ChatGPT: A Pilot Study on a Promising Tool for Mental Health Support in Psychiatric Inpatient Care

Abstract

Introduction This pilot study assesses ChatGPT’s effectiveness as an artificial intelligence (AI) chatbot in psychiatric inpatient care. Global mental health challenges highlight a significant treatment gap, mainly due to restricted service access and mental health professional shortages. AI chatbots like ChatGPT offer innovative solutions, providing services such as self-help advice, coaching, psychoeducation, and emotional support. Methods This study involved a group of patients receiving psychiatric inpatient care. The intervention group engaged in 3-6 ChatGPT sessions under guided prompts, while the control group received standard care. The primary outcome was based on World Health Organization Quality of Life Questionnaire – Brief Version (WHOQOL-BREF) scores, and the secondary outcome assessed patient satisfaction with ChatGPT. Results Twelve patients were included in this study, with a mean age of 27 (standard deviation of 8.57). The intervention group (7 patients) showed notable improvements in WHOQOL-BREF scores compared to the control group (5 groups), and high satisfaction levels with ChatGPT sessions were reported. Discussion These findings suggest that AI chatbots like ChatGPT can enhance patient-reported quality of life in a psychiatric setting, with high user satisfaction. However, limitations include a small sample size and the exclusion of patients with psychosis. Future studies should focus on larger, diverse patient groups for broader validation. These results support the potential of AI chatbots in mental health care, which can provide more accessible and varied treatment options. This study lays the groundwork for further exploration into the role of AI in enhancing psychiatric treatment, advocating for larger-scale investigations to establish more conclusive evidence of their effectiveness and applicability in diverse mental health scenarios.