Safety and Efficacy of Modular Digital Psychotherapy for Social Anxiety: Randomized Controlled Trial

Mona M. Garvert, Jessica McFadyen, Stuart Linke, Tayla McCloud, Sofie S. Meyer, Sandra Sobanska, Paul B. Sharp, Alex Long, Quentin J. M. Huys, M Haji AhmadiJournal of Medical Internet Research 2025 · DOI

Abstract

Social anxiety disorder (SAD) is a common mental health condition characterized by an intense fear of social situations, leading to significant daily impairment. While Cognitive Behavioral Therapy (CBT) is effective, access is often limited by therapist availability and the social anxiety inherent in seeking help. This study evaluates a novel, fully digital, modular CBT smartphone app ("Alena") that operates without a therapist in the loop. Two online randomized controlled trials (RCTs) were conducted to assess safety, acceptability, and efficacy. Results across both trials (n=102 and n=267) demonstrated significant symptom reduction (Social Phobia Inventory scores) in the active group compared to waitlist controls, with high user acceptability and no significant safety concerns. These findings suggest that modular digital interventions can provide a scalable and effective treatment for social anxiety.

Introduction

Social Anxiety Disorder (SAD) affects a large portion of the population, yet many individuals never receive treatment. Two primary barriers prevent care: the severe shortage of trained therapists and the "phobic stimulus" itself—the act of interacting with a human therapist can be a significant deterrent for those with social anxiety.

While internet-delivered CBT (iCBT) has emerged as a solution, most versions still require therapist guidance. Furthermore, standard CBT often follows a "one-size-fits-all" protocol. This research investigates whether a fully digital, modular approach—breaking treatment into specific cognitive processes like rumination or self-focus—can effectively reduce symptoms without human intervention.

Methods

The researchers developed the "Alena" app based on the Clark and Wells model of SAD. The intervention is structured into an introductory module followed by four targeted modules: Beliefs (conditional beliefs about self/others), Attention (self-focus during interactions), Avoidance (safety behaviors), and Rumination (post-event overthinking).

The Alena app interface for RCT #1 and RCT #2, illustrating the modular design and various therapeutic exercises.

Two web-based, unblinded RCTs were conducted:

  • RCT #1: Focused on safety and acceptability among 102 women (aged 18-35) over 6 weeks.
  • RCT #2: A larger efficacy study involving 267 men and women (aged 18-75) over 8 weeks.

Participants were recruited via the Prolific platform and required a Social Phobia Inventory (SPIN) score of $\ge 30$ (moderate to severe anxiety). In both trials, participants were randomized 1:1 to either immediate app access or a waitlist control group.

Key Results

Efficacy and Clinical Significance

The app demonstrated robust efficacy in reducing social anxiety symptoms across both trials:

  • RCT #1: The active group saw a mean SPIN reduction of -9.83 points compared to -4.13 in the control group (Cohen’s $d = 0.47$).
  • RCT #2: The active group saw a reduction of -12.89 points versus -7.48 in the control group (Cohen’s $d = 0.42$).

In RCT #2, 60% of the intervention group achieved "clinically significant improvement" (a drop of $\ge 10$ points). Furthermore, participants using the app were 2.7 times more likely to reach subclinical anxiety levels than those on the waitlist.

Safety and Adherence

Safety data showed no significant difference in adverse events between the active and control arms. Most reported negative effects were mild and typical of psychological therapy, such as temporary discomfort during exposure exercises. Adherence was notably high (84-91% completion rates), which the authors attribute to the "bite-sized" nature of the exercises (1-8 minutes) and the user-centric interface.

Discussion and Limitations

The study confirms that a fully digital intervention can achieve meaningful clinical results without therapist involvement. This modular approach provides a foundation for "precision psychiatry," where future iterations could tailor treatment by serving only the modules that address a user's specific cognitive profile.

However, several limitations remain. The use of a waitlist control may inflate effect sizes compared to an active placebo. Additionally, the study's reliance on tech-savvy participants from the Prolific platform may not fully represent the general clinical population. Long-term follow-up is also necessary to determine if the symptom reductions are maintained over time.

Conclusion

The Alena app represents a safe, acceptable, and efficacious tool for managing social anxiety. By removing the therapist from the loop and utilizing a modular digital framework, this intervention offers a highly scalable solution for individuals who might otherwise avoid or lack access to traditional mental health services.

Summary updated

Abstract

ClinicalTrials.gov NCT05858294; https://clinicaltrials.gov/study/NCT05858294 (RCT 1) and ClinicalTrials.gov NCT05987969; https://clinicaltrials.gov/study/NCT05987969 (RCT 2).
BibTeX
@article{garvert2025safety,
  author = {Mona M. Garvert and Jessica McFadyen and Stuart Linke and Tayla McCloud and Sofie S. Meyer and Sandra Sobanska and Paul B. Sharp and Alex Long and Quentin J. M. Huys and M Haji Ahmadi},
  title = {{Safety and Efficacy of Modular Digital Psychotherapy for Social Anxiety: Randomized Controlled Trial}},
  journal = {Journal of Medical Internet Research},
  year = {2025},
  doi = {10.2196/64138},
}
Figure