This is the breadcrumbs block. Breadcrumb will populate on actual page.
Can Large Language Models Answer Questions about Spinal Cord Injury? Risks, Challenges, and Opportunities—A Narrative Review
Rahul K. Desai, Simran Saggu, Masha Panahi, Kealey L. Nguyen, Parisa Razavi Yeganeh, Ornell Douglas, Marzieh Mussavi Rizi, Nader Fallah, John Chernesky, Vanessa K. Noonan, Abel Torres-Espín
DOI: https://doi.org/10.1177/2689288X261425900
Introduction
Access to high-quality health information (HI) is critical for everyone involved in the research and management of medical conditions such as spinal cord injury (SCI). Recently, the use of Large Language Models (LLMs) through AI-based chatbots like ChatGPT has become increasingly integral to how people seek and consume HI. While LLMs have been evaluated in various clinical and health domains, there remains a notable gap in the literature regarding their use for SCI-specific questions.
Methods
The authors conducted a narrative synthesis to identify the opportunities, challenges, and risks of using LLMs in SCI-related HI tasks, and to provide future direction for researchers, clinicians, and policymakers seeking to better understand this fast-evolving landscape. The authors searched PubMed, Embase, and Google Scholar up to December 2025 and identified nine primary articles that investigated LLMs in the context of SCI-related queries.
Results
The synthesis found that, although there are promising results, these should be interpreted with caution due to mixed evidence regarding LLMs’ ability to effectively answer SCI-related questions. In addition, LLM outputs were challenging to read, typically requiring an education level equivalent to a college-level student, grades 14–15, to be adequately understood.
Conclusion
The review recognizes that LLMs can serve as valuable tools for accessing HI in SCI. However, LLMs can also pose significant risks, including the spread of mis- or disinformation that may be inaccurate or even dangerous. This could mislead individuals and caregivers, potentially resulting in detrimental health outcomes. Finally, the authors note that methodological rigour needs to be improved to produce higher levels of evidence.
This publication is a collaboration with Praxis’researchers, Dr. Nader Fallah, Dr. Vanessa Noonan and John Chernesky. Work funded by Craig H. Neilsen Foundation.