Effectiveness of a Guided Mental Health Chatbot for Youth Living in Lithuania

Effectiveness of a Guided Mental Health Chatbot STARS for Youth Living in Lithuania: a Randomized Controlled Trial

The study aims to assess the effectiveness of a guided mental health chatbot called "Scalable Technology for Adolescents and youth to Reduce Stress (STARS)", adapted for youth living in Lithuania.

Study Overview

• Status: Active, not recruiting
• Conditions: Symptoms of Depression and Anxiety
• Intervention / Treatment: Behavioral: STARS (Scalable Technology for Adolescents and youth to Reduce Stress); Behavioral: Basic psychoeducation

Detailed Description

The intervention is a guided mental health chatbot STARS developed by the World Health Organization (WHO). It consists of 10 sessions which include: Introduction (1), Psychoeducation (2), Emotion Regulation (3, 4), Behavior Activation (5, 6), Managing Problems (7), Supportive Self-talk (8, 9), and Relapse Prevention (10). Each session consists of psychoeducation and exercise parts. Trained and supervised non-specialists (called e-helpers) provide weekly calls (five calls lasting approximately 15 minutes) to support participants individually.

The effect of the intervention will be compared against a control group which receives enhanced usual care.

• Study Type: Interventional
• Enrollment (Estimated): 262
• Phase: Not Applicable

Contacts and Locations

Study Locations

• Lithuania
  • Vilnius County
    • Vilnius, Vilnius County, Lithuania, 01131
      • Vilnius University

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult
• Accepts Healthy Volunteers: No

Description

Inclusion Criteria:

• Age of 18-25 years
• Understanding the Lithuanian language
• Having access to a device for intervention delivery
• Experiencing moderate levels of psychological distress (a cut-off score of 8 as measured with Kessler-6)

Exclusion Criteria:

• People at imminent risk of suicide
• People currently experiencing a psychotic episode
• People currently experiencing interpersonal violence
• Current dependency on alcohol/drugs

Study Plan

How is the study designed?

Design Details

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

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: Intervention group; The intervention group will get a guided mental health chatbot STARS.	Behavioral: STARS (Scalable Technology for Adolescents and youth to Reduce Stress); The intervention is a guided mental health chatbot - STARS - developed by the World Health Organization (WHO). It consists of 10 sessions which include: Introduction (1), Psychoeducation (2), Emotion Regulation (3, 4), Behavior Activation (5, 6), Managing Problems (7), Supportive Self-talk (8, 9), and Relapse Prevention (10). Each session consists of psychoeducation and exercise parts. Trained and supervised non-specialists (called e-helpers) provide weekly calls (five calls lasting approximately 15 minutes) to support participants individually and increase motivation and adherence to the intervention.
Other: Control group; The control group will receive basic psychoeducation. The content of the psychoeducation is very similar to the content of session 2 of the STARS chatbot and includes psychoeducation about emotions, a personal story about a fictional character who talks about her emotions, and information about where to access mental health support. It also includes a list of organizations and clinics providing mental health care in Lithuania.	Behavioral: Basic psychoeducation; The control group will receive basic psychoeducation. The content of the psychoeducation is very similar to the content of session 2 of the STARS chatbot and includes psychoeducation about emotions, a personal story about a fictional character who talks about her emotions, and information about where to access mental health support. It also includes a list of organizations and clinics providing mental health care in Lithuania.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Changes in The Patient Health Questionnaire-9	Changes in symptoms of depression are measured. The Patient Health Questionnaire-9 (PHQ-9, Kroenke et al., 2001) is a self-report measure comprising 9 items about symptoms of depression. All items are answered on a 4-point Likert scale that ranges from 0 (not at all) to 3 (nearly every day). A higher score indicates more pronounced symptoms. In the Lithuanian sample, the PHQ-9 had a good internal consistency with Cronbach's α of 0.85 (Nomeikaite et al., 2023).	Baseline, eight weeks after baseline, three months after baseline
Changes in The Generalized Anxiety Disorder-7	Changes in symptoms of anxiety are measured. The Generalized Anxiety Disorder-7 (GAD-7, Spitzer et al., 2006) is a self-report measure comprising 7 items about symptoms of anxiety. All items are answered on a 4-point Likert scale that ranges from 0 (not at all) to 3 (nearly every day). A higher score indicates more pronounced symptoms. In the Lithuanian sample, the GAD-7 had a good internal consistency with Cronbach's α of 0.91 (Nomeikaite et al., 2023).	Baseline, eight weeks after baseline, three months after baseline

