Development of a positive body image chatbot (KIT) with young people and parents/carers: Qualitative focus group study

Francesca Beilharz, Suku Sukunesan, Susan L. Rossell, Jayashri Kulkarni, Gemma Sharp

Research output: Contribution to journalArticleResearchpeer-review

Abstract

Background: Body image and eating disorders represent a significant public health concern; however, many affected individuals never access appropriate treatment. Conversational agents or chatbots reflect a unique opportunity to target those affected online by providing psychoeducation and coping skills, thus filling the gap in service provision. Objective: A world-first body image chatbot called “KIT” was designed. The aim of this study was to assess preliminary acceptability and feasibility via the collection of qualitative feedback from young people and parents/carers regarding the content, structure, and design of the chatbot, in accordance with an agile methodology strategy. The chatbot was developed in collaboration with Australia’s national eating disorder support organization, the Butterfly Foundation. Methods: A conversation decision tree was designed that offered psychoeducational information on body image and eating disorders, as well as evidence-based coping strategies. A version of KIT was built as a research prototype to deliver these conversations. Six focus groups were conducted using online semistructured interviews to seek feedback on the KIT prototype. This included four groups of people seeking help for themselves (n=17; age 13-18 years) and two groups of parents/carers (n=8; age 46-57 years). Participants provided feedback on the cartoon chatbot character design, as well as the content, structure, and design of the chatbot webchat. Results: Thematic analyses identified the following three main themes from the six focus groups: (1) chatbot character and design, (2) content presentation, and (3) flow. Overall, the participants provided positive feedback regarding KIT, with both young people and parents/carers generally providing similar reflections. The participants approved of KIT’s character and engagement. Specific suggestions were made regarding the brevity and tone to increase KIT’s interactivity. Conclusions: Focus groups provided overall positive qualitative feedback regarding the content, structure, and design of the body image chatbot. Incorporating the feedback of lived experience from both individuals and parents/carers allowed the refinement of KIT in the development phase as per an iterative agile methodology. Further research is required to evaluate KIT’s efficacy.

Original languageEnglish
Article numbere27807
Number of pages12
JournalJournal of Medical Internet Research
Volume23
Issue number6
DOIs
Publication statusPublished - Jun 2021

Keywords

  • Artificial intelligence
  • Body image
  • Chatbot
  • Conversational agent
  • Design
  • Digital health
  • Eating disorder
  • Mental health

Cite this