The therapeutic effectiveness of artificial intelligence-based chatbots in alleviation of depressive and anxiety symptoms in short-course treatments: A systematic review and meta-analysis

The emergence of artificial intelligence-based chatbot has revolutionized the field of clinical psychology and psychotherapy, granting individuals unprecedented access to professional assistance, overcoming time constraints and geographical limitations with cost-effective convenience. However, despite its potential, there has been a noticeable gap in the literature regarding their effectiveness in addressing common mental health issues like depression and anxiety. This meta-analysis aims to evaluate the efficacy of AI-based chatbots in treating these conditions. A systematic search was executed across multiple databases, including PubMed, Cochrane Library, Web of Science, PsycINFO, and Embase on April 4th, 2024. The effect size of treatment efficacy was calculated using the standardized mean difference (Hedge's g). Quality assessment measures were implemented to ensure trial's quality. In our analysis of 18 randomized controlled trials involving 3477 participants, we observed noteworthy improvements in depression (g = −0.26, 95 % CI = −0.34, −0.17) and anxiety (g = −0.19, 95 % CI = −0.29, −0.09) symptoms. The most significant benefits were evident after 8 weeks of treatment. However, at the three-month follow-up, no substantial effects were detected for either condition. Several limitations should be considered. These include the lack of diversity in the study populations, variations in chatbot design, and the use of different psychotherapeutic approaches. These factors may limit the generalizability of our findings. This meta-analysis highlights the promising role of AI-based chatbot interventions in alleviating depressive and anxiety symptoms among adults. Our results indicate that these interventions can yield substantial improvements over a relatively brief treatment period.