Evaluating Online Educational Resources for Osteogenesis Imperfecta: Insights from Google, Bing, and ChatGPT3.5

Poster #: 172
Session/Time: A
Author: Victoria Partin
Mentor: April Pace, D.H.Sc., M.L.S.
Co-Investigator(s): Andrea Hooberman-Pineiro, EVMS MD Program
Research Type: Quality Improvement

Abstract

Introduction: Osteogenesis Imperfecta (OI) is a rare connective tissue disorder resulting in a diverse range of musculoskeletal deformities. Patient education is an important aspect of medical care, and the internet is a popular place in which patients seek information about their medical conditions. This is a cross-sectional study aimed to evaluate the quality of online patient educational resources about OI.

Methods: The authors searched for patient education websites using Google and Bing. Twenty patient education websites about OI were collected from each. Websites meeting inclusion criteria were evaluated using the DISCERN tool, JAMA Benchmark Criteria evaluation, and a Flesch-Kincaid Readability test. T-tests were used to compare search engine results. Websites were also collected from ChatGPT3.5, however, due to hallucinations and exclusion criteria, the sample size was too small to compare to the search engines.

Results: No significant differences were found between search engines, and the quality of the websites was not high. Average overall ratings for DISCERN were mediocre, and for the JAMA criteria they were low. The average readability scores required 8-9 grade level reading comprehension, although there was variability within each search engine.

Conclusion: Neither search engine provided significantly better sources than the other, the overall quality of the websites was not high, and the low readability scores could make these difficult for patients with lower health literacy to understand.