January 20, 2026
Dear Colleagues,
The National Social Anxiety Center (NSAC) provides information about relevant and current research in service of disseminating and promoting evidence-based treatment. This month’s summary is written by Zach Pacha, LISW, A-CBT of NSAC Des Moines, and NSAC Cochair. The article, Safety Behavior Fading for Social Anxiety: A Randomized Controlled Trial of a Self-Monitoring Intervention by Zech et al (2025), shows the results of an RCT with the focused intervention of fading safety behaviors (SB) of social anxiety disorder.
Safety behaviors (SBs) are any behavior whose function is intended to avoid, escape, or minimize a feared outcome (Salkovskis, 1991). In CBT models of social anxiety disorder (SAD), these have been identified as a key factor in the development and maintenance of SAD (Hofmann, 2007).
The authors’ rationale for the importance of doing this study has several points: SBs are a viable target for treatment, current CBT treatment requires time and is at a higher cost than some other modalities of treatment, and corrective learning will take place in real-world settings outside of the therapy office setting. The authors further state that a lower burden of time commitment and cost for the treated individual can also occur through digital forms, which has shown promise in prior studies.
The study was a 28-day digital safety behavior fading treatment for social anxiety. This consisted of brief, daily checklist delivered to participants’ mobile devices. It was compared to an experientially similar control condition: daily unhealthy behavior (UB) self-monitoring and fading. The authors conceptualized UB as behaviors that maintain low activity and self-esteem (eg. snoozing your alarm; scrolling on your phone before bed). In the design of the study, the authors aimed to have the digital app be brief and mostly automated.
A total of 201 individuals took part in the study and were recruited online via advertisements on Facebook and Instagram. The criteria for admittance to the study were: 1) experience elevated social anxiety symptoms as defined by a score or 30 or higher on the Social Phobia Inventory (SPIN); 2) reside in the United States; 3) have access to a smart-phone that is connected to the internet; and 4) not receive psychosocial treatment for anxiety or depression or experience a change in psychiatric medication treatment or dosing within the last four weeks. All participants were then randomized.
The participants in both SB and UB groups were sent daily morning text message reminders to avoid their SBs and UBs. They were also sent a link to a self-monitoring checklist where they could indicate the extent to which they engaged in their target behaviors over the prior day. The authors hypothesized that compared to UB control, the SB condition group would report few SBs at post-treatment and one-month follow-up, and report lower social anxiety, depression, loneliness and maladaptive self-beliefs related to social anxiety.
Key Findings:
The findings indicated that the SB fading treatment led individuals to reduce SB frequency more than the UB fading group at post-treatment, but these differences were not maintained at one-month follow-up. Additionally, the treatment groups did not differ in social anxiety, depression, loneliness, or self-beliefs related to social anxiety either at post-treatment or follow-up, which was contrary to the author’s predictions. An important take-away is the effect sizes across conditions were large for social anxiety (Cohen’s d = 1.64-1.38). The authors point out that the effects did not hold up at a one month follow up. These findings offer important questions to consider and explore further. Ideas the authors shared were to make improvements in the SB fading treatment protocol, noting that there are benefits from unhealthy behavior fading that show promise in treatment, as well. However, the authors note that expectancy effects (the belief that the treatment will work) can also play a role.
Limitations:
Limitations of the study were the absence of a waitlist or placebo control. The sample was predominantly women, which departed from rates of social anxiety in the general population. SBs could have been “swapped” one for another that possibly would dilute the sample. Additionally, encouragement in social interactions and encouragement of “anti-phobic” actions (eg. make eye-contact instead of avoiding) could help facilitate improvement.
Future Considerations:
Future research may improve on this study by: refining the specifics of the SB protocol by including other treatment elements, testing SB fading against active and equally credible control conditions, and considering this treatment’s use with other psychosocial treatments.
Questions for clinicians:
- When treating social anxiety what do you do to specifically address safety behaviors?
- Have you attempted SB fading of any form or forms with a client(s)?
Reference:
Zech JM, Patel TA & Cougle JR (2025). Safety behavior fading for social anxiety: A randomized controlled trial of a self‑monitoring intervention. Journal of Anxiety Disorders, 115, 103068.
___
Zach Pacha, LISW, A-CBT, NSAC Cochair
Representing NSAC Des Moines
(The Anxiety and OCD Center of Iowa)