Young People Aged 18-25 With Depression or Anxiety Mood Participate in an LLM-based Digital Dialogue Intervention Study

Effect of an AI Agent Trained on a Large Language Model (LLM) as an Intervention for Depression and Anxiety Symptoms in Young Adults: a 28-day Randomized Controlled Trial

This study is a 28-day randomized controlled trial (RCT). Residents were randomly assigned to an intervention group or a waiting group according to the order in which they were successfully contacted by the staff, and each user was asked to engage in a total of 28 days of dialog intervention with the Douyin companion bot and complete three psychological questionnaires (on Days 1, 14, and 28); however, the intervention group began to receive the dialog intervention after completing the first questionnaire, and the waiting group began to receive the dialog intervention after completing the third questionnaire. During the first four weeks, the waiting group was treated as a blank control. The two groups of subjects completed the three questionnaires at exactly the same point in time. Each user's depression, anxiety, and positive and negative emotions were measured using the Patient Health Questionnaire (PHQ-9), the Generalized Anxiety Disorder Scale (GAD-7) and the Positive and Negative Affect Schedule (PANAS), respectively.

Study Overview

• Status: Completed
• Conditions: Depression; Anxiety; LLM-based AI Dialogue Bot
• Intervention / Treatment: Other: Douyin Xinqing AI dialog bot; Other: Douyin Xinqing AI dialog bot for delayed intervention

• Study Type: Interventional
• Enrollment (Actual): 657
• Phase: Not Applicable

Contacts and Locations

Study Locations

• China
  • Beijing
    • Beijing, Beijing, China
      • Institute of Psychology, Chinese Academy of Sciences

Participation Criteria

Eligibility Criteria

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

Description

Inclusion Criteria:

• 18-25 years old
• Proficient in Chinese
• Not dyslexic
• Scoring at least 5 on the PHQ-9 or GAD-7
• Not having a serious physical or diagnosed mental illness

Exclusion Criteria:

• Unwilling to continue to participate in the study
• Insufficient clocking in

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Other
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: Single

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: AI conversation intervention; Participants in the intervention group received a 28-day AI conversation intervention after completing the baseline questionnaire. We asked users in the intervention group to use the web link every day, upload the start and end times of their conversations with the AI bot and screenshots of their chats; users who chatted for more than 5 minutes were considered to have successfully punched in for that day. Participants who clocked more than 8 days in the first 14 days were allowed to complete the second questionnaire at the two-week mark. Similarly, participants who quit more than eight times between days 15 and 28 could complete the third questionnaire at the four-week mark. Therefore, all users in the intervention group who completed the third questionnaire received at least 16 AI conversation interventions.	Other: Douyin Xinqing AI dialog bot; Douyin Xinqing is an AI dialog bot. Users can start chatting with the Douyin companion bot by clicking on the "Direct Message" on the homepage of their accounts on Douyin app. After a user sends any message to the Douyin companion bot, it first asks the user about his or her mood and guides him or her to start chatting around his or her emotions. According to the situation described by the user, the companion bot expresses empathy, asks questions to guide the user to talk, and provides suggestions at the appropriate time. Users can also record their emotions every day and look back regularly to observe changes in their inner feelings. In addition, the Douyin companion bot can also identify crisis situations such as suicide and severe mental illness and, if necessary, guide users to call the hotline to obtain professional psychological assistance.
Other: Waiting group; The waiting group received the same intervention as the intervention group after a 28-day waiting period.	Other: Douyin Xinqing AI dialog bot for delayed intervention; Only the users in the waiting group had the task of waiting for the first four weeks, and they completed the three questionnaires at exactly the same point in time as those in the intervention group. Only after the first four weeks of intervention did the waiting group begin to receive the 28-day conversational intervention, and no additional questionnaires were interspersed in between.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Depression	Depression was assessed with the Patient Health Questionnaire (PHQ-9). The PHQ-9 contains nine items scored on a scale from 0 (not at all) to 3 (almost every day) based on the self-reported frequency of depressive symptoms in the past 2 weeks. The total score ranges from 0 to 27, and scores above 10 indicate probable depression symptoms.	The day before the start of the intervention, day 14 and day 28 after the start of the intervention
Anxiety	Anxiety was assessed with the Chinese version of the Generalized Anxiety Disorder Scale (GAD-7). The 7 items are rated on a 4-point scale ranging from 0 (never) to 3 (almost every day) to assess the frequency of anxiety symptoms in the past 2 weeks. The total score ranges from 0 to 21, and according to the established criteria, scores above 10 indicate probable anxiety symptoms.	The day before the start of the intervention, day 14 and day 28 after the start of the intervention
Positive and negative moods	Positive and negative moods were assessed with the Positive and Negative Affect Schedule (PANAS). The scale was categorized into two dimensions, positive (PA) and negative (NA), and each dimension was scored independently, with 9 items per dimension. All the items are rated on a 5-point scale ranging from 1 (few) to 5 (very much). The higher the subject's score on a subscale is, the stronger the associated emotion experienced.	The day before the start of the intervention, day 14 and day 28 after the start of the intervention

Collaborators and Investigators

• Sponsor: Institute of Psychology, Chinese Academy of Sciences
• Investigators: Study Chair: Zhengkui Liu, PhD, Institute of Psychology, Chinese Academy of Sciences; Study Director: Kankan Wu, PhD, Institute of Psychology, Chinese Academy of Sciences

Study record dates

Study Major Dates

• Study Start (Actual): November 20, 2023
• Primary Completion (Actual): December 18, 2023
• Study Completion (Actual): January 15, 2024

Study Registration Dates

• First Submitted: March 28, 2024
• First Submitted That Met QC Criteria: March 28, 2024
• First Posted (Actual): April 4, 2024

Study Record Updates

• Last Update Posted (Actual): April 9, 2024
• Last Update Submitted That Met QC Criteria: April 7, 2024
• Last Verified: October 1, 2023

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Behavioral Symptoms; Mental Disorders; Mood Disorders; Depression; Depressive Disorder; Anxiety Disorders
• Other Study ID Numbers: H23133

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No