AI-enabled conversational agent increases engagement with cognitive-behavioral therapy: A randomized controlled trial
Open accessAbstract
Timely support after referral to mental healthcare is crucial, yet patients often face prolonged wait times without intervention. Digital mental health interventions offer scalable solutions, but many struggle to achieve acceptable patient engagement. Tailoring and personalizing materials to individual needs is paramount for driving engagement, a task that generative artificial intelligence AI (genAI) is potentially able to achieve. To examine this promise, we conducted a randomized controlled trial using a genAI-enabled therapy app, Limbic Care, which delivers personalized cognitive behavioral therapy (CBT) materials, against PDF workbooks delivering static CBT content, as commonly used in standard care. Adults with elevated symptoms of anxiety or depression (N = 540) were randomly assigned to the app or control group for six weeks. The app group exhibited a threefold increase in engagement (2.4 times higher usage frequency, 3.8 times longer usage durations). While both groups showed similar overall symptom improvement, participants who engaged with the app’s clinical personalization capabilities experienced significantly greater reductions in anxiety symptoms and enhanced well-being than those who engaged with the standard CBT materials. Importantly, the app was safe, with no increase in adverse events compared to standard care. Our findings suggest that genAI-enabled therapy apps can safely enhance patient engagement and improve clinical outcomes through clinically persona…
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
McFadyen et al reported 0.26 SD baseline difference for anxiety.
“For example, Jeong [99] reported 0.30 SD for depression, McFadyen et al [61] reported 0.26 SD for anxiety, Sabour [64] reported 0.33 SD for depression.”
From Generative AI Mental Health Chatbots as Therapeutic Tools: Systematic Review and Meta-Analysis of Their Role in Reducing Mental Health Issues (2025) · cites this paper as [61]
Cites
The PHQ-9: Validity of a brief depression severity measure
Kroenke, K., Spitzer, R. L., & Williams, J. B. W. (2001) · Journal of General Internal Medicine · doi:10.1046/j.1525-1497.2001.016009606.x · Cited by 44,910 · see citing works
Torous, J., Nicholas, J., Larsen, M. E., Firth, J., & Christensen, H. (2018) · Evidence Based Mental Health · doi:10.1136/eb-2018-102891 · Cited by 790 · see citing works
Li, H., Zhang, R., Lee, Y.-C., Kraut, R. E., & Mohr, D. C. (2023) · npj Digital Medicine · doi:10.1038/s41746-023-00979-5 · Cited by 598 · see citing works
Abd-Alrazaq, A. A., Rababeh, A., Alajlani, M., Bewick, B. M., & Househ, M. (2020) · Journal of Medical Internet Research · doi:10.2196/16021 · Cited by 503 · see citing works
Gilbody, S., Littlewood, E., Hewitt, C., Brierley, G., Tharmanathan, P., Araya, R., Barkham, M., Bower, P., Cooper, C., Gask, L., et al. (2015) · BMJ · doi:10.1136/bmj.h5627 · Cited by 409 · see citing works
Digital interventions for the treatment of depression: a meta-analytic review
Moshe, I., Terhorst, Y., Philippi, P., Domhardt, M., Cuijpers, P., Cristea, I., Pulkki-Råback, L., Baumeister, H., & Sander, L. B. (2021) · Psychological Bulletin · doi:10.1037/bul0000334 · Cited by 390 · see citing works
The Relationship Between Homework Compliance and Therapy Outcomes: An Updated Meta-Analysis
Mausbach, B. T., Moore, R., Roesch, S., Cardenas, V., & Patterson, T. L. (2010) · Cognitive Therapy and Research · doi:10.1007/s10608-010-9297-z · Cited by 309 · see citing works
Punton, G., Dodd, A. L., & McNeill, A. (2022) · PLOS ONE · doi:10.1371/journal.pone.0265542 · Cited by 167 · see citing works
Conklin, L. R., & Strunk, D. R. (2015) · Behaviour Research and Therapy · doi:10.1016/j.brat.2015.06.011 · Cited by 59 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Zhang, Q., Zhang, R., Xiong, Y., Sui, Y., Tong, C., & Lin, F.-H. (2025) · Journal of Medical Internet Research · doi:10.2196/78238 · Cited by 56 · see citing works
“[26,50,51,53,55,58,60,61,63,64,66,67]. g: Hedges g ; PI: predictive intervals; SE: standard error; SMD: standardized mean difference.”
InfluentialHabicht, J., Dina, L.-M., McFadyen, J., Stylianou, M., Harper, R., Hauser, T. U., & Rollwage, M. (2025) · Journal of Medical Internet Research · doi:10.2196/60435 · Cited by 52 · see citing works
Self-citation
- Authors
- McFadyen, J., Habicht, J., Dina, L.-M., Harper, R., Hauser, T. U., & Rollwage, M.
- Affiliations
- University of Tübingen (DE)
- Corresponding author
- McFadyen, Jessica
- Venue
- medRxiv
- Publisher
- openRxiv
- Published
- 2024-11-02
- OpenAlex
- W4403999154
- Cited by
- 10
- References
- 9
- Citations by year
- 2026: 6 · 2025: 4
- Type
- posted content
- Language
- en
- Open access
- green
- doi.org
- Licence
- Locations
- medRxiv (acceptedVersion, cc-by-nd)
- Topics
- Social Sciences › Psychology › Applied Psychology › Digital Mental Health Interventions
- Social Sciences › Psychology › Experimental and Cognitive Psychology › Anxiety, Depression, Psychometrics, Treatment, Cognitive Processes
- Social Sciences › Psychology › Clinical Psychology › Psychotherapy Techniques and Applications
- Keywords
- generative AI, chatbots, cognitive behavioral therapy, digital mental health interventions, patient engagement, anxiety, depression, well-being
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