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Changes in The Client Services Receipt Inventory	Changes in client service are measured. The Client Services Receipt Inventory (CSRI, Beeacham and Knapp, 2001) is a self-report measure comprising questions about client service. The CSRI covers a broad range of services that may be utilized, including primary and secondary care services, other services, and informal care.	Baseline, three months after baseline
Changes in Kessler-6	Kessler-6 is used as a screener with a cut-off score of 8, which is commonly used to indicate moderate levels of psychological distress. Also, changes in psychological distress are measured. Kessler-6 (K6, Kessler et al., 2002) is a self-report measure comprising 6 questions about psychological distress. All items are answered on a 5-point Likert scale that ranges from 0 (none of the time) to 4 (all of the time). A higher score indicates more pronounced symptoms. Previous studies have shown good psychometric properties of the Kessler-6 (Cronbach α = 0.86) (Ferro, 2019).	Baseline, eight weeks after baseline, three months after baseline
Changes in The Brief Adjustment Disorder Measure-8	Changes in symptoms of adjustment disorder are measured. The Brief Adjustment Disorder Measure-8 (ADNM-8, Kazlauskas et al., 2018) is a self-report measure comprising 8 items about symptoms of adjustment disorder. All items are answered on a 4-point Likert scale that ranges from 1 (never) to 4 (often). A higher score indicates more pronounced symptoms. In the Lithuanian sample, the ADNM-8 had a good internal consistency with Cronbach's α of 0.93 (Kazlauskas et al., 2023).	Baseline, eight weeks after baseline, three months after baseline
Changes in The SIDAS	Changes in suicidal ideation are measured. The SIDAS (Van Spijker et al., 2014) is a self-report measure comprising 5 questions about suicidal ideation. All items are answered on a 10-point Likert scale that ranges from 0 (never, no control, not close at all, not at all) to 10 (always, full control, made an attempt, extremely). A higher score indicates more pronounced symptoms. In the Lithuanian sample, the SIDAS had a good internal consistency with Cronbach's α of 0.93 (Skruibis et al., 2022).	Baseline, eight weeks after baseline, three months after baseline
Changes in The WHODAS 2.0	Changes in functioning are measured. The WHODAS 2.0 (Ustun et al., 2010) is a self-report measure comprising 12 questions about functioning. All items are answered on a 5-point Likert scale that ranges from 0 (none) to 4 (extreme or cannot do). A higher score indicates more pronounced symptoms. Previous studies have shown good psychometric properties of the WHODAS 2.0 (Cronbach α = 0.96) (Saltychev et al., 2021).	Baseline, eight weeks after baseline, three months after baseline
Changes in The World Health Organization Well-being Index-5	Changes in psychological well-being are measured. The World Health Organization Well-being Index-5 (WHO-5, Bech, 2004) is a self-report measure comprising 5 items about psychological well-being. All items are answered on a 5-point Likert scale that ranges from 0 (at no time) to 4 (all of the time). A higher score indicates more pronounced psychological well-being. In the Lithuanian sample, the WHO-5 had a good internal consistency with Cronbach's α of 0.87 (Nomeikaite et al., 2023).	Baseline, eight weeks after baseline, three months after baseline
Changes in Internalized Stigma of Mental Illness	Changes in internalized stigma of psychological difficulties (the term of psychological difficulties instead of mental illness is used in this study) are measured. The Internalized Stigma of Mental Illness-Brief Version (ISMI-10, Boyd et al., 2014) is a self-report measure comprising 10 items about internalized stigma of psychological difficulties. All items are answered on a 4-point Likert scale that ranges from 1 (strongly disagree) to 4 (strongly agree). A higher score indicates more pronounced internalized stigma of psychological difficulties. Previous studies have shown good psychometric properties of the ISMI-10 (Cronbach's α = .75-.81) (Boyd et al., 2014).	Baseline, eight weeks after baseline, three months after baseline
Changes in Self-Stigma of Seeking Help	Changes in stigma of self-stigma of seeking help are measured. The Self-Stigma of Seeking Help-Ultra-Brief Version (SSOSH-3, Brenner et al., 2021) is a self-report measure comprising 3 items about self-stigma of seeking help. All items are answered on a 5-point Likert scale that ranges from 1 (strongly disagree) to 5 (strongly agree). A higher score indicates more pronounced self-stigma of seeking help. Previous studies have shown good psychometric properties of the SSOSH-3 (Cronbach's α = .82-.87) (Brenner et al., 2021).	Baseline, eight weeks after baseline, three months after baseline
Changes in The EQ-5D-5L	Changes in health are measured. The EQ-5D-5L (EuroQol Research Foundation, 2019) is a self-report measure comprising 5 items about health. All items are answered on a 5-point Likert scale that ranges from 1 (I have no problems or similar) to 5 (I am unable or similar). A higher score indicates more pronounced symptoms. Convergent validity was demonstrated by a correlation between EQ-5D-5L and the WHO-5, (r = .43, P < .001) (Janssen et al. 2013).	Baseline, eight weeks after baseline, three months after baseline

