Development of a chatbot to promote activity participation for people with serious mental illness

• Author / Creator

  Liu, Huei-Tsz

• People with serious mental illness living in the community often encounter difficulties in engaging in meaningful activities. Sustained disconnection from meaningful activities, significant others, and community create the force of “inertia” that keeps them from making positive changes in day-to-day lives. Their passive, isolated, and resource-restricted lifestyle has been associated with poor well-being and adverse recovery outcomes. Action Over Inertia (AOI) is one of the few interventions explicitly addressing activity health for people with serious mental illness. The AOI intervention emphasizes the power of enabling clients to make quick changes, which create momentum to break the force of “inertia.” However, making changes, even a small step, can be challenging for people who are long-term disengaged. Additional supports are therefore needed to promote participation.
  Given the advance of technology, the chatbot application is an emerging eHealth approach to assist people in promoting health. The most significant advantage of such technology is its ability to offer real-time feedback, facilitate access to information, promote intervention adherence, improve doctor-patient communication, and support intervention in remote areas.
  This research's primary purpose was to develop and evaluate a chatbot prototype, which was designed to serve as a companion to the AOI intervention. The specific objectives include (1) understanding the activity needs of people with serious mental illness who lives in Edmonton, Canada; (2) developing a chatbot prototype that supports people with serious mental illness to actively participate in activities; and (3) examining the usability and technology acceptance of the proposed chatbot.
  This research is comprised of four phases. In Phase One, a needs assessment was conducted to understand people with serious mental illness’ activity participation and collect activities that fit the cultural- and physical- environment in Edmonton. An activity bank, serving as the chatbot database, was created based on the results of the needs assessment. In Phase Two, the chatbot’s design structure and principles were outlined to ensure the chatbot features were consistent with the intervention goal. In Phase Three, interviews with three service providers and usability tests (followed by a questionnaire and interview) with nine people with lived experience were conducted to explore user experiences and perceptions of using the chatbot. In Phase Four, the identified issues and suggestions identified were prioritized and addressed.
  The findings of Phase One indicate that participants overall considered productive, social, community, and physical activities to be important but that they insufficiently participated in them. The analysis also demonstrates how people of different ages and gender prioritized their valued activities. Also, activities that people commonly participate in in the local context were listed. By integrating the findings, an activity bank consisting of 60 activity ideas was developed. In Phase Two, a chatbot prototype was built on Facebook Messenger. The chatbot offered users with activity ideas of categories include physical activity, social interaction, access to the community, activity ideas based on age and gender, and random activity ideas. The results of Phase Three indicate the potential of applying the chatbot application in time use intervention for people with serious mental illness. No major usability issue was found, and the results of the questionnaire show high technology acceptance (average 6.34 out of 7 points). Several themes that discuss the strengths, limitations, and other suggestions emerged from the interviews. In Phase Four, 18 issues/suggestions were prioritized based on their impacts on intervention quality and the effort required to address them. Finally, eight issues were fixed in the refined chatbot.
  In general, people with serious mental illness lack participation in important activities. Age and gender might have influences on how people value different activities. Also, people with serious mental illness overall held a positive attitude toward using the chatbot to find activities for participating in. Despite some limitations related to the content and functionality, the chatbot was overall considered as useful in inspiring activity ideas for participating, easy-to-use, and empowering people with serious mental illness to make their own activity choices. This research demonstrated the potential of utilizing a chatbot in time-use intervention for people with serious mental illness. Future research is needed to replicate the findings on a larger scale and with a more rigorous research design to validate the effectiveness and feasibility of this innovative intervention approach.

• Subjects / Keywords

  • serious mental illness
  • chatbot
  • activity participation

• Graduation date

  Fall 2021

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/r3-qjcw-3295

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.