Yet, in key measures of understandability and actionability, materials performed worse than institution-generated materials
By Lori Solomon HealthDay Reporter
MONDAY, May 1, 2023 (HealthDay News) — ChatGPT provides helpful postoperative instructions for families of pediatric otolaryngology patients with low health literacy, according to a research letter published online April 27 in JAMA Otolaryngology-Head & Neck Surgery.
Noel F. Ayoub, M.D., from the Stanford University School of Medicine in California, and colleagues assessed the value of ChatGPT (versus Google Search and a university) in augmenting patient knowledge and generating postoperative instructions for use in populations with low educational or health literacy levels. The analysis included postoperative patient instructions for eight common pediatric otolaryngologic procedures.
The researchers found that understandability scores, overall, ranged from 73 to 91 percent; actionability scores ranged from 20 to 100 percent; and procedure-specific items ranged from 0 to 100 percent. ChatGPT-generated instructions (requested for a fifth-grade reading level) were scored from 73 to 82 percent for understandability, 20 to 80 percent for actionability, and 75 to 100 percent for procedure-specific items. Institution-generated instructions had the highest understandability scores (91 percent versus 81 percent for both ChatGPT and Google Search) and actionability scores (92 percent versus 73 percent for ChatGPT and 83 percent for Google Search). For procedure-specific items, scores were highest for institution-generated and ChatGPT instructions (both 97 percent versus 72 percent for Google Search).
“Despite these findings, ChatGPT may be beneficial for patients and clinicians, especially when alternative resources are limited,” the authors write.
Copyright © 2023 HealthDay. All rights reserved.
