{"title":"Patient perspectives on AI-powered cognitive behavioral therapy tools in managing anxiety and stress: a systematic review of qualitative studies.","authors":"Ravi Shankar, Toon Hao Foo, Fiona Devi, Qian Xu","doi":"10.1080/09638237.2025.2595612","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Artificial intelligence (AI)-powered cognitive behavioral therapy (CBT) tools show promise for anxiety and stress management, but patient perspectives remain poorly understood.</p><p><strong>Aims: </strong>To synthesize qualitative evidence on patient experiences, attitudes, and preferences regarding AI-powered CBT tools for anxiety and stress management.</p><p><strong>Methods: </strong>Systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, searching eight databases through March 2025. Qualitative studies exploring patient perspectives on AI-powered CBT interventions were included. Thematic synthesis was employed with GRADE-CERQual confidence assessment.</p><p><strong>Results: </strong>Nine studies (<i>n</i> = 600+ participants) from nine countries were included. Six major themes emerged: (1) Perceived benefits and therapeutic value, including enhanced self-awareness and 24/7 accessibility; (2) Technical limitations and personalization challenges; (3) Trust, privacy, and acceptability concerns; (4) User preferences for conversational interfaces and human-AI integration; (5) Facilitators and barriers to engagement; and (6) Cultural and demographic influences on user experiences. Users consistently viewed AI tools as supplements to human therapy.</p><p><strong>Conclusions: </strong>AI-powered CBT tools provide accessible, non-judgmental support and practical coping strategies. However, technical limitations and personalization gaps remain significant challenges. Future development should enhance conversational capabilities, cultural adaptation, and integration with traditional care. Limitations of this study include English-language restriction, heterogeneity of AI interventions, and limited evidence for specific populations such as older adults.</p>","PeriodicalId":48135,"journal":{"name":"Journal of Mental Health","volume":" ","pages":"1-15"},"PeriodicalIF":3.5000,"publicationDate":"2025-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Mental Health","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1080/09638237.2025.2595612","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"PSYCHIATRY","Score":null,"Total":0}
引用次数: 0
Abstract
Background: Artificial intelligence (AI)-powered cognitive behavioral therapy (CBT) tools show promise for anxiety and stress management, but patient perspectives remain poorly understood.
Aims: To synthesize qualitative evidence on patient experiences, attitudes, and preferences regarding AI-powered CBT tools for anxiety and stress management.
Methods: Systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, searching eight databases through March 2025. Qualitative studies exploring patient perspectives on AI-powered CBT interventions were included. Thematic synthesis was employed with GRADE-CERQual confidence assessment.
Results: Nine studies (n = 600+ participants) from nine countries were included. Six major themes emerged: (1) Perceived benefits and therapeutic value, including enhanced self-awareness and 24/7 accessibility; (2) Technical limitations and personalization challenges; (3) Trust, privacy, and acceptability concerns; (4) User preferences for conversational interfaces and human-AI integration; (5) Facilitators and barriers to engagement; and (6) Cultural and demographic influences on user experiences. Users consistently viewed AI tools as supplements to human therapy.
Conclusions: AI-powered CBT tools provide accessible, non-judgmental support and practical coping strategies. However, technical limitations and personalization gaps remain significant challenges. Future development should enhance conversational capabilities, cultural adaptation, and integration with traditional care. Limitations of this study include English-language restriction, heterogeneity of AI interventions, and limited evidence for specific populations such as older adults.
期刊介绍：
The Journal of Mental Health is an international forum for the latest research in the mental health field. Reaching over 65 countries, the journal reports on the best in evidence-based practice around the world and provides a channel of communication between the many disciplines involved in mental health research and practice. The journal encourages multi-disciplinary research and welcomes contributions that have involved the users of mental health services. The international editorial team are committed to seeking out excellent work from a range of sources and theoretical perspectives. The journal not only reflects current good practice but also aims to influence policy by reporting on innovations that challenge traditional ways of working.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
