Costs and Outcomes of a Therapist-Guided Internet-Delivered Cognitive Behavioral Therapy: Multicenter Observational Study

Abstract

Background: Therapist-guided internet-delivered cognitive behavioral therapy (iCBT) has demonstrated efficacy and potential cost-effectiveness in treating depression, anxiety, and panic disorder in randomized controlled trials (RCTs). However, evidence of its outcomes and costs in routine care settings on a national level remains limited.

Objective: We aimed to assess the program costs and outcomes of therapist-guided iCBT for depression, social anxiety, and panic disorder posttreatment (T2) and at 6-month follow-up (T3) in a broad context, in addition to exploring how the program's costs and effects vary by hospital.

Methods: This single-arm observational study analyzed patient-reported data collected from routine care between 2021 and 2024 in 4 hospitals in Norway. The symptom severity of depression, social anxiety, and panic disorder was measured using the 9-item Patient Health Questionnaire (PHQ-9), the Social Phobia Inventory (SPIN), and the Panic Disorder Severity Scale (PDSS), respectively. Generic health-related quality of life (HRQoL) was measured with the EQ-5D-5L, while work and social functioning were measured with the Work and Social Adjustment Scale (WSAS) and sick leave days. Mixed effects models estimated changes in outcomes over time and between hospitals. Hospital-specific annual program costs per patient were estimated based on infrastructure and therapist guidance expenses. The economic evaluation was performed with a hospital perspective and extended to a societal perspective by examining sick leave days.

Results: Data from 565 participants showed substantial improvements across all outcomes at T2 and T3. At T2, 102 (35.1%) participants responded positively to treatment (depression: 44/140, 31.4%; social anxiety: 21/79, 26.6%; panic disorder: 37/72, 51.4%), and 66 (23.7%) achieved remission (depression: 17/139, 12.2%; social anxiety: 26/73, 35.6%; panic disorder: 23/67, 34.3%). Regarding work and social functioning, 97 (33.0%) patients responded positively to the treatment and 60 (23.0%) achieved remission, as measured by the WSAS. Patients reported a mean reduction of 3.2 sick leave days (95% CI -5.4 to -0.9) at T2 and 7.7 sick leave days (95% CI -11.5 to -3.9) at T3. EQ-5D-5L utility scores increased by a mean of 0.11 (95% CI 0.08-0.13) at T2 and 0.12 (95% CI 0.09-0.15) at T3. Patient-reported outcomes were consistent across time points and hospitals and robust to sensitivity analyses accounting for patient and hospital characteristics and missing data scenarios. The mean total program costs per patient were US $1030.12 (SD 451.6), which varied by location (US $636.41-$2152.47), mostly due to differences in patient volume.

Conclusions: This study confirms the potential of therapist-guided iCBT to relieve symptom severity and improve well-being across different health service providers when implemented as part of routine specialist health care. The observed variance in costs per patient between hospitals underscores the importance of patient volume to optimize efficient use of resources.

Keywords: MeSH; Medical Subject Headings; depression; health service delivery; panic disorder; routine care; social anxiety; specialist mental health care; telemedicine.

Figure 1

Study flowchart. T2: posttreatment; T3: 6-month follow-up.

Figure 2

Estimated patient-reported outcomes by hospital over time. GAD-7: 7-item Generalized Anxiety Disorder scale; H1: Haukeland University Hospital; H2: St Olavs Hospital; H3: Vestfold Hospital; H4: Innlandet Hospital; PHQ-9: 9-item Patient Health Questionnaire; T1: baseline; T2: posttreatment; T3: 6-month follow-up; WSAS: Work and Social Adjustment Scale.