Treatment Uptake Chatbot for Eating Disorders

Developing an Optimized Conversational Agent or "Chatbot" to Facilitate Mental Health Services Use in Individuals With Eating Disorders

Eating disorders (EDs) are serious mental illnesses associated with high morbidity and mortality, clinical impairment, and comorbid psychopathology. Although evidence-based treatments for EDs have been established, the treatment gap is wide. Indeed, <20% of individuals with EDs receive treatment. There is need for a novel solution not only to identify individuals with EDs but also to encourage mental health services use and to address treatment barriers. This study aims to implement a conversational agent or "chatbot" that is optimized to increase mental health services use among individuals with EDs through such features as: 1) education on the seriousness and consequences of EDs; 2) engaging the user in motivational interviewing to overcome barriers to care; 3) providing a personalized recommendation for seeking intervention; 4) repeated check-ins with the user to encourage follow-up with care. This study will utilize a randomized optimization trial with adults who have completed screening on the National Eating Disorders Association (NEDA) website and screen positive for an ED but are not in treatment to determine chatbot feasibility and to generate data on the effect of the chatbot on motivation for treatment and mental health services use. This trial will employ the Multiphase Optimization Strategy framework, using a 2^4 full factorial design, to randomly assign participants to a combination of the four proposed intervention components (n=16 conditions) to isolate the active ingredients.

A discussion with the chatbot, including follow ups over the course of 2 weeks to encourage help seeking for their eating disorder.

A discussion with the chatbot, including identifying a personalized recommendation for accessing help for their eating disorder.

A discussion with the chatbot, including motivational interviewing techniques to help overcome barriers to seeking care for their eating disorder.

A discussion with the chatbot, including motivational interviewing techniques to help overcome barriers to seeking care for their eating disorder, identifying a personalized recommendation for accessing help for their eating disorder, and follow ups over the course of 2 weeks to encourage help seeking for their eating disorder.

A discussion with the chatbot, including motivational interviewing techniques to help overcome barriers to seeking care for their eating disorder and follow ups over the course of 2 weeks to encourage help seeking for their eating disorder.

A discussion with the chatbot, including motivational interviewing techniques to help overcome barriers to seeking care for their eating disorder and identifying a personalized recommendation for accessing help for their eating disorder.

A discussion with the chatbot, including identifying a personalized recommendation for accessing help for their eating disorder and follow ups over the course of 2 weeks to encourage help seeking for their eating disorder.

A discussion with the chatbot, including psychoeducational content to increase knowledge related to eating disorders.

A discussion with the chatbot, including psychoeducational content to increase knowledge related to eating disorders and follow ups over the course of 2 weeks to encourage help seeking for their eating disorder.

A discussion with the chatbot, including psychoeducational content to increase knowledge related to eating disorders and identifying a personalized recommendation for accessing help for their eating disorder.

A discussion with the chatbot, including psychoeducational content to increase knowledge related to eating disorders and motivational interviewing techniques to help overcome barriers to seeking care for their eating disorder.

A discussion with the chatbot, including psychoeducational content to increase knowledge related to eating disorders, motivational interviewing techniques to help overcome barriers to seeking care for their eating disorder, and identifying a personalized recommendation for accessing help for their eating disorder.

A discussion with the chatbot, including psychoeducational content to increase knowledge related to eating disorders, identifying a personalized recommendation for accessing help for their eating disorder, and follow ups over the course of 2 weeks to encourage help seeking for their eating disorder.

A discussion with the chatbot, including psychoeducational content to increase knowledge related to eating disorders, motivational interviewing techniques to help overcome barriers to seeking care for their eating disorder, and follow ups over the course of 2 weeks to encourage help seeking for their eating disorder.

A discussion with the chatbot, including psychoeducational content to increase knowledge related to eating disorders, motivational interviewing techniques to help overcome barriers to seeking care for their eating disorder, identifying a personalized recommendation for accessing help for their eating disorder, and follow ups over the course of 2 weeks to encourage help seeking for their eating disorder.

Rates of Treatment/Services Utilization for Eating, Shape, or Weight Concerns Among All Participants Randomized to the Intervention Following Online Eating Disorder Screening

Treatment utilization was measured at each follow-up time point (2, 6, and 14 weeks), which was assessed via the prompt, "Have you tried or used any type of mental health help (e.g., self-help app, telehealth/in-person counseling, etc.) for your concerns related to your eating, shape, or weight in the past [2, 6, or 14] weeks?". We defined a participant as having utilized treatment at the first time they responded, "yes" (vs. "no") to this question. The data presented below indicate the rates of self-reported receipt of treatment/services utilization among participants assigned (component on) or not assigned (component off) to each of the four chatbot components (motivational interviewing, psychoeducation, personalized recommendations, repeated administration), as delineated by each time point during which the aforementioned item was assesed.

Participants' willingness to seek psychotherapy was assessed via an identical item in the baseline assessment (i.e., before the initial chatbot assessment) as well as each off the follow-up assessments delivered at 2, 6, and 14 weeks post-intervention. The item prompted users to rate their willingness to seek in-person or telehealth psychotherapy in subsequent weeks. Responses were rated on a Likert scale from 1 (not at all important/ready) to 7 (very important/ready). Mean values reported below indicate the average Likert scale willingness rating among all participants at each time point who did (component turned on) and did not (component turned off). Statistical analyses outline the time and component effects on changes in mean willingness to seek pyschotherapy from baseline to 14 weeks post-interventioon.

Mean Attitude Toward Change for Concerns Related to Eating, Shape, or Weight Concerns Among All Participants Randomized to the Intervention Following Online Eating Disorder Screening

Participants' attitudes toward making changes (i.e., motivation toward seeking help) to address their personal eating, shape, and weight concerns were measured via identical items in the baseline assessment, and in each of the 2, 6, and 14 week follow-up assessments. Items included, "How important is it for you to change your eating, shape, or weight concerns?" and, "How ready are you to make changes in your eating, shape, or weight concerns?". Both items were rated on a Likert scale from 1 (not at all important/ready) to 7 (very important/ready), and the sum of the two questions was used as an index of change/motivation attitudes. Values below indicate the average Likert scale attitude rating (from 2 (lowest) to 14 (highest)) among all participants at each time point who did (component turned on) and did not (component turned off) receive each component. Analyses outline the time, component, and time by component interaction effects on changes in motivation attitudes since baseline.

Inclusion Criteria: U.S. resident English-speaking Age 18 years or older Owns a personal smartphone Screens positive for a Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (DSM-5) clinical or subclinical eating disorder Not currently receiving treatment for an eating disorder Exclusion Criteria: Non-U.S. resident Younger than 18 years old Does not own a personal smartphone Does not screen positive for a DSM-5 clinical or subclinical eating disorder Screens positive for Avoidant Restrictive Food Intake Disorder Currently receiving treatment for an eating disorder