Digital Psychotherapies for Adults Experiencing Depressive Symptoms: Systematic Review and Meta-Analysis
Joanna Omylinska Thurston,Supritha Aithal,Shaun Liverpool,Rebecca Clark,Zoe Moula,January Wood,Laura Viliardos,Edgar Rodríguez-Dorans,Fleur Farish-Edwards,Ailsa Parsons,Mia Eisenstadt,Marcus Bull,Linda Dubrow-Marshall,Scott Thurston,Vicky Karkou
DOI: https://doi.org/10.2196/55500
2024-09-30
Abstract:Background: Depression affects 5% of adults and it is a major cause of disability worldwide. Digital psychotherapies offer an accessible solution addressing this issue. This systematic review examines a spectrum of digital psychotherapies for depression, considering both their effectiveness and user perspectives. Objective: This review focuses on identifying (1) the most common types of digital psychotherapies, (2) clients' and practitioners' perspectives on helpful and unhelpful aspects, and (3) the effectiveness of digital psychotherapies for adults with depression. Methods: A mixed methods protocol was developed using PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. The search strategy used the Population, Intervention, Comparison, Outcomes, and Study Design (PICOS) framework covering 2010 to 2024 and 7 databases were searched. Overall, 13 authors extracted data, and all aspects of the review were checked by >1 reviewer to minimize biases. Quality appraisal was conducted for all studies. The clients' and therapists' perceptions on helpful and unhelpful factors were identified using qualitative narrative synthesis. Meta-analyses of depression outcomes were conducted using the standardized mean difference (calculated as Hedges g) of the postintervention change between digital psychotherapy and control groups. Results: Of 3303 initial records, 186 records (5.63%; 160 studies) were included in the review. Quantitative studies (131/160, 81.8%) with a randomized controlled trial design (88/160, 55%) were most common. The overall sample size included 70,720 participants (female: n=51,677, 73.07%; male: n=16,779, 23.73%). Digital interventions included "stand-alone" or non-human contact interventions (58/160, 36.2%), "human contact" interventions (11/160, 6.8%), and "blended" including stand-alone and human contact interventions (91/160, 56.8%). What clients and practitioners perceived as helpful in digital interventions included support with motivation and accessibility, explanation of task reminders, resources, and learning skills to manage symptoms. What was perceived as unhelpful included problems with usability and a lack of direction or explanation. A total of 80 studies with 16,072 participants were included in the meta-analysis, revealing a moderate to large effect in favor of digital psychotherapies for depression (Hedges g=-0.61, 95% CI -0.75 to -0.47; Z=-8.58; P<.001). Subgroup analyses of the studies with different intervention delivery formats and session frequency did not have a statistically significant effect on the results (P=.48 and P=.97, respectively). However, blended approaches revealed a large effect size (Hedges g=-0.793), while interventions involving human contact (Hedges g=-0.42) or no human contact (Hedges g=-0.40) had slightly smaller effect sizes. Conclusions: Digital interventions for depression were found to be effective regardless of format and frequency. Blended interventions have larger effect size than those involving human contact or no human contact. Digital interventions were helpful especially for diverse ethnic groups and young women. Future research should focus on understanding the sources of heterogeneity based on intervention and population characteristics. Trial registration: PROSPERO CRD42021238462; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=238462.