Efficacy of Artificial Intelligence-Assisted Psychotherapy in Patients With Anxiety Disorders: A Prospective, National Multicenter Randomized Controlled Trial Protocol

Abstract

BackgroundAnxiety disorders have the highest prevalence of all psychiatric disorders in China. Medication and psychotherapy are two main treatment approaches for this group of disorders, and when used in combinations are significantly more beneficial than medication alone. The resources are insufficient. The availability of psychotherapy is low due to the limited resources. Artificial intelligence (AI)-assisted psychotherapy offers an opportunity to develop an efficient and standardized psychotherapy model and improve the availability of psychotherapy, which is key to improve the clinical efficacy of anxiety disorder treatments.ObjectivesThe present protocol aims to determine whether medication plus AI-assisted psychotherapy has greater efficacy than medication alone in the treatment of anxiety disorders.MethodsWe will recruit patients in eight hospitals in China. Seven hundred and eight patients with anxiety disorders will be randomly allocated on a 1:1 basis to either medication plus AI-assisted psychotherapy group, or medication alone group. We have built an AI psychotherapy robot named XIAO AN. In this study we will deliver psychotherapy to patients in the medication plus AI-assisted psychotherapy group. Patients will be assessed at baseline and at the end of week 2, 4, 8, and 12. Follow-up assessments will be conducted at 3 and 6 months posttreatment. The primary outcome is change of Hamilton Anxiety Rating Scale (HAMA) score from baseline the end of 12-week treatment. A secondary efficacy outcome will be improvement in treatment at an early stage (score reduction in HAMA ≥25% after 2 weeks of treatment). Other measurements include Hamilton Depression Scale, Clinical Global Impression, Treatment Emergent Symptom Scale, Social Disability Screening Schedule, Insomnia Severity Index and so on. Scales will be assessed by independent raters who are blind to treatment allocation and analyses will be conducted by a statistician who is also blind to treatment allocation.DiscussionThis will be the first multicentered randomized controlled single-blind trial in China to assess the efficacy of medication plus AI-assisted psychotherapy compared with medication alone for anxiety disorders. The study has the potential to address the limitations of the limited availability of psychotherapy, and to augment the efficacy of the treatment of anxiety disorders in China.