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

Abstract

Background: Social anxiety disorder is a common mental health condition characterized by an intense fear of social situations that can lead to significant impairment in daily life. Cognitive behavioral therapy (CBT) has been recognized as an effective treatment; however, access to therapists is limited, and the fear of interacting with therapists can delay treatment seeking. Furthermore, not all individuals respond. Tailoring modular treatments to individual cognitive profiles may improve efficacy. We developed a novel digital adaptation of CBT for social anxiety that is both modular and fully digital without a therapist in the loop and implemented it in the smartphone app Alena.

Objective: This study aimed to evaluate the safety, acceptability, and efficacy of the new treatment in online participants with symptoms of social anxiety.

Methods: In total, 2 web-based randomized controlled trials (RCTs) comparing individuals with access to the treatment through the app to a waitlist control group were conducted. Participants were recruited on the web and reported Social Phobia Inventory (SPIN) total scores of ≥30. Primary outcomes were safety and efficacy over 6 weeks in 102 women aged 18 to 35 years (RCT 1) and symptom reduction (SPIN scores) after 8 weeks in 248 men and women aged 18 to 75 years (RCT 2).

Results: In RCT 1, active and control arm adverse event frequency and severity were not distinguishable (intervention: 7/52, 13%; waitlist control: 8/50, 16%; χ²₁=0.007; P=.93). App acceptability was high, with a median completion rate of 90.91% (IQR 54.55%-100%). Secondary outcomes suggested greater symptom reduction in the active arm (mean SPIN score reduction -9.83, SD 12.80) than in the control arm (mean SPIN score reduction -4.13, SD 11.59; t₉₀=-2.23; false discovery rate P=.04; Cohen d=0.47). RCT 2 replicated these findings. Adverse event frequency was comparable across the 2 groups (intervention: 20/124, 16.1%; waitlist control: 21/124, 16.8%; χ²₁<0.001; P>.99). Despite a longer treatment program, median completion remained high (84.85%, IQR 51.52%-96.97%). SPIN score reduction was greater in the active arm (mean -12.89, SD 13.87) than in the control arm (mean -7.48, SD 12.24; t₂₂₇=-3.13; false discovery rate P=.008; Cohen d=0.42).

Conclusions: The web-only, modular social anxiety CBT program appeared safe, acceptable, and efficacious in 2 independent RCTs on online patient groups with self-reported symptoms of social anxiety.

Trial registration: ClinicalTrials.gov NCT05858294; https://clinicaltrials.gov/study/NCT05858294 (RCT 1) and ClinicalTrials.gov NCT05987969; https://clinicaltrials.gov/study/NCT05987969 (RCT 2).

Keywords: cognitive behavioral therapy; digital mental health; internet-delivered CBT; randomized controlled trial; social anxiety disorder.

Figure 1

User interface of the apps used in the 2 randomized controlled trials (RCTs). (A) Interfaces used in RCT 1: the sign-in screen (left), program overview with filters for displaying different modules (middle), and a list of exercises for a particular module (eg, attention; right). (B) Updated user interface used in RCT 2: the home screen showing each module, which could be tapped to show a list of exercises (left); the Recharge screen showing a list of exercises not included in the main program but still centered on alleviating social anxiety (middle); and a Community screen showing forum posts from members of the Alena community (right).

Figure 2

CONSORT (Consolidated Standards of Reporting Trials) diagram. Flow of participants through the study for randomized controlled trial (RCT) 1 (A) and RCT 2 (B). The number of participants who completed the questionnaires at each time point is visualized at the bottom. SPIN: Social Phobia Inventory.

Figure 3

Acceptability ratings of the Alena app. We measured acceptability in 4 categories: how satisfied participants were with the app, how helpful they found the app, how likely they were to recommend the app, and how easy the app was to use. Response options ranged from 1 (lowest) to 5 (highest). Measures were taken each week (see the legend for the color scale) in both randomized controlled trial (RCT) 1 (A) and RCT 2 (B). Panels C and D visualize average ratings across weeks. Error bars denote the SD.

Figure 4

Therapy completion rates. (A) The box-and-whisker plot shows the distribution of therapy completion rates across participants in each RCT, with the median at the notch, the 25th to 75th percentiles represented by the box (ie, the IQR), and the whiskers of the plot representing each box boundary –1.5 to +1.5 × the IQR. (B) The histogram shows the proportion of exercises completed across participants in the intervention group for randomized controlled trial (RCT) 1 (dark blue) and RCT 2 (green).

Figure 5

Improvement in social anxiety symptoms over time in both randomized controlled trials (RCTs). (A) Mean change in Social Phobia Inventory (SPIN) score across participants in either the intervention (blue) or the waitlist control (purple) group each week relative to the baseline assessment at week 0 in RCT 1. (B) Total change in SPIN score from week 0 to the final week of the intervention or waitlist period in RCT 1. (C) Mean change in SPIN score across participants in either the intervention (blue) or the waitlist control (purple) group each week relative to the baseline assessment at week 0 in RCT 2. (D) Total change in SPIN score from week 0 to the final week of the intervention or waitlist period in RCT 2. Error bars represent the SEM.

Figure 6

Improvement in daily functioning over time. (A) Mean change in Work and Social Adjustment Scale (WSAS) score across participants in either the intervention (blue) or the waitlist control (purple) group each week relative to the baseline assessment at week 0 in randomized controlled trial (RCT) 1. (B) Total change in WSAS score from week 0 to the final week of the intervention or waitlist period in RCT 1. (C) Mean change in WSAS score across participants in either the intervention (blue) or the waitlist control (purple) group each week relative to the baseline assessment at week 0 in RCT 2. (D) Total change in WSAS score from week 0 to the final week of the intervention or waitlist period in RCT 2. Error bars represent the SEM.