JMIR Publications

ABSTRACT

Background:

Cognitive-behavioral therapy (CBT) is the first-choice treatment for several anxiety disorders, but it’s accessibility and affordability are limited—a challenge that can be tackled by modern internet-delivered CBT programs (iCBTs). For Social anxiety disorder (SAD), these programs are demonstrably efficacious, but less is known about their effectiveness in routine clinical care.

Objective:

Cognitive-behavioral therapy (CBT) is the first-choice treatment for several anxiety disorders, but it’s accessibility and affordability are limited—a challenge that can be tackled by modern internet-delivered CBT programs (iCBTs). For Social anxiety disorder (SAD), these programs are demonstrably efficacious, but less is known about their effectiveness in routine clinical care.

Methods:

HUS-iCBT for SAD is a 7-session, flexible time-scheduled, specialized clinic-delivered, therapist-supported program for physician-referred patients. The data were extracted from the natural register of HUS-iCBT from 2019 to 2020. No diagnostic interview at screening was performed, but a cut-off score of ≥21 on the Social Phobia Inventory (SPIN, primary outcome) was used to exclude patients with subclinical symptoms and verify the diagnosis of SAD made by a referring physician. Secondary measures were the Overall Anxiety Severity and Impairment Scale (OASIS) for overall anxiety and related impairment, and the Patient Health Questionnaire (PHQ-9) for depressive symptoms.

Results:

Altogether, 1654 patients were included. For SPIN-measured severity, we observed a large effect size for both Intent-to-Treat (ITT) and completer populations (Cohen d=1.09, 95% CI 1.03- 1.16 and d=1.15, 95%, CI 1.06-1.23, respectively). Reliable symptomatic improvement was achieved by 41.1% in ITT and 57.8% in completer populations. Completers of the program demonstrated effect sizes of d=0.77, 95% (CI 0.69-0.84) for OASIS and d=0.46 (95% CI 0.39 - 0.52) for PHQ-9. Noncompleters also benefited from the treatment. Patients completed a mean of 5.74 (SD 1.99; 82%) of the 7 sessions (vs. 59%-68% in earlier reports); 54% of patients completed all sessions.

Conclusions:

This study demonstrates that HUS-iCBT is effective for SAD in routine clinical care and, hence, confirms previous effectiveness reports. Our program might, however, impose somewhat less burden on patients, plausibly yielding better adherence compared to other similar programs. More research is needed to locate outcome predictors and cost-outcome relations for further development of iCBTs for SAD. Clinical Trial: Not applicable: this is a register study.