Can readers understand findings from systematic reviews almost as well from artificial intelligence-assisted visual abstracts as from those published by The BMJ?
Systematic reviews bring together many research studies to answer important health questions. While they provide strong evidence, they are often long and full of technical terms. Visual summaries use simple graphics and brief text to show the main findings clearly, but making them by hand takes a lot of time. Artificial intelligence (AI) could help speed up this process, but computers can sometimes make mistakes or misinterpret medical data. In this study, researchers use AI to draft summaries, which are then carefully checked and corrected by human scientists. Volunteers with an interest or background in healthcare will view either the AI-assisted summaries or traditional summaries published by The BMJ. They will then answer brief questions to measure their understanding. The goal is to see whether AI-assisted designs can share complex health science clearly without losing accuracy.
At a glance
What is this study about?
Every day, thousands of medical studies are published worldwide. To make sense of this information, researchers create 'systematic reviews'. These are large reports that pool together all high-quality evidence on a single topic, such as whether a new treatment works. However, these documents are often dozens of pages long and packed with scientific jargon, making them difficult for busy doctors, policymakers, and patient advocates to digest quickly.
To solve this, journals often publish 'visual abstracts'—graphic summaries that use icons, charts, and plain bullet points to explain the main conclusions at a glance. The British Medical Journal (The BMJ) is well known for producing high-quality visual summaries. However, designing them requires significant graphic design skill and time, meaning many important studies never receive one.
Artificial intelligence (AI) tools could help make these visual summaries much faster. The concern is that AI might misread the data, exaggerate benefits, or miss key safety warnings. To prevent this, the research team developed a system where AI creates an initial draft following strict rules, and two experienced health researchers then review and correct every detail against the original paper.
This study compares these carefully checked, AI-assisted summaries directly against standard visual summaries created by The BMJ. By testing both with readers, the researchers want to find out whether the AI-assisted versions are just as easy to understand, clear, and trustworthy.
Key takeaways
- Tests whether AI can help create clear, accurate visual summaries of complex medical research.
- Compares AI-assisted visual summaries against standard graphics published by The BMJ.
- All AI drafts are carefully checked and corrected by human health researchers before use.
- Entirely online, completed in a single remote session with no clinic visits.
- Open to healthcare professionals, researchers, and active patient or carer research partners.
Who may be eligible?
This study is looking for adults aged 18 and older who have an active background or strong interest in healthcare, medical research, or health policy.
You may be eligible if you work in healthcare (such as a doctor, nurse, or allied health professional), conduct scientific research, help write clinical guidelines, or work in public health. You may also join if you are an active patient partner, carer representative, or peer researcher regularly involved in health projects.
To take part, you must be able to read and understand English fluently, have access to an internet-connected device (such as a computer, tablet, or smartphone), and agree to give online consent. You cannot take part if you were involved in creating, selecting, or testing the study materials beforehand.
Could this study suit you?
Answer these quick questions to see if you may be eligible. This is a guide only — the research team makes the final call.
- I am aged 18 or older.
- I work in healthcare, health research, policy, or I am an active patient/carer research partner.
- I can read and understand English fluently.
- I have a computer, tablet, or smartphone with a reliable internet connection.
- I was not involved in designing or pre-testing the visual summaries for this trial.
What does participation involve?
Taking part in this study is completely online and does not require any clinic visits, blood tests, or medications. You will complete the entire study in a single session using your own computer, tablet, or smartphone.
When you start, a computer programme will randomly place you into one of two groups. One group will view three visual summaries drafted by AI and checked by researchers. The other group will view three standard visual summaries published by The BMJ. Both groups will look at summaries of the exact same three medical reviews.
After viewing each image, you will answer six short questions about the findings to check how well the information came across. You will also be asked to rate how clear and trustworthy you found the summary. The whole session is completed remotely in one sitting.
Potential risks and benefits
Locations (2)
- Online recruitment through ProlificUnverifiedNA – online study, England
- International online recruitment through ProlificUnverifiedNA – online study, Ireland
Common questions
Do I have to visit a hospital or clinic?
No. The entire study takes place online, meaning you can complete it from home on your computer, tablet, or phone.
Does this trial test any medications or treatments?
No. This is a communication study looking at how information is presented. It does not involve any medicines, physical therapies, or medical procedures.
Can patients or carers participate?
Yes, provided you are an active patient partner, carer representative, or peer researcher involved in healthcare initiatives or research projects.
Is the AI working on its own?
No. The AI only creates a first draft based on specific instructions. Two experienced human researchers carefully review and correct every draft before anyone sees it.
How long will the study take?
The study consists of a single online session where you review three graphic summaries and answer short questions about each one.
How to find out more
Xuenan Pang
Always speak to your GP or specialist before deciding to take part in a study.
Interested in taking part?
Discussion
Community discussion
Powered by our forum at community.patient.info. Please be respectful — this is not medical advice.