A chatbot to improve adherence to internet-based cognitive-behavioural therapy among workers with subthreshold depression: a randomised controlled trial

Background Internet-based cognitive-behavioural therapy (iCBT) is effective for subthreshold depression, but adherence is a challenge, particularly without human guidance. Objective To examine whether an automated chatbot can improve the adherence rate of an iCBT programme. Methods A chatbot was developed to support users of an existing iCBT programme. In an open-label randomised controlled trial, full-time employees in Japan with subthreshold depression were allocated to iCBT plus chatbot or iCBT alone (n=149; mean age=41.4, SD=11.1). The primary endpoint was the completion rate of the iCBT programme at 8 weeks. Findings Primary outcome analysis included 142 participants. Completion was 34.8% (24/69; 95% CI 23.5 to 46.0) in the iCBT+chatbot group and 19.2% (14/73; 95% CI 10.2 to 28.2) in the iCBT group; risk ratio=1.81 (95% CI 1.02 to 3.21). Conclusions Adding a chatbot increased iCBT completion rates. Clinical implications Encouraging messages delivered by a chatbot may improve engagement with iCBT. Further studies should assess long-term adherence and impacts on depression, anxiety and well-being.