Exploring the Potential of Large Language Models for Automated Safety Plan Scoring in Outpatient Mental Health Settings

crisispapercliniciantier 3link ok

Open source

Overview

AI-generated summary, not a substitute for reading the source.

This paper introduces the Safety Planning Intervention Fidelity Rater (SPIFR), an automated tool utilizing large language models to assess the quality of Safety Planning Interventions in outpatient mental health settings. The study evaluates GPT-4, LLaMA 3, and o3-mini on their ability to analyze key components of safety plans using deidentified data from New York. Results indicate that LLaMA 3 and o3-mini outperformed GPT-4, suggesting potential for providing clinicians with timely feedback on intervention fidelity.

ID
3e9b75dcfbda4eea
Source org
PubMed Central
Author
Not recorded
Published
Not recorded
OA status
Not recorded
DOI
10.1101/2025.03.26.25324610
PMID
Not recorded
PMCID
Not recorded
Citations
Not recorded

Tags

Verification

Automated checks only, not clinical endorsement or advice.

Cite this entry

BibTeX

@article{allodium:3e9b75dcfbda4eea,
  title = {Exploring the Potential of Large Language Models for Automated Safety Plan Scoring in Outpatient Mental Health Settings},
  journal = {PubMed Central},
  doi = {10.1101/2025.03.26.25324610},
  url = {https://doi.org/10.1101/2025.03.26.25324610}
}

RIS

TY  - JOUR
TI  - Exploring the Potential of Large Language Models for Automated Safety Plan Scoring in Outpatient Mental Health Settings
PB  - PubMed Central
DO  - 10.1101/2025.03.26.25324610
UR  - https://doi.org/10.1101/2025.03.26.25324610
ER  -

APA

Exploring the Potential of Large Language Models for Automated Safety Plan Scoring in Outpatient Mental Health Settings. (n.d.). PubMed Central. https://doi.org/10.1101/2025.03.26.25324610

Related entries

More like this