The HIET-1 trial protocol has been published in The Journal of Clinical Epidemiology: Comparison of AI-assisted and human-generated plain language summaries for Cochrane reviews: protocol for a randomised trial (HIET-1)
Led by ESI team members Prof. Declan Devane, KM Saif-Ur-Rahman, Dr. Nikita Burke PhD and Dr. Tom Conway, this protocol outlines a study exploring how artificial intelligence can support the creation of accessible, plain-language summaries of Cochrane reviews, helping to make health evidence more understandable for all audiences.
Key points:
This protocol compares AI-assisted and human-generated plain language summaries of Cochrane reviews in a randomised trial to establish whether AI assistance yields non-inferior summaries regarding comprehension, readability, quality, safety, and trustworthiness. Plain Language Summaries (PLS) are crucial for making complex health evidence accessible to the general public, but creating them requires significant expertise and resources. Large language models (LLMs) show promise for enhancing health information communication, but their effectiveness in creating accurate, comprehensible, and trustworthy summaries has not been rigorously evaluated in randomised trials. This study will provide empirical evidence to guide the integration of AI technologies in health communication and inform future practices in making health evidence more accessible. Findings will inform future practices in disseminating evidence-based health information to the public.
View the full protocol here.
