Behavior Emotion Therapy System and You: Co-Design and Evaluation of a Mental Health Chatbot and Digital Human for Mild to Moderate Anxiety in Healthy Participants
Authors:
- Almira Osmanovic ThunströmATAlmira Osmanovic Thunström
- Lilas AliLALilas Ali
- Hanne Krage CarlsenHCHanne Krage Carlsen
- Maria BohmMBMaria Bohm
- Linda WesénLWLinda Wesén
- Olof WredeOWOlof Wrede
- Iris Sarajlic VukovicIVIris Sarajlic Vukovic
- Andreas HellströmAHAndreas Hellström
- Tomas LarsonTLTomas Larson
- Steinn SteingrimssonSteinn Steingrimsson
Open Access
- 30 October 2025
- journal article
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 9, e66163
Abstract
Background: Co-design is a collaborative approach involving end users, stakeholders, and designers in creating digital tools for health care. This study focuses on the co-design and evaluation of Behavior, Emotion, Therapy System, and You (BETSY), a mental health chatbot and digital human for mild to moderate anxiety. Objective: This study aims to develop and evaluate BETSY through a co-design process involving potential users. Methods: The study used a mixed-methods approach across 3 phases. Phase 1 involved recruiting 87 volunteer participants through social media for initial end-user requirements. Phase 2 focused on the design process based on end-user requirements using the expertise of 10 stakeholders in health care and health service user. Phase 3 evaluated the 2 prototyped interfaces (a text-only chatbot and a voice-activated digital human chatbot) with 45 healthy volunteers. Results: For phase 1, 61% (n=51) of participants had previous experience with chatbots and 86% expressed a willingness or potential willingness to use a chatbot for mental health support. Thematic analysis in phase 2 revealed key user preferences for a personalized, nonjudgmental chatbot with a clear identity and a focus on empathy. Privacy concerns and the need for clear interaction guidelines were highlighted. In phase 3, the users showed a strong preference for discussing anxiety related to work, relationships, and health. Both text-only and voice-activated digital human users considered BETSY valuable for managing mild to moderate anxiety and providing self-help exercises. Conclusions: The co-design process yielded valuable insights for developing BETSY. While users recognized its potential as an accessible first step before seeking professional help, they also identified areas for improvement, including more nuanced conversation capabilities and a broader range of exercises. BETSY’s potential as a screening tool for health care was consistently acknowledged, suggesting its capability for alleviation of health care system burdens.Keywords
This publication has 27 references indexed in Scilit:
- Unlocking human-like conversations: Scoping review of automation techniques for personalized healthcare interventions using conversational agentsInternational Journal of Medical Informatics, 2024
- ChatGPT and mental healthcare: balancing benefits with risks of harmsBMJ Mental Health, 2023
- Chatbots to Support Mental Wellbeing of People Living in Rural Areas: Can User Groups Contribute to Co-design?Journal of Technology in Behavioral Science, 2021
- The human side of human-chatbot interaction: A systematic literature review of ten years of research on text-based chatbotsInternational Journal of Human-Computer Studies, 2021
- User Perception of Text-Based Chatbot Personality2021
- “I’m Here for You”: Can Social Chatbots Truly Support Their Users? A Literature Review2021
- GPT-3: What’s it good for?Natural Language Engineering, 2020
- How Should My Chatbot Interact? A Survey on Social Characteristics in Human–Chatbot Interaction DesignInternational Journal of Human–Computer Interaction, 2020
- Practitioner review: Co‐design of digital mental health technologies with children and young peopleJournal of Child Psychology and Psychiatry, 2020
- A Mental Health Chatbot for Regulating Emotions (SERMO) - Concept and Usability TestIEEE Transactions on Emerging Topics in Computing, 2020