Mental health chatbot for young adults with depression symptoms during the COVID-19 pandemic: a single-blind, three-arm, randomized controlled trial (Preprint)

BACKGROUND Depression has a high detection ratio among young adults, especially during the COVID-19 pandemic, which, however, occupies a low utilization of health services. Mental health chatbot is a novel digital technology to provide fully automated intervention for depression symptoms. OBJECTIVE The purpose of this study is to test the clinical effectiveness and nonclinical performance of the cognitive behavioral therapy (CBT)-based mental health chatbot (XiaoE) for young adults with depression symptoms during the COVID-19 pandemic. METHODS In a single-blind, three-arm, randomized controlled trial, participants manifesting depression symptoms aged 17-34 years recruited from a university in China were randomly assigned to mental health chatbot (XiaoE; n = 49), e-book (n = 49) or general chatbot (Xiaoai; n = 50), in a ratio of 1:1:1. The primary outcome was the reduction of depression symptoms according to the 9-item Patient Health Questionnaire (PHQ-9) at 1 week later (T1) and 1 month later (T2). Both intention-to-treat analyses and per-protocol analyses were conducted under analysis of covariance (ANCOVA) models adjusting for baseline data. Controlled multiple imputation and δ-based sensitivity analysis were performed for missing data. The secondary outcome was the level of working alliance measured using Working Alliance Questionnaire (WAQ), usability measured using the Usability Metric for User Experience-LITE (UMUX-LITE) and acceptability measured using Acceptability Scale (AS). RESULTS Intent-to-treat analysis revealed a moderate short-term effect of group diversity on the reduction of depression symptoms (PHQ-9) at T1 (F2, 136 = 17.011, P < .001, d = 0.51), while a light long-term effect at T2 (F2, 136 = 5.477, P= .005, d = 0.31). Better working alliance (WAQ, F2, 145 = 3.407, P = .036) and acceptability (AS, F2, 145 = 4.322, P = .015) was discovered with XiaoE, while no significant difference among arms was found on usability (UMUX-LITE, F2, 145 = 0.968, P = .382). CONCLUSIONS This novel intervention conducting CBT provides a feasible and engaging digital therapeutic that allows easy accessibility and self-guided mental health assistance for young adults with depression symptoms. A systematic evaluation of nonclinical metrics for mental health chatbot is established in this study. In the future, concern with both clinical outcomes and nonclinical metrics is necessary to explore the mechanism by which the mental health chatbots work on patients. Further evidence is required to confirm the long-term effectiveness via trails replicated with a longer dose as well as exploration on its stronger efficacy in comparison with other active controls. CLINICALTRIAL Chinese Clinical Trial Registry ChiCTR2100052532; http://www.chictr.org.cn/showproj.aspx?proj=135744.