AI-Powered Counseling for Depression and Anxiety

(SPEAC Trial)

Recruiting at 1 trial location

Overseen By Jun Ma, MD, PhD

Age: 18+

Sex: Any

Trial Phase: Academic

Sponsor: University of Illinois at Chicago

Must not be taking: Antidepressants, Antianxiety

Disqualifiers: Bipolar, Psychotic disorder, Severe medical, others

Compensation: Varies

Number of Visits: 11

Duration: 12 weeks

200 Participants Needed

Trial Summary

Do I have to stop taking my current medications to join the trial?

Yes, you must not be on current pharmacotherapy for depression or anxiety to join the trial. However, if you start taking medications after joining, you won't be withdrawn from the study.

Do I need to stop my current medications to join the trial?

The trial requires that participants do not currently take medications for depression or anxiety. However, if you start taking these medications after joining, you won't be removed from the study.

What data supports the idea that AI-Powered Counseling for Depression and Anxiety is an effective treatment?

The available research shows that AI-Powered Counseling for Depression and Anxiety can be effective. For example, a study on a computer-based Problem Solving Treatment program found that participants with moderate to severe depression symptoms reported significant improvement over time. They also felt a strong connection with the program, similar to what they might experience in live therapy. Another study on a digital mental health app called Wysa showed that users developed a positive relationship with the AI, which is important for therapy success. These findings suggest that AI-powered treatments can be as effective as traditional therapy methods. ^{1 2 3 4 5}

What data supports the effectiveness of the treatment AI-Powered Counseling for Depression and Anxiety?

Research shows that digital mental health services, like AI-powered counseling, can establish a strong therapeutic alliance (a supportive relationship between therapist and patient) similar to traditional face-to-face therapy. Additionally, computer-based Problem Solving Treatment (PST) programs have been found to significantly improve depressive symptoms, suggesting that AI-guided PST interventions could be effective for treating depression and anxiety. ^{1 2 3 4 5}

What safety data exists for AI-Powered Counseling for Depression and Anxiety?

The safety data for AI-Powered Counseling for Depression and Anxiety, under various names like Human Coached PST Intervention and Lumen, is not directly addressed in the provided research. However, related studies indicate that AI and machine-learning technologies are being used to enhance psychotherapy feedback, improve therapeutic alliance, and manage depression through digital interventions. These studies suggest that AI-based systems are generally well-received, usable, and comparable to traditional therapy in terms of therapeutic alliance. While these findings are promising, they primarily focus on feasibility, usability, and therapeutic alliance rather than direct safety outcomes. Larger clinical trials are recommended to further evaluate safety and efficacy. ^{1 2 3 6 7}

Is AI-Powered Counseling for Depression and Anxiety safe for humans?

The studies reviewed do not provide specific safety data for AI-Powered Counseling for Depression and Anxiety, but they suggest that AI-based mental health interventions are generally well-received and comparable to traditional therapy in terms of user satisfaction and therapeutic alliance. ^{1 2 3 6 7}

Is the treatment 'Human Coached PST Intervention, Lumen Coached PST Intervention' a promising treatment for depression and anxiety?

Yes, the treatment is promising because AI-powered counseling, like chatbots and digital mental health services, can effectively support people with depression and anxiety. They help build a strong connection with users, similar to traditional therapy, and can be used anytime, making mental health care more accessible and efficient. ^{1 2 4 8 9}

How is the AI-Powered Counseling for Depression and Anxiety treatment different from other treatments?

This treatment is unique because it uses AI-powered conversational agents to provide problem-solving therapy (PST) for depression and anxiety, offering real-time, personalized support that can enhance the therapeutic alliance and improve patient engagement compared to traditional methods. ^{1 2 4 8 9}

What is the purpose of this trial?

