Internet-based interventions (IBIs) include unguided web-based self-help without any therapist involvement, guided self-help with minimal professional input, and therapies in which the internet functions primarily as a communication medium between therapist and patient (email, chat, videoconferencing). IBIs have the potential to increase access to psychotherapy or therapeutic interventions. They have also been tested in numerous randomized controlled trials, suggesting their efficacy in treating a variety of disorders. However, questions have been raised about conflicts of interest when trials are funded by those with a stake in the technology’s effectiveness, and about the unacceptably low engagement and high dropout rates for self-guided interventions. Also, therapists report concerns with online therapy reducing emotional engagement, breaching privacy, increasing distractions, and challenging professional boundaries. One issue that requires more understanding is the role of the therapeutic alliance in IBIs. The alliance is the collaborative agreement between the therapist and the patient on the goals of therapy, the tasks needed to reach those goals, and the relational bond between the patient and the therapist. Meta-analyses have shown a moderately large positive association between the therapeutic alliance and improvement (r = .278), with the alliance accounting for almost 8% of the variance in patient outcomes for face-to-face therapy. However, far less is known about the role of the therapeutic alliance in IBIs. In this meta-analysis, Fluckiger and colleagues examined the role of the alliance in IBIs. They found 40 independent studies reporting 82 effect sizes. The average age of patients was 36.83 years, 55% of studies did not involve any face-to-face contact, CBT was the most frequent therapeutic orientation studied, over 70% of patients had a depressive or anxiety disorder, and therapy lasted between 5 and 36 weeks. The authors found a small-to-moderate association between the alliance and outcomes (r = .211, p < .001, 95% CI: 0.162, 0.259), accounting for about 4.3% of the variance in outcomes. Therapeutic orientation and patient diagnosis did not moderate the alliance – outcome association. The most interesting finding was that agreement on goals (r = .19) and agreement on therapy tasks (r = .24) were more predictive of outcomes than the relational bond between therapist and patient (r = .12).
Practice Implications
Internet-based interventions (IBIs) are increasingly popular and have the potential to improve access to therapies for mental health problems. However, concerns have been raised about low engagement and high dropout rates when therapist involvement is lower or non-existent. This meta-analysis suggests that patients’ agreement with the tasks and goals of therapy remains important when receiving structured modules of IBI. However, the role of bonding with the “therapist” may be diminished in IBI by reduced social and emotional cues and presence. The alliance in IBIs may have more to do with patients' confidence in the treatment program than with trust in the particular therapist.
Flückiger, C., Schweizer, A.-M., Boehme, M.-S., Goméz Penedo, J. M., Berger, T., & Wampold, B. E. (2026). Alliance in internet-based interventions: A systematic review and correlation multilevel meta-analysis. Journal of Consulting and Clinical Psychology, 94(4), 187–196. https://doi.org/10.1037/ccp0001005