Exploring the Credibility of Large Language Models for Mental Health Support: Protocol for a Scoping Review
Bibliographic record
Abstract
BACKGROUND: The rapid evolution of large language models (LLMs), such as Bidirectional Encoder Representations from Transformers (BERT; Google) and GPT (OpenAI), has introduced significant advancements in natural language processing. These models are increasingly integrated into various applications, including mental health support. However, the credibility of LLMs in providing reliable and explainable mental health information and support remains underexplored. OBJECTIVE: This scoping review systematically maps the factors influencing the credibility of LLMs in mental health support, including reliability, explainability, and ethical considerations. The review is expected to offer critical insights for practitioners, researchers, and policy makers, guiding future research and policy development. These findings will contribute to the responsible integration of LLMs into mental health care, with a focus on maintaining ethical standards and user trust. METHODS: This review follows PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines and the Joanna Briggs Institute (JBI) methodology. Eligibility criteria include studies that apply transformer-based generative language models in mental health support, such as BERT and GPT. Sources include PsycINFO, MEDLINE via PubMed, Web of Science, IEEE Xplore, and ACM Digital Library. A systematic search of studies from 2019 onward will be conducted and updated until October 2024. Data will be synthesized qualitatively. The Population, Concept, and Context framework will guide the inclusion criteria. Two independent reviewers will screen and extract data, resolving discrepancies through discussion. Data will be synthesized and presented descriptively. RESULTS: As of September 2024, this study is currently in progress, with the systematic search completed and the screening phase ongoing. We expect to complete data extraction by early November 2024 and synthesis by late November 2024. CONCLUSIONS: This scoping review will map the current evidence on the credibility of LLMs in mental health support. It will identify factors influencing the reliability, explainability, and ethical considerations of these models, providing insights for practitioners, researchers, policy makers, and users. These findings will fill a critical gap in the literature and inform future research, practice, and policy development, ensuring the responsible integration of LLMs in mental health services. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/62865.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.156 | 0.222 |
| Meta-epidemiology (narrow) | 0.007 | 0.006 |
| Meta-epidemiology (broad) | 0.017 | 0.021 |
| Bibliometrics | 0.024 | 0.020 |
| Science and technology studies | 0.006 | 0.007 |
| Scholarly communication | 0.010 | 0.012 |
| Open science | 0.006 | 0.009 |
| Research integrity | 0.011 | 0.009 |
| Insufficient payload (model declined to judge) | 0.071 | 0.013 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".