Due to necessary scheduled maintenance, the JMIR Publications website will be unavailable from Wednesday, July 01, 2020 at 8:00 PM to 10:00 PM EST. We apologize in advance for any inconvenience this may cause you.
Who will be affected?
Readers: No access to all 28 journals. We recommend accessing our articles via PubMed Central
Authors: No access to the submission form or your user account.
Reviewers: No access to your user account. Please download manuscripts you are reviewing for offline reading before Wednesday, July 01, 2020 at 7:00 PM.
Editors: No access to your user account to assign reviewers or make decisions.
Copyeditors: No access to user account. Please download manuscripts you are copyediting before Wednesday, July 01, 2020 at 7:00 PM.
Challenges in Designing an Eating Disorders Prevention Chatbot
William W. Chan;
Ellen E. Fitzsimmons-Craft;
Arielle C. Smith;
Marie-Laure Firebaugh;
Lauren A. Fowler;
Bianca DePietro;
Naira Topooco;
Denise E. Wilfley;
C. Barr Taylor;
Nicholas C. Jacobson
ABSTRACT
Background:
Chatbots have the potential to provide mental health intervention programs at scale cost-effectively and increase interactivity, ease of use, and accessibility of intervention programs.
Objective:
Mental health prevention chatbot development is still in its infancy. This paper aims to present the challenges in designing and refining a mental health prevention chatbot program and to report data related to the impact of program modifications on user outcomes.
Methods:
The research team reviewed transcripts including over 150,000 user comments and chatbot responses to identify and modify problematic chatbot responses and technical glitches. To assess the impact of the modifications on program performance, we analyzed changes in recruitment, helpfulness ratings, time of program usage, and body image across time.
Results:
The most common conversational problem of the chatbot was a general limitation in understanding and responding appropriately to non-scripted user responses. A number of modifications were implemented to limit these problems while retaining user interactivity. However, there was no evidence that fixing bugs in the program and reducing inappropriate conversations were associated with within-subject improvement in outcomes.
Conclusions:
Chatbots have the potential to reach large populations at low cost but are limited in understanding and responding appropriately to non-scripted user responses. Workarounds can reduce “conversation errors” while retaining some interactivity but may not improve outcome. Clinical Trial: Open Science Framework (OSF) 7ZMBV; doi.org/10.17605/OSF.IO/7ZMBV
Citation
Please cite as:
Chan WW, Fitzsimmons-Craft EE, Smith AC, Firebaugh ML, Fowler LA, DePietro B, Topooco N, Wilfley DE, Taylor CB, Jacobson NC
The Challenges in Designing a Prevention Chatbot for Eating Disorders: Observational Study