#Chatbot #ED #SingleSessionIntervention
#eHealth #EatingDisorder
Eating disorders (ED) are severe psychiatric
conditions characterized by a persistent disturbance in eating behavior, weight
regulation, and body image. They are associated with
major physical
complications and a
profound psychosocial impact, affecting quality of life,
social relationships, and mental health. The longer treatment is delayed, the
worse the prognosis becomes. Yet, quick access to care remains a major
challenge.
Among the main current limitations are the overload of mental health services,
a shortage of trained professionals, and waiting times that can extend for
several months. This wait prolongs suffering, reinforces symptoms, and
undermines motivation to engage in structured therapy. Furthermore, the
available care is not always accessible, particularly outside major cities or
for young, vulnerable, or minority populations.
In this context, the major challenge is to offer effective, accessible, and
engaging transitional interventions that can help prevent the disorder from
worsening during the wait. Single-Session Interventions (SSIs) have emerged as
a promising approach, especially when delivered digitally via chatbots. These
digital tools, based on validated protocols, can provide an initial
psychoeducational response and help foster subsequent therapeutic engagement.
This study was initiated to evaluate the clinical effectiveness of a chatbot,
based on the CBT-E (enhanced cognitive behavioral therapy) approach, in
reducing ED symptoms and facilitating access to care for patients on treatment
waitlists.
Can a chatbot really ease eating disorders?
This study included 60 participants aged 16 and older, all on a waitlist for
outpatient eating disorder treatment. Participants were randomly assigned to
two groups:
- A group using an interactive chatbot named ED
ESSI, lasting around 30 minutes and based on principles of enhanced cognitive
behavioral therapy (CBT-E);
- A group receiving standard digital
psychoeducational materials with the same content but without interactivity or
personalization.
The intervention, delivered in a single remote session via a digital platform,
aimed to compare the chatbot’s effects with those of traditional
psychoeducation. The primary outcome was the self-reported severity of eating
disorder symptoms. Secondary indicators included psychosocial distress, levels
of anxiety and depression, motivation for change, and the actual rate of
treatment engagement three months after the intervention.
The results show that the chatbot group experienced a significant reduction in
ED symptoms within the first month, with effects maintained at 3 months.
Depressive symptoms, psychosocial distress, and anxiety also significantly
decreased, with small to moderate effect sizes. Although food cravings did not
significantly decline, there was a marked increase in confidence in one’s
ability to change immediately after using the chatbot, suggesting a strong
motivational effect.
Finally, at 3 months, 93% of the chatbot group had entered
a treatment pathway, compared to 70% in the control group, despite similar
access to healthcare services. These findings highlight the chatbot’s
effectiveness in facilitating treatment initiation, in addition to reducing
psychological symptoms.
A digital ally to activate care
Eating disorders are complex conditions, involving psychological suffering,
physical disturbances, and impaired social functioning. Given the saturation of
specialized services and prolonged waiting times, many individuals are left
without therapeutic support at a critical moment when they are seeking help.
One of the main challenges is to provide effective, accessible, and engaging
transitional solutions that can prevent symptom worsening and prepare patients
for treatment. In this context, this study was designed to assess the effectiveness of a
chatbot, ED ESSI, delivering a one-time CBT-E-based intervention to individuals
awaiting treatment.
The results show that ED ESSI is well tolerated, easy to use, and clinically
beneficial. While it does not replace long-term psychotherapy, it can
significantly reduce ED symptoms, alleviate psychological distress, and
accelerate entry into care. It therefore serves as a relevant priming tool,
particularly useful during therapeutic waiting periods.
Future research is needed to test the chatbot’s effectiveness in more diverse
populations, including different age groups, genders, and diagnoses. Long-term
follow-up beyond three months would help assess the durability of observed
effects. It would also be valuable to explore integrating this intervention
into blended approaches, such as CBT or neuromodulation. Finally, the
development of more advanced versions of the chatbot, based on more
interactive, secure, and personalized conversational models, could further
enhance its clinical impact and accessibility.
Read next: Binge eating under high tension
