• Posted March 29, 2024

Chat GPT Can Produce Medical Records Ten Times Faster Than Doctors

Artificial intelligence programs could be an effective way to relieve the paperwork burden that keeps doctors from seeing more patients, a new study finds.

The AI program Chat GPT can write administrative medical notes up to ten times faster than doctors without compromising the quality of the reports, Swiss researchers report.

Human orthopedic doctors and Chat GPT separately prepared discharge papers and notes for six fake patient cases that closely mimicked real cases, researchers said.

An expert panel of 15 people unaware of the source of the documents couldn't tell the difference between them, results show.

What's more, the AI program ground out the paperwork in a tenth of the time it took doctors to do the same.

“The results show that ChatGPT-4 and human-generated notes are comparable in quality overall, but ChatGPT-4 produced discharge documents ten times faster than the doctors,” said researcher Dr. Cyrus Brodén, an orthopedic physician at Uppsala University Hospital in Basel.

The new study was published recently in the journal ActaOrthopaedica.

Researchers plan to follow up this pilot study with a larger test of AI involving 1,000 authentic patient medical records.

“This will be an interesting and resource-intensive project involving many partners,” Broden said. “We are already working actively to fulfill all data management and confidentiality requirements to get the study under way.”

Paperwork has become a bane for doctors, eating away time that could be spent seeing patients and increasing doctors' stress levels, Broden said.

“For years, the debate has centered on how to improve the efficiency of healthcare,” Broden said. “Thanks to advances in generative AI and language modeling, there are now opportunities to reduce the administrative burden on healthcare professionals. This will allow doctors to spend more time with their patients.”

More information

The American Medical Association has more about AI in healthcare.

SOURCE: Uppsala University, news release, March 26, 2024