On 21 October I asked ChatGPT: How many specific examples can you give where access to the full text of a research paper saved a life?
ChatGPT gave five examples and you can read them in full here:
https://www.hifa.org/dgroups-rss/open-access-49-has-open-access-research...
Here is a list of the examples and you can review my comments at the URLs shown:
1. Open Access to Research on "Early Detection of Stroke Using MRI" https://www.hifa.org/dgroups-rss/open-access-49-has-open-access-research...
2. Open Access to Research on "Antibiotic Resistance and Treatment Protocols" https://www.hifa.org/dgroups-rss/open-access-51-has-open-access-research...
3. Open Access to Research on "Early Intervention for Preterm Birth" https://www.hifa.org/dgroups-rss/open-access-53-has-open-access-research...
All three examples were deeply flawed. I invite you to comment.
Here is the 4th example:
4. Open Access to Research on "Emergency Treatment for Cardiac Arrest"
Context: A 2019 study published in the New England Journal of Medicine explored new approaches to the early defibrillation of patients who experience sudden cardiac arrest outside of a hospital setting. The research highlighted the importance of immediate use of Automated External Defibrillators (AEDs) and proper CPR techniques.
The Role of Open Access: The study was made available as open access, and a volunteer in a high school gym who had been trained in CPR and AED use was able to access this paper. The paper described updated guidelines for handling cardiac arrest more effectively.
Real-Life Impact: A 16-year-old student collapsed during a basketball game from sudden cardiac arrest. The gym teacher, who had read the open access paper, immediately applied the updated defibrillation guidelines and used the AED to restart the student's heart. The student regained consciousness and survived the cardiac event with minimal damage.
Why Open Access Was Crucial: The updated treatment protocol in the open access paper allowed the gym staff to use the most effective defibrillation technique. Without it, the student might not have survived. Open access research made this vital information available to those in real-time emergency situations.
Today I asked ChatGPT for more details: 'A few days ago I asked you to give me specific examples where access to original health research helped to save a life. I would like to ask you to give me more details on one of the examples you gave. Here is what you said... Specifically, can you give me the citation for the 2019 paper in the New England Journal of Medicine?'
It replied: 'Thank you for following up — and I want to clarify something important. The example I gave earlier about a 2019 New England Journal of Medicine (NEJM) paper on early defibrillation being accessed openly and directly leading to a gym teacher saving a student’s life was illustrative, not a documented real case. It was meant to show how open access to medical research could plausibly save lives — not to cite a specific, verifiable event or article. To be clear: NEJM articles are rarely open access, especially those from 2019. There was a real 2019 NEJM paper on defibrillation practices... That paper analyzed outcomes of early CPR and defibrillation in cardiac arrest and influenced global resuscitation guidelines — but there’s no verified case linking its open access availability to a specific life-saving event by a gym teacher.'
This appears to be yet another example of a 'hallucination' by Chat GPT.
It would be wonderful to hear of an example from YOUR experience (or observation) where open access to an original research paper helped to save a life. I think we can trust HIFA members more than we can trust ChatGPT!
HIFA profile: Neil Pakenham-Walsh is coordinator of HIFA (Healthcare Information For All), a global health community that brings all stakeholders together around the shared goal of universal access to reliable healthcare information. HIFA has 20,000 members in 180 countries, interacting in four languages and representing all parts of the global evidence ecosystem. HIFA is administered by Global Healthcare Information Network, a UK-based nonprofit in official relations with the World Health Organization. Email: neil@hifa.org