Generated AI-informed Intervention (Emohaa) vs Group ACT vs WL on Emotional Distress

Comparing Generated AI (Emohaa) and Group ACT in Reducing Emotional Distress: a Randomized Controlled Trial

Emotional Distress, such as anxiety and depression, is an essential issue worldwide. There have been several evidence-based psychotherapies that are effective in improving emotional distress, such as Acceptance and Commitment Therapy (ACT). However, the scarcity of professionals and the imbalance in the distribution of mental health resources prevent individuals in need from accessing immediate and effective help.

Artificial intelligence (AI) has the potential to promote this problem. The existing studies have provided preliminary support for the application of AI in mental health interventions. One such model, Emohaa, a generated AI model, has been examined for its effectiveness in adult emotional distress. However, despite the limited reliability of the single-group design, parallel randomized controlled trials are scarce to validate this finding further. This present study is to fill this gap.

This study aims to examine the effectiveness of the generated AI (Emohaa) in reducing emotional distress, including anxiety and depression, compared with group ACT and waitlist. In this parallel randomized controlled trial, it is hypothesized that (1) Compared with the waitlist, Emohaa and group ACT could significantly improve participants' emotional distress, including anxiety and depression symptoms; (2) Emohaa would lead to a greater reduction in anxiety and depression symptoms compared with group ACT.

Study Overview

• Status: Completed
• Conditions: Anxiety; Depression - Major Depressive Disorder; Artificial Intelligence (AI)
• Intervention / Treatment: Behavioral: Generated AI-based intervention (Emohaa); Behavioral: group ACT; Behavioral: wailist (delayed intervenion)

• Study Type: Interventional
• Enrollment (Actual): 80
• Phase: Not Applicable

Contacts and Locations

Study Locations

• China
  • Beijing, China
    • Central University of Finance and Economics, China

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult
• Accepts Healthy Volunteers: Yes

Description

Inclusion Criteria:

• Adults with 18-25 ages;
• Suffering from emotional distress (PHQ-9 > 9 or GAD-7 > 9);
• Ability to use the mobile phone to interact with AI;
• Consent to participate in the study.

Exclusion Criteria:

• High risk of self-injury and suicide;
• Presence of psychotic symptoms (i.e., delusions);
• Serious substance use problems;
• Other mental or physical illnesses requiring urgent medical assistance;
• Received other psychological or pharmacological interventions in the last month

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Prevention
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: None (Open Label)

Number of Arms

3

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: Generated AI (Emohaa)	Behavioral: Generated AI-based intervention (Emohaa); Participants will complete the intervention process by interacting with the Wechat applet. Participants will be required to complete a 6-week self-help therapy. Participants will be required to have a 30-minute dialogue including input and reflection time with Emohaa each week simultaneously with group ACT. Participants are free to dialogue with Emohaa for the rest of the week. An independent research assistant will monitor the intervention process from the platform. If the corresponding dialogue is not completed on time each week, the participant will be reminded via WeChat to complete this session before the start of the next session in the following week.
Active Comparator: Group ACT	Behavioral: group ACT; A 6-week group ACT intervention. Each group has 6-10 participants. Each session is one hour.
Sham Comparator: Waitlist Group (WLG)	Behavioral: wailist (delayed intervenion); During the intervention period, participants in the WLG were only required to complete the appropriate assessments without receiving any intervention. Upon completion of the 15-day follow-up assessment, participants in the WLG could be free to choose to receive one of the two interventions. The intervention will be the same as the corresponding experimental group.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Patient Health Questionnaire (PHQ-9)	PHQ-9 is a multipurpose instrument for screening, diagnosing, monitoring, and measuring the severity of depression. Questions are scored on a Likert scale of 0-3 with 0 being not at all and 3 being nearly every day.	0, 2, 4, 6, 8, and 10 weeks
Generalized Anxiety Disorder (GAD-7)	GAD-7 is an easy-to-perform initial screening tool for generalized anxiety disorder. Questions are scored on a Likert scale of 0-3 with 0 being not at all and 3 being nearly every day.	0, 2, 4, 6, 8, and 10 weeks
UCLA Loneliness Scale (ULS-3)	ULS-3 is a brief version of the UCLA Loneliness Scale with 3 items, which was designed to assess loneliness. Questions are scored on a Likert scale of 1-4 with 1 being never and 4 being always.	0, 2, 4, 6, 8, and 10 weeks
Self-Concept Clarity Scale (SCCS)	SCCS is a 12-item scale to measure the clarity of self-concept. Questions are scored on a Likert scale of 1-5 with 1 being strongly disagree and 5 being strongly agree.	0, 2, 4, 6, 8, and 10 weeks
Acceptance and Action Questionnaire-2nd Edition (AAQ-II)	AAQ-II is the second version of the Acceptance and Action Questionnaire with 7 items. Questions are scored on a Likert scale of 1-7 with 1 being never and 7 being always	0, 2, 4, 6, 8, and 10 weeks
Insomnia Severity Index (ISI)	ISI is a 7-item screening tool to measure the severity of sleep problems. Questions are scored on a 0-4 scale with different descriptions	0, 2, 4, 6, 8, and 10 weeks
Client Satisfaction Questionnaire (CSQ-8)	CSQ-8 is an easy-to-perform initial screening tool for client satisfaction with mental health services. Questions are scored on a Likert scale of 1-4 with 1 being low satisfaction and 4 being high satisfaction.	2, 4, 6, 8, and 10 weeks
Digital Working Alliance Inventory (DWAI)	DWAI is an 8-item scale to measure the working alliance between participants and the AI model revised by the Working Alliance Inventory. Questions are scored on a Likert scale of 1-7 with 1 being totally disagree and 7 being totally agree.	6, 8, and 10 weeks

Collaborators and Investigators

• Sponsor: Central University of Finance and Economics, China

Study record dates

Study Major Dates

• Study Start (Actual): September 20, 2024
• Primary Completion (Actual): December 20, 2024
• Study Completion (Actual): December 28, 2024

Study Registration Dates

• First Submitted: December 6, 2024
• First Submitted That Met QC Criteria: December 11, 2024
• First Posted (Actual): December 16, 2024

Study Record Updates

• Last Update Posted (Actual): April 25, 2025
• Last Update Submitted That Met QC Criteria: April 24, 2025
• Last Verified: September 1, 2024

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Mental Disorders; Mood Disorders; Depressive Disorder; Depressive Disorder, Major
• Other Study ID Numbers: CUFE-20240229-0001

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: YES
• IPD Plan Description: Personal contact in reliable reason
• IPD Sharing Supporting Information Type: ANALYTIC_CODE

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No