Approximately 200 Participants with mild-to-moderate, untreated depression and/or anxiety will be randomly assigned (by chance, like flipping a coin) to 1 of 3 study groups: Lumen Coached Problem-Solving Treatment (PST) (n=100), Human Coached PST (n=50), and optional (delayed) Lumen Coached PST as waitlist control (n=50) to improve emotional health. All participants will complete assessments at baseline and at 18 weeks post randomization. Depending on the group assignment the PST program will be delivered by Lumen, a virtual voice-based coach on a study iPad, or by a human coach in person for the first session and then via videoconference or phone for the remaining 7 sessions. Participants assigned to the waitlist control group can receive the Lumen coached PST on a study iPad after completing their 18-week follow-up assessment. Participants will receive 8 coaching sessions to learn problem-solving skills and work on unresolved problems in daily living that may be interfering with their emotional well-being and contributing to depression and anxiety symptoms.

Research Team

Jun Ma, MD

Principal Investigator

University of Illinois at Chicago

Eligibility Criteria

Adults with mild-to-moderate, untreated depression or anxiety can join. They must score within certain ranges on the PHQ-9 and GAD-7 scales, be over 18 years old, and able to give informed consent in English. Exclusions include severe medical conditions, brain injuries or abnormalities, current heavy treatments for cancer, pregnancy or recent childbirth, other clinical trials participation that could interfere with this study's results.

Inclusion Criteria

Emotional distress defined by elevated depressive (PHQ-9 scores 10-19) and/or anxious symptoms (GAD-7 scores 10-14)

Willing and able to provide informed eConsent and HIPAA authorization

Exclusion Criteria

Unable to speak, read, or understand English for informed consent

I weigh 325 pounds or more.

I was diagnosed with cancer, other than non-melanoma skin cancer, in the last year.

Timeline

Screening

Participants are screened for eligibility to participate in the trial

2-4 weeks

Intervention Orientation

Participants attend an orientation session to receive training and materials for the PST program

1 week

1 visit (in-person)

Treatment

Participants complete 8 PST sessions over 12 weeks, either with Coach Lumen or a human coach

12 weeks

8 sessions (4 weekly, 4 biweekly)

Follow-up

Participants are monitored for safety and effectiveness after treatment, including assessments at 18 weeks

18 weeks

2 visits (in-person)

Optional Extension

Participants in the delayed Lumen Coached Group can choose to start PST sessions after the 18-week follow-up

Treatment Details

Interventions

• Human Coached PST Intervention
• Lumen Coached PST Intervention
• Optional (delayed) Lumen Coached PST (Waitlist control)

Trial Overview The trial is testing a virtual voice-based coach called Lumen against human coaches to deliver Problem-Solving Treatment (PST) via an iPad or videoconference/phone. Participants will learn problem-solving skills over eight sessions to help manage daily problems contributing to their emotional distress.

Participant Groups

Group I: Group 1: Lumen Coached Group Active Control 1 Intervention

Participants in this group will attend an Intervention Orientation session to receive the Coach Lumen tutorial, a study iPad, and a Lumen Intervention Workbook. They will complete 8 PST sessions with Coach Lumen using their study assigned iPad at home over 12 weeks. The first 4 sessions occur weekly and then the last 4 occur every other week. Participants will receive automated reminders on the day before and on the day of their scheduled session, and have the opportunity to reschedule sessions through the iPad. Participants will work with Coach Lumen to learn problem-solving skills to address current life challenges, plan activities, and complete home activities. Participants will also complete surveys before and after each PST session: a mood assessment survey before and a user experience survey after. At the end of study, Participants may be invited to provide their perspective regarding their Lumen use experience.

Group II: Group 2: Human Coached Group Active Control 1 Intervention

Participants in this group will receive a study iPad and Human-Coached Intervention Workbook and will complete the first of 8 PST sessions in person with a trained health coach. Participants' first PST session will be in person for approximately 1 hour. Participants will complete the remaining sessions remotely via Zoom (or by phone, if necessary) using their study iPad. The first 4 sessions occur weekly and then the last 4 occur every other week. Participants will receive automated reminder notifications on the day before and on the day of their scheduled session, and have the opportunity to reschedule sessions with their coach. Participants will also complete a mood assessment survey at the beginning of each PST session. Participants will work with their health coach to learn problem-solving skills to address current life challenges, plan activities, and complete home activities.

