Mobile app-based chatbot to deliver cognitive behavioral therapy and psychoeducation for adults with attention deficit: A development and feasibility/usability study

Abstract

Background: Attention deficit is a growing problem in adults, and early diagnosis and treatment are needed. Previous studies have shown that cognitive behavioral therapy (CBT) is effective in improving attention deficit symptoms. However, many patients are not receiving adequate treatment due to time, space, and cost constraints. Recently, in other mental illnesses, mobile-based chatbots delivering CBT and psychoeducation have been used for symptom mitigation and treatment.

Objective: This study aimed to investigate the feasibility and usability of a short-term intervention, specifically a mobile-based interactive chatbot application, in alleviating attention deficit symptoms.

Methods: This was a randomized, non-blind parallel-group pilot study conducted from September 2019 to March 2020. Forty-six individuals with attention deficit aged 19-60 were randomly allocated to the chatbot (n = 23) and information-only control groups (n = 23) for 4 weeks. The former group was instructed to use the chatbot application "Todaki," while the latter group was provided with a book on managing attention deficit symptoms. Participants were administered questionnaires to assess their symptoms of attention deficit, depression, and anxiety and evaluated at baseline and 4 weeks after the intervention. The post-intervention survey assessed the chatbot's usability, acceptability, and side effects.

Results: The average age of the participants was 25.1 years (standard deviation [SD] 7.5 years), and 56.5 % (26/46) participants were female. Intention-to-treat analysis (chatbot, n = 23; control, n = 23) revealed a significant reduction of attention deficit symptoms only in the chatbot group, which is represented by group-by-time interaction in Conner's Adult ADHD Rating Scale subscales of Diagnostic and Statistical Manual-IV Attention-Deficit/Hyperactivity Disorder (ADHD) Hyperactive-Impulsive symptoms (F = 4.39; p = .04) and ADHD symptoms total (F = 6.74, p = .01). Further, the results of the paired t-test were significant only in the chatbot group. The average number of times the chatbots were used in 4 weeks was 20.32 (SD 12.89). The total average usage time was 1 h 15 min (SD 1 h 20 min). The degree of improvement in the ADHD symptoms total score was correlated with the number of times the psychoeducation program was used. According to the participants, the empathic/friendly character and unnatural flow of conversation were the best and worst features of the chatbot, respectively.

Conclusions: This study identified the feasibility and usability of using the mobile-based chatbot to improve attention deficit and its associated psychiatric symptoms. Using this novel intervention to conduct CBT would provide a useful digital therapeutic tool that allows easy accessibility and self-guided management for people with attention deficit, which should be verified through the large scale randomized controlled trial.