Other Outcome Measures

Outcome Measure	Measure Description	Time Frame
User satisfaction after intervention	The usability of the intervention is measured. Participants are asked to evaluate the usability of the intervention by indicating how useful (from 1 "not useful at all" to 5 "very useful"), satisfactory (from 1 "I did not like it at all" to 5 "I liked it a lot"), easy to use (from 1 "it was not easy at all" to 5 "it was very easy"), etc., the program is. Participants are also asked to report their subjective impression regarding the improvement of mental well-being (from 1 "worsened a lot" to 5 "improved a lot"), physical health (from 1 "worsened a lot" to 5 "improved a lot"), general understanding of oneself and one's well-being (from 1 "not at all" to 5 "definitely improved"), recommending the program to others (from 1 "not at all" to 5 "definitely would recommend"), etc.	Eight weeks after baseline
International Trauma Exposure Measure before intervention	Number and type of traumatic events before intervention are measured. International Trauma Exposure Measure (ITEM, Hyland et al., 2021) is a self-report measure comprising of 21 items. All items are answered on a dichotomous scale (yes/no). A higher score indicates higher number of traumatic events experienced. ITEM will be used as a predictor in the analysis.	Baseline
The International Trauma Questionnaire before intervention	Symptoms of post-traumatic stress disorders before intervention are measured. International Trauma Questionnaire (ITQ, Cloitre et al., 2018) is a self-report measure comprising 18 items. All items are answered on a 5-point Likert scale that ranges from 0 (not at all) to 4 (extremely). A higher score indicates more pronounced symptoms. ITQ will be used as a predictor in the analysis. In the Lithuanian sample, the ITQ had a good internal consistency with Cronbach's α of 0.86 (Dumarkaite et al., 2023).	Baseline
The Resilience scale before intervention	Resilience before intervention is measured. The Resilience scale (RS-14, Wagnild, 2011) is a self-report measure comprising 14 items about resilience. All items are answered on a 7-point Likert scale that ranges from 1 (strongly disagree) to 7 (strongly agree). A higher score indicates more pronounced resilience. RS-14 will be used as a predictor in the analysis. The analysis of the psychometric properties of the RS-14 in the sample of Lithuanian adolescents demonstrated high reliability of the scale (Zelviene, 2020).	Baseline

Collaborators and Investigators

• Sponsor: Vilnius University
• Collaborators: World Health Organization
• Investigators: Principal Investigator: Evaldas Kazlauskas, Ph.D., Vilnius University

Study record dates

Study Major Dates

• Study Start (Actual): September 8, 2025
• Primary Completion (Estimated): September 30, 2026
• Study Completion (Estimated): June 30, 2028

Study Registration Dates

• First Submitted: May 16, 2025
• First Submitted That Met QC Criteria: June 4, 2025
• First Posted (Actual): June 10, 2025

Study Record Updates

• Last Update Posted (Actual): May 4, 2026
• Last Update Submitted That Met QC Criteria: April 27, 2026
• Last Verified: April 1, 2026

More Information

Terms related to this study

• Keywords: Intervention; Mental health; Effectiveness; Youth; Chatbot; STARS
• Additional Relevant MeSH Terms: Mental Disorders; Behavioral Symptoms; Behavior; Personal Satisfaction; Anxiety Disorders; Depression; Psychological Well-Being
• Other Study ID Numbers: (1.13E)250000-KT-56/ERC0004137

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: YES
• IPD Plan Description: Individual participant data can be shared upon a reasonable request.

Drug and device information, study documents

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