The objective of this study was to determine if a novel game-based digital therapeutic intervention reduced anxiety symptom severity in adults with clinically elevated symptoms of social anxiety disorder (SAD). Participants were randomly allocated (1:1) to receive four weeks of either the active intervention, a game-based form of attention bias modification (Active ABM) for anxiety, or the sham control training (Control). Between June 2022 to June 2023, 104 participants were enrolled with 93 completing the trial per-protocol and 104 included in the final intention-to-treat analysis (54 intervention, 50 control)
mean age was 38.08 (10.56) years and 79 were female. The dependent variable was reduction in SAD symptoms, measured via the Liebowitz SAD Scale (LSAS). Participants were classified as having clinically elevated SAD symptoms only, or both SAD and Generalized Anxiety Disorder (GAD) symptoms (comorbid). The Active condition induced significantly greater reductions in SAD symptoms compared to the Control condition between Baseline and Post-Treatment [Active: M = -29.71, SD = 23.68
Control: M = -14.59, SD = 21.52, d = .67, t(102) = -3.40, p <
.001] across the four-week study period. While no significant between-groups differences emerged at each timepoint individually, the Active condition induced significantly greater change over time in SAD symptoms compared to the Control condition. Use of this game-based digital ABM intervention showed benefits in the reduction of anxiety symptoms.