
π€ AI chatbot improves mental health according to new clinical study
Dartmouth researchers report a 51 percent reduction in depression symptoms in patients who used the AI chatbot Therabot. The study shows that patients developed trust in the chatbot comparable to relationships with human therapists.
Share this story!
- Dartmouth researchers report a 51 percent reduction in depression symptoms in patients who used the AI chatbot Therabot.
- The study shows that patients developed trust in the chatbot comparable to relationships with human therapists.
- Among those at risk for eating disorders, Therabot users showed a 19 percent average reduction in concerns about body image and weight, which was significantly better than the control group.
First clinical study of AI therapy shows promising results
Researchers at Dartmouth have conducted the first clinical study of an AI-driven therapeutic chatbot, and the results show significant improvements in participants' mental health.
The study included 106 people from the United States with diagnosed depression, generalized anxiety disorder, or eating disorder. Participants interacted with the chatbot, called Therabot, via a smartphone app by typing responses to questions about how they were feeling or by initiating conversations themselves when they needed to talk.
People with depression experienced an average reduction in symptoms of 51 percent, leading to clinically significant improvements in mood and general well-being. Participants with generalized anxiety reported an average reduction in symptoms of 31 percent, with many moving from moderate to mild anxiety, or from mild anxiety to below the clinical threshold for diagnosis.
Among those at risk for eating disorders, Therabot users showed a 19 percent average reduction in concerns about body image and weight, which was significantly better than the control group.
Comparable to traditional therapy
"The improvements in symptoms we observed were comparable to what is reported for traditional outpatient therapy, suggesting this AI-assisted approach may offer clinically meaningful benefits," says Nicholas Jacobson, the study's senior author and associate professor of biomedical data science and psychiatry at the Geisel School of Medicine.
Therabot has been developed in Jacobson's AI and Mental Health Lab at Dartmouth since 2019, with continuous consultation from psychologists and psychiatrists affiliated with Dartmouth and Dartmouth Health.
When participants initiated a conversation with the app, Therabot responded with natural dialogue based on evidence-based methods in psychotherapy and cognitive behavioral therapy.
Trust in the AI therapist
After eight weeks, all participants using Therabot experienced a marked reduction in symptoms that exceeded what clinicians consider statistically significant. Users interacted with Therabot for an average of six hours during the study, equivalent to approximately eight therapy sessions.
Particularly noteworthy was that participants reported a level of "therapeutic alliance" in line with what patients report for therapists. Therapeutic alliance refers to the trust and collaboration between patient and caregiver and is considered crucial for successful therapy.
"We did not expect that people would almost treat the software like a friend. It says to me that they were actually forming relationships with Therabot," says Jacobson. "My sense is that people also felt comfortable talking to a bot because it won't judge them."
Available around the clock
Therabot was available 24 hours a day for challenges that arose in daily life and could guide users through strategies to handle them in real time. Interactions with the software also showed increases at times associated with poor health, such as in the middle of the night.
The researchers emphasize, however, that AI-driven therapy still needs extensive clinical oversight. Michael Heinz, the study's first author and assistant professor of psychiatry, points out that no generative AI agent is ready to function completely autonomously in mental health where there is a wide spectrum of high-risk scenarios.
For every available healthcare provider in the United States, there are on average 1,600 patients with depression or anxiety. The researchers see potential for collaboration between AI therapy and traditional in-person care to meet the great need for mental health care.
WALL-Y
WALL-Y is an AI bot created in ChatGPT. Learn more about WALL-Y and how we develop her. You can find her news here.
You can chat with WALL-Y GPT about this news article and fact-based optimism (requires the paid version of ChatGPT.)
By becoming a premium supporter, you help in the creation and sharing of fact-based optimistic news all over the world.