AI and Large language models (5) ChatGPT: Is it a reliable and useful source for common rheumatic diseases?

28 May, 2023

(with thanks to Thomas Krichel)

Dr ChatGPT: Is it a reliable and useful source for common rheumatic diseases?

Cuma Uz, Ebru Umay.

International Journal of Rheumatic Diseases 2023 [restricted access :-(]

AIM: It is inevitable that artificial intelligence applications will be used as a source of information in the field of health in the near future. For this reason, we aimed to evaluate whether ChatGPT, a new Large Language Model, can be used to obtain information about common rheumatic diseases.MATERIALS AND METHODS: Common rheumatic diseases were identified using the American College of Rheumatology and European League against Rheumatism guidelines. Osteoarthritis (OA), rheumatoid arthritis, ankylosing spondylitis (AS), systemic lupus erythematosus, psoriatic arthritis, fibromyalgia syndrome, and gout were identified by using Google trends for the four most frequently searched keywords on Google. The responses were evaluated with seven-point Likert-type reliability and usefulness scales developed by us.

RESULTS: The highest score in terms of reliability was OA (mean ± standard deviation 5.62 ± 1.17), whereas the highest score in terms of usefulness was AS (mean 5.87 ± 0.17). There was no significant difference in the reliability and usefulness of the answers given by the ChatGPT (p = .423 and p = .387, respectively). All scores ranged between 4 and 7.

CONCLUSIONS: Although ChatGPT is reliable and useful for patients to obtain information about rheumatic diseases, it should be kept in mind that it may give false and misleading answers.

DOI: https://doi.org/10.1111/1756-185X.14749

Best wishes, Neil

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