Group III: Group 3: Optional (Delayed) Lumen Coached Group Placebo Group 1 Intervention

Participants assigned to this group can choose to attend a Lumen Orientation session to receive training and a study iPad to complete 8 PST sessions with Coach Lumen after their follow-up assessment at 18 weeks. They will receive 8 Sunday mood assessment surveys to complete (4 every Sunday and then 4 every other Sunday).

Who Is Running the Clinical Trial?

University of Illinois at Chicago

Lead Sponsor

Trials

653

Recruited

1,574,000+

Penn State University

Collaborator

Trials

380

Recruited

131,000+

Washington University School of Medicine

Collaborator

Trials

2,027

Recruited

2,353,000+

Stanford University

Collaborator

Trials

2,527

Recruited

17,430,000+

National Institute of Mental Health (NIMH)

Collaborator

Trials

3,007

Recruited

2,852,000+

Findings from Research

In a study involving 1,205 users of the Wysa app for mental health, users reported a strong therapeutic alliance with the AI conversational agent, with scores comparable to traditional therapy methods like CBT.

Over a brief period of 3 days, users' perceptions of their bond with Wysa improved, indicating that digital mental health services can effectively foster a therapeutic relationship similar to that found in face-to-face therapy.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Evaluating the Therapeutic Alliance With a Free-Text CBT Conversational Agent (Wysa): A Mixed-Methods Study. Beatty, C., Malik, T., Meheli, S., et al.[2022]

This study will evaluate the effectiveness of an innovative internet-based cognitive behavioral therapy (iCBT) program enhanced by artificial intelligence (AI) technologies to improve depression among 1400 healthy workers during the COVID-19 pandemic.

The program, called SMART-CBT, consists of a 6-week course with machine-guided exercises and aims to provide a cost-effective mental health solution that can be widely disseminated, marking the first randomized controlled trial of its kind.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Effectiveness of an Internet-Based Machine-Guided Stress Management Program Based on Cognitive Behavioral Therapy for Improving Depression Among Workers: Protocol for a Randomized Controlled Trial. Kawakami, N., Imamura, K., Watanabe, K., et al.[2021]

The ePST™ program, a stand-alone multimedia interactive treatment for depression, was well-received by participants, who found it highly usable, acceptable, and credible, with a therapeutic alliance comparable to live therapy.

Among the 29 participants, those who completed at least 4 sessions showed significant improvement in depressive symptoms over the 10-week study, suggesting that ePST™ may be an effective treatment option for depression.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Feasibility study of an interactive multimedia electronic problem solving treatment program for depression: a preliminary uncontrolled trial. Berman, MI., Buckey, JC., Hull, JG., et al.[2021]

A computer-aided self-help clinic effectively increased the number of anxious and depressed patients treated per clinician while reducing the time each patient spent with a clinician, without compromising treatment effectiveness.

Patients showed improvement using one of four computer-aided cognitive behavior therapy (CBT) systems for various anxiety and mood disorders, highlighting the potential for these systems to enhance accessibility and reduce stigma associated with mental health treatment.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Computer-aided CBT self-help for anxiety and depressive disorders: experience of a London clinic and future directions. Gega, L., Marks, I., Mataix-Cols, D.[2022]

The implementation of the ItFits-toolkit in the online clinic improved the intake procedure for therapists, providing them with more relevant patient information and leading to better alignment of expectations between patients and therapists.

Therapists reported that the new automated screening process resulted in a better fit of patients for treatment, increased patient inclusion in the program, and overall satisfaction, indicating that the ItFits-toolkit effectively enhanced the delivery of guided internet-based cognitive behavioral therapy (iCBT) for anxiety and depression.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Therapist perceptions of the implementation of a new screening procedure using the ItFits-toolkit in an iCBT routine care clinic: A mixed-methods study using the consolidated framework for implementation research. Tarp, K., Nielsen, SL., Holmberg, TT., et al.[2023]

An automated feedback system using machine learning and speech processing technologies was developed to provide therapists with real-time quality metrics on their motivational interviewing skills, making the feedback process more efficient and scalable.

