J Neurosurg Pediatr. 2025 Sep 26:1-6. doi: 10.3171/2025.5.PEDS25122. Online ahead of print.
ABSTRACT
OBJECTIVE: This study investigates the potential of artificial intelligence (AI), specifically ChatGPT 4o, to revolutionize the readability of patient education materials in pediatric neurosurgery. The American Medical Association and the National Institutes of Health recommend that educational materials be written at a 3rd- to 7th-grade reading level for accessibility. However, existing resources often exceed this range, hindering comprehension for many patients.
METHODS: This study analyzed 38 patient education materials on hydrocephalus, spina bifida, tethered cord syndrome, cerebral palsy, Chiari malformation, and craniosynostosis from 7 top-ranked US children’s hospitals. The Flesch-Kincaid grade level calculator was used to assess readability before and after AI modification.
RESULTS: ChatGPT effectively reduced the mean reading level from 10.60 (SD 0.57) to 6.18 (SD 0.28; p < 0.05), achieving the target 6th-grade level across all conditions.
CONCLUSIONS: Despite some limitations in maintaining word count and precise grade-level control, the results demonstrate the promising potential of AI in significantly enhancing the accessibility of pediatric neurosurgical education materials, which may lead to more inclusive patient communication and understanding.
PMID:41004846 | DOI:10.3171/2025.5.PEDS25122
