorder: A Randomized Clinical Trial

Generalized anxiety disorder (GAD) is a chronic and disabling condition. Although cognitive-behavioral therapy (CBT) is a first-line treatment for GAD, access to this treatment is limited due to systematic barriers such as limited numbers of trained therapists, burden of in-person therapy, cost, and stigma. Digital CBT (dCBT) programs have the potential to overcome barriers to effective treatment. Yet, early meta-analytic evaluation of smartphone-delivered interventions for anxiety yielded effect sizes below the efficacy identified for therapist-guided CBT, indicating ample room for improvement that may be achieved by better designed or more engaging smartphone-based programs. Daylight, a smartphone-delivered digital CBT (dCBT), was designed to address this gap and has shown promise for achieving the improved outcomes for digital treatment of GAD. The goal the current study is to evaluate the effectiveness of Daylight compared to an active psychoeducation control for the treatment of GAD.

The current study was a single-blind, parallel-group, decentralized, randomized clinical trial (1:1 allocation) conducted between March 2023 and February 2024. A total of 351 adults (259 women, 88 men, 4 other; mean age 40.6 years) who scored 15 or greater on the GAD-7 and met current diagnostic criteria for GAD consented, completed the baseline assessment, and were randomized. Following randomization, participants were asked to complete online self-report questionnaires at Week 6 (mid-intervention), Week 10 (end of treatment), and Week 24 (follow-up) and independent evaluator assessments at Week 10 and Week 24. Primary outcomes were remission status as determined by independent evaluator rated Clinical Global Impression of improvement (CGI-I) ratings and self-reported anxiety symptom severity (assessed with the Generalized Anxiety Disorder-7 scale; GAD-7) evaluated at the primary endpoint at 10 weeks, as well as the secondary endpoint at 24 weeks.

Significant treatment advantages for dCBT were found. Specifically, significantly lower GAD-7 scores were found for dCBT than the psychoeducation condition, reflecting large effect sizes (d = 1.09 and d = 0.96 at weeks 10 and 24 respectively). Additionally, 71% of the dCBT sample met remission status (assessed by the CGI-I) at week 10 compared to 35% of the psychoeducation condition (OR = 4.63; 95% CI: 2.85, 7.54; P < .0001); remission rates were 78% for dCBT and 52% for psychoeducation at week 24 (OR = 3.22; 95% CI 1.95, 5.32; P < 0.001).

The dCBT program Daylight offered significant and sustained benefits to adults with GAD. Given the longstanding limitations on access to empirically supported CBT, Daylight has the potential for dramatic public health impact, particularly given cost savings estimated for this approach.