LLMs vs Consultant-Provided Information for Rotator Cuff Surgery
Research type
Research Study
Full title
Evaluating the Effectiveness and Comprehensibility of Large Langauge Models vs Clinician-generated Patient Information Leaflets for Rotator Cuff Surgery: A Survey-Based Study
IRAS ID
359595
Contact name
Timothy P Davis
Contact email
Sponsor organisation
Lewisham & Greenwich NSH Trust
Duration of Study in the UK
0 years, 2 months, 3 days
Research summary
This study aims to improve the quality of information given to patients undergoing rotator cuff surgery, a common shoulder operation. Patient Information Leaflets (PILs) are often used to explain procedures, but many patients still feel unsure or confused about what to expect. We want to better understand how patients feel about the current materials they are given and how these can be improved to meet their needs.
We are also interested in how well artificial intelligence (AI), such as large language models, can answer common patient questions compared to senior doctors. As more patients turn to the internet or AI tools for health advice, it's important to know whether these sources provide useful and accurate information.
The study will involve two groups: patients who have had rotator cuff surgery in the last two years, and patients who are currently waiting for the surgery. Around 15-20 people from each group will be invited to take part. Participants will be asked to review a small selection of written answers to common questions - some written by a consultant surgeon and some generated by AI and give feedback on how clear, helpful and easy to understand the responses are.
The study will take place at Lewisham and Greenwich NHS Trusts and will use surveys sent by post or email, depending on each patient’s preference. Participation is voluntary and all responses will be anonymous. Consent for participation will be performed at the start of the survey - patients will be able to indicate if they are happy to proceed and for their answers to be used for research purposes.
This research is managed locally within the Trust. It will help identify how to make patient information more understandable, whether AI can support this and how patients feel about the quality of information they receive before and after surgery.
REC name
North West - Greater Manchester West Research Ethics Committee
REC reference
25/NW/0277
Date of REC Opinion
17 Sep 2025
REC opinion
Favourable Opinion