Barriers to and Facilitators of Implementation of Internet-Delivered Therapist-Guided Therapy in Child and Adolescent Mental Health Services: Systematic Review and Bayesian Meta-Analysis

Background: Implementation of internet-delivered therapist-guided therapy (e-therapy) can increase accessibility and resource efficiency in youth mental health services. At present, there is limited evidence to guide mental health researchers and practitioners on critical barriers and facilitators of successful e-therapy implementation in child and adolescent mental health services (CAMHS). Objective: This systematic review assesses the current state of knowledge regarding Barriers and Facilitators to the Implementation of Therapist-Guided Internet-Delivered Psychotherapy in Child and Adolescent Mental Health Services. Methods: This systematic literature review includes all peer-reviewed quantitative and qualitative studies that assess factors affecting implementation of e-therapy in the context of outpatient CAMHS (8-18 yrs). A PRISMA-compliant systematic literature search of primary research studies was performed in the PsycINFO, MEDLINE, Web of Science, CINAHL, Embase, Cochrane and OpenGrey databases. ASReview was utilized for screening. The Consolidated Framework for Implementation Research (CFIR) was applied to categorize the barriers and facilitators. We used Bayesian meta-analyses with weakly informative priors and multiple imputation of missing data to assess the implementation outcomes. Results: The database search returned 50 026 reports which were screened. 50 studies were included. We identified common barriers and facilitators related to characteristics of the intervention, organization, therapist and patient. Pooled estimates (95% credible intervals) of implementation outcomes in terms of fidelity were: 0.20 probability for dropout (0.14-0.27), and 68% of the e-therapy program modules were completed by the patients on average (60-75%); in terms of cost-effectiveness: mean therapist time per patient per week was 24 minutes (19-28 minutes); and in terms of acceptability: 24 mean CSQ-8 satisfaction level (22-27 score) and 76% mean satisfaction rate (62 -87%). Conclusions: Overall, the studies showed great heterogeneity in patient fidelity. The reporting of implementation outcomes and barriers and facilitators in the literature is inconsistent and needs a unified framework for measurement. To date, there are not enough quantitative studies on predictors or factors associated with implementation outcomes. In addition, we found a lack of studies exploring implementation processes and implementation determinants systematically and prospectively. Clinical Trial: PROSPERO 2024 CRD42024502578.