Systematic Scoping Review of 29 Studies on LLM Applications in Mental Health

29 articles from a systematic scoping review exploring the application of large language models (LLMs) in mental health. The review, authored by Jinhua Yang and last updated in March 2026, synthesizes findings from eleven academic databases. It identifies key research trends, application boundaries, and ethical considerations for using LLMs in clinical mental health contexts.

Use Cases

• Identify research trends and gaps in LLM applications for mental health based on the synthesized review of 29 studies.
• Analyze the shift towards specialized BERT models for clinical accuracy based on the described domain specialization trend.
• Investigate ethical and practical boundaries for AI in mental health based on the review's conclusions on model value alignment and cultural sensitivity.
• Guide future research on multimodal integration for patient assessment based on the review's identified future priorities.

Strengths

• Follows PRISMA-ScR guidelines for systematic scoping reviews, indicating a structured methodology.
• Synthesizes findings from 29 articles sourced from eleven major academic databases.
• Explicitly addresses critical limitations like cultural sensitivity and model value alignment.

Limitations

• The dataset is a single 18.2 KB document, representing a summary review rather than raw research data.
• Column-level documentation for any underlying data is absent; field semantics must be inferred from the review text.
• The review's conclusions are based on a specific set of 29 articles and may not encompass the entire field.

Provenance

Source

Jinhua Yang via figshare

Collection Method

Systematic scoping review adhering to PRISMA-ScR guidelines, with searches across eleven academic databases.

Time Range

Review covers literature up to its publication date; specific temporal coverage of included studies is not detailed.

Freshness

Last updated 2026-03-18 07:45:40

Geography

Global, but the review notes insufficient cultural sensitivity in non-Western contexts.