Health Care Professionals' Perspectives on Conversational Mental Health Chatbots: Protocol for a Systematic Review

Mental health disorders (MHDs) represent a growing global challenge and pose a significant risk to public health. Alongside developments in the field of large language models (LLMs), conversational mental health chatbots (CMHBs) have emerged and are increasingly being used by individuals in self-directed and independent ways for mental health support. Although users' perspectives on CMHBs have been extensively examined and systematically synthesized, relatively little research has focused on how health care professionals (HCPs) perceive these tools. To develop a more comprehensive understanding of the implications of using CMHBs, the perspectives of HCPs should also be considered. As HCPs' views are informed by their clinical expertise and professional responsibility, they may point to underexplored implications related to the safety, ethical use, and implementation of these tools.

This paper presents the protocol for a systematic review that aims to identify, synthesize, and critically appraise the available evidence on HCPs' perspectives regarding the use of CMHBs for mental health support, including perceptions of their therapeutic role, trustworthiness, safety, risks and benefits, ethical concerns, and implementation barriers and facilitators. Where relevant, the completed review will discuss its findings in relation to the existing literature on user perspectives to contextualize possible areas of convergence and divergence.

A systematic review of the literature will be conducted in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020 guidelines. Peer-reviewed qualitative, quantitative, and mixed methods studies will be identified through searches of PubMed (MEDLINE), PsycInfo, Embase, CINAHL, Web of Science, and Scopus, with no restrictions on publication date. Study screening will be supported by AI-assisted active learning using ASReview, following the SAFE stopping procedure, with independent quality-assurance (QA) screening by a second reviewer. Data will be synthesized using a convergent integrated mixed methods approach, and the findings will be reported narratively. Methodological quality will be appraised using the Mixed Methods Appraisal Tool (MMAT; version 2018).

The search for this review was conducted and completed in late November 2025 and identified 24,905 records before deduplication and 18,535 records after deduplication. The initial title and abstract screening began in January 2026 and is ongoing. Data extraction is expected to be completed by August 2026.

This protocol outlines a systematic review that will synthesize the available empirical evidence on HCPs' perspectives on the use of CMHBs for mental health support. The completed review aims to identify aspects such as perceived benefits, barriers and facilitators, and ethical concerns. By integrating qualitative, quantitative, and mixed methods evidence, the review will contribute to a more comprehensive understanding of the implementation and broader implications of CMHBs in mental health care.