In a study with 21 therapists, the majority found the system easy to use and reported that the feedback aligned with their self-assessment, indicating strong potential for integrating this technology into regular psychotherapy training and practice.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Design feasibility of an automated, machine-learning based feedback system for motivational interviewing. Imel, ZE., Pace, BT., Soma, CS., et al.[2022]

Project AFFECT aims to create an AI-based software, LyssnCBT, to automatically assess the quality of cognitive behavioral therapy (CBT) sessions, which could enhance training and supervision for therapists and improve treatment outcomes for clients.

In a planned Phase II trial involving 1,875 clients, the effectiveness of LyssnCBT will be evaluated to see if it can improve therapist skills, client outcomes, and reduce client drop-out rates, marking a significant step towards scalable quality assurance in psychotherapy.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Enhancing the quality of cognitive behavioral therapy in community mental health through artificial intelligence generated fidelity feedback (Project AFFECT): a study protocol. Creed, TA., Salama, L., Slevin, R., et al.[2023]

Chatbots can effectively support patients with anxiety and depression, serving as affordable and efficient virtual assistants, especially in areas with limited mental health resources.

Out of 42 studies reviewed, about 60% of chatbots targeted both anxiety and depression, indicating their versatility in addressing multiple mental health issues, while their role in mental health care is expected to grow.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Chatbot features for anxiety and depression: A scoping review. Ahmed, A., Hassan, A., Aziz, S., et al.[2023]

A systematic review of 10 studies found that using artificial intelligence (AI), particularly conversational AI agents like chatbots, can significantly enhance psychotherapy outcomes and reduce symptoms in patients with emotional disorders.

Patients reported high satisfaction and engagement when AI was integrated into therapy, suggesting that AI can make psychological interventions more personalized and effective, although further robust research is needed.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Using Artificial Intelligence to Enhance Ongoing Psychological Interventions for Emotional Problems in Real- or Close to Real-Time: A Systematic Review. Gual-Montolio, P., Jaén, I., Martínez-Borba, V., et al.[2022]

References

1. Switzerland pubmed.ncbi.nlm.nih.gov

Evaluating the Therapeutic Alliance With a Free-Text CBT Conversational Agent (Wysa): A Mixed-Methods Study. [2022]

2. Canada pubmed.ncbi.nlm.nih.gov

Effectiveness of an Internet-Based Machine-Guided Stress Management Program Based on Cognitive Behavioral Therapy for Improving Depression Among Workers: Protocol for a Randomized Controlled Trial. [2021]

3. England pubmed.ncbi.nlm.nih.gov

Feasibility study of an interactive multimedia electronic problem solving treatment program for depression: a preliminary uncontrolled trial. [2021]

4. United States pubmed.ncbi.nlm.nih.gov

Computer-aided CBT self-help for anxiety and depressive disorders: experience of a London clinic and future directions. [2022]

5. Switzerland pubmed.ncbi.nlm.nih.gov

Therapist perceptions of the implementation of a new screening procedure using the ItFits-toolkit in an iCBT routine care clinic: A mixed-methods study using the consolidated framework for implementation research. [2023]

6. United States pubmed.ncbi.nlm.nih.gov

Design feasibility of an automated, machine-learning based feedback system for motivational interviewing. [2022]

7. England pubmed.ncbi.nlm.nih.gov

Enhancing the quality of cognitive behavioral therapy in community mental health through artificial intelligence generated fidelity feedback (Project AFFECT): a study protocol. [2023]

8. England pubmed.ncbi.nlm.nih.gov

Chatbot features for anxiety and depression: A scoping review. [2023]

9. Switzerland pubmed.ncbi.nlm.nih.gov

Using Artificial Intelligence to Enhance Ongoing Psychological Interventions for Emotional Problems in Real- or Close to Real-Time: A Systematic Review. [2022]