X.5.PK.2. AI for health services research: building skills for the responsible use of large language models
Abstract
Large language models (LLMs) are rapidly transforming research workflows, yet their application in health services research remains largely unexplored. While these tools offer considerable potential for tasks such as document analysis, literature reviews, and mixed-methods research, most public health researchers lack the knowledge or skills to use them in a structured way. Critical questions around data protection, compliance, and ethics processes must also be addressed before LLMs can be responsibly integrated into health services research. This seminar provides participants with examples of where and how LLMs can be applied in health services research, and to navigate the regulatory, ethical landscape and transparency considerations involved in their responsible use. The workshop is designed for a broad audience, including researchers, practitioners, and policymakers, regardless of prior experience with LLMs.
Objectives: (1) introduce the landscape of LLM capabilities relevant to health services research, including tasks they can support and their current limitations; (2) examine the regulatory (e.g. GDPR), ethical and transparency considerations of integrating LLMs into health services research (3) facilitate discussion about the utility of LLMs in health services research and what’s next.
Format: This Pecha Kucha workshop will open with three presentations, and close with a facilitated discussion:1) LLM-assisted literature reviews: processes and tools to support LLM-supported screening, extraction and synthesis for literature reviews.2) LLMs for qualitative and textual analysis: practical examples and tips of using LLMs to analyse interview transcripts and other unstructured text data, illustrating what works well and where they introduce risk, and how their performance can be evaluated. 3) Governance and reporting: the regulatory and ethical scaffolding for responsible use, covering data protection (GDPR), the EU AI Act (Regulation 2024/1689), and transparent reporting of LLM use in research methods. We will then have ∼15 minutes for open discussion, questions and thoughts around what’s next for LLMs. Participants will leave with a clearer understanding of how LLMs can support health services research workflows and of the regulatory and ethical considerations involved in their use.
Key messages: • Public health researchers need practical skills to use LLMs responsibly; this workshop provides examples of LLMs for realistic health services research and opens a discussion of their responsible use.
• Integrating LLMs into health services research demands not just technical skills but collective critical debate on transparency, ethics, and trustworthiness of outputs.
Speakers/Panellists: Saoirse Moriarty
EUPHANxt
Stefan Buttigieg
Malta Association of Public Health Medicine, St. Paul’s Bay, Malta
Patrícia Pita Ferreira
University Medical Center Groningen, Groningen, Netherlands
Related
- Responsible use of large language models in digital health: an equity-first governance framework
- Enabling Responsible Industry–Academic Collaboration in Mental Health Research and the Involvement of People with Lived Experience
- Challenges of Responsible Research and Open science in Health Emergencies and Disasters
- Building recovery capital: co-creating an assets-based community engagement toolkit for people accessing drug and alcohol treatment services
- Building better citation practices in public health research: Practical considerations for student researchers, from source discovery to crafting and positioning claims
BibTeX
@article{euphanxt2026x5pk2,
title = {{X.5.PK.2. AI for health services research: building skills for the responsible use of large language models}},
author = {EUPHAnxt, EUPHA Digital Health and AI section, EUPHA Health Services Research section, ELSA-NN and Saoirse Moriarty and Laura Maaß},
journal = {The European Journal of Public Health},
year = {2026},
month = oct,
doi = {10.1093/eurpub/ckag169.696},
url = {https://doi.org/10.1093/eurpub/ckag169.696},
note = {CC BY-NC 4.0},
}
RIS
TY - JOUR
TI - X.5.PK.2. AI for health services research: building skills for the responsible use of large language models
AU - EUPHAnxt, EUPHA Digital Health and AI section, EUPHA Health Services Research section, ELSA-NN
AU - Saoirse Moriarty
AU - Laura Maaß
PY - 2026
DA - 2026/10/01
JO - The European Journal of Public Health
DO - 10.1093/eurpub/ckag169.696
UR - https://doi.org/10.1093/eurpub/ckag169.696
AB - Large language models (LLMs) are rapidly transforming research workflows, yet their application in health services research remains largely unexplored. While these tools offer considerable potential for tasks such as document analysis, literature reviews, and mixed-methods research, most public health researchers lack the knowledge or skills to use them in a structured way. Critical questions around data protection, compliance, and ethics processes must also be addressed before LLMs can be responsibly integrated into health services research. This seminar provides participants with examples of where and how LLMs can be applied in health services research, and to navigate the regulatory, ethical landscape and transparency considerations involved in their responsible use. The workshop is designed for a broad audience, including researchers, practitioners, and policymakers, regardless of prior experience with LLMs. Objectives: (1) introduce the landscape of LLM capabilities relevant to health services research, including tasks they can support and their current limitations; (2) examine the regulatory (e.g. GDPR), ethical and transparency considerations of integrating LLMs into health services research (3) facilitate discussion about the utility of LLMs in health services research and what’s next. Format: This Pecha Kucha workshop will open with three presentations, and close with a facilitated discussion:1) LLM-assisted literature reviews: processes and tools to support LLM-supported screening, extraction and synthesis for literature reviews.2) LLMs for qualitative and textual analysis: practical examples and tips of using LLMs to analyse interview transcripts and other unstructured text data, illustrating what works well and where they introduce risk, and how their performance can be evaluated. 3) Governance and reporting: the regulatory and ethical scaffolding for responsible use, covering data protection (GDPR), the EU AI Act (Regulation 2024/1689), and transparent reporting of LLM use in research methods. We will then have ∼15 minutes for open discussion, questions and thoughts around what’s next for LLMs. Participants will leave with a clearer understanding of how LLMs can support health services research workflows and of the regulatory and ethical considerations involved in their use. Key messages: • Public health researchers need practical skills to use LLMs responsibly; this workshop provides examples of LLMs for realistic health services research and opens a discussion of their responsible use. • Integrating LLMs into health services research demands not just technical skills but collective critical debate on transparency, ethics, and trustworthiness of outputs. Speakers/Panellists: Saoirse Moriarty EUPHANxt Stefan Buttigieg Malta Association of Public Health Medicine, St. Paul’s Bay, Malta Patrícia Pita Ferreira University Medical Center Groningen, Groningen, Netherlands
SN - 1101-1262
SN - 1464-360X
ER -