Assessing the Quality, Readability, and Comprehensibility of Fibromyalgia and Treatment Information Provided by ChatGPT

Abstract

This study aimed to investigate the quality, readability, and comprehensibility of responses provided by ChatGPT to the most frequently searched topics using the keywords “fibromyalgia” and “fibromyalgia treatment.” Popular keywords for “fibromyalgia” and “fibromyalgia treatment” were obtained using Google Trends and entered into ChatGPT 3.5. The Ensuring Quality Information for Patients (EQIP), Flesch- Kincaid Reading Ease (FKRE), and Flesch-Kincaid Grade Level (FKGL) scores were calculated for the generated responses. These scores were assessed by two experts, and comparisons were made between the fibromyalgia and fibromyalgia treatment keywords. General fibromyalgia information had a mean FKRE score of -2.98 ± 13.53. The mean FKGL score was 19.18 ± 2.36. The mean EQIP score was 41.60 ± 9.18, reflecting high-quality information. Statistically significant differences were observed in EQIP scores among different categories, with the “Drug, Medication, or Product” group scoring lower than other groups. Comparisons between fibromyalgia and fibromyalgia treatment keywords showed that general fibromyalgia information had higher FKRE scores, while treatment-related content had higher FKGL scores, indicating greater complexity. The findings highlight that while general information about fibromyalgia provided by ChatGPT is more readable, texts concerning fibromyalgia treatment are more complex. Bangladesh Journal of Medical Science Vol. 24 No. 01 January’25 Page : 111-118

Ähnliche Arbeiten